-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathatom.xml
More file actions
383 lines (220 loc) · 341 KB
/
Copy pathatom.xml
File metadata and controls
383 lines (220 loc) · 341 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
<title>Notes</title>
<link href="/atom.xml" rel="self"/>
<link href="https://github-9233.github.io/"/>
<updated>2020-03-14T07:59:42.398Z</updated>
<id>https://github-9233.github.io/</id>
<author>
<name>Xu</name>
</author>
<generator uri="http://hexo.io/">Hexo</generator>
<entry>
<title>优酷路由宝YK-L1路由器刷老毛子</title>
<link href="https://github-9233.github.io/2020/03/14/%E4%BC%98%E9%85%B7%E8%B7%AF%E7%94%B1%E5%AE%9DYK-L1%E8%B7%AF%E7%94%B1%E5%99%A8%E5%88%B7%E8%80%81%E6%AF%9B%E5%AD%90/"/>
<id>https://github-9233.github.io/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/</id>
<published>2020-03-14T07:55:55.000Z</published>
<updated>2020-03-14T07:59:42.398Z</updated>
<content type="html"><![CDATA[<h1 id="优酷路由宝yk-l1路由器刷老毛子">优酷路由宝YK-L1路由器刷老毛子</h1><h2 id="刷入固件">1 刷入固件</h2><p>思路:</p><ol type="1"><li><p>查询路由器参数</p></li><li><p>根据路由器参数选择刷入不同<a href="https://openwrt.org/docs/techref/bootloader" target="_blank" rel="noopener">bootloader</a></p><blockquote><p><strong>引导</strong>(或称<strong>引导程序</strong>)在<a href="https://zh.wikipedia.org/wiki/计算机" target="_blank" rel="noopener">计算机</a>中是计算控制系统的一个初始化过程。引导过程可以是“硬引导”,如:开机通电后硬件诊断;也可以是“软引导”,此时会跳过开机自启。一些系统中软启动过程中,<a href="https://zh.wikipedia.org/wiki/随机存取存储器" target="_blank" rel="noopener">RAM</a>可以不清零。软启动和硬启动都可以通过硬件发起,如按下电源开关;也可以通过软件命令来进行。当正常、有效的运行环境达到后,启动完成。 <strong>引导装载程序</strong>是计算机开机自检完成后装载<a href="https://zh.wikipedia.org/wiki/操作系统" target="_blank" rel="noopener">操作系统</a>或者其他系统软件的<a href="https://zh.wikipedia.org/wiki/计算机程序" target="_blank" rel="noopener">计算机程序</a>。自检完成后运行引导装载程序,然后再加载并运行软件。引导装载程序可以从硬盘装载到主内存中。一些老计算机中,引导装载程序可以从打孔卡、穿孔纸带、或者磁带中加载到主内存中。启动装载程序加载且执行完成启动的程序。和POST程序相似,启动设备代码可以存储在永久存储体位置中,也可以为硬连线的形式。</p><p>当代通用计算机中,启动过程通常包含开机自检(POST)、初始化<a href="https://zh.wikipedia.org/wiki/周邊裝置" target="_blank" rel="noopener">周边设备</a>、然后加载<a href="https://zh.wikipedia.org/wiki/操作系统" target="_blank" rel="noopener">操作系统</a>。一些嵌入式系统直接运行存储在<a href="https://zh.wikipedia.org/wiki/ROM" target="_blank" rel="noopener">ROM</a>中的可执行程序。</p><p>引导在英文中为“boot”,是bootstrap的缩写,源自于短语“Pull oneself up by one's bootstraps”,即“靠自己振作起来”。如果大部分软件需要通过其他已运行的程序加载到计算机中,则必须存在一个将初始软件加载到计算机中的机制。</p></blockquote><ul><li>PC:<a href="https://en.wikipedia.org/wiki/BIOS" target="_blank" rel="noopener">BIOS</a>, <a href="https://en.wikipedia.org/wiki/Extensible%20Firmware%20Interface" target="_blank" rel="noopener">UEFI</a>, <a href="https://en.wikipedia.org/wiki/coreboot" target="_blank" rel="noopener">coreboot</a>, <a href="https://en.wikipedia.org/wiki/GNU%20GRUB" target="_blank" rel="noopener">GNU GRUB</a></li><li>Embedded Devices:<strong><a href="https://openwrt.org/docs/techref/bootloader/uboot" target="_blank" rel="noopener">Das U-Boot</a></strong>, <a href="https://openwrt.org/docs/techref/bootloader/redboot" target="_blank" rel="noopener">RedBoot</a>, <a href="https://openwrt.org/docs/techref/bootloader/cfe" target="_blank" rel="noopener">CFE</a>, <a href="https://openwrt.org/docs/techref/bootloader/adam2" target="_blank" rel="noopener">Adam2</a>, <a href="https://openwrt.org/docs/techref/bootloader/brnboot" target="_blank" rel="noopener">brnboot</a>, <a href="https://openwrt.org/docs/techref/bootloader/bootbase" target="_blank" rel="noopener">bootbase</a>, <a href="https://openwrt.org/docs/techref/bootloader/jboot" target="_blank" rel="noopener">jboot</a>, <a href="https://openwrt.org/docs/techref/bootloader/myloader" target="_blank" rel="noopener">myloader</a>, <a href="https://openwrt.org/docs/techref/bootloader/pp_boot" target="_blank" rel="noopener">pp_boot</a>, <a href="https://openwrt.org/docs/techref/bootloader/yamon" target="_blank" rel="noopener">yamon</a>, <a href="https://openwrt.org/docs/techref/bootloader/breed" target="_blank" rel="noopener">Breed</a></li></ul></li><li><p>根据路由器参数选择刷入不同<a href="https://zh.wikipedia.org/zh-cn/客制路由器韌體清單" target="_blank" rel="noopener">第三方路由器固件</a></p></li></ol><p>下面以优酷路由宝YOUKU YK1为例子</p><h3 id="查询yk-l1参数">1.1 查询YK-L1参数</h3><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/优酷路由宝.png" alt=""><figcaption>优酷路由宝</figcaption></figure><p><a href="https://openwrt.org/toh/hwdata/youku/youku_yk1" target="_blank" rel="noopener">Openwrt Techdata: YOUKU YK1</a></p><h3 id="刷入breed">1.2 刷入Breed</h3><p>Boot and Recovery Environment for Embedded Devices (BREED) A multi-task bootloader with real-time firmware upgrading progress.</p><p><a href="https://breed.hackpascal.net" target="_blank" rel="noopener">下载地址</a></p><p>YK-L1选择<a href="https://breed.hackpascal.net/breed-mt7620-youku-yk1.bin" target="_blank" rel="noopener">breed-mt7620-youku-yk1.bin</a></p><h3 id="padavan">1.3 Padavan</h3><p>国内可使用恩山无线论坛维护的<a href="https://opt.cn2qq.com/padavan/" target="_blank" rel="noopener">padavan</a></p><p>YK-L1选择<a href="https://opt.cn2qq.com/padavan/RT-N14U-GPIO-1-youku1-128M_3.4.3.9-099.trx" target="_blank" rel="noopener">RT-N14U-GPIO-1-youku1-128M_3.4.3.9-099.trx</a></p><h2 id="如何配置entware">2 <a href="https://bitbucket.org/padavan/rt-n56u/wiki/EN/HowToConfigureEntware" target="_blank" rel="noopener">如何配置Entware</a></h2><p>⚠️<strong>该部分主要针对Padavan老毛子用户</strong></p><p><a href="https://github.com/Entware-ng/Entware-ng" target="_blank" rel="noopener">Entware</a>是一个高级软件包管理器,它使您可以轻松地将交叉编译的Linux应用程序通过Internet安装和卸载到设备上。这是<a href="https://openwrt.org/" target="_blank" rel="noopener">OpenWRT</a>软件存储库,使用相当现代的工具链进行了编译。它使用了<a href="http://en.wikipedia.org/wiki/Optware" target="_blank" rel="noopener">Optware</a>中不提供的新功能。 <a href="http://entware.zyxmon.org/binaries/mipsel/Packages.html" target="_blank" rel="noopener">这是Entware可用的软件包列表</a>(也可使用opkg list查询)。</p><h3 id="终端应用程序的安装和使用">2.1 终端应用程序的安装和使用</h3><p>首先,打开浏览器,在左下方窗格中选择高级设置->管理->系统,然后选择要使用的服务器。请注意,强烈建议仅将SSH用作外部连接的服务器。因为Telnet是不安全的协议,并且SSH是它的替代品。但是,对于专用(内部)网络使用,安全性不是必需的。</p><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/adm-system-en.png" alt=""><figcaption>adm-system-en</figcaption></figure><p>在上面的示例中,选择了SSH作为首选方法。选择您要使用的选项,然后应用以将设置保存到路由器。</p><p>您需要一个终端客户端才能与路由器建立连接。终端应用程序是用于文本输入和显示输出的串行计算机接口。有许多不同的应用程序。您应该根据自己的操作系统选择一个。</p><h4 id="windows-xpvista和windows-7用户的指导">2.1.1 Windows XP,Vista和Windows 7用户的指导</h4><p><a href="http://www.putty.org/" target="_blank" rel="noopener"><strong>PuTTY</strong></a>是Windows和Unix的Telnet和SSH的免费实现。下载安装程序,然后安装应用程序。安装后,只需很少的配置即可访问路由器。要开始会话,您将需要输入一些基本参数。在“Session”中,选择“Host name”,输入<strong>router的IP地址</strong>。从与路由器配置的协议匹配的“Connection type”中选择要使用的登录协议。然后,选择“Translation”,并确保为字符集字段配置了UTF-8。</p><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/putty_932x448.png" alt=""><figcaption>putty_932x448</figcaption></figure><p><strong>重要说明:使用终端连接时,切勿将登录名和密码保存到任何客户端</strong></p><p>完成正确的配置后,选择配置窗口右下方的“打开”按钮,PuTTY将使用正确的协议启动与路由器的连接。</p><h4 id="macos用户的指导">2.1.2 MacOS用户的指导</h4><p>打开Finder,然后转到Applications => Utilities。双击Terminal<img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/terminal_in_osx.jpg" alt="terminal_in_osx"></p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">ssh admin@ip_address</span><br></pre></td></tr></table></figure><h3 id="usb磁盘格式化">2.2 USB磁盘格式化</h3><p>基于Linux的系统使用<a href="http://en.wikipedia.org/wiki/Extended_file_system" target="_blank" rel="noopener">EXTended文件系统</a>。建议使用容量超过512MB的外部磁盘(HDD,闪存),并使用EXT2,EXT3或EXT4文件系统进行格式化。推荐使用EXT4文件系统。</p><p><strong>注意:可以但不建议在Linux上使用NTFS磁盘。这些说明将基于推荐的Linux文件系统</strong></p><h4 id="对于windows-xpvista或windows-7用户">2.2.1 对于Windows XP,Vista或Windows 7用户</h4><p>下载并安装<a href="http://www.partition-tool.com/download.htm" target="_blank" rel="noopener"><strong>EaseUS Partition Master Home Edition</strong></a>。</p><p><a href="http://www.partition-tool.com/easeus-partition-manager/linux-partition-manager.htm" target="_blank" rel="noopener"><strong>在这里,您可以找到</strong></a>带有说明和屏幕截图的详细过程。</p><h4 id="对于macos用户">2.2.2 对于MacOS用户</h4><p>您可以使用<a href="http://code.google.com/p/macfuse" target="_blank" rel="noopener"><strong>MacFuse</strong></a>。请按照本<a href="http://code.google.com/p/macfuse/wiki/AUTOINSTALL" target="_blank" rel="noopener"><strong>指南</strong></a>安装程序。</p><h4 id="使用路由器格式化磁盘驱动器">2.2.3 使用路由器格式化磁盘驱动器</h4><p>插入USB磁盘,然后等待USB指示灯点亮。指示灯指示驱动器已安装后,您就可以开始使用USB驱动器了。</p><p>思路:</p><ol type="1"><li>删除旧分区表;</li><li>创建“ DOS”分区表;</li><li>添加一个分区;</li><li>用'ext4'格式化它;</li><li>提交所有结果。</li></ol><p>打开一个终端会话,然后连接到您的路由器。系统将提示您输入路由器的密码。输入用户密码。</p><p>建立连接后,您将看到BusyBox问候语和提示。</p><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/term_greet.png" alt=""><figcaption>term_greet</figcaption></figure><p>在使用提示符输入命令之前,请检查是否未通过Web界面安装磁盘。</p><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/webiface1-en.png" alt=""><figcaption>webiface1-en</figcaption></figure><p>在此示例中,磁盘仍处于安装状态,在继续操作之前需要将其卸下。在“安全移除磁盘”选项中按“移除”以卸载连接的驱动器。</p><p><strong>警告!如果光盘驱动器仍安装在网络接口中,请勿继续!</strong></p><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/webiface2-en.png" alt=""><figcaption>img</figcaption></figure><p>要检查驱动器上存在哪些分区:</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br></pre></td><td class="code"><pre><span class="line">cat /proc/partitions</span><br><span class="line">major minor #blocks name</span><br><span class="line"></span><br><span class="line"> 31 0 192 mtdblock0</span><br><span class="line"> 31 1 64 mtdblock1</span><br><span class="line"> 31 2 64 mtdblock2</span><br><span class="line"> 31 3 1228 mtdblock3</span><br><span class="line"> 31 4 6579 mtdblock4</span><br><span class="line"> 31 5 64 mtdblock5</span><br><span class="line"> 31 6 7872 mtdblock6</span><br><span class="line"> 8 0 7910400 sda</span><br><span class="line"> 8 1 7908626 sda1</span><br></pre></td></tr></table></figure><p>由于可以将两个USB驱动器连接到不同路由器 USB端口,因此USB驱动器可以是sda或sdb。 正如您在示例中看到的那样,一个驱动器是sda,它有一个分区是sda1。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br></pre></td><td class="code"><pre><span class="line">fdisk -lu /dev/sda</span><br><span class="line"></span><br><span class="line">Disk /dev/sda: 8100 MB, 8100249600 bytes</span><br><span class="line">175 heads, 12 sectors/track, 7533 cylinders, total 15820800 sectors</span><br><span class="line">Units = sectors of 1 * 512 = 512 bytes</span><br><span class="line"></span><br><span class="line"> Device Boot Start End Blocks Id System</span><br><span class="line">/dev/sda1 2048 15819299 7908626 83 Linux</span><br></pre></td></tr></table></figure><p>创建一个新的分区表并将其格式化。注意,您将需要在卷volume(fdisk /dev/sda)而不是分区partition(fdisk /dev/sda1)上执行磁盘操作!</p><blockquote><p>补充知识</p><p><a href="https://en.wikipedia.org/wiki/Cylinder-head-sector" target="_blank" rel="noopener"><strong>Cylinder-head-sector</strong></a></p><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/550px-Hard_drive_geometry_-_English_-_2019-05-30.svg.png" alt=""><figcaption>550px-Hard_drive_geometry_-<em>English</em>-_2019-05-30.svg</figcaption></figure></blockquote><h5 id="如果不需要交换分区">2.2.3.1 如果不需要交换分区</h5><p>请使用“ fdisk”创建一个新分区。(输入<code>fdisk -u /dev/sda</code>用于扇区sector而不是圆柱体cylinders)</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br><span class="line">20</span><br><span class="line">21</span><br><span class="line">22</span><br><span class="line">23</span><br><span class="line">24</span><br><span class="line">25</span><br><span class="line">26</span><br><span class="line">27</span><br><span class="line">28</span><br><span class="line">29</span><br><span class="line">30</span><br><span class="line">31</span><br><span class="line">32</span><br><span class="line">33</span><br><span class="line">34</span><br><span class="line">35</span><br><span class="line">36</span><br><span class="line">37</span><br><span class="line">38</span><br><span class="line">39</span><br><span class="line">40</span><br><span class="line">41</span><br><span class="line">42</span><br><span class="line">43</span><br><span class="line">44</span><br><span class="line">45</span><br><span class="line">46</span><br><span class="line">47</span><br><span class="line">48</span><br><span class="line">49</span><br><span class="line">50</span><br><span class="line">51</span><br></pre></td><td class="code"><pre><span class="line">fdisk /dev/sda</span><br><span class="line"></span><br><span class="line">The number of cylinders for this disk is set to 7533.</span><br><span class="line">There is nothing wrong with that, but this is larger than 1024,</span><br><span class="line">and could in certain setups cause problems with:</span><br><span class="line">1) software that runs at boot time (e.g., old versions of LILO)</span><br><span class="line">2) booting and partitioning software from other OSs</span><br><span class="line"> (e.g., DOS FDISK, OS/2 FDISK)</span><br><span class="line"></span><br><span class="line">Command (m for help): m</span><br><span class="line">Command Action</span><br><span class="line">a toggle a bootable flag</span><br><span class="line">b edit bsd disklabel</span><br><span class="line">c toggle the dos compatibility flag</span><br><span class="line">d delete a partition</span><br><span class="line">l list known partition types</span><br><span class="line">n add a new partition</span><br><span class="line">o create a new empty DOS partition table</span><br><span class="line">p print the partition table</span><br><span class="line">q quit without saving changes</span><br><span class="line">s create a new empty Sun disklabel</span><br><span class="line">t change a partition's system id</span><br><span class="line">u change display/entry units</span><br><span class="line">v verify the partition table</span><br><span class="line">w write table to disk and exit</span><br><span class="line">x extra functionality (experts only)</span><br><span class="line"></span><br><span class="line">Command (m for help): d</span><br><span class="line">Selected partition 1</span><br><span class="line"></span><br><span class="line">Command (m for help): n</span><br><span class="line">Command action</span><br><span class="line"> e extended</span><br><span class="line"> p primary partition (1-4)</span><br><span class="line">p</span><br><span class="line">Partition number (1-4): 1</span><br><span class="line">First cylinder (1-7533, default 1): Using default value 1</span><br><span class="line">Last cylinder or +size or +sizeM or +sizeK (1-7533, default 7533): Using default value 7533</span><br><span class="line"></span><br><span class="line">Command (m for help): p</span><br><span class="line"></span><br><span class="line">Disk /dev/sda: 8100 MB, 8100249600 bytes</span><br><span class="line">175 heads, 12 sectors/track, 7533 cylinders</span><br><span class="line">Units = cylinders of 2100 * 512 = 1075200 bytes</span><br><span class="line"></span><br><span class="line"> Device Boot Start End Blocks Id System</span><br><span class="line">/dev/sda1 1 7533 7908626 83 Linux</span><br><span class="line"></span><br><span class="line">Command (m for help): w</span><br><span class="line">The partition table has been altered.</span><br><span class="line">Calling ioctl() to re-read partition table</span><br></pre></td></tr></table></figure><p>检查分区以查看是否已创建。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br></pre></td><td class="code"><pre><span class="line">fdisk -lu /dev/sda</span><br><span class="line"></span><br><span class="line">Disk /dev/sda: 8100 MB, 8100249600 bytes</span><br><span class="line">175 heads, 12 sectors/track, 7533 cylinders, total 15820800 sectors</span><br><span class="line">Units = sectors of 1 * 512 = 512 bytes</span><br><span class="line"></span><br><span class="line"> Device Boot Start End Blocks Id System</span><br><span class="line">/dev/sda1 2048 15819299 7908626 83 Linux</span><br></pre></td></tr></table></figure><p>现在卸载驱动器:<code>ejusb</code></p><p>现在使用卷标“ Main”格式化分区。将来,该分区将使用该标签安装在系统中。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br><span class="line">20</span><br><span class="line">21</span><br></pre></td><td class="code"><pre><span class="line">mkfs.ext4 -m 0 -L Main /dev/sda1</span><br><span class="line">mke2fs 1.42.8 (20-Jun-2013)</span><br><span class="line">Filesystem label=Main</span><br><span class="line">OS type: Linux</span><br><span class="line">Block size=4096 (log=2)</span><br><span class="line">Fragment size=4096 (log=2)</span><br><span class="line">Stride=0 blocks, Stripe width=0 blocks</span><br><span class="line">7808 inodes, 1977156 blocks</span><br><span class="line">0 blocks (0.00%) reserved for the super user</span><br><span class="line">First data block=0</span><br><span class="line">Maximum filesystem blocks=2025848832</span><br><span class="line">61 block groups</span><br><span class="line">32768 blocks per group, 32768 fragments per group</span><br><span class="line">128 inodes per group</span><br><span class="line">Superblock backups stored on blocks:</span><br><span class="line"> 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632</span><br><span class="line"></span><br><span class="line">Allocating group tables: done</span><br><span class="line">Writing inode tables: done</span><br><span class="line">Creating journal (32768 blocks): done</span><br><span class="line">Writing superblocks and filesystem accounting information: done</span><br></pre></td></tr></table></figure><p><strong>注意:</strong>如果格式化容量超过256 TB的USB驱动器,则可以使用命令<code>mkfs.ext4 -m 0 -T largefile -L Main /dev/sda1</code></p><p><strong>注意:</strong>超级用户使用了5%的磁盘容量。</p><p>拔下磁盘,然后再次插入以创建以下目录,或者使用下面的这两个命令。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">mkdir /media/Main</span><br><span class="line">mount -t ext4 /dev/sda1 /media/Main</span><br></pre></td></tr></table></figure><h5 id="如果需要交换分区">2.2.3.2 如果需要交换分区</h5><p>并且打算使用BitTorrent客户端等应用程序或需要大量内存的应用程序(下载管理器,“微型服务器”等),则可能需要SWAP分区。另外,<strong>当使用媒体服务器UPnP / DLNA时</strong>,<strong>建议存在SWAP分区</strong>,这是在数据库创建媒体内容的过程中消耗了大量RAM的。</p><p><strong>重要的提示!不建议在USB硬盘驱动器上创建交换文件或分区。如果USB磁盘未正确拔出</strong>(例如,路由器突然断电或其他原因),则<strong>可能导致系统崩溃</strong>。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br><span class="line">20</span><br><span class="line">21</span><br><span class="line">22</span><br><span class="line">23</span><br><span class="line">24</span><br><span class="line">25</span><br><span class="line">26</span><br><span class="line">27</span><br><span class="line">28</span><br><span class="line">29</span><br><span class="line">30</span><br><span class="line">31</span><br><span class="line">32</span><br><span class="line">33</span><br><span class="line">34</span><br><span class="line">35</span><br><span class="line">36</span><br><span class="line">37</span><br><span class="line">38</span><br><span class="line">39</span><br><span class="line">40</span><br><span class="line">41</span><br><span class="line">42</span><br><span class="line">43</span><br><span class="line">44</span><br><span class="line">45</span><br><span class="line">46</span><br><span class="line">47</span><br><span class="line">48</span><br><span class="line">49</span><br><span class="line">50</span><br><span class="line">51</span><br><span class="line">52</span><br><span class="line">53</span><br><span class="line">54</span><br><span class="line">55</span><br><span class="line">56</span><br><span class="line">57</span><br><span class="line">58</span><br><span class="line">59</span><br><span class="line">60</span><br><span class="line">61</span><br></pre></td><td class="code"><pre><span class="line">fdisk -u /dev/sda</span><br><span class="line"></span><br><span class="line">Command (m for help): m</span><br><span class="line">Command Action</span><br><span class="line">a toggle a bootable flag</span><br><span class="line">b edit bsd disklabel</span><br><span class="line">c toggle the dos compatibility flag</span><br><span class="line">d delete a partition</span><br><span class="line">l list known partition types</span><br><span class="line">n add a new partition</span><br><span class="line">o create a new empty DOS partition table</span><br><span class="line">p print the partition table</span><br><span class="line">q quit without saving changes</span><br><span class="line">s create a new empty Sun disklabel</span><br><span class="line">t change a partition's system id</span><br><span class="line">u change display/entry units</span><br><span class="line">v verify the partition table</span><br><span class="line">w write table to disk and exit</span><br><span class="line">x extra functionality (experts only)</span><br><span class="line"></span><br><span class="line">Command (m for help): d</span><br><span class="line">Selected partition 1</span><br><span class="line"></span><br><span class="line">Command (m for help): n</span><br><span class="line">Command action</span><br><span class="line"> e extended</span><br><span class="line"> p primary partition (1-4)</span><br><span class="line">p</span><br><span class="line">Partition number (1-4): 1</span><br><span class="line">First sector (2048-15820799, default 2048): Using default value 2048</span><br><span class="line">Last sector or +size or +sizeM or +sizeK (2048-15820799, default 15820799): +256M</span><br><span class="line"></span><br><span class="line">Command (m for help): n</span><br><span class="line">Command action</span><br><span class="line"> e extended</span><br><span class="line"> p primary partition (1-4)</span><br><span class="line">p</span><br><span class="line">Partition number (1-4): 2</span><br><span class="line">First sector (502049-15820799, default 502049): Using default value 502049</span><br><span class="line">Last sector or +size or +sizeM or +sizeK (502049-15820799, default 15820799): Using default value 15820799</span><br><span class="line"></span><br><span class="line">Command (m for help): t</span><br><span class="line">Partition number (1-4): 1</span><br><span class="line">Hex code (type L to list codes): 82</span><br><span class="line">Changed system type of partition 1 to 82 (Linux swap)</span><br><span class="line"></span><br><span class="line">Command (m for help): p</span><br><span class="line"></span><br><span class="line">Disk /dev/sda: 8100 MB, 8100249600 bytes</span><br><span class="line">175 heads, 12 sectors/track, 7533 cylinders, total 15820800 sectors</span><br><span class="line">Units = sectors of 1 * 512 = 512 bytes</span><br><span class="line"></span><br><span class="line"> Device Boot Start End Blocks Id System</span><br><span class="line">/dev/sda1 2048 502048 250000+ 82 Linux swap</span><br><span class="line">Partition 1 does not end on cylinder boundary</span><br><span class="line">/dev/sda2 502049 15820799 7659375+ 83 Linux</span><br><span class="line">Partition 2 does not end on cylinder boundary</span><br><span class="line"></span><br><span class="line">Command (m for help): w</span><br><span class="line">The partition table has been altered.</span><br><span class="line">Calling ioctl() to re-read partition table</span><br></pre></td></tr></table></figure><p>现在卸载驱动器:<code>ejusb</code></p><p>现在使用卷标“ Main”格式化分区。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br><span class="line">20</span><br><span class="line">21</span><br></pre></td><td class="code"><pre><span class="line">mkfs.ext4 -m 0 -L Main /dev/sda2</span><br><span class="line">mke2fs 1.42.8 (20-Jun-2013)</span><br><span class="line">Filesystem label=Main</span><br><span class="line">OS type: Linux</span><br><span class="line">Block size=4096 (log=2)</span><br><span class="line">Fragment size=4096 (log=2)</span><br><span class="line">Stride=0 blocks, Stripe width=0 blocks</span><br><span class="line">7552 inodes, 1914843 blocks</span><br><span class="line">0 blocks (0.00%) reserved for the super user</span><br><span class="line">First data block=0</span><br><span class="line">Maximum filesystem blocks=1962934272</span><br><span class="line">59 block groups</span><br><span class="line">32768 blocks per group, 32768 fragments per group</span><br><span class="line">128 inodes per group</span><br><span class="line">Superblock backups stored on blocks:</span><br><span class="line"> 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632</span><br><span class="line"></span><br><span class="line">Allocating group tables: done</span><br><span class="line">Writing inode tables: done</span><br><span class="line">Creating journal (32768 blocks): done</span><br><span class="line">Writing superblocks and filesystem accounting information: done</span><br></pre></td></tr></table></figure><p><strong>注意:</strong>如果格式化容量超过256TB 的USB驱动器,则可以使用命令<code>mkfs.ext4 -m 0 -T largefile -L Main /dev/sda1</code></p><p>并创建交换空间:</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">mkswap /dev/sda1</span><br><span class="line">Setting up swapspace version 1, size = 255996416 bytes</span><br></pre></td></tr></table></figure><p>交换分区将在启动时或插入USB磁盘时自动安装。</p><h3 id="使用entware">2.3 使用Entware</h3><p>如果尚未插入USB驱动器,请立即进行操作。打开终端(PuTTY),然后登录路由器。下一步是创建一个新目录。</p><p>检查是否通过终端安装了磁盘。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br></pre></td><td class="code"><pre><span class="line">mount</span><br><span class="line">rootfs on / type rootfs (rw)</span><br><span class="line">/dev/root on / type squashfs (ro,relatime)</span><br><span class="line">proc on /proc type proc (rw,relatime)</span><br><span class="line">sysfs on /sys type sysfs (rw,relatime)</span><br><span class="line">usbfs on /proc/bus/usb type usbfs (rw,relatime)</span><br><span class="line">tmpfs on /dev type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /etc type tmpfs (rw,noatime,size=2048k)</span><br><span class="line">tmpfs on /home type tmpfs (rw,relatime,size=1024k)</span><br><span class="line">tmpfs on /media type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /mnt type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /tmp type tmpfs (rw,relatime,size=24576k)</span><br><span class="line">tmpfs on /var type tmpfs (rw,relatime,size=4096k)</span><br><span class="line">devpts on /dev/pts type devpts (rw,relatime,mode=600)</span><br><span class="line">/dev/sda1 on /media/Main type ext4 (rw,noatime,data=ordered)</span><br></pre></td></tr></table></figure><p>现在创建新目录<strong>opt</strong></p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">mkdir /media/Main/opt</span><br></pre></td></tr></table></figure><p>现在,卸下磁盘驱动器:<code>ejusb</code></p><p>转到<strong>高级设置-> USB应用程序->通用设置</strong>。将“允许运行Optware”切换为“ Entware”,然后最后按“应用”以保存设置。</p><figure><img src="/2020/03/14/优酷路由宝YK-L1路由器刷老毛子/webiface3-en.png" alt=""><figcaption>img</figcaption></figure><p>拔下USB驱动器,然后再次插入。等待USB指示灯指示可以使用路由器上的驱动器,然后在终端中使用“ mount”命令检查驱动器的输出。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br></pre></td><td class="code"><pre><span class="line">mount</span><br><span class="line">rootfs on / type rootfs (rw)</span><br><span class="line">/dev/root on / type squashfs (ro,relatime)</span><br><span class="line">proc on /proc type proc (rw,relatime)</span><br><span class="line">sysfs on /sys type sysfs (rw,relatime)</span><br><span class="line">usbfs on /proc/bus/usb type usbfs (rw,relatime)</span><br><span class="line">tmpfs on /dev type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /etc type tmpfs (rw,noatime,size=2048k)</span><br><span class="line">tmpfs on /home type tmpfs (rw,relatime,size=1024k)</span><br><span class="line">tmpfs on /media type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /mnt type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /tmp type tmpfs (rw,relatime,size=24576k)</span><br><span class="line">tmpfs on /var type tmpfs (rw,relatime,size=4096k)</span><br><span class="line">devpts on /dev/pts type devpts (rw,relatime,mode=600)</span><br><span class="line">/dev/sda1 on /media/Main type ext4 (rw,noatime,data=ordered)</span><br><span class="line">/dev/sda1 on /opt type ext4 (rw,noatime,data=ordered)</span><br></pre></td></tr></table></figure><p>如您所见,/ dev / sda1分区已安装到/ media / Main,下一行显示/ media / Main / opt已安装到/ opt。</p><p>同样,如果您有交换分区/ dev / sda1,您将看到另一个输出:</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br></pre></td><td class="code"><pre><span class="line">mount</span><br><span class="line">rootfs on / type rootfs (rw)</span><br><span class="line">/dev/root on / type squashfs (ro,relatime)</span><br><span class="line">proc on /proc type proc (rw,relatime)</span><br><span class="line">sysfs on /sys type sysfs (rw,relatime)</span><br><span class="line">usbfs on /proc/bus/usb type usbfs (rw,relatime)</span><br><span class="line">tmpfs on /dev type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /etc type tmpfs (rw,noatime,size=2048k)</span><br><span class="line">tmpfs on /home type tmpfs (rw,relatime,size=1024k)</span><br><span class="line">tmpfs on /media type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /mnt type tmpfs (rw,relatime,size=8k)</span><br><span class="line">tmpfs on /tmp type tmpfs (rw,relatime,size=24576k)</span><br><span class="line">tmpfs on /var type tmpfs (rw,relatime,size=4096k)</span><br><span class="line">devpts on /dev/pts type devpts (rw,relatime,mode=600)</span><br><span class="line">/dev/sda2 on /media/Main type ext4 (rw,noatime,data=ordered)</span><br><span class="line">/dev/sda2 on /opt type ext4 (rw,noatime,data=ordered)</span><br></pre></td></tr></table></figure><p><strong>警告!除非将其安全地移除或通过端子拆卸,否则切勿拔出磁盘的插头!</strong></p><h3 id="安装应用程序的方法">2.4 安装应用程序的方法</h3><p><strong>注意!在安装任何应用程序之前,强烈建议您更新有关存储库的信息!</strong>:</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">opkg update</span><br><span class="line">opkg upgrade</span><br></pre></td></tr></table></figure><hr><p>要从Entware存储库安装应用程序,请使用:</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">opkg install { app_name }</span><br></pre></td></tr></table></figure><p>此命令将安装具有其依赖项的应用程序。</p><hr><p><strong>注意:切勿自己从磁盘上的/opt目录中删除任何东西。采用:</strong></p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">opkg remove { app_name }</span><br></pre></td></tr></table></figure><hr><p>例如,如果您需要Midnight Commander,请使用:</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">opkg update</span><br><span class="line">opkg install mc</span><br></pre></td></tr></table></figure><p>这将安装Midnight Commander。</p><hr><p>用<code>opkg list</code>看到的应用程序库avaliable整个列表。要获取更多信息,请使用<code>opkg --help</code>。</p><hr><p><strong>Entware具有按名称和软件包描述进行搜索的功能:</strong></p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">opkg find "*game*"</span><br></pre></td></tr></table></figure><hr><h3 id="安装xupnpd">2.4 安装xUpnpd</h3><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">opkg update</span><br><span class="line">opkg install xupnpd</span><br></pre></td></tr></table></figure><p>启动xUpnpd:</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">/opt/etc/init.d/S94xupnpd start</span><br></pre></td></tr></table></figure><p>安装xUpnpd之后,您可以在/opt/share/xupnpd目录中编辑它的配置(可以通过<a href="http://my.router:4044/" target="_blank" rel="noopener">WEB-interface</a>进行编辑)。</p><p><strong>注意:任何应用程序的所有配置文件都必须保存为UNIX风格!!!!</strong> 例如,您可以安装Midnight Commander并从中编辑所有配置文件。</p><hr><p>如果您希望某个应用程序在启动时启动,请检查脚本的第一个字母,该脚本在/opt/etc/init.d中对其进行调用。应该是“ S”。</p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">ls -a /opt/etc/init.d</span><br><span class="line">. .. S01system S10iptables S94xupnpd rc.func rc.unslung</span><br></pre></td></tr></table></figure><hr><p><strong>重要!!当您安装任何应用程序时,请始终使用脚本中的内容检查该应用程序是否可以完全停止<code>/opt/etc/init.d</code> 它。如果该应用程序不能停止运行,则您在磁盘(例如文件系统)上的信息可能会严重损坏!</strong></p>]]></content>
<summary type="html">
<h1 id="优酷路由宝yk-l1路由器刷老毛子">优酷路由宝YK-L1路由器刷老毛子</h1>
<h2 id="刷入固件">1 刷入固件</h2>
<p>思路:</p>
<ol type="1">
<li><p>查询路由器参数</p></li>
<li><p>根据路由器参数选
</summary>
</entry>
<entry>
<title>XRF-X射线荧光分析设备</title>
<link href="https://github-9233.github.io/2020/03/10/XRF-X%E5%B0%84%E7%BA%BF%E8%8D%A7%E5%85%89%E5%88%86%E6%9E%90%E8%AE%BE%E5%A4%87/"/>
<id>https://github-9233.github.io/2020/03/10/XRF-X射线荧光分析设备/</id>
<published>2020-03-10T00:52:49.000Z</published>
<updated>2020-03-10T00:58:30.975Z</updated>
<content type="html"><![CDATA[<h1 id="xrf-x射线荧光分析设备">XRF-X射线荧光分析设备</h1><p><strong>Hitachi High-Tech Analytical Science</strong></p><h3 id="检测范围">1 检测范围</h3><p>X-MET8000 系列可以分析镁12和铀92之间的所有元素。</p><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/Simple_Periodic_Table_Chart-en.png" alt=""><figcaption>Simple_Periodic_Table_Chart-en</figcaption></figure><h3 id="警告️">2 警告⚠️</h3><h4 id="警告有毒材料-铍">2.1 警告:有毒材料-铍</h4><p>检测器有薄铍窗口。铍是有毒化合物,但当铍窗口完好无损时,不会对健康造成任何危害。</p><p><strong>警告-有毒材料</strong>:任何情况下都不要刺穿、打破或损坏铍窗口。这会生成尘埃粒子。长时间吸入铍会导致癌症。</p><p>切勿使检测器接触湿气或因湿度过高生成的冷凝液。这会腐蚀铍窗口;如果还存在氯、硫酸盐、铜或铁,则会更易于造成腐蚀。</p><h4 id="警告电流">2.2 警告:电流</h4><p>X-MET8000 系列内含极高电压,触电危险性极高。这可导致严重的人身伤害。</p><h4 id="警告x射线">2.3 警告:X射线</h4><p>测试期间,请勿操作触发键进行测量。如果 X-MET8000 系列发出 X 光辐射时没有与样品紧密接触,辐射会出现分散。长时间直接暴露在 X 射线辐射中会导致严重的人身伤害。</p><h4 id="警告射频暴露">2.4 警告:射频暴露</h4><p>本产品发射无线电射频能量,但该设备的辐射输出功率低于 FCC 和 IC 射频暴露限值。本设备符合对 非受控环境的 FCC RF 和 IC 辐射暴露限制。尽管如此,在该设备正常操作过程中,仍应尽量减少天线 与人体接触的可能性。</p><blockquote><p><strong>Specific Absorption Rate</strong> (<strong>SAR</strong>) is a measure of the rate at which <a href="https://en.wikipedia.org/wiki/Energy" target="_blank" rel="noopener">energy</a> is <a href="https://en.wikipedia.org/wiki/Absorption_(electromagnetic_radiation)" target="_blank" rel="noopener">absorbed</a> by the human body when exposed to a <a href="https://en.wikipedia.org/wiki/Radio_frequency" target="_blank" rel="noopener">radio frequency</a> (RF) <a href="https://en.wikipedia.org/wiki/Electromagnetic_field" target="_blank" rel="noopener">electromagnetic field</a>. It can also refer to absorption of other forms of energy by tissue, including <a href="https://en.wikipedia.org/wiki/Ultrasound" target="_blank" rel="noopener">ultrasound</a>. It is defined as the <a href="https://en.wikipedia.org/wiki/Power_(physics)" target="_blank" rel="noopener">power</a> absorbed per <a href="https://en.wikipedia.org/wiki/Mass" target="_blank" rel="noopener">mass</a> of <a href="https://en.wikipedia.org/wiki/Body_tissue" target="_blank" rel="noopener">tissue</a> and has units of <a href="https://en.wikipedia.org/wiki/Watt" target="_blank" rel="noopener">watts</a> per <a href="https://en.wikipedia.org/wiki/Kilogram" target="_blank" rel="noopener">kilogram</a> (W/kg).</p></blockquote><h3 id="执行测量">3 执行测量</h3><h4 id="打开-x-met8000-系列">3.1 打开 X-MET8000 系列</h4><p>点触数字键入密码。 出厂设置为:操作员:1111</p><h4 id="结果屏幕">3.2 结果屏幕</h4><p>【<strong>+/-</strong> 含量范围】 这指示测量精度(2 西格玛)。+/- 值越低,精度越高。</p><blockquote><p>在<a href="https://zh.wikipedia.org/wiki/統計" target="_blank" rel="noopener">统计</a>上,<strong>68–95–99.7原则</strong>(68–95–99.7 rule)是在<a href="https://zh.wikipedia.org/wiki/正态分布" target="_blank" rel="noopener">正态分布</a>中,距平均值小于一个<a href="https://zh.wikipedia.org/wiki/標準差" target="_blank" rel="noopener">标准差</a>、二个标准差、三个标准差以内的百分比,更精确的数字是68.27%、95.45%及99.73%。</p></blockquote><p>可在标准结果屏幕上查看“固定”元素。“固定”元素显示在已测量元素后面,并使用分隔线和文本“固定元素”隔开。如果一个元素在“固定”元素和“标准”测量同时存在,则仅在“固定”元素列表中显示。如果结果显示为“PPM”格式,则“固定”元素也会进行转换并显示为“PPM” 格式。</p><h4 id="测试方法">3.3 测试方法</h4><blockquote><p>【ALLOY合金应用】使用经验法校正以提高准确度。可关闭轻元素以提高速度。</p><p>【ALLOY+合金+应用】以 FP 计算为基础,可为样品自动选择正确的参数。可关闭轻元素以提高速度。</p><p>有时,单一模式无法对样品进行完全测量,因为样品中的某些含量超出该方法的限制。如果发生这种情况,则会在含量旁边显示大于 (>) 或小于 (<) 标志。</p><p>如果发生这种情况,操作员应选择适用的“FP”方法。这些方法适用于范围广泛的含量和样品,是操作员的第二选择。</p></blockquote><h5 id="合金应用将经验法校正用于不锈钢">3.3.1 合金应用,将经验法校正用于不锈钢</h5><blockquote><p>这些步骤显示合金+应用如何使用经验法校正分析不锈钢样品并返回牌号。</p><ol type="1"><li><p>X-MET8000 系列获得可识别样品的光谱图。</p></li><li><p>它将标识光谱图与所有经验法校正进行比较来获取匹配。</p></li><li><p>它查找与不锈钢经验法校正结果(以绿色显示)匹配的结果。</p></li><li><p>它使用不锈钢校正分析样品,然后显示结果。</p></li><li><p>它将结果与牌号库进行比较,发现 SS304 是最佳匹配。</p></li></ol><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/合金应用,将经验法校正用于不锈钢.png" alt=""><figcaption>合金应用,将经验法校正用于不锈钢</figcaption></figure></blockquote><h5 id="合金应用将基本参数校正用于黄金">3.3.2 合金<strong>+</strong>应用,将基本参数校正用于黄金</h5><blockquote><p>这些步骤显示合金+应用如何使用基本参数校正分析黄金样品并返回牌号。</p><ol type="1"><li>X-MET8000 系列获得可识别样品的光谱图。</li><li>它将标识光谱图与所有经验法校正进行比较来获取匹配。</li><li>它没有使用任何经验法校正(全部以红色显示)来查找匹配。</li><li>它切换到基本参数校正,并使用内部计算器分析样品、进行计算,然后显示结果。</li><li><ol start="5" type="1"><li>它将结果与牌号库进行比较,发现重量为 24 克拉的黄金是最佳匹配。</li></ol></li></ol><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/合金+应用,将基本参数校正用于黄金.png" alt=""><figcaption>合金+应用,将基本参数校正用于黄金</figcaption></figure></blockquote><h4 id="测量时间">3.4 测量时间</h4><p>出厂设置的 10 秒测量时间适用于大多数测量。铁26、铜29、镍28和类似合金的测量时间更短,仅需 5 秒。镁12和铝13合金所需的测量时间较长,需要 15 秒或更长。多元合金也需要较长测量时间,以分析所有微量元素。</p><p>时间较长的测量将始终提供更为精确的结果。但 X-MET8000 系列可以仅在数秒时间内提供优异结果。标准偏差(显示屏上显示为 +/-)数字表示测量精度。测量时间越长,标准偏差数字越低,精度越高。</p><p>测量时间结束后,X-MET8000 系列会发出“砰”音。也可使用内部计时器完全控制测量。将其称为“定时测量”。操作员扣动触发键,然后松开,开始进行测量。内部计时器会自动停止测量。</p><p>如果定时测量已关闭,则操作员必须在听到“砰”响声后松开触发键停止测量。“结果”屏幕大约每 2 秒钟刷新一次,操作员可以决定是否在听到“砰”响声前松开触发键立即停止测量。</p><p>对时间较长的测量和使用台式支架的测量而言,“定时测量”很有帮助。它还可帮助确保一批测量中的所有测量均具有相同的测量时间。</p><p>可以将测量时间设置为 0。由于没有“砰”响声,操作员必须决定何时停止测量。如果定时测量已关闭, 操作员可以扣动触发键开始测量,然后松开触发键停止测量。如果定时测量已开启,操作员可以扣动并 松开触发键开始测量,然后再次扣动并松开触发键停止测量。</p><h3 id="进行安全测量">4 进行安全测量</h3><h4 id="使用正确附件">4.1 使用正确附件</h4><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/使用正确附件1.png" alt=""><figcaption>使用正确附件1</figcaption></figure><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/使用正确附件2.png" alt=""><figcaption>使用正确附件2</figcaption></figure><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/使用正确附件3.png" alt=""><figcaption>使用正确附件3</figcaption></figure><h3 id="维护">5 维护</h3><h4 id="控样校准">5.1 控样校准</h4><blockquote><p>管理员应对控样进行测量。如果合金 CRM 参考测量有明显变化,请联系当地 Hitachi High-Tech Analytical Science 代表获取帮助。</p><p>X-MET8000 系列可包含特定于各种应用情况的其他控样。管理员也可以使用这些控样,根据代表性样品对 X-MET8000 系列进行检查校正。</p></blockquote><blockquote><p>【扩展】光谱标样和光谱控样的作用与区别</p><p>「光谱标样」是由于测量值偏差引起的时间变化引起的仪器工作曲线的校正, 仪器根据标准样品的标准化是进行两点校准, 在整体曲线范围内, 可接受的结果, 样品和用于分析标准样品的冶炼过程、结构 , 成分会不同, 结果确定的几个元素在分析中也可以相同, 导致系统误差。光谱标样又称实物标准。是根据实施和制定文字标准的需要研制,作为文字标准补充的实物。如果测量值与它提供的参考值互相匹配, 则测量方法被认为是合格和可靠的, 相反, 如果不匹配, 则假定测量方法是不合格的且不可行。</p><p>「光谱控样」样品通常是由定点单位专门制作的。通过控样其实是标准化的设备类型之前的样品分析, 通常是为了减少分析误差。样品控制校准是单点校准, 控制样品一般是单元自控, 并分析样品冶炼过程是相同的, 结构、成分基本相同, 通过对样品的控制校准是, 可以减少分析错误。通过控样分析组织的成分, 越接近对照样品, 误差就越小。</p><p>在光电直接读数谱中, 为了使分析结果更加准确, 有必要使用光谱标样和光谱控制样品。光谱标样主要用于绘制设备曲线, 在使用时间过长或清洗和维护时, 设备参数漂移必然会引起工作曲线偏移, 需要通过标准化进行调整。光谱控制样本是通过测定化学成分含量值, 确定光谱仪的工作状态, 并进行调整。</p></blockquote><h4 id="电池维护">5.2 电池维护</h4><blockquote><p>要经常对电池充电。在放电状态存储电池会损坏电池,如果电池长时间处于放电状态,电池会进入深度放电状态,这种情况下,内部保护电路出于安全原因阻止充电。</p><p>即使设备已关闭,也不要将电量用尽的电池放入设备,设备会消耗一小部分电量,而且电池将进入深度放电状态,进而阻止再次充电。</p></blockquote><h3 id="认证标志">6 <a href="https://en.wikipedia.org/wiki/Certification_mark" target="_blank" rel="noopener">认证标志</a></h3><table><colgroup><col style="width: 31%"><col style="width: 34%"><col style="width: 34%"></colgroup><thead><tr class="header"><th>美国 <a href="https://en.wikipedia.org/wiki/FCC_mark" target="_blank" rel="noopener">FCC mark</a></th><th>欧洲经济区 <a href="https://en.wikipedia.org/wiki/CE_marking" target="_blank" rel="noopener">CE marking</a></th><th>中国 <a href="https://en.wikipedia.org/wiki/China_Compulsory_Certificate" target="_blank" rel="noopener">CCC Mark</a></th></tr></thead><tbody><tr class="odd"><td>Federal Communications Commission</td><td>"CE" ="Conformité Européenne"* (<a href="https://en.wikipedia.org/wiki/French_language" target="_blank" rel="noopener">French</a> for "European Conformity")</td><td>China Compulsory Certificate</td></tr><tr class="even"><td><img src="/2020/03/10/XRF-X射线荧光分析设备/FCC_New_Logo.png" alt="FCC_New_Logo"></td><td><img src="/2020/03/10/XRF-X射线荧光分析设备/../../../source/实验室设备/Conformité_Européenne_(logo).png" alt="Conformité_Européenne_(XRF-X射线荧光分析设备/Conformité_Européenne_(logo).png)"></td><td><img src="/2020/03/10/XRF-X射线荧光分析设备/C.C.C.-Logo.png" alt="C.C.C.-Logo"></td></tr></tbody></table><h3 id="扩展xrf">7 扩展<a href="https://en.wikipedia.org/wiki/X-ray_fluorescence" target="_blank" rel="noopener">XRF</a></h3><h4 id="xrf构成">7.1 XRF构成</h4><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/gun-diagram.jpg" alt=""><figcaption>gun-diagram</figcaption></figure><h4 id="ed-xrf-wd-xrf">7.2 ED-XRF & WD-XRF</h4><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/Dmedxrfschematic.jpg" alt=""><figcaption>Dmedxrfschematic</figcaption></figure><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/Dmwdxrfschematic.jpg" alt=""><figcaption>Dmwdxrfschematic</figcaption></figure><h4 id="x-ray-detection-by-eds">7.3 <a href="https://myscope.training/legacy/analysis/eds/xraydetection/" target="_blank" rel="noopener">X-ray detection by EDS</a></h4><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/detection-silicon-2-3741607.png" alt=""><figcaption>detection-silicon-2</figcaption></figure><figure><img src="/2020/03/10/XRF-X射线荧光分析设备/detection-channels-3741612.png" alt=""><figcaption>detection-channels</figcaption></figure>]]></content>
<summary type="html">
<h1 id="xrf-x射线荧光分析设备">XRF-X射线荧光分析设备</h1>
<p><strong>Hitachi High-Tech Analytical Science</strong></p>
<h3 id="检测范围">1 检测范围</h3>
<p>X-MET800
</summary>
</entry>
<entry>
<title>睡眠革命读书笔记</title>
<link href="https://github-9233.github.io/2020/03/08/%E7%9D%A1%E7%9C%A0%E9%9D%A9%E5%91%BD%E8%AF%BB%E4%B9%A6%E7%AC%94%E8%AE%B0/"/>
<id>https://github-9233.github.io/2020/03/08/睡眠革命读书笔记/</id>
<published>2020-03-08T13:36:44.000Z</published>
<updated>2020-03-08T13:37:09.306Z</updated>
<content type="html"><![CDATA[<h1 id="读书笔记睡眠革命">读书笔记:《睡眠革命》</h1><figure><img src="/2020/03/08/睡眠革命读书笔记/cover.png" alt=""><figcaption>cover</figcaption></figure><h2 id="引言-不要浪费宝贵的时间睡觉">★引言 不要浪费宝贵的时间睡觉</h2><h2 id="第一部分-睡眠修复的关键指标">★第一部分 睡眠修复的关键指标</h2><h3 id="时钟在嘀嗒昼夜节律">1.1 时钟在嘀嗒——昼夜节律</h3><figure><img src="/2020/03/08/睡眠革命读书笔记/昼夜节律.png" alt=""><figcaption>昼夜节律</figcaption></figure><blockquote><p>昼夜节律:智慧睡眠的7个要点</p><p>1.走到户外!让日光调整你的生物钟,而不是人工光线。</p><p>2.花时间了解一下你的昼夜节律,以及它将如何影响你的生活。让你的亲友也加入进来。</p><p>3.了解你的各项高峰和低谷数值,监控那些会在你身上自然而然发生的变化。使用穿戴式健身追踪器进行测量。</p><p>4.睡眠质量最高的时间出现在凌晨2-3点。如果在日出之后才去睡觉,你就违背了你的生物钟。</p><p>5.早上放慢速度。醒来之后立马陷入一团忙乱之中,早晚会毁了你的身体。睡眠质量和我们早上醒来之后的所作所为息息相关。</p><p>6.在晚上,蓝光是一种不合时宜的光线,尽量让蓝光变得暗淡。不妨多利用红光和黄光,甚至是蜡烛光。</p><p>7.想象自己此刻在一个岛屿上、坐在篝火边的场景:在你目前正在做的事情中,哪些和这一场景起了冲突?你打算如何纠正过来?对目前的日常安排做出一些简单的改变,使你自己的身体状态更加吻合人体昼夜节律。和我们早上醒来之后的所作所为息息相关。</p><p>8.在晚上,蓝光是一种不合时宜的光线,尽量让蓝光变得暗淡。不妨多利用红光和黄光,甚至是蜡烛光。</p><p>9.想象自己此刻在一个岛屿上、坐在篝火边的场景:在你目前正在做的事情中,哪些和这一场景起了冲突?你打算如何纠正过来?对目前的日常安排做出一些简单的改变,使你自己的身体状态更加吻合人体昼夜节律。</p></blockquote><h3 id="走慢与走快睡眠类型">1.2 走慢与走快——睡眠类型</h3><blockquote><p>睡眠类型:智慧睡眠的7个要点</p></blockquote><blockquote><p>1.了解你和你的亲朋好友属于哪种睡眠类型。如果你不能确定,可以使用慕尼黑大学的问卷调查表。</p><p>2.巧妙规划好你一天的日程,在你状态最佳时做最重要的工作。</p><p>3.把咖啡因当成高效的表现增强剂使用,而不是出于习惯去喝咖啡,并且一天的咖啡因摄入量不要超过400毫克。</p><p>4.晚睡星人——如果你想要克服社交时差,就不要在周末睡懒觉。</p><p>5.在会议室、办公室和办公桌配置日光灯,提高员工的灵敏度、工作效率,并改善他们的工作情绪。</p><p>6.知道何时该上前、何时该退后。如果你是一个早起型的人,那么你该不该自告奋勇地在深夜比赛中参加点球大战呢?</p><p>7.如果你和你的伴侣属于不同的睡眠类型,你们应该学会彼此协调、彼此适应。</p></blockquote><h3 id="分钟睡眠法睡眠周期">1.3 90分钟睡眠法——睡眠周期</h3><figure><img src="/2020/03/08/睡眠革命读书笔记/睡眠周期.png" alt=""><figcaption>睡眠周期</figcaption></figure><blockquote><p>睡眠周期:智慧睡眠的7个要点</p><p>1.固定的起床时间就像一根铁锚,支撑着整个R90方案。设定一个固定的起床时间,并持之以恒地坚持下去。如果你和伴侣同床共眠,让他也这样做,你们的起床时间保持一致是最理想的。</p><p>2.用90分钟时长的睡眠周期衡量睡眠,而不是睡了多少小时。</p><p>3.你可以自行选择入睡时间,但入睡时间取决于你的起床时间。从起床时间出发,根据90分钟时长的睡眠周期,向后推算。</p><p>4.把睡眠放在更长一段时间中考虑,减少不必要的压力。一个“没睡好的糟糕晚上”不会要了你的命,不妨试着去考虑,你每周一共获得了多少睡眠周期。</p><p>5.尽量避免连续三个晚上睡眠不足(少于理想的睡眠周期)的情况发生。</p><p>6.睡眠不是一个简单的数量或质量的问题,试着了解你究竟需要多少睡眠时间。对大多数人来说,每周获得35个睡眠周期是最理想的。28(每晚睡上6小时)~30个睡眠周期也比较理想。如果你睡得比计划中更少,也许会过于疲劳。</p><p>7.争取每周至少有4个晚上能获得理想的睡眠时间。</p></blockquote><h3 id="热身与舒缓睡眠前后的例行程序">1.4 热身与舒缓——睡眠前后的例行程序</h3><blockquote><p>睡眠前后的例行程序:智慧睡眠的7个要点</p><p>1.睡眠前后的例行程序将直接影响你睡眠的质量,以及你清醒的一天:如果予以重视,整个白天和整个晚上就会更加高效。</p><p>2.在日间偶尔脱离电子设备,将此作为一种犒劳,并通过这种方式训练你的身体和心灵。</p><p>3.对于晚睡星人来说,睡眠后例行程序至关重要,如果他们不想输给早起星人的话。不要因为青睐于止闹按钮,而轻易放弃这些睡眠后例行程序。</p><p>4.不要在头脑不清醒的时候发送消息!先让自己清醒一点,再去拿手机。</p><p>5.让你的身体感受从温暖到凉爽的变化,有利于带来自然的体温下降。快速洗个温水浴,然后选择较凉爽的睡眠环境,能帮助你达到目的。</p><p>6.在上床前整理你的卧室,放空你的大脑,“下载”你的一天,就不会在该睡觉的时候想东想西、难以入眠。</p><p>7.执行睡眠前例行程序,是为了结束过去的一天——用鼻子呼吸、放松心情、实现从明亮到昏暗的过渡。执行睡眠后例行程序,是为了开启不匆不忙的新一天。这两段时间都只属于你,不属于别人。</p></blockquote><h3 id="暂停片刻该休息了日间小睡">1.5 暂停片刻,该休息了!——日间小睡</h3><figure><img src="/2020/03/08/睡眠革命读书笔记/睡眠冲动和睡眠需求变化.png" alt=""><figcaption>睡眠冲动和睡眠需求变化</figcaption></figure><blockquote><p>日间小睡:智慧睡眠的7个要点</p><p>1.利用午后休憩时机(下午1—3点)给自己安排一个可控修复期,是弥补夜间睡眠周期的完美方法,这也与你的昼夜节律彼此协调。</p><p>2.黄昏(下午5—7点)是次优的休憩时机,因为此时人的睡眠需求极高。但这一时段的日间小睡应控制在30分钟之内,这样才不会影响晚上的睡眠。</p><p>3.白天睡不着?没关系。只要花30分钟放松一会儿,暂时脱离周围的世界片刻。</p><p>4.至少每隔90分钟休息一会儿,消除大脑的疲劳,提高注意力的集中水平。在休息时不要使用电子设备,你无需让自己自始至终地受到电子设备的控制。</p><p>5.不要受到你所在的企业文化的影响,切莫先入为主地给那些白天睡觉的人贴上“懒惰”的标签,而应着手构建一种接受日间小睡和休憩的企业文化——你若不打盹儿,就会输得很惨。</p><p>6.使用冥想或正念应用软件,或者把玩某个珍爱的私人物品,暂时脱离当前的环境。</p><p>7.如果你真的无法脱身,那就巧妙地安排一天的工作。在午后倦怠期,不要让自己困于太费神的工作。</p></blockquote><h3 id="改造你的床铺寝具套装">1.6 改造你的床铺——寝具套装</h3><figure><img src="/2020/03/08/睡眠革命读书笔记/如何选择床垫.png" alt=""><figcaption>如何选择床垫</figcaption></figure><blockquote><p>寝具套装:智慧睡眠的7个要点</p><p>1.学会以胎儿姿势睡眠,躺向非主要的身体一侧(惯用左手的人向右侧睡,惯用右手的人向左侧睡)。</p><p>2.检测床垫,了解什么样的床垫对你来说软硬适中。让你的伴侣也这样做。</p><p>3.循序渐进:7年间分两次各花500英镑在你的床垫上,而不是一次花1000英镑。可以考虑购买能够经常换洗的床垫。</p><p>4.使用低过敏性、透气舒适的床上用品,无论你是否属于过敏体质。避免潜在的睡眠障碍因素,并注意调节好温度。</p><p>5.床的尺寸很重要——能买多大,就买多大。特大号床垫是值得一对儿夫妇考虑的最小型号(只要卧室中放得下)。双人床是给一个人睡的。</p><p>6.不要盲目购置寝具!可以根据销售人员的介绍,了解你能买到什么样的寝具。但在做出最后决定时,记得运用你在本章中学到的知识。</p><p>7.记住床垫和床架重要性比率:你可以把100%的预算全部花在购置床垫上,因为床架主要是装饰性的。</p></blockquote><h3 id="修复室睡眠环境">1.7 修复室——睡眠环境</h3><blockquote><p>睡眠环境:智慧睡眠的7个要点</p><p>1.你的卧室不应该成为你生活空间的延伸。如果可以的话,把它重新命名为你的身心修复室。</p><p>2.清空你的房间(哪怕只在你的脑海中),只把睡眠、修复、休息所必需的东西拿回房间。</p><p>3.让你的房间隔离外界光源,这样室外的光线就不会干扰你的睡眠。</p><p>4.让你的房间比家中其他地方更凉爽,但不是更寒冷。</p><p>5.你的房间能给你带来安全感——放上你最爱的泰迪熊或者爱人的照片,或者再次去查看一下门窗是否锁好,都有助于营造安全感。</p><p>6.让房间的装饰风格保持中性,保持房间洁净,避免任何可能刺激大脑的东西(色彩鲜明的图片,或让你产生可怕联想的书籍)。</p><p>7.控制房间中电子设备的使用——在夜间关闭待机灯。把手机拿出房间,或至少放在你看不见的地方(并设置成静音)。</p></blockquote>]]></content>
<summary type="html">
<h1 id="读书笔记睡眠革命">读书笔记:《睡眠革命》</h1>
<figure>
<img src="/2020/03/08/睡眠革命读书笔记/cover.png" alt=""><figcaption>cover</figcaption>
</figure>
<h2 id
</summary>
</entry>
<entry>
<title>结构化写作</title>
<link href="https://github-9233.github.io/2020/03/08/%E7%BB%93%E6%9E%84%E5%8C%96%E5%86%99%E4%BD%9C/"/>
<id>https://github-9233.github.io/2020/03/08/结构化写作/</id>
<published>2020-03-08T13:08:09.000Z</published>
<updated>2020-03-08T13:08:22.712Z</updated>
<content type="html"><![CDATA[<h1 id="结构化写作">结构化写作</h1><h2 id="清晰传递信息短时间内快速组织信息并传递">清晰传递信息:短时间内快速组织信息并传递</h2><h3 id="设计标题">设计标题</h3><p>设计一个吸引读者眼球的标题</p><p>好标题要符合SPA原则</p><ul><li>Simple 简单明了 取决于受众是否能快速听明白</li><li>Profit 利益相关<ul><li>得到的结果(好的效果、好的状态、好的结局)</li><li>产生的影响(好的影响、直接间接的影响、近期长远的影响)</li></ul></li><li>Accurate 准确客观<ul><li>基本要素(时间、地点、人物)</li><li>采取的行动(措施、方法、手段)</li></ul></li></ul><h3 id="撰写序言">撰写序言</h3><p>写一段让读者愿意看下去的精彩开头</p><p>SCQA(情境situation、冲突complication、疑问question、回答answer)讲故事的方法</p><h3 id="展开内容">展开内容</h3><p>快速构建职场常见场景下的行文脉络</p><p>WHAT WHY HOW 并列式/层进式</p><h2 id="准确总结工作写一份有思想有高度的工作总结">准确总结工作:写一份有思想有高度的工作总结</h2><h3 id="成果分类">成果分类</h3><p>发掘为自己加分的关键工作成果,并进行分类</p><ul><li>开放式分类:核心原则MECE(mutually exclusive —— no overlaps 相互独立;collectively exhaustive —— no gaps 完全穷尽)</li><li>封闭式分类:调用成熟思考框架先建立分类</li></ul><h3 id="排序整理">排序整理</h3><p>让工作总结清晰、一目了然</p><ul><li>时间顺序</li><li>结构顺序</li><li>重要顺序</li></ul><h3 id="概括总结">概括总结</h3><p>将自己的工作业绩写出高度</p><p>关键是形成【结论/判断】</p><ul><li>信息摘要法(抽取共通的属性/特征)归纳</li><li>逻辑推论法(导向共同的结果/目的)论证</li></ul><h2 id="充分说服他人有理有据地进行职场说服">充分说服他人:有理有据地进行职场说服</h2><h3 id="明确观点">明确观点</h3><p>说服他人的前提是一个鲜明的观点</p><p>制定目标的WA法(who action)</p><h3 id="疑问回答">疑问回答</h3><p>用充分地依据回答读者心中的疑问</p><p>自上而下的疑问/问答式对话</p><h3 id="逻辑归整">逻辑归整</h3><p>用逻辑论证使说服更有力量</p><ul><li>演绎论证(从普遍性的理论知识出发,去认识个别的、特殊的现象的一种论证推理方法)</li><li>归纳论证(从许多个别的事物中概括出一般性概念、原则或结论的推理方法)</li></ul><h2 id="有力汇报方案写出有价值有力度的方案">有力汇报方案:写出有价值有力度的方案</h2><p>基于三层次模型的说服性写作5个步骤</p><ul><li>理解<ul><li>第一步:描述问题定方向 描述问题5W2H</li></ul></li><li>重构<ul><li>第二步:基于目标定主题 确定目标WA-->设计序言SCQA-->设定主题SPA</li><li>第三步:纵向结构分层次 自上而下(疑问-回答) + 自下而上(共性分类)</li><li>第四步:横向结构选顺序 演绎结构 or 归纳结构</li></ul></li><li>呈现<ul><li>第五步: 形象表达做演示<ul><li>卖点包装(配得上)<ul><li>配关系:流动、作用、比较、关系</li><li>得图示</li><li>上包装:简化、整合、引用、类比</li></ul></li><li>形象表达</li></ul></li></ul></li></ul>]]></content>
<summary type="html">
<h1 id="结构化写作">结构化写作</h1>
<h2 id="清晰传递信息短时间内快速组织信息并传递">清晰传递信息:短时间内快速组织信息并传递</h2>
<h3 id="设计标题">设计标题</h3>
<p>设计一个吸引读者眼球的标题</p>
<p>好标题要符合SPA原则
</summary>
</entry>
<entry>
<title>手环测心率</title>
<link href="https://github-9233.github.io/2020/03/08/%E6%89%8B%E7%8E%AF%E6%B5%8B%E5%BF%83%E7%8E%87/"/>
<id>https://github-9233.github.io/2020/03/08/手环测心率/</id>
<published>2020-03-08T13:01:29.000Z</published>
<updated>2020-03-08T13:06:24.829Z</updated>
<content type="html"><![CDATA[<h1 id="心率监测的原理">心率监测的原理</h1><h2 id="光体积变化描记图法photoplethysmographyppg">1 <a href="https://zh.wikipedia.org/zh-cn/光體積變化描記圖法" target="_blank" rel="noopener">光体积变化描记图法</a>Photoplethysmography(PPG)</h2><blockquote><p><a href="https://www.sekorm.com/news/8674.html" target="_blank" rel="noopener">以下详细介绍Si1144心率传感器如何通过光电式PPG信号实现心率监测:</a></p><figure><img src="/2020/03/08/手环测心率/Si1144心率传感器如何通过光电式PPG信号实现.png" alt=""><figcaption>Si1144心率传感器如何通过光电式PPG信号实现</figcaption></figure><p>1、首先Silicon Labs Si1144 心率传感器采用PPG光电式测量,内置绿色LED是525nm光波段,采用绿光LED是因为其穿透力强,同时容易被血红蛋白吸收。</p><p>2、LED发出的绿光经过人体的皮肤进入皮下组织,皮下组织吸收后漫反射至SI1144传感器的接收端。通过人体血红蛋白吸收绿光的特性,皮下组织血液脉动使血液密度发生改变,随着血流量的变化,光的反射量发生了变化。</p><p>3、Si1144传感器接收到皮下组织漫反射回来的光电信号后,接收的光电脉冲信号比发光LED的强度小100倍左右,信号比较微弱,经过滤波和信号放大处理后最终转化为数字信号,通过I2C接口与MCU连接,输出接收光强度的数字信号,然后经过MCU算法处理后计算出心率值。</p></blockquote><blockquote><p>互补色原理</p><figure><img src="/2020/03/08/手环测心率/互补色.png" alt=""><figcaption>互补色</figcaption></figure></blockquote><h2 id="心电图ecg">2 心电图ECG</h2><p>一是基于生物电势测量的方法如ECG(心电图),通过测量ECG的方法获得心率,这是最准确的方法。这需要两个或三个电极,一个放在手环/手表的上面(使用手指接触),其他的一个或两个放在下面(接触皮肤),使用时将一只手搭在另一只手的手环/手表电极上完成测量。</p><blockquote><h3 id="通过心脏的电传导如何产生ecg"><a href="https://en.wikipedia.org/wiki/Electrocardiography#How_the_Electrical_Conduction_Through_the_Heart_Creates_the_ECG" target="_blank" rel="noopener">通过心脏的电传导如何产生ECG</a></h3><figure><img src="/2020/03/08/手环测心率/Animation_of_ECG_Limb_Leads_and_Electrical_Conduction_Through_the_Heart.gif" alt=""><figcaption>Animation_of_ECG_Limb_Leads_and_Electrical_Conduction_Through_the_Heart</figcaption></figure><h3 id="肢体导联">肢体导联</h3><figure><img src="/2020/03/08/手环测心率/EKG_leads.png" alt=""><figcaption>EKG_leads</figcaption></figure><h3 id="临床导联组">临床导联组</h3><figure><img src="/2020/03/08/手环测心率/临床导联组.png" alt=""><figcaption>临床导联组</figcaption></figure><figure><img src="/2020/03/08/手环测心率/EKG-Animation.gif" alt=""><figcaption>EKG-Animation</figcaption></figure></blockquote><blockquote><h3 id="apple-watch-如何测量心率"><a href="https://support.apple.com/zh-cn/guide/watch/apda88aefe4c/watchos" target="_blank" rel="noopener">Apple Watch 如何测量心率</a></h3><p>Apple Watch 的光学心率传感器采用的是光体积描记法。这个技术名词念起来貌似复杂难懂,实际上原理很简单:血液之所以呈现红色,是因为它反射红光并吸收绿光。Apple Watch 使用绿色 LED 灯,配合对光敏感的光电二极管,检测任意时间点流经手腕的血液流量。手腕血流量和绿光吸收量会在心脏跳动时增加,在心跳间隔期间减少。通过 LED 灯每秒数百次的闪烁,Apple Watch 可以算出您每分钟的心跳次数,也就是您的心率。光学心率传感器的测量范围为每分钟心跳 30 至 210 次。此外,特别设计的光学心率传感器可以提高 LED 亮度和采样率,进而弥补信号弱的问题。</p><p>光学心率传感器还会利用红外线。当 Apple Watch 在后台测量心率以及为提供心率通知而测量心率时,它就会使用这种模式。Apple Watch 使用绿色 LED 灯测量您在体能训练期间和“呼吸”时段的心率,并计算步行平均心率和心率变异性 (HRV)。</p><figure><img src="/2020/03/08/手环测心率/watch-series4-measure-ecg-sensors.png" alt=""><figcaption>watch-series4-measure-ecg-sensors</figcaption></figure><p>Apple Watch Series 4 或更新机型的数码表冠内和背面都有内建电极,这些电极可以在与“心率”应用或“心电图”应用搭配使用时测量通过您心脏的电信号。当您将手指放在数码表冠上时,它会在您的心脏和双臂之间形成一个闭合电路,从而捕捉到通过您胸部的电脉冲。</p><p>要使用电极式心率传感器测量您的心率,请打开“心率”应用,然后将您的手指放在数码表冠上。您获得读数的速度将更快,而且精度更高,因为是每 1 秒而不是每 5 秒获得一次测量值。在“健康”应用中查看记录的心率数据时,您会在“心率情境”中看到“心电图”。您还可以通过电极式心率传感器来<a href="https://support.apple.com/zh-cn/HT208955" target="_blank" rel="noopener">使用“心电图”应用生成心电图</a>。</p></blockquote><h2 id="心冲击图ballistocardiograph-bcg">3 心冲击图<a href="https://en.wikipedia.org/wiki/Ballistocardiography" target="_blank" rel="noopener">ballistocardiograph (BCG)</a></h2><p>一种测量由心脏产生的<a href="https://en.wikipedia.org/wiki/Ballistics" target="_blank" rel="noopener">ballistic</a> forces 的方法。血液通过<a href="https://en.wikipedia.org/wiki/Descending_aorta" target="_blank" rel="noopener">降主动脉</a>的向下运动会产生<a href="https://en.wikipedia.org/wiki/Newton's_third_law" target="_blank" rel="noopener">向上的后坐力</a>,每次心跳都会使身体向上运动。当主动脉的不同部分扩张和收缩时,身体继续以重复的方式上下移动。Ballistocardiography是一种产生图形表示的人体重复运动的技术,该运动是由于每次心跳使血液突然喷射到大血管中而引起的。这是一个在1–20 <a href="https://en.wikipedia.org/wiki/Hertz" target="_blank" rel="noopener">Hz的</a> <a href="https://en.wikipedia.org/wiki/Frequency" target="_blank" rel="noopener">频率</a>范围内的生命体征,这是由<a href="https://en.wikipedia.org/wiki/Heart" target="_blank" rel="noopener">心脏</a>的机械运动引起的,可以通过<a href="https://en.wikipedia.org/wiki/Noninvasive" target="_blank" rel="noopener">无创</a>方法从<a href="https://en.wikipedia.org/wiki/Human_body" target="_blank" rel="noopener">身体</a>表面记录下来。在<a href="https://en.wikipedia.org/wiki/Isaac_Starr" target="_blank" rel="noopener">艾萨克·斯塔尔</a>(<a href="https://en.wikipedia.org/wiki/Isaac_Starr" target="_blank" rel="noopener">Isaac Starr)</a>博士进行了广泛的研究工作之后,首次显示出可以通过观察和分析BCG信号来识别主要心脏功能障碍的影响。[<a href="https://en.wikipedia.org/wiki/Ballistocardiography#cite_note-5" target="_blank" rel="noopener">5]</a>最近的工作还验证了可以使用相机以非接触方式监视BCG 。</p><h2 id="小米手环数据导出">4 小米手环数据导出</h2><p>小米账号分为国际版账号和国内账号,国内帐号才使用公交卡、NFC模拟服务 如果登录的是非国内的帐号,“关于”里面会有行使用户权利一项,可以清空数据,注销帐号,撤回同意,导出数据。 如果登录的是国内的帐号,登录 http://mifit.huami.com/t/account_mifit</p>]]></content>
<summary type="html">
<h1 id="心率监测的原理">心率监测的原理</h1>
<h2 id="光体积变化描记图法photoplethysmographyppg">1 <a href="https://zh.wikipedia.org/zh-cn/光體積變化描記圖法" target="_blank"
</summary>
</entry>
<entry>
<title>科学上网</title>
<link href="https://github-9233.github.io/2020/02/23/%E7%A7%91%E5%AD%A6%E4%B8%8A%E7%BD%91/"/>
<id>https://github-9233.github.io/2020/02/23/科学上网/</id>
<published>2020-02-23T02:53:56.000Z</published>
<updated>2020-02-26T16:26:25.357Z</updated>
<content type="html"><![CDATA[<h1 id="v2raywebsockettlswebcdn安装配置教程含ipv6mtproto">V2Ray(WebSocket+TLS+Web+CDN)安装配置教程含ipv6/MTProto</h1><h2 id="背景知识">1. 背景知识</h2><p>以下示例中的域名均为<code>www.example.tk</code>,域名、证书、UUID以及MTProto的密钥请记得替换成自己的</p><h3 id="代理原理">1.1 代理原理</h3><blockquote><figure><img src="/2020/02/23/科学上网/pc-firewall-cdn-vps.png" alt=""><figcaption>pc-firewall-cdn-vps</figcaption></figure><p><img src="/2020/02/23/科学上网/v2ray%20的代理原理和配置项的关系.png" alt="v2ray 的代理原理和配置项的关系">上图为 v2ray v4.0 之前的代理原理图。v2ray v4.1+ 合并了输入输出配置项:</p><ul><li><code>inbounds = inbound + inboundDetour</code></li><li><code>outbounds = outbound + outboundDetour</code></li><li>为便于理解,你可以将上图 <code>inbound</code>、<code>inboundDetour</code> 视为各种不同的代理方式,全部定义在新版配置文件的 <code>inbounds</code> 配置项中,<code>outbounds</code> 亦同理。</li></ul></blockquote><h3 id="vmess协议">1.2 VMess协议</h3><blockquote><p>V2Ray 使用 inbound(传入) 和 outbound(传出) 的结构,这样的结构非常清晰地体现了数据包的流动方向,同时也使得 V2Ray 功能强大复杂的同时而不混乱,清晰明了。形象地说,我们可以把 V2Ray 当作一个盒子,这个盒子有入口和出口(即 inbound 和 outbound),我们将数据包通过某个入口放进这个盒子里,然后这个盒子以某种机制(这个机制其实就是路由,后面会讲到)决定这个数据包从哪个出口吐出来。以这样的角度理解的话,V2Ray 做客户端,则 inbound 接收来自浏览器数据,由 outbound 发出去(通常是发到 V2Ray 服务器);V2Ray 做服务器,则 inbound 接收来自 V2Ray 客户端的数据,由 outbound 发出去(通常是如 Google 等想要访问的目标网站)。</p><p>无论是客户端还是服务器,配置文件都由两部分内容组成: <code>inbounds</code> 和 <code>outbounds</code>。V2Ray 没有使用常规代理软件的 C/S(即客户端/服务器)结构,它既可以当做服务器也可以作为客户端。于是我们可以从另一个角度来理解,认为每一个 V2Ray 都是一个节点,<code>inbound</code> 是关于如何与上一个节点连接的配置,<code>outbound</code> 是关于如何与下一个节点连接的配置。对于第一个节点,<code>inbound</code> 与浏览器连接;对于最后一个节点,<code>outbound</code>与目标网站连接。<code>inbounds</code> 和 <code>outbounds</code> 是 <code>inbound</code> 和 <code>outbound</code>的集合,意味着每一个 V2Ray 节点都可以有多个入口和出口。</p></blockquote><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">{浏览器} <--(socks)--> {V2Ray 客户端 inbound <-> V2Ray 客户端 outbound} <--(VMess)--> {V2Ray 服务器 inbound <-> V2Ray 服务器 outbound} <--(Freedom)--> {目标网站}</span><br></pre></td></tr></table></figure><h2 id="参考文档">2. 参考文档</h2><p><a href="https://www.v2ray.com/" target="_blank" rel="noopener">V2Ray官方文档</a> <a href="https://toutyrater.github.io/" target="_blank" rel="noopener">V2Ray白话文教程</a> <a href="https://github.com/TelegramMessenger/MTProxy" target="_blank" rel="noopener">MTProxy官方文档</a></p><h2 id="安装部署websockettlswebcdn">3. 安装部署WebSocket+TLS+Web+CDN</h2><h3 id="准备">3.1 准备</h3><h4 id="端口规划">3.1.1 端口规划</h4><table><thead><tr class="header"><th style="text-align: left;">Port</th><th style="text-align: left;">Listen IP</th><th style="text-align: left;">Use</th></tr></thead><tbody><tr class="odd"><td style="text-align: left;">443</td><td style="text-align: left;">0.0.0.0</td><td style="text-align: left;">Nginx Https</td></tr><tr class="even"><td style="text-align: left;">10000</td><td style="text-align: left;">127.0.0.1</td><td style="text-align: left;">V2Ray WebSocket</td></tr><tr class="odd"><td style="text-align: left;">8000</td><td style="text-align: left;">0.0.0.0</td><td style="text-align: left;">V2Ray MTProto</td></tr></tbody></table><h4 id="服务器准备">3.1.2 服务器准备</h4><p><a href="https://bandwagonhost.com/" target="_blank" rel="noopener">BandwagonHOST搬瓦工</a></p><p><a href="https://www.bandwagonhost.net" target="_blank" rel="noopener">搬瓦工非官方中文网</a></p><p>Plan:<strong>SPECIAL 20G KVM PROMO V5 - LOS ANGELES - CN2 GIA ECOMMERCE</strong></p><p>OS:Centos-7-x86_64-bbr(自带bbr加速)</p><p>VPS Type: KiwiVM</p><h4 id="申请免费域名">3.1.3 申请免费域名</h4><p><a href="https://www.freenom.com" target="_blank" rel="noopener">Freenom - A Name for Everyone</a> <code>.tk</code> <code>.ml</code>不需要实名 免费一年</p><blockquote><h4 id="域名基本概念">域名基本概念</h4><p>首先说明一下域名的基本概念,你购买的域名是一个二级域名,形式类似于 <strong>xxx.后缀</strong>,比如 <strong>baidu.com</strong>、<strong>google.com</strong> 就是二级域名,com 是后缀。</p><p>你可以在你的二级域名基础上添加三级域名,比如 <strong>haha.baidu.com</strong>、<strong>heihei.google.com</strong>。甚至是四级或者更多级,比如 <strong>haha.heihei.baidu.com</strong>、<strong>hehe.hehe.hehe.google.com</strong>,因为二级域名已经是你的了,所以你想添加什么就添加什么。</p><p>光设置一个三级域名还不行,你得把你的域名和你的 IP 地址结合起来,这个 IP 就叫做域名的解析记录,解析记录有许多不同的类型,常用的有 A、AAAA、CNAME 等等。</p><ul><li>A 类型:解析到一个 IPv4 地址</li><li>AAAA 类型:解析到一个 IPv6 地址</li><li>CNAME 类型:解析到另一个域名,由该域名来提供 IP 地址</li></ul><p>假设要将三级域名 haha.baidu.com 解析到 IP 地址 123.123.123.123 上,那么通常的做法是,选择添加解析记录,名称设置为 haha,类型设置为 A,解析记录填入 123.123.123.123,设置好之后 haha.baidu.com 就会解析到 123.123.123.123 了</p></blockquote><h4 id="使用免费cdn">3.1.4 使用免费CDN</h4><p><a href="https://www.cloudflare.com" target="_blank" rel="noopener">Cloudflare</a></p><blockquote><h4 id="localhost0.0.0.0本机ip地址区别">127.0.0.1、localhost、0.0.0.0、本机IP地址区别</h4><h5 id="ip类型"><a href="https://zh.wikipedia.org/zh-cn/IP地址" target="_blank" rel="noopener">IP类型</a></h5><figure><img src="/2020/02/23/科学上网/IP%20address%20class.png" alt=""><figcaption>IP address class</figcaption></figure><p>在A类、B类、C类IP地址中,如果主机号是全1,那么这个地址为直接广播地址,它是用来使路由器将一个分组以广播形式发送给特定网络上的所有主机。<a href="https://zh.wikipedia.org/wiki/32位" target="_blank" rel="noopener">32位</a>全为1的<strong>IP地址</strong>“255.255.255.255”为<strong>受限广播地址</strong>("limited broadcast" destination address),用来将一个分组以广播方式发送给本<a href="https://zh.wikipedia.org/wiki/电脑网络" target="_blank" rel="noopener">网络</a>中的所有<a href="https://zh.wikipedia.org/wiki/主機" target="_blank" rel="noopener">主机</a>,<a href="https://zh.wikipedia.org/wiki/路由器" target="_blank" rel="noopener">路由器</a>则阻挡该分组通过,将其广播功能限制在本网内部。</p><figure><img src="/2020/02/23/科学上网/IPv4地址分类.png" alt=""><figcaption>IPv4地址分类</figcaption></figure><h5 id="localhost">localhost</h5><p>首先,localhost 就是一个域名,域名指向 127.0.0.1,两者是一样的,如果是 mac,可以在 <code>/etc/hosts</code> 文件中看到这一行。</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">> 127.0.0.1 localhost</span><br><span class="line">></span><br></pre></td></tr></table></figure></blockquote><blockquote><h5 id="section">127.0.0.1</h5><p>127.0.0.1 这个地址通常分配给 loopback 接口,用来测试本机的 TCP/IP 协议栈。loopback 是一个特殊的网络接口(可理解成虚拟网卡),用于本机中各个应用之间的网络交互。只要操作系统的网络组件是正常的,loopback 就能工作。只要使用这个地址发送数据,则数据包不会出现在网络传输过程中。</p><h5 id="section-1">0.0.0.0</h5><p>这个 IP 相当于 java 中的 this,代表当前设备的 IP,可以代表本机的所有 IP 地址。</p><p>比如你的一个 tomcat 配置文件中,如果监听的 IP 地址设置了 0.0.0.0,就表示你的这个 tomcat 服务器监听在你本机的所有 IP 地址上,通过任何一个 IP地址 都可以访问到。</p><p>如果你本地的 IP 地址有 192.168.1.10,172.16.2.10,那么访问你这个 tomcat 就可以。 http://192.168.1.10:8080/,http://172.16.2.10:8080/ 都可以访问。</p><p>如果你的监听地址设置了 192.168.1.10 ,那么将无法通过 http://172.16.2.10:8080/ 进行访问。</p><h5 id="本机-ip">本机 IP</h5><p>至于本机 IP,你可以理解为本机有三块网卡,一块网卡叫做 loopback(这是一块虚拟网卡),另外一块网卡叫做 ethernet (这是你的有线网卡),另外一块网卡叫做 wlan(这是你的无线网卡)。</p><p>你的本机 IP 是你真实网卡的 IP,具体来说有线无线各有一个,而 127.0.0.1 是那块叫做 loopback 的虚拟网卡的 IP。</p></blockquote><h4 id="开启ssl">3.1.5 开启SSL</h4><p><a href="https://www.freessl.com" target="_blank" rel="noopener">SSL For Free - Free SSL Certificate in Minutes</a></p><p>生成<code>www.example.tk_chain.crt</code> <code>www.example.tk_key.key</code></p><h4 id="免费获得ipv6访问byr北邮人bt站">3.1.6 免费获得IPv6访问<a href="https://bt.byr.cn" target="_blank" rel="noopener">byr北邮人bt站</a></h4><p>工具</p><ul><li><a href="https://tunnelbroker.net" target="_blank" rel="noopener">Hurricane Electric - IPv6 Tunnel Broker</a></li><li><a href="https://www.qbittorrent.org" target="_blank" rel="noopener">qBittorrent</a></li><li><a href="https://www.utorrent.com" target="_blank" rel="noopener">uTorrent</a></li></ul><p>参考<a href="https://www.bandwagonhost.net/512.html" target="_blank" rel="noopener">搬瓦工VPS KVM 方案如何使用IPv6 思路]</a></p><h3 id="安装部署v2ray">3.2 安装部署V2Ray</h3><h4 id="安装nginx">3.2.1 安装<a href="https://www.nginx.com" target="_blank" rel="noopener">Nginx</a></h4><p><code>sudo yum install nginx</code></p><p>使用 <code>systemctl start|stop|restart|status nginx</code> 控制 Nginx 的运行</p><h4 id="配置nginx">3.2.2 配置Nginx</h4><p><code>/etc/nginx/conf.d/example.tk.conf</code></p><figure class="highlight shell"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br></pre></td><td class="code"><pre><span class="line">server {</span><br><span class="line"> listen 443 ssl;</span><br><span class="line"> ssl on;</span><br><span class="line"> ssl_certificate /etc/nginx/conf.d/www.example.tk_chain.crt;</span><br><span class="line"> ssl_certificate_key /etc/nginx/conf.d/www.example.tk_key.key;</span><br><span class="line"> ssl_protocols TLSv1 TLSv1.1 TLSv1.2;</span><br><span class="line"> ssl_ciphers HIGH:!aNULL:!MD5;</span><br><span class="line"> server_name example.tk;</span><br><span class="line"> location /ray/ { # 与 V2Ray 配置中的 path 保持一致</span><br><span class="line"> proxy_redirect off;</span><br><span class="line"> proxy_pass http://127.0.0.1:10000;#假设WebSocket监听在环回地址的10000端口上</span><br><span class="line"> proxy_http_version 1.1;</span><br><span class="line"> proxy_set_header Upgrade $http_upgrade;</span><br><span class="line"> proxy_set_header Connection "upgrade";</span><br><span class="line"> proxy_set_header Host $http_host;</span><br><span class="line"> proxy_set_header X-Real-IP $remote_addr;</span><br><span class="line"> proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;</span><br><span class="line"> }</span><br><span class="line">}</span><br></pre></td></tr></table></figure><h4 id="安装v2ray">3.2.3 安装V2Ray</h4><p>参考<a href="https://www.v2ray.com/chapter_00/install.html#linuxscript" target="_blank" rel="noopener">V2Ray官方文档</a>:</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br></pre></td><td class="code"><pre><span class="line">bash <(curl -L -s https://install.direct/go.sh)</span><br></pre></td></tr></table></figure><p>此脚本会配置自动运行脚本。自动运行脚本会在系统重启之后,自动运行 V2Ray。目前自动运行脚本只支持带有 Systemd 的系统,以及 Debian / Ubuntu 全系列。</p><p>脚本运行完成后,你需要:</p><ol type="1"><li>编辑 <code>/etc/v2ray/config.json</code> 文件来配置你需要的代理方式;</li><li>运行 <code>service v2ray start</code>来启动 V2Ray 进程;</li><li>之后可以使用 <code>service v2ray start|stop|status|reload|restart|force-reload</code> 控制 V2Ray 的运行。</li></ol><h4 id="配置v2ray">3.2.4 配置V2Ray</h4><h5 id="为v2ray生成一个uuid">3.2.4.1 为V2Ray生成一个UUID</h5><ul><li><p><a href="https://www.uuidgenerator.net/" target="_blank" rel="noopener">UUID Generator</a></p></li><li><p>Linux 使用命令 <code>cat /proc/sys/kernel/random/uuid</code> 生成</p></li><li><p>使用V2Ray自带的<code>v2ctl</code>工具生成一个新的uuid。</p></li><li><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br></pre></td><td class="code"><pre><span class="line">$ /usr/bin/v2ray/v2ctl uuid</span><br><span class="line">27e0efcc-8e13-fef1-9e82-febebc469b2b</span><br></pre></td></tr></table></figure></li></ul><h5 id="为mtproto生成一个密钥">3.2.4.2 为MTProto生成一个密钥</h5><p>使用此命令生成 MTProto 代理所需要的用户密钥:<code>openssl rand -hex 16</code></p><h5 id="服务器-v2ray-配置">3.2.4.3 服务器 V2Ray 配置</h5><p>修改<code>/etc/v2ray/config.json</code></p><p>可以使用 <code>$ /usr/bin/v2ray/v2ray -test -config /etc/v2ray/config.json</code> 检查配置文件</p><p>完整配置文件如下:</p><figure class="highlight"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br><span class="line">20</span><br><span class="line">21</span><br><span class="line">22</span><br><span class="line">23</span><br><span class="line">24</span><br><span class="line">25</span><br><span class="line">26</span><br><span class="line">27</span><br><span class="line">28</span><br><span class="line">29</span><br><span class="line">30</span><br><span class="line">31</span><br><span class="line">32</span><br><span class="line">33</span><br><span class="line">34</span><br><span class="line">35</span><br><span class="line">36</span><br><span class="line">37</span><br><span class="line">38</span><br><span class="line">39</span><br><span class="line">40</span><br><span class="line">41</span><br><span class="line">42</span><br><span class="line">43</span><br><span class="line">44</span><br><span class="line">45</span><br><span class="line">46</span><br><span class="line">47</span><br><span class="line">48</span><br><span class="line">49</span><br><span class="line">50</span><br><span class="line">51</span><br><span class="line">52</span><br><span class="line">53</span><br><span class="line">54</span><br><span class="line">55</span><br><span class="line">56</span><br><span class="line">57</span><br><span class="line">58</span><br><span class="line">59</span><br><span class="line">60</span><br><span class="line">61</span><br><span class="line">62</span><br><span class="line">63</span><br><span class="line">64</span><br><span class="line">65</span><br><span class="line">66</span><br><span class="line">67</span><br><span class="line">68</span><br><span class="line">69</span><br><span class="line">70</span><br><span class="line">71</span><br><span class="line">72</span><br><span class="line">73</span><br><span class="line">74</span><br><span class="line">75</span><br><span class="line">76</span><br><span class="line">77</span><br><span class="line">78</span><br><span class="line">79</span><br><span class="line">80</span><br><span class="line">81</span><br></pre></td><td class="code"><pre><span class="line">{</span><br><span class="line"> <span class="attr">"log"</span>: {</span><br><span class="line"> "loglevel": "warning",//日志级别debug|info|warning|error|none</span><br><span class="line"> "access": "/var/log/v2ray/access.log",//将访问的记录保存到文件中,这个选项的值是要保存到的文件的路径</span><br><span class="line"> "error": "/var/log/v2ray/error.log"//将错误的记录保存到文件中,这个选项的值是要保存到的文件的路径</span><br><span class="line"> },</span><br><span class="line"> "inbounds": [</span><br><span class="line"> {//主端口配置</span><br><span class="line"> "port": 10000,// 监听端口</span><br><span class="line"> "listen": "127.0.0.1",//只监听 127.0.0.1,避免除本机外的机器探测到开放了 10000 端口</span><br><span class="line"> "protocol": "vmess",// 入口协议为vmess</span><br><span class="line"> "settings": {</span><br><span class="line"> "clients": [</span><br><span class="line"> {</span><br><span class="line"> "id": "27e0efcc-8e13-fef1-9e82-febebc469b2b",//换成自己的UUID</span><br><span class="line"> "alterId": 64</span><br><span class="line"> }</span><br><span class="line"> ]</span><br><span class="line"> },</span><br><span class="line"> "streamSettings": {</span><br><span class="line"> "network": "ws",</span><br><span class="line"> "wsSettings": {</span><br><span class="line"> "path": "/ray/"</span><br><span class="line"> }</span><br><span class="line"> }</span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"tag"</span>: <span class="string">"tg-in"</span>,</span><br><span class="line"> "port": 8000,//telegram访问端口</span><br><span class="line"> "protocol": "mtproto",</span><br><span class="line"> "settings": {</span><br><span class="line"> "users": [</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"secret"</span>: <span class="string">"80e2e037610bac1444ac02979364f666"</span>//用户密钥</span><br><span class="line"> }</span><br><span class="line"> ]</span><br><span class="line"> }</span><br><span class="line"> }</span><br><span class="line">],</span><br><span class="line"> "outbounds": [</span><br><span class="line"> {</span><br><span class="line"> "protocol": "freedom",// 出口协议</span><br><span class="line"> "settings": {}</span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"protocol"</span>: <span class="string">"blackhole"</span>,</span><br><span class="line"> <span class="attr">"settings"</span>: {</span><br><span class="line"> <span class="attr">"response"</span>: {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"none"</span></span><br><span class="line"> }</span><br><span class="line"> },</span><br><span class="line"> <span class="attr">"tag"</span>: <span class="string">"blocked"</span></span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"tag"</span>: <span class="string">"tg-out"</span>,</span><br><span class="line"> <span class="attr">"protocol"</span>: <span class="string">"mtproto"</span>,</span><br><span class="line"> <span class="attr">"settings"</span>: {}</span><br><span class="line"> }</span><br><span class="line"> ],</span><br><span class="line"> "routing": {</span><br><span class="line"> "domainStrategy": "IPOnDemand",</span><br><span class="line"> "settings": {</span><br><span class="line"> "rules": [</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"field"</span>,</span><br><span class="line"> <span class="attr">"ip"</span>: [</span><br><span class="line"> <span class="string">"geoip:private"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outboundTag"</span>: <span class="string">"blocked"</span></span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"field"</span>,</span><br><span class="line"> <span class="attr">"inboundTag"</span>: [</span><br><span class="line"> <span class="string">"tg-in"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outboundTag"</span>: <span class="string">"tg-out"</span></span><br><span class="line"> }</span><br><span class="line"> ]</span><br><span class="line"> }</span><br><span class="line"> }</span><br><span class="line">}</span><br></pre></td></tr></table></figure><h5 id="客户端配置文件供参考">3.2.4.4 客户端配置文件(供参考)</h5><figure class="highlight"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br><span class="line">20</span><br><span class="line">21</span><br><span class="line">22</span><br><span class="line">23</span><br><span class="line">24</span><br><span class="line">25</span><br><span class="line">26</span><br><span class="line">27</span><br><span class="line">28</span><br><span class="line">29</span><br><span class="line">30</span><br><span class="line">31</span><br><span class="line">32</span><br><span class="line">33</span><br><span class="line">34</span><br><span class="line">35</span><br><span class="line">36</span><br><span class="line">37</span><br><span class="line">38</span><br><span class="line">39</span><br><span class="line">40</span><br><span class="line">41</span><br><span class="line">42</span><br><span class="line">43</span><br><span class="line">44</span><br><span class="line">45</span><br><span class="line">46</span><br><span class="line">47</span><br><span class="line">48</span><br><span class="line">49</span><br><span class="line">50</span><br><span class="line">51</span><br><span class="line">52</span><br><span class="line">53</span><br><span class="line">54</span><br><span class="line">55</span><br><span class="line">56</span><br><span class="line">57</span><br><span class="line">58</span><br><span class="line">59</span><br><span class="line">60</span><br><span class="line">61</span><br><span class="line">62</span><br><span class="line">63</span><br><span class="line">64</span><br><span class="line">65</span><br><span class="line">66</span><br><span class="line">67</span><br><span class="line">68</span><br><span class="line">69</span><br><span class="line">70</span><br><span class="line">71</span><br><span class="line">72</span><br><span class="line">73</span><br><span class="line">74</span><br><span class="line">75</span><br><span class="line">76</span><br><span class="line">77</span><br><span class="line">78</span><br><span class="line">79</span><br><span class="line">80</span><br><span class="line">81</span><br><span class="line">82</span><br><span class="line">83</span><br><span class="line">84</span><br><span class="line">85</span><br><span class="line">86</span><br><span class="line">87</span><br><span class="line">88</span><br><span class="line">89</span><br><span class="line">90</span><br><span class="line">91</span><br><span class="line">92</span><br><span class="line">93</span><br><span class="line">94</span><br><span class="line">95</span><br><span class="line">96</span><br><span class="line">97</span><br><span class="line">98</span><br><span class="line">99</span><br><span class="line">100</span><br><span class="line">101</span><br><span class="line">102</span><br><span class="line">103</span><br><span class="line">104</span><br><span class="line">105</span><br><span class="line">106</span><br><span class="line">107</span><br><span class="line">108</span><br><span class="line">109</span><br><span class="line">110</span><br><span class="line">111</span><br><span class="line">112</span><br><span class="line">113</span><br><span class="line">114</span><br><span class="line">115</span><br><span class="line">116</span><br><span class="line">117</span><br><span class="line">118</span><br><span class="line">119</span><br><span class="line">120</span><br><span class="line">121</span><br><span class="line">122</span><br><span class="line">123</span><br><span class="line">124</span><br><span class="line">125</span><br><span class="line">126</span><br><span class="line">127</span><br><span class="line">128</span><br><span class="line">129</span><br><span class="line">130</span><br><span class="line">131</span><br><span class="line">132</span><br><span class="line">133</span><br><span class="line">134</span><br><span class="line">135</span><br><span class="line">136</span><br><span class="line">137</span><br><span class="line">138</span><br><span class="line">139</span><br><span class="line">140</span><br><span class="line">141</span><br><span class="line">142</span><br><span class="line">143</span><br><span class="line">144</span><br><span class="line">145</span><br><span class="line">146</span><br><span class="line">147</span><br><span class="line">148</span><br><span class="line">149</span><br><span class="line">150</span><br><span class="line">151</span><br><span class="line">152</span><br><span class="line">153</span><br><span class="line">154</span><br><span class="line">155</span><br><span class="line">156</span><br><span class="line">157</span><br><span class="line">158</span><br><span class="line">159</span><br><span class="line">160</span><br></pre></td><td class="code"><pre><span class="line">{</span><br><span class="line"> <span class="attr">"inbounds"</span>: [</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"port"</span>: <span class="number">1087</span>,</span><br><span class="line"> <span class="attr">"listen"</span>: <span class="string">"127.0.0.1"</span>,</span><br><span class="line"> <span class="attr">"protocol"</span>: <span class="string">"http"</span>,</span><br><span class="line"> <span class="attr">"settings"</span>: {</span><br><span class="line"> <span class="attr">"allowTransparent"</span>: <span class="literal">true</span></span><br><span class="line"> }</span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"port"</span>: <span class="number">1086</span>,</span><br><span class="line"> <span class="attr">"listen"</span>: <span class="string">"127.0.0.1"</span>,</span><br><span class="line"> <span class="attr">"protocol"</span>: <span class="string">"socks"</span>,</span><br><span class="line"> <span class="attr">"domainOverride"</span>: [</span><br><span class="line"> <span class="string">"tls"</span>,</span><br><span class="line"> <span class="string">"http"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"settings"</span>: {</span><br><span class="line"> <span class="attr">"auth"</span>: <span class="string">"noauth"</span>,</span><br><span class="line"> <span class="attr">"udp"</span>: <span class="literal">true</span></span><br><span class="line"> }</span><br><span class="line"> }</span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outbounds"</span>: [</span><br><span class="line"> {</span><br><span class="line"> "protocol": "vmess",// 出口协议为vmess</span><br><span class="line"> "settings": {</span><br><span class="line"> "vnext": [</span><br><span class="line"> {</span><br><span class="line"> "address": "www.example.tk",// 服务器地址,请修改为你自己的服务器 IP 或域名</span><br><span class="line"> "port": 443,// 服务器端口</span><br><span class="line"> "users": [</span><br><span class="line"> {</span><br><span class="line"> "id": "27e0efcc-8e13-fef1-9e82-febebc469b2b",// 用户 ID,必须与服务器端配置相同</span><br><span class="line"> "alterId": 64,// 此处的值也应当与服务器相同</span><br><span class="line"> "security": "auto"</span><br><span class="line"> }</span><br><span class="line"> ]</span><br><span class="line"> }</span><br><span class="line"> ]</span><br><span class="line"> },</span><br><span class="line"> "streamSettings": {</span><br><span class="line"> "network": "ws",</span><br><span class="line"> "wsSettings": {</span><br><span class="line"> "path": "/wss"// 此处的值也应当与服务器相同</span><br><span class="line"> },</span><br><span class="line"> "security": "tls"</span><br><span class="line"> },</span><br><span class="line"> "mux": {</span><br><span class="line"> "enabled": false,</span><br><span class="line"> "concurrency": 8</span><br><span class="line"> },</span><br><span class="line"> "tag": "proxy"</span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"protocol"</span>: <span class="string">"freedom"</span>,</span><br><span class="line"> <span class="attr">"settings"</span>: {},</span><br><span class="line"> <span class="attr">"tag"</span>: <span class="string">"direct"</span></span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"protocol"</span>: <span class="string">"blackhole"</span>,</span><br><span class="line"> <span class="attr">"settings"</span>: {},</span><br><span class="line"> <span class="attr">"tag"</span>: <span class="string">"block"</span></span><br><span class="line"> }</span><br><span class="line"> ],</span><br><span class="line"> "log": {</span><br><span class="line"> "loglevel": "none",</span><br><span class="line"> "access": "D:/v2ray_access.log",</span><br><span class="line"> "error": "D:/v2ray_error.log"</span><br><span class="line"> },</span><br><span class="line"> "dns": {</span><br><span class="line"> "hosts": {</span><br><span class="line"> "example.com": "127.0.0.1"</span><br><span class="line"> },</span><br><span class="line"> "servers": [</span><br><span class="line"> "223.5.5.5",</span><br><span class="line"> "8.8.8.8",</span><br><span class="line"> <span class="string">"localhost"</span></span><br><span class="line"> ]</span><br><span class="line"> },</span><br><span class="line"> "routing": {</span><br><span class="line"> "strategy": "rules",</span><br><span class="line"> "settings": {</span><br><span class="line"> "domainStrategy": "IPIfNonMatch",</span><br><span class="line"> "rules": [</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"field"</span>,</span><br><span class="line"> <span class="attr">"domain"</span>: [</span><br><span class="line"> <span class="string">"dropbox"</span>,</span><br><span class="line"> <span class="string">"github"</span>,</span><br><span class="line"> <span class="string">"google"</span>,</span><br><span class="line"> <span class="string">"instagram"</span>,</span><br><span class="line"> <span class="string">"netflix"</span>,</span><br><span class="line"> <span class="string">"pinterest"</span>,</span><br><span class="line"> <span class="string">"pixiv"</span>,</span><br><span class="line"> <span class="string">"tumblr"</span>,</span><br><span class="line"> <span class="string">"twitter"</span>,</span><br><span class="line"> <span class="string">"domain:facebook.com"</span>,</span><br><span class="line"> <span class="string">"domain:fbcdn.net"</span>,</span><br><span class="line"> <span class="string">"domain:fivecdm.com"</span>,</span><br><span class="line"> <span class="string">"domain:ggpht.com"</span>,</span><br><span class="line"> <span class="string">"domain:gstatic.com"</span>,</span><br><span class="line"> <span class="string">"domain:line-scdn.net"</span>,</span><br><span class="line"> <span class="string">"domain:line.me"</span>,</span><br><span class="line"> <span class="string">"domain:medium.com"</span>,</span><br><span class="line"> <span class="string">"domain:naver.jp"</span>,</span><br><span class="line"> <span class="string">"domain:pximg.net"</span>,</span><br><span class="line"> <span class="string">"domain:t.co"</span>,</span><br><span class="line"> <span class="string">"domain:twimg.com"</span>,</span><br><span class="line"> <span class="string">"domain:youtube.com"</span>,</span><br><span class="line"> <span class="string">"domain:ytimg.com"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outboundTag"</span>: <span class="string">"proxy"</span></span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"field"</span>,</span><br><span class="line"> <span class="attr">"ip"</span>: [</span><br><span class="line"> <span class="string">"125.209.222.0/24"</span>,</span><br><span class="line"> <span class="string">"149.154.167.0/24"</span>,</span><br><span class="line"> <span class="string">"149.154.175.0/24"</span>,</span><br><span class="line"> <span class="string">"91.108.56.0/24"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outboundTag"</span>: <span class="string">"proxy"</span></span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"field"</span>,</span><br><span class="line"> <span class="attr">"domain"</span>: [</span><br><span class="line"> <span class="string">"cctv"</span>,</span><br><span class="line"> <span class="string">"geosite:cn"</span>,</span><br><span class="line"> <span class="string">"umeng"</span>,</span><br><span class="line"> <span class="string">"domain:apple.com"</span>,</span><br><span class="line"> <span class="string">"domain:crashlytics.com"</span>,</span><br><span class="line"> <span class="string">"domain:icloud.com"</span>,</span><br><span class="line"> <span class="string">"domain:ixigua.com"</span>,</span><br><span class="line"> <span class="string">"domain:pstatp.com"</span>,</span><br><span class="line"> <span class="string">"domain:snssdk.com"</span>,</span><br><span class="line"> <span class="string">"domain:toutiao.com"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outboundTag"</span>: <span class="string">"direct"</span></span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"field"</span>,</span><br><span class="line"> <span class="attr">"ip"</span>: [</span><br><span class="line"> <span class="string">"geoip:cn"</span>,</span><br><span class="line"> <span class="string">"geoip:private"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outboundTag"</span>: <span class="string">"direct"</span></span><br><span class="line"> },</span><br><span class="line"> {</span><br><span class="line"> <span class="attr">"type"</span>: <span class="string">"field"</span>,</span><br><span class="line"> <span class="attr">"domain"</span>: [</span><br><span class="line"> <span class="string">"domain:doubleclick.net"</span></span><br><span class="line"> ],</span><br><span class="line"> <span class="attr">"outboundTag"</span>: <span class="string">"block"</span></span><br><span class="line"> }</span><br><span class="line"> ]</span><br><span class="line"> }</span><br><span class="line"> }</span><br><span class="line">}</span><br></pre></td></tr></table></figure>]]></content>
<summary type="html">
<h1 id="v2raywebsockettlswebcdn安装配置教程含ipv6mtproto">V2Ray(WebSocket+TLS+Web+CDN)安装配置教程含ipv6/MTProto</h1>
<h2 id="背景知识">1. 背景知识</h2>
<p>以下示例中的
</summary>
</entry>
<entry>
<title>hack-miband4-nfc</title>
<link href="https://github-9233.github.io/2020/02/22/hack-miband4-nfc/"/>
<id>https://github-9233.github.io/2020/02/22/hack-miband4-nfc/</id>
<published>2020-02-22T07:15:54.000Z</published>
<updated>2020-02-29T10:54:47.110Z</updated>
<content type="html"><![CDATA[<h1 id="小米手环复制mifare-门禁卡"><a href="https://blog.czbix.com/clone-MIFARE-Classic-card.html" target="_blank" rel="noopener">小米手环复制Mifare 门禁卡</a></h1><h2 id="背景知识">背景知识</h2><h3 id="市场情况">市场情况</h3><p>门禁卡种类:</p><ul><li>ID卡,是早期的非接触式电子标签,工作频段在 125kHz,芯片中只记录了一个ID号,不可以存储任何数据卡,无需任何权限即可读取。由于太容易复制已经基本退出市场。</li><li><a href="https://www.mifare.net/en/products/chip-card-ics/mifare-classic/" target="_blank" rel="noopener">MIFARE Classic</a>,又叫 M1 卡,是 NXP 半导体公司的产品,属于IC卡的一种,由于成本低廉也是目前在中国市场上使用最广泛的门禁卡技术,使用场景包括且不限于门禁卡,还有饭卡、水卡等。工作频段在 13.56MHz,符合 ISO/IEC 14443 Type A (国内二代身份证是 Type B)标准,有数据存储空间并可反复擦写数据,数据读写都需要密码验证。</li></ul><p>白卡类型:</p><table><thead><tr class="header"><th style="text-align: center;">类型</th><th style="text-align: left;">说明</th></tr></thead><tbody><tr class="odd"><td style="text-align: center;">M1</td><td style="text-align: left;">普通 IC 卡,除 0 扇区不可修改,其他扇区可重复擦写</td></tr><tr class="even"><td style="text-align: center;">UID</td><td style="text-align: left;">普通 IC 复制卡(Chinese clones),可重复擦写所有扇区</td></tr><tr class="odd"><td style="text-align: center;">CUID</td><td style="text-align: left;">用于绕过系统对 UID 卡的拦截</td></tr><tr class="even"><td style="text-align: center;">FUID</td><td style="text-align: left;">用于绕过系统对 CUID 卡的拦截,0 扇区写入一次后变成 M1 卡</td></tr><tr class="odd"><td style="text-align: center;">UFUID</td><td style="text-align: left;">UID 和 FUID 的合成卡,需要封卡操作</td></tr></tbody></table><h3 id="安全问题">安全问题</h3><p>从 1994 年就面世的 MIFARE Classic 已经算的上是一个古老的产品,在 2008 年就已经被宣告能用普通 PC 在几分钟之内通过与读卡器通信破解出密钥,2009 年更是公布了仅需卡片的离线破解。NXP 官方也已经建议使用更安全的 <a href="https://www.mifare.net/products/chip-card-ics/mifare-plus/" target="_blank" rel="noopener">MIFARE Plus</a> 和 <a href="https://www.mifare.net/products/chip-card-ics/mifare-desfire/" target="_blank" rel="noopener">MIFARE DESFire</a>。有兴趣的朋友可以看看 Black Hat 上的 slides:<a href="https://www.blackhat.com/docs/sp-14/materials/arsenal/sp-14-Almeida-Hacking-MIFARE-Classic-Cards-Slides.pdf" target="_blank" rel="noopener">Hacking Mifare Classic Cards</a>。</p><h3 id="基本结构">基本结构</h3><p>MIFARE Classic 一般使用的是 MIFARE Classic 1K(M1卡),1K = 16 sectors ✖️ 4 blocks ✖️ 16B = 1024字节</p><p>16 个 sector (扇区)每个 sector 再划分为 4 个 block(块),编号为block 0 - 63,sector 0的block 0用于存放厂商代码,已经固化无法更改,包括卡片 ID(官方名称叫 UID,注意不要和后文的 UID 卡弄混)和类型信息。其余sector的第block 0 - 2 用于存放数据,block 3 为控制块用于存放密钥A、控制访问权限的 Access Bits、密钥B,结构如下:</p><table><thead><tr class="header"><th style="text-align: center;">A0 A1 A2 A3 A4 A5</th><th style="text-align: center;">FF 07 80 69</th><th style="text-align: center;">B0 B1 B2 B3 B4 B5</th></tr></thead><tbody><tr class="odd"><td style="text-align: center;">(Key A 6字节)</td><td style="text-align: center;">(存储控制 4字节)</td><td style="text-align: center;">(Key B 6字节)</td></tr></tbody></table><p>并且每个扇区块3中的密码和存储控制全部都是独立的,独立控制本扇区的各种操作,每个扇区都能实现不同的功能,所以广泛用作一卡通。</p><p>卡中每个块包含16字节,所以M1卡的容量=16扇区✖️4块✖️16字节=1024字节=1k (M1卡的由来)。</p><p>Wiki 上有直观的布局图:<a href="https://en.wikipedia.org/wiki/File:MiFare_Byte_Layout.png" target="_blank" rel="noopener">MiFare Byte Layout.png</a>。</p><h3 id="破解原理">破解原理</h3><h3 id="nfc原理">NFC原理</h3><p>NFC是现在主流的非接触式IC卡的读卡技术,其全称是近场通信(near field communication),是基于无线射频识别RFID(Radio Frequency Identification)技术发展起来的一种近距离无线通信技术,与RFID一样,NFC也是通过频谱中无线频率部分的电磁感应耦合方式传递,但两者之间还是存在很大的区别。NFC的传输范围比RFID小,RFID的传输范围可以达到0~1m,但由于NFC采取了独特的信号衰减技术,相对于RFID来说NFC具有成本低、带宽高、能耗低等特点。</p><p>近场通信技术(Near-field communication,NFC)由非接触式射频识别(<a href="https://zh.wikipedia.org/wiki/RFID" target="_blank" rel="noopener">RFID</a>)演变而来,由<a href="https://zh.wikipedia.org/wiki/飞利浦" target="_blank" rel="noopener">飞利浦</a>半导体(现<a href="https://zh.wikipedia.org/wiki/恩智浦半导体" target="_blank" rel="noopener">恩智浦半导体</a>)、<a href="https://zh.wikipedia.org/wiki/諾基亞" target="_blank" rel="noopener">诺基亚</a>和<a href="https://zh.wikipedia.org/wiki/索尼" target="_blank" rel="noopener">索尼</a>共同于2004年研制开发,其基础是RFID及互连技术。近场通信是一种短距高频的无线电技术,在13.56MHz频率运行于20厘米距离内。其传输速度有106 Kbit/秒、212 Kbit/秒或者424 Kbit/秒三种。当前近场通信已通过成为ISO/IEC IS 18092国际标准、EMCA-340标准与ETSI TS 102 190标准。NFC采用主动和被动两种读取模式。</p><p><a href="https://zh.wikipedia.org/wiki/近場通訊" target="_blank" rel="noopener">近场通信是一种短距高频的无线电技术,在13.56MHz频率运行于20厘米距离内。其传输速度有106 Kbit/秒、212 Kbit/秒或者424 Kbit/秒三种</a>。</p><p>这段话也说明了它的技术指标,<strong>13.56MHz</strong>,传输距离,传输速度。</p><h2 id="准备操作">准备操作</h2><h3 id="确认门禁卡类型">确认门禁卡类型</h3><h3 id="硬件准备">硬件准备</h3><h4 id="nfc-模块和-usb-转-ttl-模块">NFC 模块和 USB 转 TTL 模块</h4><ul><li>一档(30RMB):[PN532 NFC RFID 模块 V3] 和 「CH340 USB 转 TTL 模块」</li><li>二档(100RMB): [ACR122U]</li><li>三档(300RMB):<a href="https://github.com/Proxmark/proxmark3/" target="_blank" rel="noopener">Proxmark3</a>,支持高低频卡,可通过配套软件实现破解、嗅探、模拟、复制等功能,被誉为 RFID 界的瑞士军刀</li></ul><h4 id="uid-卡">UID 卡</h4><p>不同的门禁验证的内容也不一样。</p><ul><li>只验证卡片 ID :UID 卡</li><li>只验证卡片内容:使用 NFC 模块配套赠送的白卡就行了</li><li>同时验证卡片 ID 和内容:UID 卡</li><li>其它:可能需要用 CUID/FUID 卡才行</li></ul><h4 id="购买备注">购买备注</h4><p>购买 USB 转 TTL 模块时注意避免 PL 2303 HX 型号,该型号的早期版本 Rev A 由于山寨问题已经被官方在新版本驱动中<a href="http://www.prolific.com.tw/US/ShowProduct.aspx?p_id=155&pcid=41" target="_blank" rel="noopener">封杀</a>,无法在 Win10 系统中自动安装驱动使用。如果你不幸购买到了此(山寨)版本,请使用1.5.0 (2011-10-21) 版本的<a href="https://drive.czbix.com/NFC/PL2303_Prolific_DriverInstaller_v1.5.0.zip" target="_blank" rel="noopener">驱动包</a>或者询问卖家索取。</p><h3 id="软件准备">软件准备</h3><ul><li><p>Windows</p><p>Windows10 <a href="https://docs.microsoft.com/windows/wsl/install-win10" target="_blank" rel="noopener">启用 WSL</a> 后从商店安装 <a href="https://www.microsoft.com/p/kali-linux/9pkr34tncv07" target="_blank" rel="noopener">Kali Linux</a>,用 <code>apt</code> 命令从源里安装 <code>libnfc-bin</code> 和 <code>libnfc-examples</code>。如果碰到加密卡,还需要安装 <code>mfoc</code> 和 <code>mfcuk</code>。</p></li><li><p>Linux</p><p>基本同上,不过源里可能没有 <code>mfoc</code> 和 <code>mfcuk</code>,需要自行下载源码编译。具体编译过程请参考官方 GitHub( <a href="https://github.com/nfc-tools/mfoc" target="_blank" rel="noopener">mfoc</a>,<a href="https://github.com/nfc-tools/mfcuk" target="_blank" rel="noopener">mfcuk</a>)。Arch 用户可直接从 AUR 安装 <code>libnfc-git mfoc-git mfcuk-git</code></p></li><li><p>Mac</p><p>用 brew 安装 <code>libnfc mfoc mfcuk</code></p></li></ul><h2 id="操作步骤">操作步骤</h2><p><em>以 PN532 + Windows 10 + WSL + Kali Linux为例</em></p><h3 id="连接-nfc-模块">连接 NFC 模块</h3><table><thead><tr class="header"><th style="text-align: center;">PN532</th><th style="text-align: center;">USB 转 TTL 模块</th></tr></thead><tbody><tr class="odd"><td style="text-align: center;">GND</td><td style="text-align: center;">GND</td></tr><tr class="even"><td style="text-align: center;">VCC</td><td style="text-align: center;">5.0V</td></tr><tr class="odd"><td style="text-align: center;">TXD/SDA</td><td style="text-align: center;">RXD(接收端)</td></tr><tr class="even"><td style="text-align: center;">RXD/SCL</td><td style="text-align: center;">TXD(传送端)</td></tr></tbody></table><p>打开“设备管理器”,将 USB 插上电脑,等待 Windows 自动安装驱动。(PL 2303 HXA 注意先安装老版本驱动再插入 USB 模块)。驱动安装完成后,在“端口”里找到 USB 转 TTL 模块的端口号,例如 <code>COM3</code>。</p><p>修改 <code>/etc/nfc/libnfc.conf</code>,在文件末尾加入:</p><figure class="highlight apacheconf"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br></pre></td><td class="code"><pre><span class="line"><span class="attribute">allow_autoscan</span> = false</span><br><span class="line"><span class="attribute">log_level</span>=2</span><br><span class="line"><span class="attribute">allow_intrusive_scan</span> = true</span><br><span class="line"><span class="attribute">device</span>.name = <span class="string">"PN532"</span></span><br><span class="line"><span class="attribute">device</span>.connstring = <span class="string">"pn532_uart:/dev/ttyS3"</span> #Linux 下为 /dev/ttyUSB0</span><br></pre></td></tr></table></figure><p>执行 <code>nfc-scan-device -v</code>(需要 <code>sudo</code>,后文略) 确认 NFC 模块能正常识别。正常输出如下:</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br></pre></td><td class="code"><pre><span class="line">nfc-scan-device uses libnfc 1.7.1</span><br><span class="line">1 NFC device(s) found:</span><br><span class="line">- pn532_uart:/dev/ttyS3:</span><br><span class="line"> pn532_uart:/dev/ttyS3</span><br><span class="line">chip: PN532 v1.6</span><br><span class="line">initator mode modulations: ISO/IEC 14443A (106 kbps), FeliCa (424 kbps, 212 kbps), ISO/IEC 14443-4B (106 kbps), Innovision Jewel (106 kbps), D.E.P. (424 kbps, 212 kbps, 106 kbps)</span><br><span class="line">target mode modulations: ISO/IEC 14443A (106 kbps), FeliCa (424 kbps, 212 kbps), D.E.P. (424 kbps, 212 kbps, 106 kbps)</span><br></pre></td></tr></table></figure><h3 id="dump-卡片">Dump 卡片</h3><h4 id="卡片基本信息">卡片基本信息</h4><p>执行 <code>nfc-poll</code>,将门禁卡在 NFC 模块上贴一下,可以看到卡片信息输出如下:</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br></pre></td><td class="code"><pre><span class="line">nfc-poll uses libnfc 1.7.1</span><br><span class="line">NFC reader: pn532_uart:/dev/ttyS3 opened</span><br><span class="line">NFC device will poll during 30000 ms (20 pollings of 300 ms for 5 modulations)</span><br><span class="line">ISO/IEC 14443A (106 kbps) target:</span><br><span class="line"> ATQA (SENS_RES): 00 04</span><br><span class="line"> UID (NFCID1): 01 23 45 67</span><br><span class="line"> SAK (SEL_RES): 08</span><br><span class="line">nfc_initiator_target_is_present: Target Released</span><br><span class="line">Waiting for card removing...done.</span><br></pre></td></tr></table></figure><p>从上可以看到卡片 ID 为 <code>01 23 45 67</code>,ATQA 为 <code>00 04</code>,SAK 为 <code>08</code>。如果 ATQA 和 SAK 不是这两个值,则门卡可能是其他类型,很有可能无法复制。</p><h4 id="无密码卡">无密码卡</h4><p>现在将卡放在 NFC 模块上,并执行 <code>nfc-mfclassic r a dump.mfd</code> 尝试无密码复制卡片数据,成功的输出如下:</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br></pre></td><td class="code"><pre><span class="line">NFC reader: pn532_uart:/dev/ttyS3 opened</span><br><span class="line">Found MIFARE Classic card:</span><br><span class="line">ISO/IEC 14443A (106 kbps) target:</span><br><span class="line"> ATQA (SENS_RES): 00 04</span><br><span class="line"> UID (NFCID1): 01 23 45 67</span><br><span class="line"> SAK (SEL_RES): 08</span><br><span class="line">Guessing size: seems to be a 1024-byte card</span><br><span class="line">Reading out 64 blocks |................................................................|</span><br><span class="line">Done, 64 of 64 blocks read.</span><br><span class="line">Writing data to file: dump.mfd ...Done.</span><br></pre></td></tr></table></figure><p>如果能成功复制出所有数据,这就意味着卡片是没有任何密码保护的。 我们再用 <code>xxd dump.mfd</code> 确认下卡片内容。 如果除了前4行,其他的内容都和下面类似,则门禁可能是只校验卡片 ID 的低级门禁。</p><figure class="highlight plain"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br></pre></td><td class="code"><pre><span class="line">00000040: 0000 0000 0000 0000 0000 0000 0000 0000 ................</span><br><span class="line">00000050: 0000 0000 0000 0000 0000 0000 0000 0000 ................</span><br><span class="line">00000060: 0000 0000 0000 0000 0000 0000 0000 0000 ................</span><br><span class="line">00000070: ffff ffff ffff ff07 8069 0000 0000 0000 .........i......</span><br></pre></td></tr></table></figure><h4 id="部分加密卡">部分加密卡</h4><p>如果卡片有加密没法读出,我们就需要用到 mfoc 工具来尝试破解密钥了。它利用了在 2009 年研究出来的 offline nested attack,可以在拥有任意 sector 密码的情况下,猜解出其他 sector 的密码。如果卡片不是完全加密,仅需要几分钟就可以猜解出密钥。 我们执行 <code>mfoc -O dump.mfd</code> 开始破解,如果成功的话会提示以下内容,并将卡片内容和密钥一同显示出来。</p><blockquote><p>Auth with all sectors succeeded, dumping keys to a file!</p></blockquote><h4 id="完全加密卡">完全加密卡</h4><p>如果门禁卡是全加密卡,上面使用 mfoc 就会有失败提示:</p><blockquote><p>No sector encrypted with the default key has been found, exiting..</p></blockquote><p>这时我们就需要用到 mfcuk 了。它利用了同样是 2009 年研究出的 Dark Side attack,利用认证协议上的漏洞来猜解密码,可能需要一个小时才能破解出密钥。当拿到其中一个密钥之后,就可以可上面部分加密卡一样,使用 mfoc 工具猜解其它的密钥。 先执行 <code>mfcuk -C -R 0:A -s 250 -S 250</code>,然后就是漫长的等待。当输出如下内容时,表示破解成功:</p><blockquote><p>INFO: block 3 recovered KEY: 123456789ABC</p></blockquote><p>将拿到的 Key 传给 mfoc,执行 <code>mfoc -k 123456789ABC -O dump.mfd</code>,就可破解出完整的卡片数据。</p><h3 id="写入卡片">写入卡片</h3><p>将 <strong>UID</strong> 卡放在 NFC 模块上,执行 <code>nfc-mfclassic W a dump.mfd</code> 即可写入卡片 ID 和卡片数据。参数中的 <code>W</code> 表示同时写入 ID 号(需要 UID 系列卡片),也可以用 <code>w</code> 仅写入卡片内容。如果手头没有 UID 卡,可以仅写入卡片内容试试看是不是仅校验卡内容的门禁。</p><h3 id="懒人模式图形界面">懒人模式(图形界面)</h3><p>步骤类似,读卡,破解,写卡</p><ul><li><p>MifareOneTool</p></li><li><p>NFC上位机(NFC_READER_crack.exe+PN532破解全加密GUI.exe)</p></li></ul><h2 id="小米手环nfc-模拟加密门禁卡">小米手环NFC 模拟加密门禁卡</h2><ol type="1"><li>使用设备(如 PN532)读加密门卡内容到 PC,dump</li><li>将block 0写到一张空白UID卡上,给手环复制(手环只模拟了 block 0 的卡号部分,而厂商代码是固定值且不可修改,所以碰到检查卡号和厂商的门禁就没用了)</li><li>再写入加密卡片内容的其它部分到手环(剩余63 blocks)</li></ol><p><a href="https://docs.ioin.in/writeup/www.91ri.org/_16081_html/index.html" target="_blank" rel="noopener">进监狱全攻略:Mifare1 Card破解</a></p><p><a href="https://hceng.cn/2019/07/12/NFC手机模拟加密门禁卡/" target="_blank" rel="noopener">NFC手机模拟加密门禁卡</a></p>]]></content>
<summary type="html">
<h1 id="小米手环复制mifare-门禁卡"><a href="https://blog.czbix.com/clone-MIFARE-Classic-card.html" target="_blank" rel="noopener">小米手环复制Mifare 门禁卡</a
</summary>
</entry>
<entry>
<title>反脆弱读书笔记</title>
<link href="https://github-9233.github.io/2020/02/22/%E5%8F%8D%E8%84%86%E5%BC%B1%E8%AF%BB%E4%B9%A6%E7%AC%94%E8%AE%B0/"/>
<id>https://github-9233.github.io/2020/02/22/反脆弱读书笔记/</id>
<published>2020-02-22T07:01:12.000Z</published>
<updated>2020-02-22T07:02:10.181Z</updated>
<content type="html"><![CDATA[<h1 id="读书笔记反脆弱从不确定性中获益"><a href="https://program-think.blogspot.com/2018/12/Book-Review-Antifragile-Things-That-Gain-from-Disorder.html" target="_blank" rel="noopener">读书笔记:《反脆弱——从不确定性中获益》</a></h1><h2 id="反脆弱一书的简介">★《反脆弱》一书的简介</h2><figure><img src="/2020/02/22/反脆弱读书笔记/cover.png" alt=""><figcaption>cover</figcaption></figure><p> 原作名《Antifragile——Things That Gain from Disorder》,作者是<a href="https://zh.wikipedia.org/zh-cn/納西姆·尼可拉斯·塔雷伯" target="_blank" rel="noopener">纳西姆·尼可拉斯·塔雷伯</a>(阿拉伯语:نسيم نيقولا نجيب طالب,英语:Nassim Nicholas Taleb,1960年1月1日-),<a href="https://zh.wikipedia.org/w/index.php?title=黎巴嫩裔美國人&action=edit&redlink=1" target="_blank" rel="noopener">黎巴嫩裔美国人</a>。知名思想家,<a href="https://zh.wikipedia.org/wiki/經驗主義" target="_blank" rel="noopener">经验主义</a>、怀疑经验论者,金融业人士,风险工程学教授、哲学随笔作家,以《<a href="https://zh.wikipedia.org/wiki/黑天鵝效應" target="_blank" rel="noopener">黑天鹅效应</a>》一书闻名于世,诺贝尔经济学奖得主<a href="https://zh.wikipedia.org/wiki/丹尼尔·卡内曼" target="_blank" rel="noopener">丹尼尔·卡内曼</a>称其“改变了世界对于‘<a href="https://zh.wikipedia.org/wiki/不確定性" target="_blank" rel="noopener">不确定性</a>’的想法”。塔雷伯的主要研究兴趣在于“<a href="https://zh.wikipedia.org/wiki/随机性" target="_blank" rel="noopener">随机性</a>”、“<a href="https://zh.wikipedia.org/wiki/不確定性" target="_blank" rel="noopener">不确定性</a>”、“稀有事件”(黑天鹅事件)等问题。著有《随机骗局》、《<a href="https://zh.wikipedia.org/wiki/黑天鵝效應" target="_blank" rel="noopener">黑天鹅效应</a>》、《反脆弱》等书,合称“不确定性三部曲”。文字风格慧黠机智、妙趣横生,穿插丰富的哲学、历史、数学、科学与金融知识,且带有浓厚的自传色彩。成名作《随机骗局》被《<a href="https://zh.wikipedia.org/wiki/財星" target="_blank" rel="noopener">财星</a>》杂志选为“有史以来最聪明的书”(Smartest Books of All Time),而《<a href="https://zh.wikipedia.org/wiki/黑天鵝效應" target="_blank" rel="noopener">黑天鹅效应</a>》更是长踞<a href="https://zh.wikipedia.org/wiki/紐約時報暢銷書榜" target="_blank" rel="noopener">纽约时报畅销书榜</a>31周,且被《<a href="https://zh.wikipedia.org/wiki/星期日泰晤士報" target="_blank" rel="noopener">星期日泰晤士报</a>》选为“二战后12本最具影响力的书”。</p><h2 id="第零卷-前言">★第零卷 前言</h2><h3 id="三元结构或根据三个特性绘制世界万物的分布图">三元结构,或根据三个特性绘制世界万物的分布图</h3><blockquote><p>现在,经过一定的努力后,我们的目标是用一根线索将读者脑中看似割裂的元素联系起来,如老加图、尼采、米利都的泰勒斯、城邦制度的效力、手工业者的可持续性、发现的过程、不透明的片面性、金融衍生品、抗生素的耐药性、自下而上的系统、苏格拉底过度理性化的倾向、教鸟儿如何飞行、新事物狂热症、达尔文的进化论、詹森不等式的数学概念、可选择性和选择权理论、古人的启发法、约瑟夫·迈斯特和埃德蒙·伯克的作品、维特根斯坦的反理性主义、经济机构的欺骗性理论、自由探索和反复试错、恐怖组织成员的死亡进一步加剧了恐怖主义阴霾、手工业者行会的辩解、中产阶级的道德缺陷、古老的方法(和营养学)、医源性损伤的概念、宏伟事物的光荣概念、我对凸性理念的痴迷(以及我对凹性的恐惧)、2008年的银行和经济危机、对冗余的误解、游客和漫游者之间的区别,等等。所有这些要用单独的,我确信也是简单的一根线索串联起来。</p><p>三元结构根据特征将事物分为三类: <strong>脆弱类</strong> <strong>强韧类</strong> <strong>反脆弱类</strong></p></blockquote><figure><img src="/2020/02/22/反脆弱读书笔记/图0-1.jpg" alt=""><figcaption>图0-1</figcaption></figure><h2 id="第一卷-反脆弱性介绍">★第一卷 反脆弱性介绍</h2><figure><img src="/2020/02/22/反脆弱读书笔记/图1-0.jpg" alt=""><figcaption>图1-0</figcaption></figure><h3 id="第1章-达摩克利斯之剑和九头蛇怪">◇第1章 达摩克利斯之剑和九头蛇怪</h3><h4 id="生活中的一半事物未被命名">1.1 生活中的一半事物未被命名</h4><figure><img src="/2020/02/22/反脆弱读书笔记/图1-1.jpg" alt=""><figcaption>图1-1</figcaption></figure><h4 id="请砍我的头">1.2 请砍我的头</h4><ol type="1"><li>达摩克利斯——脆弱</li><li>凤凰涅槃——超越强韧</li><li>九头蛇怪——反脆弱</li></ol><h5 id="命名的必要性">命名的必要性</h5><blockquote><p>我们知道的事情其实比我们认为的更多,比我们能说清楚的要更多。比如原始族群虽然并非色盲,却只能用语言表达两三种颜色。</p></blockquote><h4 id="反脆弱性的原型">1.3 反脆弱性的原型</h4><p>两种初级的反脆弱性(医疗领域)</p><blockquote><p>1.【米特拉达梯式解毒法】:当人们不断地接受小剂量的某种物质,随着时间的推移,对额外的或更大剂量的同类物质逐步产生免疫力的结果。这是在疫苗接种和药物过敏测试中使用的方法。这还【不】能称作【完全的反脆弱性】,仍处于【较温和的强韧性阶段】,但已经逐步接近反脆弱性了。我们已经得到了一个暗示,一旦被剥夺了摄取毒物的权力,我们将变得脆弱,强韧化的道路常始于一点点的伤害。 2.【毒物兴奋效应】:摄取一定剂量的有毒物质使你变得更为强壮,也就是在强韧性的基础上更进一步。小剂量的有害物质实际上会有利于机体健康,起到药物的治疗作用。一点点的危害,只要不是很多,只会造福于机体,使其成为更好的整体,因为它会触发一些【过度反应】。</p></blockquote><h4 id="领域独立就是领域依赖">1.4 领域独立就是领域依赖</h4><blockquote><p>【领域依赖性】所谓“领域”就是某一类别的活动。有些人可以在某一个领域,比如医疗领域,很好的理解某一想法,但在另一个领域,比如社会、经济、生活领域,却未必能理解同一个想法。</p></blockquote><h3 id="第2章-随处可见的过度补偿和过度反应">◇第2章 随处可见的“过度补偿”和“过度反应”</h3><p>【创伤后成长(相对于创伤后压力综合征、创伤后应激障碍)】:人们在受到过去事件的伤害后重新振作并超越自己。</p><blockquote><p>1.【古代创新】对挫折的过度反应所释放出来的多余能量成就了创新。古代创新和成熟源自最初的创新必要性,但却超越了满足这种必要性的状态(即最初发明或发明尝试带来的意外副作用) 2.【现代创新】倾向于认为来自于体制的资金支持,并通过规划来实现</p></blockquote><h4 id="如何在跑马比赛中取胜">2.1 如何在跑马比赛中取胜</h4><p>压力源、反毒物兴奋效应与挑战的缺乏都会导致应激反应不足,从而降低最佳表现的水准。</p><blockquote><p>【过度反应机制或类似的机制】会使我们在有一点点背景噪声的地方更好地集中精力,就好像对抗这些噪声的行为可以帮助我们集中注意力。</p></blockquote><h5 id="以冗余的形式呈现的反脆弱性反应">以冗余的形式呈现的反脆弱性反应</h5><h4 id="论暴乱爱和其他意料之外压力受益者的反脆弱性">2.2 论“暴乱、爱和其他意料之外压力受益者”的反脆弱性</h4><blockquote><p>【反脆弱性】唤醒了肌体的应激反应,对压力和伤害做出了过度反应。如果这样界定反脆弱性,那么在经济生活之外你能看到的最具反脆弱性的事情莫过于难以泯灭的爱或恨,即对于距离、家庭矛盾以及压抑爱憎情绪的企图等做出的【过度反应】。</p></blockquote><h5 id="请将我的书列为禁书信息的反脆弱性">请将我的书列为禁书:信息的反脆弱性</h5><blockquote><p>书籍和想法都具有反脆弱性,它们都因遭查禁而广为传播。借用罗马帝国皇帝<a href="https://zh.wikipedia.org/wiki/馬爾庫斯·奧列里烏斯" target="_blank" rel="noopener">马可·奥勒留</a>(同时也是一位<a href="https://zh.wikipedia.org/wiki/斯多葛派" target="_blank" rel="noopener">斯多葛学派</a>作家)的话:“有了障碍物,烈火才烧得更旺。”——这就是禁书的吸引力,它们在禁令面前显示出了反脆弱性。</p></blockquote><h5 id="换份工作">换份工作</h5><blockquote><p>比起控制你的声誉或公众的看法,换份工作要容易得多。</p></blockquote><h3 id="第3章-猫与洗衣机">◇第3章 猫与洗衣机</h3><blockquote><p>任何生命体都在一定程度上都具有反脆弱性(当然,反过来说并不成立,有反脆弱性的物体并不一定都有生命)。生命的奥秘似乎就在于这种【强韧性】。</p></blockquote><h4 id="压力源即信息">3.1 压力源即信息</h4><blockquote><p>人类在急性刺激下会比在慢性刺激下表现得更出色,尤其是在急性刺激后给予较长的恢复期,这将使得这些压力源成为信息的传导渠道。急性刺激压力源肯定比一个温和但连续不断的压力源有益。后者大多是让你在生活中感到压抑的东西,包括按揭贷款、税务问题、因拖欠报税而产生的内疚、考试压力、琐碎事务、电子邮件回复、填写表格、每天上下班通勤,等等。 <strong>第一种类型的压力源是【必要的】,而第二种类型的压力源对人的健康却是【有害的】。</strong></p></blockquote><h4 id="针对儿童的犯罪教育的弊端">3.2 针对儿童的犯罪——教育的弊端</h4><blockquote><p>我们不仅厌恶压力,也不理解压力,殊不知,彻底消除波动和变化只会危害生命、生活、科学和智慧</p><ol type="1"><li>美国高中年龄段以上的人口中有1/10的人在服用某种抗抑郁药,如百忧解。</li><li>还有另外一种危险:除了伤害孩子,我们还会危害社会和我们的未来。旨在减少儿童生命中的变化和波动的举措却也会降低我们这个所谓的“伟大的全球化社会”中的多元性和差异性</li></ol></blockquote><h5 id="受到翻译的惩罚">受到翻译的惩罚</h5><blockquote><p>压力源的另一个被遗忘的特征隐藏在语言习得中。我们学习新单词并不需要死记硬背,而是利用另一种方式——与人沟通,不得不揣测他人心思,而暂时把“害怕犯错误的心理”放在一边。学习一门语言的最好办法可能就是在国外被“囚禁”一段时间。</p></blockquote><h5 id="观光化">观光化</h5><p>【观光化】好像人只能按照详细的用户手册作出机械的反应。电子日历将我们休闲时间约束起来。这一【目标驱动】型态度深深地伤害了内在的自我。</p><blockquote><p>我们用系统性的方法清除食物的不确定性和随机性,以便在最细节的层面确保高度的可预测性,其目的就是为了舒适性、便利性和效率。</p></blockquote><h5 id="对机遇的秘密渴望">对机遇的秘密渴望</h5><blockquote><p>如果你还活着,你的内心深处就会喜欢一定程度的随机性和混乱。<strong>如果我能预测我未来每一天的轨迹,那我会觉得自己身体的一部分已经死了。</strong></p></blockquote><h3 id="第4章-杀死我的东西却让其他人更强壮">◇第4章 杀死我的东西却让其他人更强壮</h3><h4 id="反脆弱性的层级">4.1 反脆弱性的层级</h4><blockquote><p>由于存在多个层级和层次,反脆弱性就变得更复杂,也更有趣了。天然有机体都不会是单独存在的终极单元;它总是由子单元组成的,而它本身也可能是更大集合体的子单元。这些子单元可能会相互竞争。以一个行业为例,餐馆往往是脆弱的,它们会相互竞争。但正因为如此,当地的餐馆群才具备了反脆弱性。</p></blockquote><h5 id="进化和不可预测性">进化和不可预测性</h5><ol type="1"><li>米特拉达梯式解毒法和毒物兴奋效应都是较弱的反脆弱性的表现形式,食物从波动、意外或伤害中获得的益处有限,如果使用剂量超过限额,其保护或有益作用就将逆转。</li><li>反脆弱性有更强的形式——【进化】。最有趣的一面是,它是依赖【反脆弱性】实现的;它喜欢压力、随机性、不确定性和混乱——而个体生物则相对脆弱。基因库正是利用冲击来确保优胜劣汰,提高整体的适应力。</li></ol><p>大自然和个体生物之间存在着一定的紧张关系:一切生物或者本质上类似于有机体的事物的生命都是有限的,最终都会死亡。大自然更喜欢在【信息层面】,也就是通过遗传密码让游戏继续下去。因此,<strong>有机体需要死亡,以确保大自然的反脆弱性</strong>。</p><h5 id="有机体即群体群体即有机体">有机体即群体,群体即有机体</h5><p>【分形自相似】:细胞是由众多细胞间的分子构成的,有机体内则由众多细胞构成,物种又是由众多生物体构成的。</p><h4 id="错误谢谢你">4.2 错误,谢谢你</h4><p>简单地叙述脆弱性、错误和反脆弱性之间的关系:</p><blockquote><p>当你脆弱的时候,你往往倾向于墨守成规,尽量减少变化——因为变化往往弊大于利。这就是为什么脆弱的事物需要明确的预测方法,反过来说,预测体系带来的只能是脆弱性。如果你想做出改变,并且不关心未来结果的多种可能性,认为大多数结果都会对你有利,那么你就具有反脆弱性。 如果能够理性地进行试错,将错误当作一种信息源,那么,试错过程中出现的随机要素其实并没有那么随机。如果每次试错都能让你了解到什么是行不通的,渐渐地,你就接近有效的解决方案了——这样,每次努力都变得更有价值,更像是一笔投资而非一个错误。显然,在此过程中你将不断地有所发现。</p></blockquote><h4 id="为何整体厌恶个体">4.3 为何整体厌恶个体</h4><blockquote><p><strong>较高层级事物的反脆弱性有赖于较低层级事物的脆弱性,或者较低层级事物的牺牲。</strong></p></blockquote><h4 id="杀不死我的会杀死其他人">4.4 杀不死我的,会杀死其他人</h4><blockquote><p>系统的反脆弱性是以牺牲个人为代价取得的(这是毒物兴奋效应和自然选择之间的差异</p><p>尼采有句名言:“<strong>杀不死我的,会让我更强</strong>”,这很容易被误解为米特拉达梯式解毒法或读物兴奋效应。......但它也可能意味着“<strong>杀不死我的,并未使我变得更坚强,但它让我幸存下来,因为我比别人更强壮;由于它杀死了弱者,我们种群的平均素质变强了</strong>”。</p></blockquote><h2 id="第二卷-现代化与对反脆弱性的否定">★第二卷 现代化与对反脆弱性的否定</h2><figure><img src="/2020/02/22/反脆弱读书笔记/图5-0.jpg" alt=""><figcaption>图5-0</figcaption></figure><h3 id="第5章-露天市场与办公楼">◇第5章 露天市场与办公楼</h3><h4 id="两种职业">5.1 两种职业</h4><blockquote><p><strong>生活中的核心【错觉】,即认为随机性是有风险的,是一桩坏事,消除随机性,就可以消除风险。</strong></p><ol type="1"><li>人为消除随机性所带来的结果相当于约翰的收入——清晰、稳定,但很脆弱。这样的收入经受不住大的冲击,随时可能变成零。</li><li>自然随机性带来的结果则更像是乔治的收入——经得起非常大的冲击,但也必须承受每天的波动性。</li><li>约翰有一个大雇主,乔治则有许多小雇主。后者可以选择对他更有利的最佳雇主,在任何时候都有“更多选择”【可选择性】</li></ol></blockquote><h5 id="列宁在苏黎世">列宁在苏黎世</h5><blockquote><p>事实上,<strong>瑞士是没有强大的中央政府</strong>,或者我们通常意义上的“政府”。这个国家完全是【自下而上】地由自治市或区域行政区(称为“州”)治理着,就好像是通过“准主权的迷你国家”所结成的联邦。<strong>这种自下而上的专政抵御了【乌托邦式浪漫主义】的侵入</strong>。</p></blockquote><h4 id="自下而上的变动">5.2 自下而上的变动</h4><blockquote><p>一群自治市之间有些领地的纷争,本身也不乏内讧。但就是这种人与人之间的争斗,凝聚形成了一个相当良性和稳定的国家。</p></blockquote><h4 id="远离极端斯坦">5.3 远离极端斯坦</h4><blockquote><p><img src="/2020/02/22/反脆弱读书笔记/图5-1.png" alt="图5-1"> <strong>图5-1</strong> 注释:该图对比了自治市的噪声或露天市场分散的变动(图A)与集中管理或人为管理的系统(图B)的情况,或者说,对比了一个出租车司机的收入(图A)与一个银行职员的收入(图B)的情况。图B显示了急转直下式的变动,或者说从“黑天鹅”到“黑天鹅”。人类对比了平稳而控制流程或过度干扰,导致系统从平均斯坦转变为极端斯坦。这种效果在所有受到约束的系统中都很常见,比如健康系统、政治系统、经济系统,甚至也会因有没有百忧解而不同的人类情绪。创业者驱动的硅谷(图A)和银行系统(图B)也存在这种差别。</p></blockquote><h5 id="重大的火鸡问题">重大的“火鸡”问题</h5><blockquote><p>从火鸡的故事中,我们可以看到那些有害性错误的根源:<strong>将“证明有危害的证据较匮乏”视为“有证据证明不存在危害”</strong>。</p></blockquote><h4 id="年">5.4 12000年</h4><p>黎凡特北部(大约今天叙利亚和黎巴嫩的北部)从陶器出现之前的新石器时代到现代的20世纪中叶,繁荣了<strong>12000</strong>年。【城邦制】但是在一战之后,黎凡特北部地区的一部分划入叙利亚,其他部分则是黎巴嫩的一部分。叙利亚执政党“<a href="https://zh.wikipedia.org/wiki/阿拉伯復興社會黨" target="_blank" rel="noopener">阿拉伯复兴社会党</a>”推行的乌托邦政策(中央集权统治)终结了该地区的繁荣。</p><h5 id="战争还是监狱或者两者都选">战争还是监狱,或者两者都选</h5><p>黎巴嫩管理过于松散,导致激烈内战。人们宁愿选择战争也不选监狱,这里【监狱】指的是政治和经济自由的丧失,经济生活似乎也更喜欢战争而非监狱</p><h5 id="罗马式和平">罗马式和平</h5><ol type="1"><li>古埃及王朝,集权制国家依赖于中央集权官僚制度</li><li>帝国,如罗马帝国和奥斯曼王朝,则更依赖于当地的精英,事实上帝国允许城邦的繁荣发展并保留一定的有效自治权——对和平更有利的是,这些自治是商业自治,而不是军事自治</li><li>中国,实践中与上面帝国治理方式差别不大,不同的是中国通过供养文士集中了智慧,通过文官制度垄断了知识。</li></ol><h3 id="第6章-告诉他们我爱随机性">◇第6章 告诉他们我爱随机性</h3><h4 id="饥饿的驴">6.1 饥饿的驴</h4><blockquote><p>到目前为止,我们的论点是,阻止一个反脆弱性系统出现随机性,并不总是一个好主意。现在,让我们看看一些以【添加随机性】为标准操作方法的情况。<strong>因为随机性作为一种必需燃料,是反脆弱性系统必不可少的。</strong></p></blockquote><p>“布里丹之驴的典故”、“随机共振机制”和“冶金业的退火工艺”。这些都是将随机性噪声注入系统以改善其功能的想法,通过引入某种随机性来提升整个系统的健壮性。</p><h5 id="政治退火">政治退火</h5><blockquote><p>雅典议会的成员是通过抽签的方式决定的,这一方法是为了保护整个系统免于退化。</p></blockquote><h4 id="一枚名叫稳定性的定时炸弹">6.2 一枚名叫“稳定性”的定时炸弹</h4><p>在缺乏危机的情况洗下,隐藏的风险会逐步积累,因为人们倾向于将越来越多的风险隐藏在统计分布的“尾部”,事实上,这提高了遭受罕见恶性事件打击的风险</p><h5 id="小规模战争能拯救生命吗">(小规模)战争能拯救生命吗?</h5><blockquote><p>冲突强化了国家的力量,某种强迫的、受限的、非自然的和平在生活中也可能导致人们付出昂贵的代价</p></blockquote><h5 id="要告诉外交政策制定者的话">要告诉外交政策制定者的话</h5><blockquote><p>人为地压制波动性不仅会导致系统变得极其脆弱,同时,系统也不会呈现出明显的风险。我们说过,<strong>波动性就是信息</strong>。事实上,这些系统往往过于风平浪静,而其表面之下却暗流涌动。虽说政治领导人和经济决策者的公开意图是<strong>通过抑制波动来稳定系统,但其结果往往适得其反</strong>。这些人为制约的系统更容易招致“黑天鹅”事件的光临。这样的环境最终会遭遇重创。在每个人都因适应了多年的稳定期而放松警惕后,灾难往往会不期而至,带来的损失将远超不稳定状态所带来的损失。</p></blockquote><h4 id="在这里我们把什么叫作现代化">6.3 在这里,我们把什么叫作现代化?</h4><p>【现代化】:人类大规模地治理环境,系统性平整世界的凹凸不平,以及控制波动和排除压力。</p><blockquote><p>现代化就是系统地将人类从充满随机性的生态环境中驱逐出去,包括物理的、社会的,甚至认识论的生态环境。“现代”并不只是像社会学教科书所定义的那样,指的是中世纪、土地改革,以及封建社会结束后的那个历史时期。它更是指以理性化(天真的理性主义)为标志的一个时代的精神,它依据的理念是,社会是可以理解的,也是可以设计的,由人来设计。在这种理念下诞生了统计学理论,以及可憎的“钟形曲线”。也是在这种理念的引导下,诞生了线性科学以及“效率”的概念或优化的概念。</p><p>现代化是【普罗克拉斯提斯之床】,有利有弊——它是对人性的一种削弱,目的是为了增强效率和效用。它在某些方面确实有效,普罗克拉斯提斯之床并非总是起到削足适履的作用,有些也带来了益处,但非常罕见。</p></blockquote><h3 id="第7章-天真的干预">◇第7章 天真的干预</h3><h4 id="干预和医源性损伤">7.1 干预和医源性损伤</h4><p>【医源性损伤】净亏损或超过治疗益处的伤害(通常被隐藏或延迟)</p><p>【代理问题(委托代理问题)】因一方(代理人)的个人利益与使用其服务的另一方(委托人)的利益相脱节所造成的。常见于股票经纪人、医生以及为“自己的政治生涯”奋斗的政治家</p><h5 id="首先不要造成伤害希波克拉底誓言">首先,不要造成伤害——希波克拉底誓言</h5><p>伤害来源于否认反脆弱性——我们总有这样的错觉,即认为人类总是有必要做些什么才能让万物正常运转。</p><h5 id="非天真的干预">非天真的干预</h5><blockquote><p>形成一套系统性准则,以决定何时可以干预,何时让系统自行运作</p></blockquote><h5 id="赞美拖延像费边社那样">赞美拖延——像费边社那样</h5><blockquote><p>拖延是我们的自然防御本能,是让事情顺其自然地发展、行事其反脆弱性的本能;它源于某些生态或自然的智慧,结果也并不总是坏的。</p></blockquote><h4 id="大规模的神经过敏">7.2 大规模的神经过敏</h4><blockquote><p>现代化的发展为我们提供了大量的信息,这些信息却把人类从从容镇静的那类人变成神经过敏的人。出于讨论的目的,我们暂且把从容镇静的人视为懂得对真实信息做出反应的人,而神经过敏的人则大多在对噪声做出反应。两类人之间的差异让我们看到了<strong>噪声和信号之间的差异——噪声是你应该忽略的,而信号才是你需要注意的</strong>。</p><p>因个人能力或智力的欠缺无法区别噪声和信号,往往是过度干预背后的原因。</p></blockquote><h5 id="媒体激发的神经过敏">媒体激发的神经过敏</h5><blockquote><p>如果你想加速一个人的死亡,就给他请一位私人医生。你越频繁地寻找数据,你会发现,你找到的噪声(而非被称为信号的宝贵数据)的比例就越高,信噪比就越差。在一个自然的环境中,压力源就是信息。因此,信息过多意味着压力过大,超过了反脆弱性的临界点。</p></blockquote><h5 id="媒体激发的神经过敏-1">媒体激发的神经过敏</h5><p>减少干预的最好方法是限量供应信息,并以尽可能自然的方式来获取信息。如今,人们仍抱有“科学”就意味着更多数据的错觉。</p><h4 id="国家无能的时候反而大有益">7.3 国家无能的时候反而大有益</h4><blockquote><p>但事实往往是,【国家的无能才能帮助我们摆脱中央集权和现代化的桎梏】——实现【反医源性损伤】。富有真知灼见的作者德米特里·奥尔洛夫阐述了苏联那种低效落后、无意识重复栽种的粮食生产,如何在苏联解体后帮助人民逃避灾难,最终促进了国家稳定。斯大林想按自己的主张规划农业,反而导致了饥荒。</p></blockquote><h4 id="将催化剂与起因混为一谈">7.4 将催化剂与起因混为一谈</h4><blockquote><p>当渴望自然波动性的系统受到压制,它们最终会崩溃,这是不可避免的结果,因为它们是脆弱的。我们(公众)从未将失败视为脆弱性的结果;相反的,此类失败往往被我们(公众)错误归因。如果大桥有一个摇摇欲坠的沙柱,那么把这座脆弱的桥梁的倒塌归咎于最后一辆通过的卡车是十分不明智的,更愚蠢的是试图提前预测哪辆卡车会将大桥压垮。</p></blockquote><h3 id="第8章-预测是现代化的产物">◇第8章 预测是现代化的产物</h3><blockquote><p><strong>我们预测政治和经济领域的显著罕见事件的成功记录接近于零,或者说就是零</strong>。我们要避免预测带来的医源性损伤</p></blockquote><h2 id="第三卷-非预测性的世界观">★第三卷 非预测性的世界观</h2><h3 id="第9章-胖子托尼与脆弱推手">◇第9章 胖子托尼与脆弱推手</h3><h4 id="懒惰的朋友">9.1 懒惰的朋友</h4><blockquote><p>好奇心是具有反脆弱性的,就像上瘾症一样,你越是满足它,这种感觉就越强烈——书籍有一种神秘的传播使命和能力,这一点对于整个房间满是图书的人来说并不陌生。</p></blockquote><h4 id="愚蠢的人和不愚蠢的人">9.2 愚蠢的人和不愚蠢的人</h4><blockquote><p>通常说来,你无法准确地预测未来;但你可以预测到,那些依靠预测行事的人往往会承受更大的风险,遇到一些麻烦,甚至可能失败。为什么呢?因为预测者在预测误差面前是脆弱的。</p></blockquote><h3 id="第10章-塞内加的不利因素和有利因素">◇第10章 塞内加的不利因素和有利因素</h3><blockquote><p><a href="https://zh.wikipedia.org/wiki/塞內卡" target="_blank" rel="noopener">卢修斯·安内乌斯·塞内加</a>是<a href="https://zh.wikipedia.org/wiki/斯多亞學派" target="_blank" rel="noopener">斯多葛主义</a>的一位优秀的诠释者。决策过程中所用到的智慧比书面知识重要得多——不仅仅在实践层面,也在哲学层面。</p></blockquote><h5 id="斯多葛学派的情绪强韧法">斯多葛学派的情绪强韧法</h5><blockquote><p>塞内加用以对抗这种脆弱性的实用性方法就是——<strong>通过心理练习来弱化财产在心目中的地位。因此当损失发生时,他就不会受到刺激。</strong>这是从外界环境中夺回个人自由的方式。</p></blockquote><h5 id="驯化情绪">驯化情绪</h5><blockquote><p>斯多葛主义的主旨就是【情绪的驯化】,而不一定是情绪的消除。它不是要把人类变成植物,而是将他们“对情绪的关注”转移到“对产生情绪的核心根源的关注”上,同时保持对情感的掌控力。在我看来,现代的斯多葛主义践行者就是能够——<strong>将恐惧转化为谨慎,将痛苦转化为信息,将错误转化为启示,将欲望转变为事业</strong>的人。</p></blockquote><h5 id="如何成为主人">如何成为主人</h5><blockquote><p>保留好的、剔除坏的;摈弃不利,留住有利。这实际上是一种有利与不利结果的不对称形式。其实,这就是最纯粹的反脆弱性。</p></blockquote><h5 id="基础的不对称性">★★★基础的不对称性</h5><blockquote><p><strong>脆弱性等于失去的比得到的更多,等于不利因素比有利因素更多,即等于【不利的】不对称性。</strong> <strong>反脆弱性等于得到的比失去的更多,等于有利因素比不利因素更多,即等于【有利的】不对称性。</strong> 如果潜在收益大于潜在损失,那么你对波动源就具有反脆弱性(反之亦然)。 如果潜在的有利因素多于不利因素,那么你可能会因为波动和压力不足而受到伤害。</p></blockquote><h3 id="第11章-千万别嫁给摇滚明星">◇第11章 千万别嫁给摇滚明星</h3><blockquote><p>【杠铃(或双峰)策略】是一个实现反脆弱性、向三元结构的第三类转移的方法。</p></blockquote><h4 id="破损的包裹无法恢复原状">11.1 破损的包裹无法恢复原状</h4><blockquote><p>【降低脆弱性】不是可有可无的选择,而是一种【强制性要求】。这听上去可能显而易见,但似乎总是被忽略。</p></blockquote><h4 id="塞内加的杠铃">11.2 塞内加的杠铃</h4><blockquote><p>杠铃的意思是什么?我用杠铃(一个杠轴,两端加重,供举重者使用)来将极端情况分隔开。在我们的语境中,杠铃策略不一定是对称的,只是说,它由两个极端条件组成,中间空无一物。你也可以用更技术性的语言称其为【双峰策略】,因为它有两个截然不同的模式,而不是单独的中庸模式。</p><p>反脆弱性是积极主动加上保守偏执的组合——消除不利因素,保护自己免受极端伤害,同时让有利因素或正面的“黑天鹅”顺其自然地发挥效用。我们已经看到塞内加的不对称性:要让有利因素大于不利因素,只需要减少极端不利因素的侵害(情绪伤害),而不是改善中间因素。</p></blockquote><h2 id="第四卷-可选择性技术与反脆弱性的智慧">★第四卷 可选择性、技术与反脆弱性的智慧</h2><figure><img src="/2020/02/22/反脆弱读书笔记/图12-0.jpg" alt=""><figcaption>图12-0</figcaption></figure><h3 id="第12章-泰勒斯的甜葡萄">◇第12章 泰勒斯的甜葡萄</h3><blockquote><p>这可能是人类历史上第一个有记录的期权。【期权】就是反脆弱性的一种代表。</p></blockquote><h4 id="选择权与不对称性">12.1 选择权与不对称性</h4><blockquote><p>反脆弱性等于获得的比失去的更多,等于有利因素比不利因素更多,即等于(有利的)不对称性,也等于偏好波动性。如果决策正确所得到的利益大于决策错误所受到的伤害,那么从长远来看,你就将从波动性中受益(反之亦然)。</p></blockquote><h5 id="不对称性">不对称性</h5><blockquote><p><img src="/2020/02/22/反脆弱读书笔记/图12-1.png" alt="图12-1"> <strong>图12–1</strong> 泰勒斯的反脆弱性。他付出的很少,潜在收益却很大。我们可以看出有利与不利因素之间的不对称性</p></blockquote><h5 id="喜欢分散的食物">喜欢分散的食物</h5><blockquote><p>选择权的一个属性是:它并不关心平均结果,而只关心有利因素(因为不利因素并不会超过某一界限)。</p></blockquote><h4 id="泰勒斯主义和亚里士多德主义">12.2 泰勒斯主义和亚里士多德主义</h4><blockquote><p>亚里士多德犯了一个错误,他认为了解某个事件(未来的作物收成或榨油机的租金,即横轴所显示的数据)和从中赚钱(即纵轴所显示的数据)是同样的事情。但在这个(泰勒斯)案例中,由于存在不对称性,所以这两者是不同的。</p></blockquote><h5 id="变得无知">变得无知</h5><blockquote><p>【可选择性】允许我们变得无知,或者说,允许我们获得比知识所能带给我们的更多的收益。我称之为炼金石,或者【凸性偏差】</p></blockquote><h5 id="大自然与选择权">大自然与选择权</h5><blockquote><p>法国伟大的生物学家弗朗索瓦·雅各布将【自然生态系统中的选择权概念】(或类似选择权式的特点)引入了科学领域,这就是【试错】的概念。</p><p>【试错】过程中我们会犯小错误,但却能获得大收益,我们不妨称其为“自由探索”(tinkering)。事实上,有一个词可以更准确地描述这种积极的不对称性,这就是【凸性】。</p><p><img src="/2020/02/22/反脆弱读书笔记/图12-2.png" alt="图12-2"> <strong>图12–2</strong> 这是一种与期权类似的试错机制(快速失败模型),又名“凸性自由探索”。在这一机制下,错误的成本低,最大损失是已知的,而潜在回报则是巨大的(无限)。正面“黑天鹅”的一个重要特征:收益是无限的(这与彩票不同),或者更确切地说,其界限是未知的;但错误所带来的损失则是有限和已知的。</p><p><img src="/2020/02/22/反脆弱读书笔记/图12-3.png" alt="图12-3"> <strong>图12–3</strong> 与图12–2情况相同,但在极端斯坦下,回报可能十分巨大</p></blockquote><h5 id="理性">理性</h5><blockquote><p>具体来说,我们可以这样描述选择权: 【<strong>选择权 = 不对称性+理性</strong>】</p></blockquote><h3 id="第13章-教鸟儿如何飞行">◇第13章 教鸟儿如何飞行</h3><h5 id="再谈少即是多">再谈少即是多</h5><blockquote><p>一项发现越是简单和明显,我们就越不容易通过复杂的方法将其弄明白。</p><p>政府和大学在推动创新和发现方面的贡献非常少,原因除了令他们盲目的理性主义外,他们还总喜欢寻找复杂的、耸人听闻的、有新闻价值的、有故事性的、科学主义的和宏伟的东西,却很少关注旅行箱的轮子。我意识到,简单很难给人戴上桂冠。</p></blockquote><h5 id="找寻与错误如何能成为投资">找寻与错误如何能成为投资</h5><blockquote><p>试错法有一种人们还未理解的首要价值:它其实【不是】随机的!由于可选择性的存在,它需要一定的理性:它需要你以一定的智慧来识别有利的结果,以及知道该放弃什么。<strong>我们从失败的尝试中能够逐渐摸索出正确的方向</strong></p></blockquote><h5 id="创造性和非创造性破坏">创造性和非创造性破坏</h5><p>【创造性破坏】有些事情需要毁灭才能促进整个系统的改善</p><h4 id="苏联哈佛鸟类学系">13.1 苏联–哈佛鸟类学系</h4><blockquote><p>让我们来看看两种类型的知识。</p><p>第一类不完全是“知识”,其模糊的特性让我们很难将其与严格定义的知识联系起来。这是一种我们无法用明确和直接的语言表达的做事方式——有时我们称其为否定法——但是,我们却切切实实地身体力行,而且做得很好。(学术→应用科学与技术→实践)</p><p>第二类更像我们平常所说的“知识”,它是你在学校里学会的东西,是你用来获得分数的东西,可以写成文章、可以解释、可以做学术、可以理性化、可以形式化、可以理论化、可以苏联化、可以官僚化、可以哈佛化,也是可以证明的。(数学 → 鸟类飞翔和羽翼扇动技术 → (忘恩负义的)鸟类会飞)</p></blockquote><h4 id="副现象">13.2 副现象</h4><blockquote><p>苏联–哈佛派谬见(给鸟类开设飞行课,并相信这些课程正是这些鸟类具备精湛的飞行技能的原因)属于一种被称为【副现象】的因果错觉。</p><p>人们很容易掉进这个陷阱:副现象引发社会行为,事后又为行为寻找合理化的解释。独裁者——就像一国政府——会觉得自己是不可缺少的,因为其他的替代方案我们不容易看见,或者被特殊利益集团隐藏起来了。例如,美联储会给经济带来巨大破坏,但人们仍旧确信它的有效性。人们害怕替代方案。同样的,在医源性损伤极度泛滥的年代,医疗仍被视为不可或缺。</p></blockquote><h5 id="贪婪是罪魁祸首">贪婪是罪魁祸首</h5><blockquote><p>每当经济危机发生时,贪婪很容易被当作罪魁祸首,这留给我们一个假象——如果我们可以找到它的根源并将它从生活中连根拔出,那经济危机就会被消灭。此外,我们倾向于认为贪婪是新生事物,因为那些肆虐全球的经济危机是新近发生的。始于2007~2008年的危机被解释为“贪婪”的产物,而不是源自系统的脆弱性,这让我们一厢情愿地(再次)做起了“消除贪婪的乌托邦之梦”。</p><p>这是一个【副现象】:贪婪比系统的脆弱性更古老。</p><p>贪婪总是以惊人的规律性反复被人视为新的与能够治愈的东西。消灭贪欲不过是一个类似普罗克拉斯提斯之床的方法;我们很难改变人类,所以应该建立一个抗贪婪的系统,但却没有人想到这个简单的解决方案。</p></blockquote><h5 id="揭穿副现象的真面目">揭穿副现象的真面目</h5><blockquote><p>通过观察事件的先后顺序,看看某件事是否总是发生在另一件事之前,我们往往就可以从文化话语和意识中挖掘出副现象。</p></blockquote><h5 id="净挑好的来说或证实谬误">“净挑好的来说”(或证实谬误)</h5><blockquote><p>“净挑好的来说”是有可选择性的:说故事(并出版)的人的优势在于能够展示用于证实的例子,而完全忽略其他情况——事实上,波动性和分散性越大,其成功的例子就越吸引人(其失败的例子也更惨痛)。有些拥有可选择性的人(即有权挑选自己要讲的故事的人)只会报告最符合其目的的东西。把故事好的一面呈现出来,把糟糕的一面隐藏起来,只要能打动听众就行了。</p></blockquote><h3 id="第14章-当两件事不是同一回事时">◇第14章 当两件事不是“同一回事”时</h3><h4 id="压力在哪里">14.1 压力在哪里?</h4><h5 id="为了学习而学习">为了学习而学习</h5><blockquote><p>教育 → 财富和经济增长</p><p>或者,</p><p>财富和经济增长 → 教育</p></blockquote><h5 id="晚餐的最佳搭档">晚餐的最佳搭档</h5><blockquote><p>教育可以让人成为举止更得体的晚餐伙伴,这一点不可忽略。通过教育来改善经济的想法还是比较新颖的。早在50年前的英国政府文件中,就提到了与我们今天的教育目标所不同的目的:改善价值观,培养好公民,促进“学习的内在价值”,而不是经济增长。</p></blockquote><h4 id="绿色木材谬误">14.2 绿色木材谬误</h4><blockquote><p>【绿色木材谬误】某个【很成功】的木材商人专售“绿色木材”。此人一直【误以为】“绿色木材”是——漆成绿色的木材。虽然此人对“绿色木材”的概念完全理解错了,但不妨碍此人经营木材生意。</p><p>除了看问题要排除【叙述性谬误】,我们还要接受另一个教训。大脑中装满太多复杂的技巧和方法,往往会使人忽略基本的东西。但是,现实世界中的人是绝对不能忽略这些事情的,否则他们会遭遇失败。与研究人员不同,他们被置于生死存亡的境地,而不只是面临一个难题。所以,少即是多:一方面,研究得越多,就越容易忽略基础但根本的事情;另一方面,行动能将事情剥丝抽茧,直至剩下尽可能简单的模型。</p></blockquote><h4 id="混为一谈">14.3 混为一谈</h4><blockquote><p>世界上存在着某种东西(在此指感觉、思想、理论)和某种东西的函数(在此指价格、现实或一些真实的东西)。混为一谈的问题就是指错把一个当作另一个,从而忘记了存在“函数”的情况,而事实上函数有不同的特性。 某种东西和某种东西的函数之间的不对称性越大,两者之间的差异就越大。最后,它们之间可能变得毫无关系。</p></blockquote><h5 id="普罗米修斯和埃庇米修斯">普罗米修斯和埃庇米修斯</h5><blockquote><p>普罗米修斯的意思是“先知先觉者”,而埃庇米修斯的意思是“后知后觉者”,相当于有人以事后叙述的方式寻找解释过去事件的恰当理论,并导致回溯性失真。普罗米修斯给人类带来了火种,代表人类文明的进步,而埃庇米修斯则代表思想落后、墨守成规和缺乏智慧。埃庇米修斯接受了潘多拉的礼物,一个大盒子,结果造成了不可逆转的后果。 【可选择性】是普罗米修斯那类人所拥有的,而【叙述性】则是埃庇米修斯那类人所拥有的——一个犯下的是可逆转的和良性的错误,另一个则象征打开潘多拉的盒子后无可挽回的严重后果。</p></blockquote><h3 id="第15章-失败者撰写的历史">◇第15章 失败者撰写的历史</h3><blockquote><p>实践者是不写文章的,他们只会放手去做;飞鸟果然会飞,但那些教导它们飞行的人,正是写故事的人。所以不难看出,历史实际上是由失败者撰写的,他们有大把的时间,学术地位也得到保护。</p></blockquote><h5 id="这与烹饪一样吗">这与烹饪一样吗?</h5><blockquote><p>一代代的集体自由探索成就了食谱的演变。这些食谱是扎根于文化中的。厨艺学校完全采取学徒制。</p></blockquote><h5 id="工业革命">工业革命</h5><blockquote><p>工业革命源于“技术人员建立起技术”,或者基利所说的“业余爱好者的科学”。</p></blockquote><h5 id="政府应该把钱花在非目的论的自由探索上而非科研上">政府应该把钱花在非目的论的自由探索上,而非科研上</h5><h4 id="医疗领域的案例">15.1 医疗领域的案例</h4><blockquote><p>在复杂领域,知识或所谓的“知识”阻碍了研究的发展。</p><p>医学领域中存在明显的漂移现象,可能为了某种目的而发明出来的药物,却在其他地方找到新的应用,这就是经济学家约翰·凯伊所称的间接发明——比如阿司匹林的用途就曾多次变更;犹大·福克曼关于限制对肿瘤的血液供应(血管生成抑制剂)的理论推动了黄斑变性治疗(贝伐单抗,俗称阿瓦斯汀)的诞生,效果比原先计划的效用更好。</p></blockquote><h5 id="马特里德利的反目的论观点">马特·里德利的反目的论观点</h5><blockquote><p>在怀疑论信仰主义者(即有宗教信仰的怀疑论者)阿尔加惹尔的眼中,知识不是掌握在人类的手中,而是掌握在上帝的手中,亚当·斯密称之为市场规律,而一些现代理论家将其理解为自我组织。</p></blockquote><h5 id="公司目的论">公司目的论</h5><blockquote><p>不管是理性的还是机会主义的企业漂移,下列公司可以告诉我们其中的真相。可口可乐起初是一种医药产品。蒂芙尼珠宝公司是做文具生意起家的。上面的两个例子中,公司的前后业务还算有关联,但再来看看下面这个例子:制造出第一枚导弹制导系统的雷神公司本来是一个冰箱制造商,而创办人之一正是范内瓦·布什,他构建了我们早先看到的目的论科学线性模型。再看更离谱儿的例子:曾经的顶级手机制造商诺基亚,是从造纸起家的(还做过一段时间的橡胶鞋)。杜邦,现在是一家以特富龙不粘锅、可丽耐台面和耐用的卡夫拉纤维闻名的企业,最早其实是一个经营炸药的公司。雅芳化妆品公司,一开始只是一个挨门推销图书的公司。最令人吃惊的是,奥奈达银器本来是搞社区宗教崇拜的,但出于监管合规的要求,它需要以合资公司的名义作为掩护。</p></blockquote><h4 id="反火鸡问题">15.2 反“火鸡”问题</h4><blockquote><p>将缺乏证明某事的证据错当成证明某事不存在的证据,这一简单的观点意味着:对反脆弱性的事物而言,好消息从过去的数据中是找不到的,而对于脆弱性的事物而言,坏消息倒是很罕见。</p><p>根据本章到目前为止所说的来总结一些规则: (1)寻找【可选择性】;根据可选择性对你要从事的业务进行排名; (2)最好选择【开放式】而非“封闭式”回报的业务; (3)切勿以商业计划为投资对象,而是以【人】为投资对象,寻找一个在其职业生涯中能够换6种或7种,甚至更多种工作的人——这是风险投资家马克·安德森运用的方法;以人为投资对象,有助于我们对商业计划书之类的文字叙述产生免疫力。这样做才可以确保提高你的强韧性; (4)不管你的业务是什么,请确保你选择了【杠铃策略】。</p></blockquote><h4 id="江湖郎中学者和爱出风头的人">15.3 江湖郎中、学者和爱出风头的人</h4><blockquote><p>“江湖郎中”被认为是经验主义者的代名词。“经验主义者”往往依靠试验和经验来确定哪些是正确的。换句话说,通过试错法和自由探索。他们被认为是劣等的——无论从职业、社会地位还是智力层面来说,他们仍然被认为不够“聪明”。</p><p>现代化带来了医学的学术化与制度化,从而引发了医源性损伤的初次大爆发,直到最近这种情况才开始扭转。此外,从历史的角度来看,正式的学者并不比那些所谓的江湖郎中更高明,他们只是将他们的欺骗行为掩藏在看似更有说服力的理性之下。他们只是有组织的骗子。我希望能够改变这种局面。</p></blockquote><h3 id="第16章-混乱中的秩序">◇第16章 混乱中的秩序</h3><h4 id="生态与游戏">16.1 生态与游戏</h4><blockquote><p>由于一个领域内的技能无法移植到另一个领域,这促使我对课堂上学习的任何技能,或者任何以非生态形式掌握的东西普遍心存怀疑,我更相信从街头斗殴和现实生活中学到的东西。</p></blockquote><h4 id="反脆弱性的杠铃式教育">16.2 反脆弱性的(杠铃式)教育</h4><blockquote><p>我读了陀思妥耶夫斯基、屠格涅夫、契诃夫、波舒哀主教、司汤达、但丁、普鲁斯特、博尔赫斯、卡尔维诺、塞林、舒尔茨、茨威格(不喜欢)、亨利·米勒、马克思·布洛德、卡夫卡、超现实主义作家尤涅斯库、福克纳、马尔罗(还有其他野外冒险家,如康拉德和梅尔维尔;我读的第一本英文书是《白鲸》)和类似的文学作家的作品,其中很多人的文字非常晦涩,还有黑格尔、叔本华、尼采、马克思、雅斯贝尔斯、胡塞尔、列维–斯特劳斯、列维纳斯、肖勒姆、便雅悯,以及类似的哲学家的作品,因为他们拥有不被列入学校课程的崇高地位,而且我坚决不看学校规定的书籍,所以迄今为止我没有读过拉辛、高乃依和其他无聊作家的书。有一年夏季,我决定在20天内读20本左拉的小说,每天读一本,最后我成功地做到了,虽然付出了很大的代价。我还读了马克思所著的书,也许是参加了一个组织促使我开始研究马克思主义。我对黑格尔的了解主要是通过亚历山大·柯热夫的作品间接得到的。</p><p>迄今,我仍然本能地认为,宝贵的知识,也就是你从事某一职业所需了解的知识,一定不在数据库中,所以要尽可能地远离中心。我在学校被灌输的东西,我已经忘光了;而我自己决定读的东西,我仍然记得很牢。追随自己的爱好来选择阅读的重要意义,莫过于此。</p></blockquote><h3 id="第17章-胖子托尼与苏格拉底辩论">◇第17章 胖子托尼与苏格拉底辩论</h3><blockquote><p>事实上,在生活中最严重的错误莫过于将不可叙述的事物误认为是愚蠢的东西——这是尼采最终得出的结论。在某种程度上,这种错误酷似火鸡问题,将我们没有看见的东西疑为不存在,这与将缺乏证明某事的证据错当成证明某事不存在的证据如出一辙。</p></blockquote><h4 id="欧蒂弗罗篇">17.1 《欧蒂弗罗篇》</h4><blockquote><p>苏格拉底使用的技巧是,先从一个论点出发,让谈话者同意一系列的陈述,然后告诉他,他所认同的这些陈述如何与他最初认同的那个论点不同,因而确定他对于自己要做的事情毫无头绪。苏格拉底使用这个技巧主要是向人们展示,他们的思考是多么缺乏清晰的逻辑,他们对自己日常运用的概念了解得多么少,以及他们多么需要哲学来帮助他们理清思路。</p></blockquote><h4 id="胖子托尼对话苏格拉底">17.2 胖子托尼对话苏格拉底</h4><p>塔勒布在这个章节假想了胖子托尼对苏格拉底的反驳。托尼的核心观点是——有些知识我们知道但不一定能够表达。</p><h4 id="定义性知识的优越性">17.2 定义性知识的优越性</h4><blockquote><p>对定义性知识的重视导致柏拉图得出了这样的结论,如果你不能认识事物的形式,你就不能说你了解事物,因为任何事物都是以定义来说明的。如果我们不能根据特殊性来界定虔诚,那么就让我们从普遍性开始,因为其中包含了特殊性。换句话说,如果你不能根据地形绘出地图,那么就根据地图造出地形。</p></blockquote><h5 id="误将难以理解当作愚蠢">误将难以理解当作愚蠢</h5><blockquote><p>“有些事情我不理解未必表示我无知”,这也许是尼采所在的时代最有力的一句话。</p><p>我花了很长的时间才搞清楚尼采在《悲剧的诞生》一书中谈到的核心问题。他看到了两股力量:太阳神阿波罗的精神和酒神狄俄尼索斯的精神。一种是可衡量的、平衡的、理性的,充满了逻辑和自我约束;另一种则是黑暗的、本能的、野性的、未驯服的、难以理解的,源自我们不为人知的另一面。古希腊文化代表了这两者的平衡,直到苏格拉底对欧里庇得斯的影响导致天平在很大程度上倾向了阿波罗,而中断了狄俄尼索斯的力量,导致理性主义的过度膨胀。它相当于通过注入激素破坏了你身体中自然的化学作用。有太阳神而没有酒神,就像中国人常说的,有阳无阴、阴阳失调。</p><p>作为一个思想家,尼采的力量给我带来了更多的惊喜:他琢磨出了反脆弱性。虽然许多人都(错误地)将“创造性破坏”的概念归功于经济学家约瑟夫·熊彼特;正如我们所看到的,更博学的人则将其归功于卡尔·马克思;然而,其实是尼采第一个创造出了这个词来形容狄俄尼索斯的,他称其为“创造性破坏”和“破坏性创造”。尼采确实以自己的方式想出了反脆弱性。</p></blockquote><h4 id="愚蠢的人和不愚蠢的人的区别">17.4 愚蠢的人和不愚蠢的人的区别</h4><blockquote><p>真与假(我们所谓的“信仰”)在人类的决策中扮演着糟糕的、辅助性的角色;起主宰作用的是真与假所带来的回报,而这几乎总是不对称的,一个结果会比另一个结果更大,也就是存在积极和消极的不对称性(脆弱性和反脆弱性)。</p></blockquote><h5 id="脆弱性而非概率">脆弱性,而非概率</h5><blockquote><p>你的决策主要基于脆弱性,而不怎么取决于概率的高/低。<strong>要知道即使是小概率的极端事件也能带来十分糟糕的结果</strong>。(对于极端事件)概率的大小在现实世界中并不重要,重要的是事件带来的影响。</p></blockquote><h5 id="事件与风险的混淆">事件与风险的混淆</h5><blockquote><p>“一个黑天鹅事件” 和 “它如何影响你”——并不是“同一回事儿”。问题的根源在于人们的标准反应——每当我们指出预测者的预测失误,他们的回答通常是“我们需要更精确的计算”,以更好地预测事件、找出概率;而不是更有效地“调整你的风险敞口”,并了解如何摆脱困境。宗教和传统的启发法比天真的伪科学更管用。</p></blockquote><h4 id="第四卷小结">17.5 第四卷小结</h4><blockquote><p>请记住,没有经验证据表明,如今所宣传的有组织的研究能够带来大学所承诺实现的成就。苏联–哈佛派观念的推动者从不使用可选择性,或二阶效应——他们的叙述中可选择性的缺乏,导致他们对技术科学的作用产生错误的看法。他们需要改写技术的历史了。</p><p>本节的结论认为,你的行动往往比你倾向于相信的更加明智,也更加理性。我在这里所做的只是揭穿了“教鸟儿飞行”这种副现象和“线性模型”,这无须特别的知识或智慧,只需在选择时有理性。</p></blockquote><h2 id="第五卷-非线性与非线性">★第五卷 非线性与非线性</h2><figure><img src="/2020/02/22/反脆弱读书笔记/图18-0.jpg" alt=""><figcaption>图18-0</figcaption></figure><h5 id="阁楼的重要性">阁楼的重要性</h5><blockquote><p>学者的工作就是忽略不重要的时事,致力于写书而不是写电子邮件,不用抛头露面到处演讲;他有更重要的事情要做,比如早上躺在床上看书,在窗前的写字台上写作,长距离散步(慢慢地走),喝浓咖啡(上午)、甘菊茶(下午)、黎巴嫩葡萄酒(傍晚)和马斯喀特酒(晚饭后),更长时间的散步(慢慢地走),与朋友和家人争论(但不要在早上),睡前阅读(再次),不要为了陌生人以及互联网上没读过你的书和理念的人去反复修改自己的书和想法。</p></blockquote><h3 id="第18章-一块大石头与一千颗小石子的区别">◇第18章 一块大石头与一千颗小石子的区别</h3><h4 id="发现脆弱性的简单法则">18.1 发现脆弱性的简单法则</h4><blockquote><p>【非线性】是指反应无法直接估计的、不呈直线分布的效应,所以如果你将药的剂量加倍,药效可能大大高于或者低于两倍。</p><p>【识别脆弱性】<strong>对于脆弱的事物来说,冲击带来的伤害会随着冲击强度的增加而以更快的速度增长(直到达到某一水平)。</strong></p></blockquote><h5 id="为什么脆弱性是非线性的">为什么脆弱性是非线性的?</h5><blockquote><p>对于脆弱的物体而言,温和冲击的累积效应低于等量的单一严重冲击所造成的单一影响。</p><p><strong>对于反脆弱性物体来说,在一定限度内,冲击越强,带来的益处越大(相应的,伤害也更小)。</strong></p></blockquote><h5 id="为什么凹性会受黑天鹅事件的伤害">为什么凹性会受黑天鹅事件的伤害?</h5><blockquote><p><img src="/2020/02/22/反脆弱读书笔记/图18-5.png" alt="图18-5"> <strong>图18–5</strong> 痛苦多于收益,或者收益多于痛苦。假设你从“你在这里”这一点开始,在第一种情况下,当变量x增加,即在横轴上向右移动,获得的收益(纵轴)将比变量x向左移动,即减少相同幅度时所遭受的损失更大。该图说明了正面不对称性(左图)会带来凸性效应(曲线向内),而负面不对称性(右图)会带来凹性效应(曲线向外)。再重申一遍,当变量在两个方向产生同等幅度的偏差时,凸性效应带来的收益会大于其损失,而凹性效应带来的收益则会小于损失</p><p><img src="/2020/02/22/反脆弱读书笔记/图18-6.png" alt="图18-6"> <strong>图18-6</strong> 每个图中有两类风险,一种是线性的,一种是非线性的。左图显示的是负凸性,也就是凹性,右图是正凸性。突发事件会对非线性产生不成比例的严重影响。事件越严重,两类风险所致影响的差别就越大</p></blockquote><h4 id="更多就是不同之处">18.2 更多就是不同之处</h4><blockquote><p>科学家们研究复杂性时发现的所谓“新兴属性”,就是指增加单元所导致的非线性结果,因为加总的单元越多,其总和就变得与各组成部分越不同。</p></blockquote><h4 id="小的可能是丑陋的但肯定不那么脆弱">18.3 小的可能是丑陋的,但肯定不那么脆弱</h4><blockquote><p>忍痛会因规模加大而恶化。规模大的东西,面对某些错误容易受到伤害,特别是在可怕的忍痛情绪下。随着规模进一步增大,其代价会非线性地增加。</p></blockquote><h4 id="预估与预测">18.4 预估与预测</h4><blockquote><p>随着复杂性的增强、各部分之间相互依存度的增加、全球化的推进,以及所谓“效率”这种让人们违背规律行事的野蛮概念的出现,“黑天鹅”效应势必增加。对了,咨询师和商学院的出现也加剧了这一趋势。一个地方出现问题就可能导致整个项目的停顿——项目最薄弱的一环往往决定了项目的成败(一种负凸性效应)。世界变得越来越难以预测,我们越来越多地依赖于错误的高科技技术,这些技术的相互影响很难估计,更不用说预测了。</p></blockquote><h4 id="强调效率的地方偏偏缺乏效率">18.5 强调“效率”的地方偏偏缺乏效率</h4><blockquote><p>小结:不管在什么领域,从瓷杯子到组织,到政治制度,再到公司的规模,或机场的延误,脆弱性总是隐藏在非线性中。此外,我们的发现可能会被视为我们反对赤字。想想与飞机延误或项目超支完全相反的情况,即可以从不确定性中受益的情况,我们发现的事情呈现出与我们所看到的脆弱的、厌恶随机性的情况截然相反的镜像。</p></blockquote><h3 id="第19章-炼金石与反炼金石">◇第19章 炼金石与反炼金石</h3><blockquote><p>如果你拥有有利的不对称性,或正凸性(选择权是特例),从长远来看,你会做得相当不错,在不确定的情况下表现优于平均数。不确定性越强,可选择性的作用越大,你的表现就越好。这个属性对人生来说非常重要。</p></blockquote><h2 id="第六卷-否定法">★第六卷 否定法</h2><figure><img src="/2020/02/22/反脆弱读书笔记/图20-0.jpg" alt=""><figcaption>图20-0</figcaption></figure><h3 id="第20章-时间与脆弱性">◇第20章 时间与脆弱性</h3><h4 id="从西蒙尼德斯到詹森">20.1 从西蒙尼德斯到詹森</h4><blockquote><p>古希腊诗人西蒙尼德斯:“时间有撕碎一切的锋利牙齿,连最坚固的东西也不例外。” 古罗马诗人奥维德:“时间吞噬一切。” 20世纪俄裔法国女诗人爱尔莎·特丽奥莱:“时间燃尽了,却没有留下任何灰烬。”</p></blockquote><p>脆弱的事物很容易在时间长河中毁灭/消逝;反脆弱的东西当然也会,但存在的时间要长得多。</p><h4 id="学习做减法">20.2 学习做减法</h4><blockquote><p>那些致力于描述未来的人往往会患上新事物狂热症(不治之症),一切只是因为他们喜欢现代。通过自我削减的方式,技术便可以消除不良技术带来的影响。</p></blockquote><h4 id="越活越年轻林迪效应">20.3 越活越年轻——林迪效应</h4><blockquote><p>伟大的伯努瓦·曼德布罗特对林迪效应的诠释: 对于会自然消亡的事物,生命每增加一天,其预期寿命就会缩短一些。 而对于不会自然消亡的事物,生命每增加一天,则可能意味着更长的预期剩余寿命。</p></blockquote><h4 id="一些心理偏差">20.4 一些心理偏差</h4><blockquote><ol type="1"><li>信息有一个讨厌的特性:它会将失败隐藏起来。</li><li>另一种导致我们过分宣扬某种技术的心理偏见,来自我们只关注动态而非静态的事实。我们对变化的注意,远远多过扮演重要角色但不变的事物。【新事物狂热症和跑步机效应】</li></ol></blockquote><h4 id="建筑和不可逆转的新事物狂热症">20.5 建筑和不可逆转的新事物狂热症</h4><blockquote><p>此外,无论是城市或是楼房,只要是以自然的方式增长的,就都具有分形的特征。就像一切生物和有机体一样,如肺或树木,它们以自我引导的形式生长,同时驯服了随机性。什么是分形?回忆一下我们在第3章中讲的曼德布罗特的洞见:“分形”既需要杂乱的延伸扩展,也需要自相似性(曼德布罗特喜欢用“自仿射性”的概念),如树木会生出很多枝条,每个枝条看起来都像一棵小树,枝条上又生出更小的枝条,看起来仍与整体相仿,就像稍加修饰但仍识别得出的整体。这些分形基于嵌套模式重复的规律,产生了某种丰富的细节。分形需要一些杂乱的表象,但是你会有一些方法来识别其杂乱背后的规律。大自然中的一切从本质上来看都是分形结构——杂乱、富含细节,但是遵循一定的模式。相比而言,有序则属于我们在学校学习的欧几里得几何,简化了形状,但失去了丰富的内涵。</p><p>建筑物自诞生后常常会出现变化,就好像它们需要缓慢地进化以与周围变化的环境相契合:它们会改变颜色、形状、窗户和特征。斯图尔特·布兰德在他的书《建筑物如何进行学习》中登出了照片,显示建筑是怎样随时间而改变的,就好像它们需要蜕变成识别不出的形状——奇怪的是,这些建筑自兴建之时,就从未考虑到未来改变的可选择性。</p></blockquote><h4 id="把科学变成新闻">20.6 把科学变成新闻</h4><blockquote><p>把脆弱性和强韧性的标准应用到【信息的处理方面】——这种情况与技术类似,脆弱的东西就是经不起时间考验的东西。所以,最好的过滤性启发法,就是看书籍和科学论文的流传年数。只出版一年的书籍通常是不值得一读的(它具有“流芳百世”的质量的概率非常低),不管炒作得多么厉害,或者它们看上去是多么“惊世骇俗”。所以在选择读什么书时,我以林迪效应为指导:已经流传10年的书将再流传10年;流传了2000年的书籍还将流传更多时间,以此类推。许多人认识到了这一点,但并未将其应用到学术工作中;学术工作的许多现代实践,与新闻工作没有多少区别(除了偶尔的原创作品)。学术工作因为有寻求关注的倾向,所以很容易受制于林迪效应:想想看,数以百万计的论文不管在出版时如何大肆宣传,本质上也只是噪声。</p></blockquote><h4 id="应该会消失的东西">20.7 应该会消失的东西</h4><blockquote><p>大多数存活了25年的技术应该还能再存活25年——再次声明,我指的是大多数,不是全部。我预测脆弱的事物将消失,或者被削弱。现在,让我们看看什么是脆弱的呢?大的、优化的、过度依赖于技术的东西。脆弱的事物过度依赖于所谓的科学方法,而不是经时间验证的启发法。今天的大型公司到那时应该都消失了,因为它们将规模视为自己的实力,结果却被规模所误:规模之所以是公司的敌人,是因为它会导致公司在“黑天鹅”面前呈现不相称的脆弱性。城邦制国家和小型公司更容易在未来幸存,甚至茁壮成长。集权制国家、印发货币的央行,以及那些被称为经济部门的机构,名义上可能仍然存在,但它们的权力将被严重侵蚀。换句话说,我们在三元结构图“脆弱类”一栏中所看到的东西将消失——取而代之以其他脆弱的事物。</p></blockquote><h4 id="先知与现在">20.8 先知与现在</h4><blockquote><p>我们无法从过去的行为中学习的现象。学习时缺乏递归思维,也就是缺乏二阶思维的问题是:如果那些传递一些从长远来看有价值信息的人在过去的历史中受到迫害。人们就会预期,应存在一个纠错机制,使聪明的人最终从这类历史经验中吸取教训,而传递信息的那个人也终将得以正名。但<strong>事实并不是这样</strong>。</p></blockquote><h4 id="恩培多克勒的狗">20.9 恩培多克勒的狗</h4><blockquote><p>如果某个东西对你来说没有任何意义(比如宗教——如果你是一个无神论者的话,或者某些不合理的古老习惯或做法);如果这个东西已经存在很长一段时间,那么,不管你认为它合理或不合理,你都可以预期它还会存在更长的时间,比那些预言它会死亡的人存活的时间更长。</p></blockquote><h3 id="第21章-医疗凸性和不透明">◇第21章 医疗、凸性和不透明</h3><p>凸性效应的概念和医疗领域的举证责任,以及对医源性损伤的风险评估。</p><blockquote><p>我们可以列举很多得到验证的和潜在的医源性损伤的惨痛例子(对危重病人之外的患者来说弊大于利,不管这样的弊端是否已被验证):消炎药伟克适造成的副作用是延后引发心脏问题。抗抑郁药(在不必要情况下的使用)。减肥手术(取代了超重糖尿病患者的饥饿疗法)。可的松。用消毒剂清洁产品可能导致自身的免疫性疾病。激素替代疗法。子宫切除术。不是绝对必要的剖腹产。给婴儿装耳管,在耳部感染时做出即时反应。前脑叶白质切除术。补铁。对大米和小麦进行漂白,这被视为进步。防晒霜,可能会造成某种伤害。卫生(过分卫生,会因拒绝了毒物兴奋效应,即我们自身的反脆弱性,而使你变得脆弱)。我们摄取益生菌,因为我们不再吃足量的“脏东西”了。来苏水等消毒剂杀死了那么多“细菌”,使得儿童发育中的免疫系统被剥夺了必要的锻炼机会。口腔卫生:我不知道用沾满化学物质的牙刷刷牙是否主要是为了给牙膏行业制造利润——牙刷是普通之物,而使用牙膏可能只是为了对抗我们消耗的非自然产品,如淀粉、糖、高果糖玉米糖浆。说到这,高果糖玉米糖浆也是新事物狂热症的产物,是由喜爱技术的尼克松政府赞助开发的,到头来不仅要补贴玉米种植户,也给消费者造成了伤害。II型糖尿病患者的胰岛素注射治疗,其基于这样一种假设,即糖尿病的危害来自血糖,而非胰岛素抗性(或别的与它相关联的东西)。豆浆。地中海和亚裔人喝的牛奶。海洛因,我们可以想见的最危险的上瘾毒品,也发展成吗啡替代品用作止咳药,不会有吗啡的成瘾副作用。精神病治疗,特别是儿童精神病治疗——但对它的危险性我想我不需要再说了……就此打住。</p><p>请允许我在这里用一段文字再重复一下我的论点,以使其更清晰。进化是靠无定向的、凸性的自由探索或试错来推进的,因此本质上是强韧的,因为它能从连续的、重复的、细小的、局部的错误中获得潜在的随机收益。而人们开展的自上而下、指挥控制的科研过程却完全相反:这是一种伴有负凸性效应的干预,即暴露于大量的潜在错误中而只获得一些很小的收益。人类理解复杂系统(生物、经济、气候)的风险记录一向少得可怜,再加上回顾时对事实的歪曲(我们只能在损害发生后才理解风险,但接下来却继续犯错误),现在没有证据可以让我相信,我们的风险管理能力增强了。在上述特殊案例中,由于错误的可扩大性,你面对的是最不可预测的随机性。</p></blockquote><h3 id="第22章-活得长寿但不要太长">◇第22章 活得长寿,但不要太长</h3><h4 id="预期寿命和凸性">22.1 预期寿命和凸性</h4><h5 id="减法使你的寿命增长">减法使你的寿命增长</h5><blockquote><p>换句话说,我们应该向后推理,即从医源性损伤出发来考虑选用何种治疗方法,而不是背道而驰。只要有可能,就用人类的反脆弱性来替代医生。但是,在其他情况下,则不要抗拒积极的治疗。</p><p>让我们探究一下古老的智慧。普罗提诺写道:“有时候缺乏营养会让机体恢复。”古代人很相信洁净身体的必要性(洁净的一种方式就是,往往造成伤害但也经常有益的放血)。萨勒诺医学院的养生法则是:愉悦的心情、充足的休息,以及适当缺乏营养。</p></blockquote><h5 id="金钱的医源性损伤">金钱的医源性损伤</h5><blockquote><p>钱也有它自己的医源性损伤,某些人失去财富只会让他们的生活更简单,以健康压力因子的形式令其受益匪浅。</p></blockquote><h5 id="宗教和天真的干预主义">宗教和天真的干预主义</h5><blockquote><p>宗教有它的无形目的,超越缺乏想象力的科学主义和科学至上的认同——事实上,宗教的目的之一就是保护我们免受科学主义的伤害。</p></blockquote><h5 id="凸性效应和随机营养">凸性效应和随机营养</h5><blockquote><p>这就是禁食的压力所带来的反脆弱性,它使我们所渴望的食物品尝起来更美味,也能在我们的肌体系统内产生更多的快感。禁食后进食的感觉与吃喝过量的感觉完全相反。</p></blockquote><h4 id="如何吃掉自己">22.2 如何吃掉自己</h4><blockquote><p>我们在约束的压力下只会变得更加体力充沛和健康。人体机制会因食物匮乏而被激活。</p></blockquote><h2 id="第七卷-脆弱性与反脆弱性的伦理">★第七卷 脆弱性与反脆弱性的伦理</h2><h3 id="第23章-切身利害反脆弱性和牺牲他人的可选择性">◇第23章 切身利害——反脆弱性和牺牲他人的可选择性</h3><blockquote><p>在传统社会中,一个人所能获得的尊敬与他(或她,人们对女人在这方面的期待更大或大得多)的价值取决于其愿意为他人而承受损失的多少。最英勇无畏的人将在社会中占据最高层级:骑士、将军、指挥官。......这同样适用于圣人,他们放弃和贡献了他们的生命以造福他人,比如帮助弱者、贫困者和无依无靠的人。</p><p>事实上,社会的强韧性,甚至反脆弱性,都有赖于这些人:我们之所以今天还能生活在这个世界上,就是因为某些人在某个阶段为我们承担了风险。但是,勇气和英雄主义并不等于盲目的冒险,也绝不等同于鲁莽。</p><p>当我们听到幸福是根据经济或其他微不足道的物质条件来定义时,应该停下来思考一下。你可以想象,每次听到有人宣扬与英雄主义毫无关系的“中产阶层价值观”时,我就觉得极其厌烦。得益于全球化和互联网,这个所谓的“中产阶层价值观”已经传递到了英国航空能够轻松抵达的任何一个地区,原本被高贵的阶层所深恶痛绝的东西如今却被大肆推崇——为大公司“卖力工作”;勤奋地阅读报纸;被某种企业结构束缚;一味听从老板的意见;遵守法律;依赖股市投资;去热带地区度假;住在郊区并以抵押贷款购房;养着一条非常漂亮的狗;周末的晚上品酒消遣。</p><p>取得一些成功的人便能跻身年度亿万富翁排行榜,并期待能在榜上多逗留一段时间。他们被称为英雄,但其实只是幸运儿。</p></blockquote><h4 id="汉谟拉比">23.1 汉谟拉比</h4><blockquote><p>《汉谟拉比法典》距今已有3800年的历史了,却认识到有必要重建【脆弱性的对称性】,其中有一段话是这么写的:</p><blockquote><p>如果建筑师建造了一所房子,房子倒塌了,并导致屋主死亡,那么造房子的建筑师应被处死。如果导致屋主的儿子死亡,那么建筑师的儿子应该偿命。如果造成屋主的奴隶死亡,那么建筑师应该赔偿屋主一个同等价值的奴隶。</p></blockquote><p>胖子托尼有两个启发法。 第一,千万不要登上没有飞行员的飞机。第一个启发法解决的是奖励和惩罚之间的不对称性,或个体之间脆弱性的转移。 第二,确保飞机上还有一个副驾驶。第二条启发法是说我们需要建立冗余、安全边际,避免优化,以减轻甚至消除我们的风险敏感度中的不对称性。</p></blockquote><h4 id="空谈者的免费选择权">23.2 空谈者的免费选择权</h4><blockquote><p>大自然和我们祖先的系统都是建立在惩罚措施之上的:任何人都没有永恒的免费选择权。许多事情会有明显效应的社会也是一样。</p><p>事后预测者总是在事情发生后对其进行解释——因为他们的工作就是空谈——让自己看上去总是比预测者聪明。由于事后扭曲,有些人当然没事先看到事件来临,却记得他们想过那件事产生的影响,而且会设法说服自己,是他们预测到了这些事件,随后还要说服别人相信这一点。每个事件发生后,事后预测者的人数总是多于真正的预测者。</p></blockquote><h4 id="古人和斯蒂格利茨综合征">23.3 古人和斯蒂格利茨综合征</h4><blockquote><p>【斯蒂格利茨综合征】=(可能出发点是善意的)脆弱推手+事后过滤式选择</p><p><strong>解决斯蒂格利茨综合征的良策</strong></p><ol type="1"><li>不要询问任何人的投资意见、预测或建议。只要问他们的投资组合中“有什么或没有什么”就行了。</li><li>不要问医生你该做什么;而是问医生,如果他处于你的情况下,他会做什么。你会很惊讶地发现这两者的差异。</li></ol></blockquote><h4 id="选择权反脆弱性与社会公平">23.4 选择权、反脆弱性与社会公平</h4><blockquote><p>无论是在“资本主义”体制,还是你希望的任何经济体制下,你所需要做的就是将三元结构中最左栏的人数保持在最低水平。没有人意识到苏联体制的中心问题是,它把每个负责经济生活的人都置于极其脆弱的左栏。</p></blockquote><h4 id="反脆弱性与大公司的伦理">23.5 反脆弱性与大公司的伦理</h4><blockquote><p>总而言之,企业是如此脆弱,长期来看,它终将不堪代理问题的重负而轰然倒塌,与此同时,经理人却能大口吃肉,却只给纳税人留点骨头。如果不是因为强大的游说机器,企业早就崩溃了:它们劫持了这个国家来帮助它们为消费者的消化系统注入含糖饮料。在美国,大型企业控制着一些国会议员。这样的做法牺牲了我们的利益,推迟了公司的死亡。</p></blockquote><h3 id="第24章-给职业戴上伦理光环">◇第24章 给职业戴上伦理光环</h3><blockquote><p>【跑步机效应(与新事物狂热症情况类似)】——为了停留在同一个地方,你需要赚越来越多的钱。贪婪是具有反脆弱性的,但贪婪的受害者则不然。</p></blockquote><h4 id="专业人士和集体">24.1 专业人士和集体</h4><blockquote><p>人经过一个阶段的知识灌输后,便可以迅速成为一种职业的奴隶,甚至他对任何问题的观点都变得以利己为目的,从而对集体来说不再可靠。这是希腊人对专业人士有意见的原因所在。</p><p>受雇于公民,理论上应该代表公民最佳利益的前主管官员和公务员,可以利用他们的专业知识和在工作中积累的人脉,然后加入私营企业(律师事务所等)并从系统的缺陷中谋利。进一步想想吧:法律规定越复杂,社会网络越官僚,就有越多深谙系统漏洞和缺陷的主管官员从中受益,因为他的主管优势将是其专业知识的凸性函数。这是一种特权,是以牺牲别人为代价获得的不对称性。(请注意,这种特权正在整个经济中蔓延;丰田汽车公司就曾聘请前美国主管官员,利用他们的“专业知识”来处理汽车缺陷调查案。)</p></blockquote><h4 id="大数据与研究者的选择权">24.2 大数据与研究者的选择权</h4><blockquote><p><img src="/2020/02/22/反脆弱读书笔记/图24-1.png" alt="图24-1"> <strong>图24–1</strong> 大数据的悲剧。变量越多,“老练”的研究员手中的显著相关性越多。虚假相关性的增长比信息增长得更快,数据因而呈现出非线性(凸性)</p></blockquote><h4 id="集体的专制">24.3 集体的专制</h4><blockquote><p>集体而非个体犯错是有组织的知识的标志,也是反对此类知识的最好论据。我们常常听到人们以“因为大家都在做”或“别人就是这么做的”来为自己辩护。这可不是小事:在一个人的情况下,如果感觉做某事很傻,人们就不会去做,但是在一群人的情况下,他们却会参与这样的傻事。而这正是学术界亵渎科学的原因所在。</p></blockquote><h3 id="第25章-结语">◇第25章 结语</h3><blockquote><p>所有真实的想法都可以提炼成一个核心问题,而某些具体领域内的绝大多数人因为专业分工和虚有其表而完全忽略了这一问题。比如说:一切宗教律法都可以归结为一条黄金法则的细化、应用和诠释,即“己所不欲勿施于人”。这就是我们看到的《汉谟拉比法典》背后的逻辑。这里的黄金法则是真正的精华,而不是普罗克拉斯提斯之床。核心论点从来都不只是理论概括,它更像是一台引擎。</p><p><strong>(本书)核心论点:一切事物都会从波动性中获得【收益或损失】。脆弱性就是波动性(不确定性)带来的损失。</strong></p><p>普罗米修斯偏爱无序;而埃庇米修斯则憎恶无序。我们可以根据无序的风险和对这个风险的偏好,将人们及其经验的质量加以区分:斯巴达士兵不同于博客作者,冒险家不同于文字编辑,腓尼基商人不同于文法教师,海盗不同于探戈教师。</p><p>每一件非线性的东西,要么是凸性,要么是凹性,或两者兼具——这取决于压力源的强度。我们看到“凸性”与“波动性偏好”之间的联系。<strong>因此,所有事物都在一定程度上【喜欢或讨厌】波动性,每一样东西都是如此。</strong></p></blockquote><p>参考https://zhuanlan.zhihu.com/p/79605630</p>]]></content>
<summary type="html">
<h1 id="读书笔记反脆弱从不确定性中获益"><a href="https://program-think.blogspot.com/2018/12/Book-Review-Antifragile-Things-That-Gain-from-Disorder.html" ta
</summary>
</entry>
<entry>
<title>Vibrations and Waves</title>
<link href="https://github-9233.github.io/2019/07/03/Vibrations-and-Waves/"/>
<id>https://github-9233.github.io/2019/07/03/Vibrations-and-Waves/</id>
<published>2019-07-03T08:32:26.000Z</published>
<updated>2019-07-03T08:36:15.728Z</updated>
<content type="html"><![CDATA[<h2 id="the-dispersion-of-waves">8 The Dispersion of Waves</h2><h3 id="the-superposition-of-waves-in-non-dispersive-media">8.1 THE SUPERPOSITION OF WAVES IN NON-DISPERSIVE MEDIA</h3><h4 id="beats">8.1.1 Beats</h4><figure><img src="https://upload.wikimedia.org/wikipedia/commons/e/eb/WaveInterference.gif" alt=""><figcaption>The sum (blue) of two sine waves (red, green) is shown as one of the waves increases in frequency. The two waves are initially identical, then the frequency of the green wave is gradually increased by 25%. Constructive and destructive interference can be seen.</figcaption></figure><h4 id="amplitude-modulation-of-a-radio-wave">8.1.2 Amplitude modulation of a radio wave</h4><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/a/a4/Amfm3-en-de.gif/250px-Amfm3-en-de.gif" alt=""><figcaption>An audio signal (top) may be carried by a <a href="https://en.wikipedia.org/wiki/Carrier_signal" target="_blank" rel="noopener"><em>carrier signal</em></a> using AM or FM methods.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/8d/Illustration_of_Amplitude_Modulation.png/391px-Illustration_of_Amplitude_Modulation.png" alt=""><figcaption>Illustration of amplitude modulation</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/b/b0/Amplitude_Modulated_Wave-hm-64.svg/400px-Amplitude_Modulated_Wave-hm-64.svg.png" alt=""><figcaption><em>Modulation depth</em>. In the diagram, the unmodulated carrier has an amplitude of 1.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/f/fd/Am-sidebands.png/300px-Am-sidebands.png" alt=""><figcaption>The power of an AM radio signal plotted against frequency. <strong>fc</strong> is the <a href="https://en.wikipedia.org/wiki/Carrier_frequency" target="_blank" rel="noopener"><em>carrier frequency</em></a>, <strong>fm</strong> is the maximum modulation frequency</figcaption></figure><h3 id="the-dispersion-of-waves-1">8.2 THE DISPERSION OF WAVES</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/9/9e/Commutative_diagram_of_harmonic_wave_properties.svg/300px-Commutative_diagram_of_harmonic_wave_properties.svg.png" alt=""><figcaption>Diagram illustrating the relationship between the wavenumber and the other properties of harmonic waves.</figcaption></figure><p>frequency <span class="math inline">\(\omega\)</span> is a function of the wavenumber <span class="math inline">\(k\)</span> :<span class="math inline">\(\omega=\omega(k)\)</span></p><h4 id="phase-and-group-velocities">8.2.1 Phase and group velocities</h4><figure><img src="https://upload.wikimedia.org/wikipedia/commons/b/bd/Wave_group.gif" alt=""><figcaption>Frequency dispersion of surface gravity waves on deep water. The <span class="math inline">\(\color{red}{ ▪︎ red\ dot}\)</span> moves with the phase velocity, and the <span class="math inline">\(\color{green}{ • green\ dots}\)</span> propagate with the group velocity. In this deep-water case, the phase velocity is twice the group velocity. The <span class="math inline">\(\color{red}{ ▪︎ red\ dot}\)</span> traverses the figure in the time it takes the green dot to traverse half.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/4/45/Two-frequency_beats_of_a_non-dispersive_transverse_wave_%28animated%29.gif" alt=""><figcaption>Two-frequency beats of a non-dispersive transverse wave. Since the wave is non-dispersive, <span class="math inline">\(\color{red}{ •\ phase}\)</span> and <span class="math inline">\(\color{green}{ •\ group}\)</span> velocities are equal.</figcaption></figure><ul><li><em>normal dispersion</em></li><li><em>anomalous dispersion</em></li><li><em>no dispersion</em></li></ul><h3 id="the-dispersion-relation">8.3 THE DISPERSION RELATION</h3><ul><li><p>Dispersion of electromagnetic waves occurs in the propagation of radio waves in the ionosphere <span class="math display">\[\omega^2=\omega_0^2+c^2k^2\]</span> <span class="math inline">\(\omega_0\)</span>:<em>plasma oscillation frequency</em></p></li><li><p>For waves on deep water, where the wavelength is small compared with the depth of the water, the angular frequency ω and wavenumber k are related by the dispersion relation <span class="math display">\[\omega^2=gk+\frac{Sk^3}{\rho}\]</span></p></li></ul><h3 id="wave-packets">8.4 WAVE PACKETS</h3><h4 id="formation-of-a-wave-packet">8.4.1 Formation of a wave packet</h4><p><em>bandwidth theorem</em></p><p><img src="/2019/07/03/Vibrations-and-Waves/wavenumber-width.png" alt="wavenumber-width"> <span class="math display">\[\Delta x \Delta k \approx 2 \pi\]</span> <img src="/2019/07/03/Vibrations-and-Waves/duration-frequency.png" alt="duration-frequency"> <span class="math display">\[\Delta t \Delta \omega \approx 2 \pi\]</span> <em>Heisenberg Uncertainty Principle</em></p>]]></content>
<summary type="html">
<h2 id="the-dispersion-of-waves">8 The Dispersion of Waves</h2>
<h3 id="the-superposition-of-waves-in-non-dispersive-media">8.1 THE SUPERPOS
</summary>
</entry>
<entry>
<title>尝试用最简单的方式讲乐理知识</title>
<link href="https://github-9233.github.io/2019/06/19/%E5%B0%9D%E8%AF%95%E7%94%A8%E6%9C%80%E7%AE%80%E5%8D%95%E7%9A%84%E6%96%B9%E5%BC%8F%E8%AE%B2%E4%B9%90%E7%90%86%E7%9F%A5%E8%AF%86/"/>
<id>https://github-9233.github.io/2019/06/19/尝试用最简单的方式讲乐理知识/</id>
<published>2019-06-19T00:38:14.000Z</published>
<updated>2019-07-15T13:36:27.175Z</updated>
<content type="html"><![CDATA[<h2 id="调式篇"><a href="https://www.gcores.com/radios/93658" target="_blank" rel="noopener">调式篇</a></h2><h3 id="音律">音律</h3><ul><li><p>十二平均律</p><p><strong>A</strong>或者<strong>la</strong>是唱名的第六個音。當計算<a href="https://zh.wikipedia.org/wiki/十二平均律" target="_blank" rel="noopener">十二平均律</a>時,以A4(中音A)為標準,設定為440 <a href="https://zh.wikipedia.org/wiki/Hz" target="_blank" rel="noopener">Hz</a></p></li><li><p>五度相生律 <img src="https://upload.wikimedia.org/wikipedia/commons/thumb/3/33/Circle_of_fifths_deluxe_4.svg/400px-Circle_of_fifths_deluxe_4.svg.png" alt="五度圈內各大小調的調號"></p></li><li><p>纯律</p></li></ul><h3 id="唱名">唱名</h3><ul><li><p>首调唱名法:将当前<a href="https://zh.wikipedia.org/wiki/调" target="_blank" rel="noopener">调</a>的<a href="https://zh.wikipedia.org/wiki/主音" target="_blank" rel="noopener">主音</a>唱成do。</p></li><li><p>固定唱名法:将do、re、mi、fa、sol、la、si与C、D、E、F、G、A、B 一一对应。无论哪个调,唱名都是绝对的。</p></li></ul><table><thead><tr class="header"><th style="text-align: center;">C</th><th style="text-align: center;">D</th><th style="text-align: center;">E</th><th style="text-align: center;">F</th><th style="text-align: center;">G</th><th style="text-align: center;">A</th><th style="text-align: center;">B</th></tr></thead><tbody><tr class="odd"><td style="text-align: center;">1</td><td style="text-align: center;">2</td><td style="text-align: center;">3</td><td style="text-align: center;">4</td><td style="text-align: center;">5</td><td style="text-align: center;">6</td><td style="text-align: center;">7</td></tr><tr class="even"><td style="text-align: center;">do</td><td style="text-align: center;">re</td><td style="text-align: center;">mi</td><td style="text-align: center;">fa</td><td style="text-align: center;">sol</td><td style="text-align: center;">la</td><td style="text-align: center;">si</td></tr><tr class="odd"><td style="text-align: center;">主音</td><td style="text-align: center;">上主音</td><td style="text-align: center;">中音</td><td style="text-align: center;">下属音</td><td style="text-align: center;">属音</td><td style="text-align: center;">下中音</td><td style="text-align: center;">下主音</td></tr></tbody></table><h3 id="调性">调性</h3><p><strong>调性</strong>(Tonality)是<a href="https://zh.wikipedia.org/wiki/调" target="_blank" rel="noopener">调</a>的主音和<a href="https://zh.wikipedia.org/wiki/调式" target="_blank" rel="noopener">调式</a>类别的总称,例如,以C为主音的<a href="https://zh.wikipedia.org/wiki/大調" target="_blank" rel="noopener">大调</a>式,其调性即是“C大调”,以a为主音的<a href="https://zh.wikipedia.org/wiki/小调" target="_blank" rel="noopener">小调</a>式,其调性就是“a小调”等。以此类推,一般<a href="https://zh.wikipedia.org/wiki/音乐" target="_blank" rel="noopener">音乐</a>中主要有24个调性。</p><table><thead><tr class="header"><th style="text-align: center;">調式 (18世紀前)</th><th style="text-align: center;">調式 (18世紀後)</th><th style="text-align: center;">英文名稱</th><th style="text-align: center;">中文翻譯</th><th style="text-align: center;">正格 / 變格</th><th style="text-align: center;">結束音 (final)</th><th style="text-align: center;">吟誦音 (reciting tone)</th></tr></thead><tbody><tr class="odd"><td style="text-align: center;">1</td><td style="text-align: center;">1</td><td style="text-align: center;">Dorian</td><td style="text-align: center;">多里安</td><td style="text-align: center;">正格</td><td style="text-align: center;">D</td><td style="text-align: center;">A</td></tr><tr class="even"><td style="text-align: center;">2</td><td style="text-align: center;">2</td><td style="text-align: center;">Hypodorian</td><td style="text-align: center;">變格多里安</td><td style="text-align: center;">變格</td><td style="text-align: center;">D</td><td style="text-align: center;">F</td></tr><tr class="odd"><td style="text-align: center;">3</td><td style="text-align: center;">3</td><td style="text-align: center;">Phrygian</td><td style="text-align: center;">弗利吉安</td><td style="text-align: center;">正格</td><td style="text-align: center;">E</td><td style="text-align: center;">B-C[<a href="https://zh.wikipedia.org/wiki/教會調式#cite_note-phrygian-3" target="_blank" rel="noopener">注 1]</a></td></tr><tr class="even"><td style="text-align: center;">4</td><td style="text-align: center;">4</td><td style="text-align: center;">Hypophrygian</td><td style="text-align: center;">變格弗利吉安</td><td style="text-align: center;">變格</td><td style="text-align: center;">E</td><td style="text-align: center;">G-A[<a href="https://zh.wikipedia.org/wiki/教會調式#cite_note-hypophrygian-5" target="_blank" rel="noopener">注 2]</a></td></tr><tr class="odd"><td style="text-align: center;">5</td><td style="text-align: center;">5</td><td style="text-align: center;">Lydian</td><td style="text-align: center;">利地安</td><td style="text-align: center;">正格</td><td style="text-align: center;">F</td><td style="text-align: center;">C</td></tr><tr class="even"><td style="text-align: center;">6</td><td style="text-align: center;">6</td><td style="text-align: center;">Hypolydian</td><td style="text-align: center;">變格利地安</td><td style="text-align: center;">變格</td><td style="text-align: center;">F</td><td style="text-align: center;">A</td></tr><tr class="odd"><td style="text-align: center;">7</td><td style="text-align: center;">7</td><td style="text-align: center;">Mixolydian</td><td style="text-align: center;">米索利地安</td><td style="text-align: center;">變格</td><td style="text-align: center;">G</td><td style="text-align: center;">D</td></tr><tr class="even"><td style="text-align: center;">8</td><td style="text-align: center;">8</td><td style="text-align: center;">Hypomixolydian</td><td style="text-align: center;">變格米索利地安</td><td style="text-align: center;">正格</td><td style="text-align: center;">G</td><td style="text-align: center;">B-C[<a href="https://zh.wikipedia.org/wiki/教會調式#cite_note-hypomixolydian-6" target="_blank" rel="noopener">注 3]</a></td></tr><tr class="odd"><td style="text-align: center;">9</td><td style="text-align: center;">9</td><td style="text-align: center;">Aeolian</td><td style="text-align: center;">伊奧利安</td><td style="text-align: center;">正格</td><td style="text-align: center;">A</td><td style="text-align: center;">E</td></tr><tr class="even"><td style="text-align: center;">10</td><td style="text-align: center;">10</td><td style="text-align: center;">Hypoaeolian</td><td style="text-align: center;">變格伊奧利安</td><td style="text-align: center;">變格</td><td style="text-align: center;">A</td><td style="text-align: center;">C</td></tr><tr class="odd"><td style="text-align: center;"></td><td style="text-align: center;">11</td><td style="text-align: center;">Locrian</td><td style="text-align: center;">羅克里安</td><td style="text-align: center;">正格</td><td style="text-align: center;">B</td><td style="text-align: center;">F</td></tr><tr class="even"><td style="text-align: center;"></td><td style="text-align: center;">12</td><td style="text-align: center;">Hypolocrian</td><td style="text-align: center;">變格羅克里安</td><td style="text-align: center;">變格</td><td style="text-align: center;">B</td><td style="text-align: center;">D</td></tr><tr class="odd"><td style="text-align: center;">11</td><td style="text-align: center;">13</td><td style="text-align: center;">Ionian</td><td style="text-align: center;">艾奧尼安</td><td style="text-align: center;">正格</td><td style="text-align: center;">C</td><td style="text-align: center;">G</td></tr><tr class="even"><td style="text-align: center;">12</td><td style="text-align: center;">14</td><td style="text-align: center;">Hypoionian</td><td style="text-align: center;">變格艾奧尼安</td><td style="text-align: center;">變格</td><td style="text-align: center;">C</td><td style="text-align: center;">E</td></tr></tbody></table><h4 id="泛音列">泛音列</h4><table><thead><tr class="header"><th style="text-align: center;"><a href="https://zh.wikipedia.org/w/index.php?title=分音&action=edit&redlink=1" target="_blank" rel="noopener">分音</a></th><th style="text-align: center;">与上个<a href="https://zh.wikipedia.org/wiki/十二音" target="_blank" rel="noopener">音</a>的<a href="https://zh.wikipedia.org/wiki/音程" target="_blank" rel="noopener">音程</a></th><th style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/频率" target="_blank" rel="noopener">频率</a>比</th></tr></thead><tbody><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C_(音名)" target="_blank" rel="noopener">C</a>(<a href="https://zh.wikipedia.org/wiki/基音" target="_blank" rel="noopener">基音</a>)</td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/基頻" target="_blank" rel="noopener">基音</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/基頻" target="_blank" rel="noopener">基频</a></td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C_(音名)" target="_blank" rel="noopener">c</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/纯八度" target="_blank" rel="noopener">纯八度</a></td><td style="text-align: center;">2:1</td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G_(音名)" target="_blank" rel="noopener">g</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/纯五度" target="_blank" rel="noopener">纯五度</a></td><td style="text-align: center;">3:2</td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C_(音名)" target="_blank" rel="noopener">c1</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/純四度" target="_blank" rel="noopener">纯四度</a></td><td style="text-align: center;">4:3</td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/E_(音名)" target="_blank" rel="noopener">e1</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/大三度" target="_blank" rel="noopener">大三度</a></td><td style="text-align: center;">5:4</td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G_(音名)" target="_blank" rel="noopener">g1</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/小三度" target="_blank" rel="noopener">小三度</a></td><td style="text-align: center;">6:5</td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/音名" target="_blank" rel="noopener">♭b1</a>(偏低)</td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/小三度" target="_blank" rel="noopener">小三度</a></td><td style="text-align: center;">7:6</td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C_(音名)" target="_blank" rel="noopener">c2</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/大二度" target="_blank" rel="noopener">大二度</a></td><td style="text-align: center;">8:7</td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/D_(音名)" target="_blank" rel="noopener">d2</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/大二度" target="_blank" rel="noopener">大二度</a></td><td style="text-align: center;">9:8</td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/E_(音名)" target="_blank" rel="noopener">e2</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/大二度" target="_blank" rel="noopener">大二度</a></td><td style="text-align: center;">10:9</td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/音名" target="_blank" rel="noopener">#f2</a>(严重偏低,几乎介于f2与bg2之间)</td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/大二度" target="_blank" rel="noopener">大二度</a></td><td style="text-align: center;">11:10</td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G_(音名)" target="_blank" rel="noopener">g2</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/小二度" target="_blank" rel="noopener">小二度</a></td><td style="text-align: center;">12:11</td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/A_(音名)" target="_blank" rel="noopener">a2</a>(偏低,几乎介于ba2和a2之间)</td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/大二度" target="_blank" rel="noopener">大二度</a></td><td style="text-align: center;">13:12</td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/音名" target="_blank" rel="noopener">♭b2</a>(偏低)</td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/小二度" target="_blank" rel="noopener">小二度</a></td><td style="text-align: center;">14:13</td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/B_(音名)" target="_blank" rel="noopener">b2</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/增一度" target="_blank" rel="noopener">增一度</a></td><td style="text-align: center;">15:14</td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C_(音名)" target="_blank" rel="noopener">c3</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/小二度" target="_blank" rel="noopener">小二度</a></td><td style="text-align: center;">16:15</td></tr></tbody></table><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/2/21/Table_of_Harmonics.svg/1920px-Table_of_Harmonics.svg.png" alt=""><figcaption>Table of <a href="https://en.wikipedia.org/wiki/Harmonics#Harmonics_on_stringed_instruments" target="_blank" rel="noopener">harmonics of a stringed instrument</a> with colored dots indicating which positions can be lightly fingered to generate just intervals up to the 7th harmonic. Table of harmonics, indicating in colors on which positions the same overtones occur.</figcaption></figure><h3 id="音程列表">音程列表</h3><ul><li><a href="https://zh.wikipedia.org/w/index.php?title=一度&action=edit&redlink=1" target="_blank" rel="noopener">一度</a></li><li><a href="https://zh.wikipedia.org/wiki/增一度" target="_blank" rel="noopener">增一度</a>(即<a href="https://zh.wikipedia.org/wiki/半音音階" target="_blank" rel="noopener">半音音階</a>中的半音)</li><li><a href="https://zh.wikipedia.org/wiki/減二度" target="_blank" rel="noopener">減二度</a></li><li><a href="https://zh.wikipedia.org/wiki/小二度" target="_blank" rel="noopener">小二度</a></li><li><a href="https://zh.wikipedia.org/wiki/大二度" target="_blank" rel="noopener">大二度</a></li><li><a href="https://zh.wikipedia.org/wiki/增二度" target="_blank" rel="noopener">增二度</a></li><li><a href="https://zh.wikipedia.org/wiki/減三度" target="_blank" rel="noopener">減三度</a></li><li><a href="https://zh.wikipedia.org/wiki/小三度" target="_blank" rel="noopener">小三度</a></li><li><a href="https://zh.wikipedia.org/wiki/大三度" target="_blank" rel="noopener">大三度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=增三度&action=edit&redlink=1" target="_blank" rel="noopener">增三度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=減四度&action=edit&redlink=1" target="_blank" rel="noopener">減四度</a></li><li><a href="https://zh.wikipedia.org/wiki/完全四度" target="_blank" rel="noopener">完全四度</a></li><li><a href="https://zh.wikipedia.org/wiki/增四度" target="_blank" rel="noopener">增四度</a></li><li><a href="https://zh.wikipedia.org/wiki/減五度" target="_blank" rel="noopener">減五度</a></li><li><a href="https://zh.wikipedia.org/wiki/完全五度" target="_blank" rel="noopener">完全五度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=增五度&action=edit&redlink=1" target="_blank" rel="noopener">增五度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=減六度&action=edit&redlink=1" target="_blank" rel="noopener">減六度</a></li><li><a href="https://zh.wikipedia.org/wiki/小六度" target="_blank" rel="noopener">小六度</a></li><li><a href="https://zh.wikipedia.org/wiki/大六度" target="_blank" rel="noopener">大六度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=增六度&action=edit&redlink=1" target="_blank" rel="noopener">增六度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=減七度&action=edit&redlink=1" target="_blank" rel="noopener">減七度</a></li><li><a href="https://zh.wikipedia.org/wiki/小七度" target="_blank" rel="noopener">小七度</a></li><li><a href="https://zh.wikipedia.org/wiki/大七度" target="_blank" rel="noopener">大七度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=增七度&action=edit&redlink=1" target="_blank" rel="noopener">增七度</a></li><li><a href="https://zh.wikipedia.org/w/index.php?title=減八度&action=edit&redlink=1" target="_blank" rel="noopener">減八度</a></li><li><a href="https://zh.wikipedia.org/wiki/純八度" target="_blank" rel="noopener">純八度</a>或八度</li><li><a href="https://zh.wikipedia.org/w/index.php?title=增八度&action=edit&redlink=1" target="_blank" rel="noopener">增八度</a></li></ul><h4 id="各大調中的屬音">各大調中的屬音</h4><table><thead><tr class="header"><th style="text-align: center;">調</th><th style="text-align: center;">屬音</th><th style="text-align: center;">調</th><th style="text-align: center;">屬音</th><th style="text-align: center;">調</th><th style="text-align: center;">屬音</th></tr></thead><tbody><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/降C大調" target="_blank" rel="noopener">降C大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G♭_(音名)" target="_blank" rel="noopener">G♭</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C大調" target="_blank" rel="noopener">C大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G_(音名)" target="_blank" rel="noopener">G</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/升C大調" target="_blank" rel="noopener">升C大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G♯_(音名)" target="_blank" rel="noopener">G♯</a></td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/降D大調" target="_blank" rel="noopener">降D大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/A♭_(音名)" target="_blank" rel="noopener">A♭</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/D大調" target="_blank" rel="noopener">D大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/A_(音名)" target="_blank" rel="noopener">A</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/升D大調" target="_blank" rel="noopener">升D大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/A♯_(音名)" target="_blank" rel="noopener">A♯</a></td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/降E大調" target="_blank" rel="noopener">降E大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/B♭_(音名)" target="_blank" rel="noopener">B♭</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/E大調" target="_blank" rel="noopener">E大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/B_(音名)" target="_blank" rel="noopener">B</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/升E大調" target="_blank" rel="noopener">升E大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/B♯_(音名)" target="_blank" rel="noopener">B♯</a></td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/降F大調" target="_blank" rel="noopener">降F大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C♭_(音名)" target="_blank" rel="noopener">C♭</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/F大調" target="_blank" rel="noopener">F大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C_(音名)" target="_blank" rel="noopener">C</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/升F大調" target="_blank" rel="noopener">升F大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/C♯_(音名)" target="_blank" rel="noopener">C♯</a></td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/降G大調" target="_blank" rel="noopener">降G大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/D♭_(音名)" target="_blank" rel="noopener">D♭</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G大調" target="_blank" rel="noopener">G大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/D_(音名)" target="_blank" rel="noopener">D</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/升G大調" target="_blank" rel="noopener">升G大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/D♯_(音名)" target="_blank" rel="noopener">D♯</a></td></tr><tr class="even"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/降A大調" target="_blank" rel="noopener">降A大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/E♭_(音名)" target="_blank" rel="noopener">E♭</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/A大調" target="_blank" rel="noopener">A大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/E_(音名)" target="_blank" rel="noopener">E</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/升A大調" target="_blank" rel="noopener">升A大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/E♯_(音名)" target="_blank" rel="noopener">E♯</a></td></tr><tr class="odd"><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/降B大調" target="_blank" rel="noopener">降B大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/F_(音名)" target="_blank" rel="noopener">F</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/B大調" target="_blank" rel="noopener">B大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/F♯_(音名)" target="_blank" rel="noopener">F♯</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/升B大調" target="_blank" rel="noopener">升B大調</a></td><td style="text-align: center;"><a href="https://zh.wikipedia.org/wiki/G_(音名)" target="_blank" rel="noopener">F♯♯</a></td></tr></tbody></table><p>further reading:七和弦 减七和弦 挂留和弦 九、七和弦</p><h2 id="和声篇"><a href="https://www.gcores.com/radios/94460" target="_blank" rel="noopener">和声篇</a></h2><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/d/d1/Major_triad.svg/250px-Major_triad.svg.png" alt=""><figcaption><a href="https://zh.wikipedia.org/wiki/和弦#.E4.B8.89.E5.92.8C.E5.BC.A6" target="_blank" rel="noopener">大三和弦</a>由三個不同頻率的聲音組成,而這三個頻率的比例接近6:5:4.</figcaption></figure><h2 id="节奏篇"><a href="https://www.gcores.com/radios/96660" target="_blank" rel="noopener">节奏篇</a></h2><p>当拍速超过一定速度(一般认为是300BPM)之后人耳就无法识别出节奏的变化,只能听出连成一段的声音。</p><p>关于节奏的学术文章有很多,资料来源这篇研究的是拍速的底线是多少,最终得出的结果是33BPM。</p><figure><img src="/2019/06/19/尝试用最简单的方式讲乐理知识/Beatmatching.png" alt=""><figcaption>Beatmatching</figcaption></figure><figure><img src="https://en.wikipedia.org/wiki/File:Metric_levels.svg" alt=""><figcaption>节拍的层级:中间是拍层,上面是分拍层,下面是合拍层。</figcaption></figure>]]></content>
<summary type="html">
<h2 id="调式篇"><a href="https://www.gcores.com/radios/93658" target="_blank" rel="noopener">调式篇</a></h2>
<h3 id="音律">音律</h3>
<ul>
<li><p>十二平均律
</summary>
</entry>
<entry>
<title>Modern Physics</title>
<link href="https://github-9233.github.io/2019/06/17/Modern-Physics/"/>
<id>https://github-9233.github.io/2019/06/17/Modern-Physics/</id>
<published>2019-06-17T02:03:33.000Z</published>
<updated>2019-08-14T09:20:35.360Z</updated>
<content type="html"><![CDATA[<h2 id="relativity">37 RELATIVITY</h2><h3 id="invariance-of-physical-laws">37.1 Invariance of Physical laws</h3><p><em>Einstein’s First Postulate (principle of relativity)</em>: The laws of physics are the same in every inertial frame of reference.</p><p><em>Einstein’s Second Postulate</em>: The speed of light in vacuum is the same in all inertial frames of reference and is independent of the motion of the source.</p><h3 id="relativity-of-simultaneity">37.2 Relativity of Simultaneity</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/9/96/Einstein_train_relativity_of_simultaneity.png/800px-Einstein_train_relativity_of_simultaneity.png" alt=""><figcaption>Einstein imagined a stationary observer who witnessed two lightning bolts simultaneously striking both ends of a moving train. He concluded that an observer standing on the train would see the bolts striking at different times.</figcaption></figure><h3 id="relativity-of-time-interval">37.3 <u>Relativity of Time Interval</u></h3><p><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/8e/Time-dilation-002-mod.svg/1920px-Time-dilation-002-mod.svg.png" alt="Left: Observer at rest measures time 2L/c between co-local events of light signal generation at A and arrival at A. Right: Events according to an observer moving to the left of the setup: bottom mirror A when signal is generated at time t'=0, top mirror B when signal gets reflected at time t'=D/c, bottom mirror A when signal returns at time t'=2D/c"> <span class="math display">\[Time\ dilation:\underbrace{\Delta t}_{Time\ interval\ between\ same\ events\\ measured\ in\ second\ frame\ of\ reference}=\frac{\overbrace{\Delta t_{0}}^{Proper\ time\ between\ two\ events\ (measured\ in\ rest\ frame)}}{\sqrt{1-u^{2} / c^{2}}}\]</span></p><h3 id="relativity-of-length">37.4 Relativity of Length</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/b/b7/Lorentzkontraktion.png" alt=""><figcaption><em>Length contraction</em>: Three blue rods are at rest in S, and three red rods in S'. At the instant when the left ends of A and D attain the same position on the axis of x, the lengths of the rods shall be compared. In S the simultaneous positions of the left side of A and the right side of C are more distant than those of D and F. While in S' the simultaneous positions of the left side of D and the right side of F are more distant than those of A and C.</figcaption></figure><h3 id="the-lorentz-transformations">37.5 The Lorentz Transformations</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/9/90/Standard_conf.png/300px-Standard_conf.png" alt=""><figcaption>Standard configuration of coordinate systems for Galilean transformations.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/9/98/Lorentz_boost_x_direction_standard_configuration.svg/400px-Lorentz_boost_x_direction_standard_configuration.svg.png" alt=""><figcaption>The spacetime coordinates of an event, as measured by each observer in their inertial reference frame (in standard configuration) are shown in the speech bubbles. <strong>Top:</strong> frame <em>F</em>′ moves at velocity <em>v</em> along the <em>x</em>-axis of frame <em>F</em>. <strong>Bottom:</strong> frame <em>F</em> moves at velocity −<em>v</em> along the <em>x</em>′-axis of frame <em>F</em>′</figcaption></figure><h3 id="the-doppler-effect-for-electromagnetic-waves">37.6 The Doppler Effect For Electromagnetic Waves</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/9/90/Dopplerfrequenz.gif" alt=""><figcaption>An animation illustrating how the Doppler effect causes a car engine or siren to sound higher in pitch when it is approaching than when it is receding. The pink circles represent sound waves.</figcaption></figure><h3 id="relativistic-momentum">37.7 Relativistic Momentum</h3><p><em>Relativistic momentum</em> <span class="math display">\[\vec{p}=\frac{m \vec{v}}{\sqrt{1-v^{2} / c^{2}}}\]</span></p><h3 id="relativistic-work-and-energy">37.8 Relativistic Work And Energy</h3><p><em>Relativistic kinetic energy</em> <span class="math display">\[K=\frac{m c^{2}}{\sqrt{1-v^{2} / c^{2}}}-m c^{2}=(\gamma-1) m c^{2}\]</span> <em>Total energyof a particle</em> <span class="math display">\[E=K+m c^{2}=\frac{m c^{2}}{\sqrt{1-v^{2} / c^{2}}}=\gamma m c^{2}\]</span> <em>Total energy, rest energy, and momentum</em> <span class="math display">\[\overbrace{E^{2}}^{Total\ energy}=\overbrace{\left(m c^{2}\right)^{2}}^{Rest\ energy}+\overbrace{\left(p_{2}\right)^{2}}^{Magnitude\ of\\momentum}\]</span></p><h3 id="newtonian-mechanics-and-relativity">37.9 Newtonian Mechanics And Relativity</h3><p><em>Correspondence principle</em></p><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/4/46/Hamilton_analogy_zh-hans.svg/320px-Hamilton_analogy_zh-hans.svg.png" alt=""><figcaption>薛定諤將哈密頓類比延伸至量子力學與波動光學之間。</figcaption></figure><p><em><a href="https://en.wikipedia.org/wiki/Tests_of_general_relativity" target="_blank" rel="noopener">Tests of general relativity</a></em></p><ol type="1"><li><p>One test has to do with understanding the rotation of the axes of the planet Mercury’s elliptical orbit, called the precession of the perihelion. (The perihelion is the point of closest approach to the sun.)</p><figure><img src="https://upload.wikimedia.org/wikipedia/commons/8/89/Precessing_Kepler_orbit_280frames_e0.6_smaller.gif" alt=""><figcaption>Each <a href="https://en.wikipedia.org/wiki/Planet" target="_blank" rel="noopener">planet</a> orbiting the <a href="https://en.wikipedia.org/wiki/Sun" target="_blank" rel="noopener">Sun</a> follows an <a href="https://en.wikipedia.org/wiki/Elliptic_orbit" target="_blank" rel="noopener">elliptic orbit</a>that gradually rotates over time (apsidal precession). This figure illustrates positive apsidal precession (advance of the perihelion), with the orbital axis turning in the same direction as the planet's orbital motion. The <a href="https://en.wikipedia.org/wiki/Orbital_eccentricity" target="_blank" rel="noopener">eccentricity</a> of this ellipse and the precession rate of the orbit are exaggerated for visualization. Most orbits in the <a href="https://en.wikipedia.org/wiki/Solar_System" target="_blank" rel="noopener">Solar System</a> have a much lower eccentricity and precess at a much slower rate, making them nearly <a href="https://en.wikipedia.org/wiki/Circular_orbit" target="_blank" rel="noopener">circular</a> and <a href="https://en.wikipedia.org/wiki/Stationary_orbit" target="_blank" rel="noopener">stationary</a>.</figcaption></figure></li><li><p>A second test concerns the apparent bending of light rays from distant stars when they pass near the sun.</p><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/d/da/1919_eclipse_negative.jpg/150px-1919_eclipse_negative.jpg" alt=""><figcaption>One of <a href="https://en.wikipedia.org/wiki/Sir_Arthur_Eddington" target="_blank" rel="noopener">Eddington</a>'s photographs of the <a href="https://en.wikipedia.org/wiki/Eddington_experiment" target="_blank" rel="noopener">1919 solar eclipse experiment</a>, presented in his 1920 paper announcing its success</figcaption></figure></li><li><p>The third test is the gravitational red shift, the increase in wavelength of light pro-ceeding outward from a massive source.</p><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/6/6d/Gravitational_red-shifting2.png/150px-Gravitational_red-shifting2.png" alt=""><figcaption>The gravitational redshift of a light wave as it moves upwards against a gravitational field (caused by the yellow star below).</figcaption></figure></li></ol><h2 id="photons-light-waves-behaving-as-particles">38 PHOTONS: LIGHT WAVES BEHAVING AS PARTICLES</h2><p><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/f/f5/Photoelectric_effect.svg/256px-Photoelectric_effect.svg.png"></p><ol type="1"><li>Low-energy phenomena: Photoelectric effect</li><li>Mid-energy phenomena: Thomson scattering/Compton scattering</li><li>High-energy phenomena: Pair production</li></ol><h3 id="light-absorbed-as-photons-the-photoelectric-effect">38.1 Light Absorbed As Photons: The Photoelectric Effect</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/e/e0/Versuch_zum_Fotoeffekt.png/220px-Versuch_zum_Fotoeffekt.png" alt=""><figcaption>Schematic of experimental apparatus to demonstrate the photoelectric effect. The filter passes light of certain wavelengths from the lamp at left. The light strikes the curved electrode, and electrons are emitted. The adjustable voltage can be increased until the current stops flowing. This "stopping voltage" is a function only of the electrode material and the frequency of the incident light, and is not affected by the intensity of the light.</figcaption></figure><h3 id="light-emitted-as-photons-x-ray-production">38.2 Light Emitted As Photons: X-Ray Production</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/1/1e/Bremsstrahlung.svg/200px-Bremsstrahlung.svg.png" alt=""><figcaption>Bremsstrahlung produced by a high-energy electron deflected in the electric field of an atomic nucleus.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/7/76/Ct-internals.jpg/220px-Ct-internals.jpg" alt=""><figcaption>CT scanner with cover removed to show internal components. Legend: T: X-ray tube D: X-ray detectors X: X-ray beam R: Gantry rotation</figcaption></figure><h3 id="light-scattered-as-photons-compton-scattering-and-pair-production">38.3 Light Scattered As Photons: Compton Scattering And Pair Production</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/e/e3/Compton-scattering.svg/320px-Compton-scattering.svg.png" alt=""><figcaption>A photon of wavelength λ comes in from the left, collides with a target at rest, and a new photon of wavelength λ′ emerges at an angle θ. The target recoils, carrying away an angle-dependent amount of the incident energy.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/6/64/Pair_production_Cartoon.gif" alt=""><figcaption>Diagram showing the process of electron–positron pair production. In reality the produced pair are nearly collinear.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/b/b2/Annihilation.png/480px-Annihilation.png" alt=""><figcaption>Naturally occurring electron-positron annihilation as a result of beta plus decay</figcaption></figure><h3 id="waveparticle-duality-probability-and-uncertainty">38.4 Wave–Particle Duality, Probability, And Uncertainty</h3><h4 id="uncertainty-principle">Uncertainty principle</h4><ul><li>Heisenberguncertainty principle forposition and momentum:</li></ul><p><span class="math display">\[\Delta x \Delta p_{x} \geq \hbar / 2\]</span></p><ul><li>Heisenberguncertainty principle forenergy and time:</li></ul><p><span class="math display">\[\Delta t \Delta E \geq \hbar / 2\]</span></p><h2 id="particles-behaving-as-waves">39 PARTICLES BEHAVING AS WAVES</h2><h3 id="electron-waves">39.1 Electron Waves</h3><p>Electron diffraction</p><p>Electron microscope</p><h3 id="the-nuclear-atom-and-atomic-spectra">39.2 The Nuclear Atom And Atomic Spectra</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/1/1a/Simple_spectroscope.jpg/300px-Simple_spectroscope.jpg" alt=""><figcaption>Continuous spectrum of an <a href="https://en.wikipedia.org/wiki/Incandescent_light_bulb" target="_blank" rel="noopener">incandescent lamp</a> (mid) and discrete spectrum lines of a <a href="https://en.wikipedia.org/wiki/Compact_fluorescent_lamp" target="_blank" rel="noopener">fluorescent lamp</a>(bottom)</figcaption></figure><h3 id="energy-levels-and-the-bohr-model-of-the-atom">39.3 Energy Levels And The Bohr Model Of The Atom</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/89/Jablonski_Diagram_of_Fluorescence_Only.png/220px-Jablonski_Diagram_of_Fluorescence_Only.png" alt=""><figcaption><a href="https://en.wikipedia.org/wiki/Jablonski_diagram" target="_blank" rel="noopener">Jablonski diagram</a>. After an electron absorbs a high-energy photon the system is excited electronically and vibrationally. The system relaxes vibrationally, and eventually fluoresces at a longer wavelength.</figcaption></figure><h3 id="the-laser">39.4 The Laser</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/0/09/Stimulated_Emission.svg/550px-Stimulated_Emission.svg.png" alt=""><figcaption>stimulated emission</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/en/e/e3/Population-inversion-4level.png" alt=""><figcaption>population inversion</figcaption></figure><h3 id="continuous-spectra">39.5 Continuous Spectra</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/d/d8/Black_body_realization.svg/220px-Black_body_realization.svg.png" alt=""><figcaption>An approximate realization of a black body as a tiny hole in an insulated enclosure</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/1/19/Black_body.svg/512px-Black_body.svg.png" alt=""><figcaption>As the temperature of a black body decreases, its intensity also decreases and its peak moves to longer wavelengths. Shown for comparison is the classical <a href="https://en.wikipedia.org/wiki/Rayleigh–Jeans_law" target="_blank" rel="noopener">Rayleigh–Jeans law</a> and its <a href="https://en.wikipedia.org/wiki/Ultraviolet_catastrophe" target="_blank" rel="noopener">ultraviolet catastrophe</a>.</figcaption></figure><p>Planck radiation law <span class="math display">\[I(\lambda)=\frac{2 \pi h c^{2}}{\lambda^{5}\left(e^{h c / \lambda k T}-1\right)}\]</span></p><h3 id="the-uncertainty-principle-revisited">39.6 The Uncertainty Principle Revisited</h3><h2 id="quantum-mechanics-i-wave-functions">40 QUANTUM MECHANICS I: WAVE FUNCTIONS</h2><h3 id="wave-functions-and-the-one-dimensional-schrödinger-equation">40.1 Wave Functions And The One-Dimensional Schrödinger Equation</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/2/21/Propagation_of_a_de_broglie_wave.svg/290px-Propagation_of_a_de_broglie_wave.svg.png" alt=""><figcaption>Propagation of <a href="https://en.wikipedia.org/wiki/Matter_wave" target="_blank" rel="noopener">de Broglie waves</a> in 1d - real part of the <a href="https://en.wikipedia.org/wiki/Complex_number" target="_blank" rel="noopener">complex</a> amplitude is blue, imaginary part is green. The probability (shown as the colour <a href="https://en.wikipedia.org/wiki/Opacity_(optics)" target="_blank" rel="noopener">opacity</a>) of finding the particle at a given point <em>x</em> is spread out like a waveform, there is no definite position of the particle. As the amplitude increases above zero the <a href="https://en.wikipedia.org/wiki/Curvature" target="_blank" rel="noopener">curvature</a> decreases, so the decreases again, and vice versa - the result is an alternating amplitude: a wave. Top: <a href="https://en.wikipedia.org/wiki/Plane_wave" target="_blank" rel="noopener">Plane wave</a>. Bottom: <a href="https://en.wikipedia.org/wiki/Wave_packet" target="_blank" rel="noopener">Wave packet</a>.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/d/db/Sequential_superposition_of_plane_waves.gif" alt=""><figcaption>The superposition of several plane waves to form a wave packet. This wave packet becomes increasingly localized with the addition of many waves. The Fourier transform is a mathematical operation that separates a wave packet into its individual plane waves. Note that the waves shown here are real for illustrative purposes only, whereas in quantum mechanics the wave function is generally complex.</figcaption></figure><p>General one-dimensional Schrödinger equation: <span class="math display">\[-\frac{\hbar^{2}}{2 m} \frac{\partial^{2} \Psi(x, t)}{\partial x^{2}}+U(x) \Psi(x, t)=i \hbar \frac{\partial \Psi(x, t)}{\partial t}\]</span> Time-independent one-dimensional Schrödinger equation: <span class="math display">\[-\frac{\hbar^{2}}{2 m} \frac{d^{2} \psi(x)}{d x^{2}}+U(x) \psi(x)=E \psi(x)\]</span> <img src="https://upload.wikimedia.org/wikipedia/commons/9/90/QuantumHarmonicOscillatorAnimation.gif" alt="A harmonic oscillator in classical mechanics (A–B) and quantum mechanics (C–H). In (A–B), a ball, attached to a spring, oscillates back and forth. (C–H) are six solutions to the Schrödinger Equation for this situation. The horizontal axis is position, the vertical axis is the real part (blue) or imaginary part (red) of the wavefunction. (C,D,E,F), but not (G,H), are stationary states, or standing waves. The standing-wave oscillation frequency, times Planck's constant, is the energy of the state."></p><h3 id="particle-in-a-box">40.2 Particle in a Box</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/1/13/Infinite_potential_well-en.svg/220px-Infinite_potential_well-en.svg.png" alt=""><figcaption>The barriers outside a one-dimensional box have infinitely large potential, while the interior of the box has a constant, zero potential.</figcaption></figure><h3 id="potential-wells">40.3 Potential Wells</h3><p>Finite potential well</p><h3 id="potential-barriers-and-tunneling">40.4 Potential Barriers and tunneling</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/1/1d/TunnelEffektKling1.png/300px-TunnelEffektKling1.png" alt=""><figcaption>Quantum tunnelling through a barrier. The energy of the tunnelled particle is the same but the probability amplitude is decreased.</figcaption></figure><p>applications of tunneling: tunnel diode & Josephson junction in quantum computing</p><p>Quantum dot Quantum wire</p><h3 id="the-harmonic-oscillator">40.5 The Harmonic Oscillator</h3><p>Hermite function</p><figure><img src="/2019/06/17/Modern-Physics/The%20first%20four%20wave%20functions%20for%20the%20harmonic%20oscillator..png" alt=""><figcaption>The first four wave functions for the harmonic oscillator.</figcaption></figure><figure><img src="/2019/06/17/Modern-Physics/Probability%20distribution%20functions%20.png" alt=""><figcaption>Probability distribution functions</figcaption></figure><h3 id="measurement-in-quantum-mechanics">40.6 Measurement In Quantum Mechanics</h3><p><strong>Measuring a physical property of a system can change the wave function of that system.</strong></p><ul><li><p>wave-function collapse</p></li><li><p>many-worlds interpretation</p><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/c/c8/Schroedingers_cat_film.svg/350px-Schroedingers_cat_film.svg.png" alt=""><figcaption>The quantum-mechanical "<a href="https://en.wikipedia.org/wiki/Schrödinger's_cat" target="_blank" rel="noopener">Schrödinger's cat</a>" theorem according to the many-worlds interpretation. In this interpretation, every event is a branch point; the cat is both alive and dead, even before the box is opened, but the "alive" and "dead" cats are in different branches of the universe, both of which are equally real, but which do not interact with each other.</figcaption></figure></li></ul><h2 id="quantum-mechanics-ii-atomic-structure">41 QUANTUM MECHANICS II: ATOMIC STRUCTURE</h2><h3 id="the-schrödinger-equation-in-three-dimensions">41.1 The Schrödinger Equation In Three Dimensions</h3><p>Time-independent three-dimensional Schrödinger equation: <span class="math display">\[\begin{aligned}-\frac{\hbar^{2}}{2 m}\left(\frac{\partial^{2} \psi(x, y, z)}{\partial x^{2}}\right.&+\frac{\partial^{2} \psi(x, y, z)}{\partial y^{2}}+\frac{\partial^{2} \psi(x, y, z)}{\partial z^{2}} ) +U(x, y, z) \psi(x, y, z)=E \psi(x, y, z) \end{aligned}\]</span></p><h3 id="particle-in-a-three-dimensional-box">41.2 Particle In A Three-Dimensional Box</h3><p><span class="math display">\[E_{n_{X}, n_{Y}, n_{Z}}=\frac{\left(n_{X}^{2}+n_{Y}^{2}+n_{Z}^{2}\right) \pi^{2} \hbar^{2}}{2 m L^{2}}\]</span></p><h3 id="the-hydrogen-atom">41.3 The Hydrogen Atom</h3><p>The Schrödinger equation for the hydrogen atom: <span class="math display">\[\begin{aligned}-\frac{\hbar^{2}}{2 m_{\mathrm{r}} r^{2}} \frac{d}{d r}\left(r^{2} \frac{d R(r)}{d r}\right)+\left(\frac{\hbar^{2} l(l+1)}{2 m_{\mathrm{r}} r^{2}}+U(r)\right) R(r) &=E R(r) \\ \frac{1}{\sin \theta} \frac{d}{d \theta}\left(\sin \theta \frac{d \Theta(\theta)}{d \theta}\right)+\left(l(l+1)-\frac{m_{l}^{2}}{\sin ^{2} \theta}\right) \Theta(\theta) &=0 \\ \frac{d^{2} \Phi(\phi)}{d \phi^{2}}+m_{l}^{2} \Phi(\phi) &=0 \end{aligned}\]</span> Energy levels of hydrogen:<strong>principal quantum number</strong> <span class="math display">\[E_{n}=-\frac{1}{\left(4 \pi \epsilon_{0}\right)^{2}} \frac{m_{\mathrm{r}} e^{4}}{2 n^{2} \hbar^{2}}=-\frac{13.60 \mathrm{eV}}{n^{2}}\]</span> Magnitude of orbital angular momentum, hydrogen atom:<strong>orbital quantum number</strong> <span class="math display">\[L=\sqrt{l(l+1) \hbar} \quad(l=0,1,2, \ldots, n-1)\]</span> z-component of orbital angular momentum, hydrogen atom:<strong>magnetic quantum number</strong> <span class="math display">\[L_{z}=m_{l} \hbar\left(m_{l}=0, \pm 1, \pm 2, \ldots, \pm l\right)\]</span> <img src="https://upload.wikimedia.org/wikipedia/commons/thumb/e/e7/Hydrogen_Density_Plots.png/550px-Hydrogen_Density_Plots.png" alt="Wavefunctions of a hydrogen atom, showing the probability of finding the electron in the space around the nucleus. Each stationary state defines a specific energy level of the atom."></p><h3 id="the-zeeman-effect">41.4 The Zeeman Effect</h3><p>normal Zeeman effect</p><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/9/9e/Sunzeeman1919.png/200px-Sunzeeman1919.png" alt=""><figcaption>Zeeman effect on a sunspot spectral line</figcaption></figure><p>Selection rule: orbital quantum number <span class="math inline">\(l\)</span> must change by 1 when a photon is emitted</p><h3 id="electron-spin">41.5 Electron Spin</h3><p>anomalous Zeeman effect</p><p>spin-orbit coupling</p><ul><li><p><strong>fine structure</strong> <a href="https://en.wikipedia.org/wiki/Electron_spin" target="_blank" rel="noopener">electron spin</a> and <a href="https://en.wikipedia.org/wiki/Special_relativity" target="_blank" rel="noopener">relativistic corrections</a></p><p>Energy levels of hydrogen, including fine structure: <em>fine-structure constant</em> <span class="math inline">\(\alpha\)</span></p></li></ul><p><span class="math display">\[E_{n, j}=-\frac{13.60 \mathrm{eV}}{n^{2}}\left[1+\frac{\alpha^{2}}{n^{2}}\left(\frac{n}{j+\frac{1}{2}}-\frac{3}{4}\right)\right]\]</span></p><ul><li><strong>hyperfine structure</strong> interaction between the state of the nucleus and the state of the electron clouds.</li></ul><h3 id="many-electron-atoms-and-the-exclusion-principle">41.6 Many-Electron Atoms And The Exclusion Principle</h3><ul><li><p>central-field approximation</p></li><li><p>exclusion principle</p></li><li><p>screening</p><p>Energy levels of an electron with screening:</p></li></ul><p><span class="math display">\[E_{n}=-\frac{Z_{\mathrm{eff}}^{2}}{n^{2}}(13.6\mathrm{eV})\]</span></p><h3 id="x-ray-spectra">41.7 X-Ray Spectra</h3><p>Mosele y’s law: <span class="math inline">\(f\)</span> Frequency of <span class="math inline">\(K_{\alpha}\)</span> line in characteristicx-ray spectrum of an element <span class="math display">\[f=\left(2.48 \times 10^{15} \mathrm{Hz}\right)(Z-1)^{2}\]</span></p><ul><li><em>outer shells</em>: optical spectra</li><li><em>inner shells</em>: Characteristic x rays</li></ul><p>x-ray absorption spectra absorption edges</p><h3 id="quantum-entanglement"><u>41.8 Quantum Entanglement</u></h3><figure><img src="/2019/06/17/Modern-Physics/Quantum%20Entanglement.png" alt=""><figcaption>Quantum Entanglement</figcaption></figure><h2 id="molecules-and-condensed-matter">42 MOLECULES AND CONDENSED MATTER</h2><h3 id="types-of-molecular-bonds">42.1 Types Of Molecular Bonds</h3><h4 id="strong-bonds-typical-bond-energies-of-1-to-5-ev">strong bonds (typical bond energies of 1 to 5 eV)</h4><h5 id="ionic-bonds">Ionic Bonds</h5><p>ionization energy: <span class="math display">\[\mathrm{X}+\text { energy } \rightarrow \mathrm{X}^{+}+\mathrm{e}^{-}\]</span></p><p>electron affinity: <span class="math display">\[X+e^{-} \rightarrow X^{-}+\text { energy }\]</span></p><p>binding energy: energy needed to dissociate molecule into separate neutral atoms</p><h5 id="covalent-bonds">Covalent Bonds</h5><p>hybrid wave function</p><h4 id="weaker-bonds-typical-energies-are-0.1-ev-or-less">weaker bonds (typical energies are 0.1 eV or less)</h4><h5 id="van-der-waals-bonds">van der Waals Bonds</h5><p>from dipole–dipole interactions</p><h5 id="hydrogen-bonds">Hydrogen Bonds</h5><h3 id="molecular-spectra">42.2 Molecular Spectra</h3><h5 id="rotational-energy-levels">Rotational energy levels</h5><p>allowed transitions between rotational states must satisfy the same selection rule <span class="math inline">\(l\)</span> must change by exactly one unit; that is, <span class="math inline">\(\Delta l = \pm 1\)</span></p><h5 id="vibrational-energy-levels">Vibrational energy levels</h5><h5 id="rotation-and-vibration-combined">Rotation and vibration combined</h5><table><thead><tr class="header"><th>vibrational</th><th>rotational</th></tr></thead><tbody><tr class="odd"><td>each individual band in a spectrum</td><td>each individual line in a band</td></tr><tr class="even"><td>n-value</td><td>l-value</td></tr></tbody></table><p>molecules states = excited states of the electrons + rotational + vibrational states</p><p>When there is a transition between <strong>electronic states</strong>, the <span class="math inline">\(\Delta n = \pm 1\)</span> selection rule for the vibrational levels no longer holds.</p><h5 id="complex-molecules">Complex Molecules</h5><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/81/Co2_vibrations.svg/640px-Co2_vibrations.svg.png" alt=""><figcaption><a href="https://en.wikipedia.org/wiki/Infrared_spectroscopy#Number_of_vibrational_modes" target="_blank" rel="noopener">Stretching and bending oscillations</a> of the CO2 carbon dioxide molecule. Upper left: symmetric stretching. Upper right: antisymmetric stretching. Lower line: degenerate pair of bending modes.</figcaption></figure><h3 id="structure-of-solids">42.3 Structure of Solids</h3><h5 id="crystal-lattices-and-structures">Crystal Lattices and Structures</h5><h5 id="bonding-in-solids">Bonding in Solids</h5><h5 id="types-of-crystals">Types of Crystals</h5><h3 id="energy-bands">42.4 Energy Bands</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/e/ef/Solid_state_electronic_band_structure.svg/640px-Solid_state_electronic_band_structure.svg.png" alt=""><figcaption>Showing how electronic band structure comes about by the hypothetical example of a large number of carbon atoms being brought together to form a diamond crystal. The graph <em>(right)</em> shows the energy levels as a function of the spacing between atoms. When the atoms are far apart <em>(right side of graph)</em> each atom has valence atomic orbitals p and s which have the same energy. However, when the atoms come closer together their orbitals begin to overlap. Due to the Pauli Exclusion Principle each atomic orbital splits into <em>N</em> molecular orbitals each with a different energy, where <em>N</em> is the number of atoms in the crystal. Since <em>N</em> is such a large number, adjacent orbitals are extremely close together in energy so the orbitals can be considered a continuous energy band. <em>a</em> is the atomic spacing in an actual crystal of diamond. At that spacing the orbitals form two bands, called the valence and conduction bands, with a 5.5 eV band gap between them.</figcaption></figure><h5 id="insulators-semiconductors-and-conductors">Insulators, Semiconductors, and Conductors</h5><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/1/12/Semiconductor_band_structure_%28lots_of_bands_2%29.svg/320px-Semiconductor_band_structure_%28lots_of_bands_2%29.svg.png" alt=""><figcaption><em>Semiconductor band structure</em></figcaption></figure><h3 id="free-electron-model-of-metals">42.5 Free-Electron Model Of Metals</h3><h5 id="free-electron-model">free-electron model</h5><h5 id="density-of-states">density of states</h5><h5 id="fermidirac-distribution-vs-maxwellboltzmann-distribution">fermi–dirac distribution vs (Maxwell–Boltzmann distribution)</h5><ul><li><p>exclusion principle</p></li><li><p>indistinguishability</p></li></ul><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/c/c9/FD_e_mu.svg/301px-FD_e_mu.svg.png" alt=""><figcaption><strong>Energy dependence.</strong></figcaption></figure><h5 id="electron-concentration-and-fermi-energy">electron concentration and fermi energy</h5><h5 id="average-free-electron-energy">average free-electron energy</h5><h3 id="semiconductors">42.6 Semiconductors</h3><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/8b/N-Type_Semiconductor_Bands.svg/200px-N-Type_Semiconductor_Bands.svg.png" alt=""><figcaption>Band structure of an n-type semiconductor. Dark circles in the conduction band are electrons and light circles in the valence band are holes. The image shows that the electrons are the majority charge carrier.</figcaption></figure><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/3/37/P-Type_Semiconductor_Bands.svg/200px-P-Type_Semiconductor_Bands.svg.png" alt=""><figcaption>Band structure of a p-type semiconductor. Dark circles in the conduction band are electrons and light circles in the valence band are holes. The image shows that the holes are the majority charge carrier</figcaption></figure><h3 id="semiconductor-devices">42.7 Semiconductor Devices</h3><p>1451</p><h5 id="the-p-n-junction">The p-n Junction</h5><h5 id="currents-through-a-p-n-junction">currents Through a p-n Junction</h5><h5 id="semiconductor-devices-and-light">semiconductor devices and light</h5><h5 id="transistors">Transistors</h5><h5 id="integrated-circuits">Integrated circuits</h5><h3 id="superconductivity">42.8 Superconductivity</h3>]]></content>
<summary type="html">
<h2 id="relativity">37 RELATIVITY</h2>
<h3 id="invariance-of-physical-laws">37.1 Invariance of Physical laws</h3>
<p><em>Einstein’s First Po
</summary>
</entry>
<entry>
<title>Optical Aberrations</title>
<link href="https://github-9233.github.io/2019/04/23/Optical-Aberrations/"/>
<id>https://github-9233.github.io/2019/04/23/Optical-Aberrations/</id>
<published>2019-04-23T01:27:23.000Z</published>
<updated>2019-04-23T02:41:12.541Z</updated>
<content type="html"><![CDATA[<h2 id="chromatic-aberrations">CHROMATIC ABERRATIONS</h2><p>Transverse chromatic aberration (TCA)</p><p>Longitudinal chromatic aberration (LCA)</p><h2 id="monochromatic-aberrations">MONOCHROMATIC ABERRATIONS</h2><p><span class="math display">\[W = \sum_{l + k +m = 0} ^{\infty} { \bigg[ W_{klm}\cdot H^k \cdot \rho^l \cdot \cos ^m \left( \theta \right) \bigg]}\]</span></p><table><thead><tr class="header"><th style="text-align: left;"><strong>Aberration Name</strong></th><th style="text-align: left;"><strong>Wavefront Coefficient</strong></th><th style="text-align: left;"><strong>Equation</strong></th></tr></thead><tbody><tr class="odd"><td style="text-align: left;">Tilt</td><td style="text-align: left;">W111</td><td style="text-align: left;">W111Hρcos(θ)</td></tr><tr class="even"><td style="text-align: left;">Defocus</td><td style="text-align: left;">W020</td><td style="text-align: left;">W020ρ2</td></tr><tr class="odd"><td style="text-align: left;">Spherical</td><td style="text-align: left;">W040</td><td style="text-align: left;">W040ρ4</td></tr><tr class="even"><td style="text-align: left;">Coma</td><td style="text-align: left;">W131</td><td style="text-align: left;">W131Hρ3cos(θ)</td></tr><tr class="odd"><td style="text-align: left;">Astigmatism</td><td style="text-align: left;">W222</td><td style="text-align: left;">W222H2ρ2cos2(θ)</td></tr><tr class="even"><td style="text-align: left;">Field Curvature</td><td style="text-align: left;">W220</td><td style="text-align: left;">W220H2ρ2</td></tr><tr class="odd"><td style="text-align: left;">Disortion</td><td style="text-align: left;">W311</td><td style="text-align: left;">W311H3ρcos(θ)</td></tr></tbody></table><p><strong>Table 1:</strong> Common Third Order Optical Aberrations</p><h2 id="comparison-of-optical-aberrations"><a href="https://www.edmundoptics.com/resources/application-notes/optics/comparison-of-optical-aberrations/" target="_blank" rel="noopener">Comparison of Optical Aberrations</a></h2>]]></content>
<summary type="html">
<h2 id="chromatic-aberrations">CHROMATIC ABERRATIONS</h2>
<p>Transverse chromatic aberration (TCA)</p>
<p>Longitudinal chromatic aberration
</summary>
</entry>
<entry>
<title>Learning How to Learn</title>
<link href="https://github-9233.github.io/2019/04/18/Learning-How-to-Learn/"/>
<id>https://github-9233.github.io/2019/04/18/Learning-How-to-Learn/</id>
<published>2019-04-18T06:45:13.000Z</published>
<updated>2019-04-18T10:48:58.446Z</updated>
<content type="html"><![CDATA[<h2 id="thinking-modes">Thinking Modes</h2><p><strong>Focused vs diffused mode</strong> (both are neccesary, take scheduled breaks)</p><p><img src="/2019/04/18/Learning-How-to-Learn/ucsd-learning-focused-diffuse-modes.png"></p><h2 id="chunking">Chunking</h2><p>know how smaller concepts fits in the bigger concept, like puzzle pieces</p><p><img src="/2019/04/18/Learning-How-to-Learn/ucsd-learning-chunking-puzzle-illustrated.png"></p><h3 id="formation-of-chunks">Formation of Chunks</h3><p><img src="/2019/04/18/Learning-How-to-Learn/ucsd-learning-chunking-context.png"></p><h3 id="practice-and-repetition">Practice and Repetition</h3><ul><li>write down</li><li>picture</li><li>sleep</li></ul><h3 id="pomodoro">Pomodoro</h3><p><img src="/2019/04/18/Learning-How-to-Learn/ucsd-learning-pomodoro.png"></p><h2 id="illusions-of-competence"><strong>Illusions of Competence</strong></h2><p>brief notes that summarize keys concepts are much more effective</p><p><img src="/2019/04/18/Learning-How-to-Learn/1*YbNb1pYWQhd9Zz0Dh2cbAQ.jpeg"></p><h3 id="recall">Recall</h3><p>after reading summarize in a few minutes your takeaway</p><p><img src="/2019/04/18/Learning-How-to-Learn/1*Qwv_V6ljhXZCGEubesLCLw.jpeg"></p><p><img src="/2019/04/18/Learning-How-to-Learn/The-Feynman-Technique.jpg"></p><h3 id="bite-sized-testing"><strong>Bite-Sized Testing</strong></h3><p>auto-test yourself about new material</p><h3 id="overlearning"><strong>Overlearning</strong></h3><p>avoid marathon sessions, spread learning over many sessions in different learning modes</p><h3 id="interleaving"><strong>Interleaving</strong></h3><p>know how but also know when to apply a concept, context is important</p><h3 id="methapors-and-analogies"><strong>Methapors and Analogies</strong></h3><p>use them when teaching/sharing what you’ve learned</p><h3 id="study-groups"><strong>Study Groups</strong></h3><p>share and get different points of view, learning in groups adds to the experience</p><h2 id="procrastination">Procrastination</h2><h3 id="process-over-product"><strong>Process over Product</strong></h3><p>aim for time goals</p><h3 id="zombie-mode">Zombie Mode</h3><h3 id="elements-of-habit">4 Elements of Habit</h3><p>神经科学认为 Chunking 与习惯是相关的。良好的 Chunking 使你在完成一项任务时只需要关注个别关键的因素,将剩余的交给僵尸模式,从而节省精力。显然习惯并不完全是有害的,但为了克服拖延这个坏习惯,你需要了解习惯的四个要素。</p><ol type="1"><li><strong>Cue</strong>:触发僵尸模式的外界刺激,比如 to-do list 上的第一个项目,或是来自朋友的微信消息。这类刺激大致可以分为四类:时间、地点、感受和反应。</li><li><strong>Routine</strong>:受到刺激后习惯性地做出的反应。</li><li><strong>Reward</strong>:习惯给予的奖励使它得以存在下去,最简单的例子就是你在拖延时感受到的短时间的愉悦。</li><li><strong>Belief</strong>:习惯的力量之所以强大,是因为你在内心深处往往认为它们是无法被改变的。</li></ol><h3 id="harnessing-zombies">Harnessing Zombies</h3><p>拖延很容易,而对抗拖延需要消耗大量的<strong>意志力(Will Power)</strong>。所以尽量使自己处于不需要对抗拖延的境地,或是反过来用尽量少的意志力来利用僵尸模式帮助自己。</p><p>根据上面对习惯的分析,要打破拖延的习惯链,只需<strong>消耗意志力去改变四个要素中的一个</strong>。</p><ol type="1"><li><strong>Cue</strong>:找到令自己进入拖延的刺激并避免它们。最简单的例子,在学习时远离网络和电视。</li><li><strong>Routine</strong>:有意识地改变日常的一部分,比如定新的计划,或是养成在学习前把手机关闭的习惯。</li><li><strong>Reward</strong>:尝试引导自己的正面进取情绪来替代拖延带来的快感,比如自豪感、满足感等。又或者允许自己在不拖延完成任务后尽情放松。一个小技巧是<strong>将奖励设定为跟 deadline 有关</strong>,比如“五点前做完作业就约上同学去吃大餐”。</li><li><strong>Belief</strong>:相信自己的新策略可以成功打败拖延。可以通过跟志同道合的小伙伴互相监督来促进自信。</li></ol>]]></content>
<summary type="html">
<h2 id="thinking-modes">Thinking Modes</h2>
<p><strong>Focused vs diffused mode</strong> (both are neccesary, take scheduled breaks)</p>
<p>
</summary>
</entry>
<entry>
<title>learning-resource</title>
<link href="https://github-9233.github.io/2019/04/18/learning-resource/"/>
<id>https://github-9233.github.io/2019/04/18/learning-resource/</id>
<published>2019-04-18T03:35:21.000Z</published>
<updated>2019-04-19T08:29:41.772Z</updated>
<content type="html"><![CDATA[<p>old school syllabi: <a href="http://explorer.opensyllabusproject.org/" target="_blank" rel="noopener">The Open Syllabus Project</a></p><p>modern internet syllabi:</p><ul><li><a href="https://www.susanjfowler.com/blog/2016/8/13/so-you-want-to-learn-physics" target="_blank" rel="noopener">Susan Fowler's guide for learning physics</a></li><li><a href="https://danigrant.github.io/syllabus/cryptonetworks" target="_blank" rel="noopener">How To Think About Cryptonetworks</a></li><li><a href="https://teachyourselfcs.com" target="_blank" rel="noopener">Computer Science</a></li><li><a href="https://cryptoeconomics.study//" target="_blank" rel="noopener">Cryptoeconomics</a></li><li><a href="https://myscope.training" target="_blank" rel="noopener">Microscopy Train for advanced research</a></li><li><a href="https://ctext.org/zh" target="_blank" rel="noopener">中國哲學書電子化計劃</a></li><li><a href="https://quantum.country/qcvc" target="_blank" rel="noopener">Quantumcomputing</a></li><li><a href="https://www.yuque.com/ham" target="_blank" rel="noopener">业余无线电通讯</a></li></ul><p>There are five projects I know of doing something in this space: <a href="https://learn-anything.xyz/" target="_blank" rel="noopener">Learn Anything</a>, which lets people upvote the best resources for learning any topic, <a href="https://gooru.org/" target="_blank" rel="noopener">Gooru</a>, which lets teachers use existing content on the web to create courses for K-12 content, <a href="https://www.holloway.com/" target="_blank" rel="noopener">Holloway</a> and <a href="https://golden.com/" target="_blank" rel="noopener">Golden</a>, which are creating open source guides, and <a href="https://samiz-dat.github.io/hyperreadings/" target="_blank" rel="noopener">Hyperreadings</a>, which is a permanent archival library plus curated reading guides on top. I am sure there are more.</p>]]></content>
<summary type="html">
<p>old school syllabi: <a href="http://explorer.opensyllabusproject.org/" target="_blank" rel="noopener">The Open Syllabus Project</a></p>
<
</summary>
</entry>
<entry>
<title>MicroscopyBasics</title>
<link href="https://github-9233.github.io/2019/04/13/MicroscopyBasics/"/>
<id>https://github-9233.github.io/2019/04/13/MicroscopyBasics/</id>
<published>2019-04-13T08:38:22.000Z</published>
<updated>2019-07-08T09:17:09.447Z</updated>
<content type="html"><![CDATA[<p>MyScope was developed by Microscopy Australia to provide an online learning environment for those who want to learn about microscopy. The platform provides insights into the fundamental science behind different microscopes, explores what can and cannot be measured by different systems and provides a realistic operating experience on high end microscopes.</p><a id="more"></a><h1 id="basis">Basis</h1><h2 id="resolution">Resolution</h2><p><span class="math display">\[d=\frac{0.61\lambda}{nsin\alpha}\]</span></p><table><thead><tr class="header"><th style="text-align: left;">Where</th><th style="text-align: left;">Equals</th></tr></thead><tbody><tr class="odd"><td style="text-align: left;"><span class="math inline">\(d\)</span></td><td style="text-align: left;">resolution (minimum resolvable distance)</td></tr><tr class="even"><td style="text-align: left;"><span class="math inline">\(\lambda\)</span></td><td style="text-align: left;">wavelength of the energy source</td></tr><tr class="odd"><td style="text-align: left;"><span class="math inline">\(n\)</span></td><td style="text-align: left;">refractive index of the medium</td></tr><tr class="even"><td style="text-align: left;"><span class="math inline">\(\alpha\)</span></td><td style="text-align: left;">aperture angle</td></tr></tbody></table><p>Note: The term <em>n</em>sin<em>α</em> is named numerical aperture.</p><h2 id="electromagnetic-spectrum">Electromagnetic Spectrum</h2><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/c/cf/EM_Spectrum_Properties_edit.svg/1280px-EM_Spectrum_Properties_edit.svg.png" alt=""><figcaption>A diagram of the electromagnetic spectrum, showing various properties across the range of frequencies and wavelengths</figcaption></figure><h2 id="problems-with-lenses-aberrations"><a href="https://myscope.training/legacy/tem/background/concepts/problems/" target="_blank" rel="noopener">Problems with lenses: aberrations</a></h2><ul><li><p><a href="https://myscope.training/legacy/tem/background/concepts/problems/spherical.php" target="_blank" rel="noopener">Spherical aberration</a></p><p><img src="/2019/04/13/MicroscopyBasics/tem_spherical_abberation1.png"></p></li><li><p><a href="https://myscope.training/legacy/tem/background/concepts/problems/chromatic.php" target="_blank" rel="noopener">Chromatic aberration</a></p><p><img src="/2019/04/13/MicroscopyBasics/tem_chromatic_abberation.png"></p></li><li><p><a href="https://myscope.training/legacy/tem/background/concepts/problems/astigmatism.php" target="_blank" rel="noopener">Astigmatism</a></p><p><img src="/2019/04/13/MicroscopyBasics/tem_astigmatism-5488775.png"></p></li></ul><h2 id="microscope-comparisons"><a href="https://myscope.training/#/M101level_1_13" target="_blank" rel="noopener">Microscope Comparisons</a></h2><p><img src="/2019/04/13/MicroscopyBasics/figure_09.svg"></p><table><colgroup><col style="width: 10%"><col style="width: 4%"><col style="width: 42%"><col style="width: 42%"></colgroup><thead><tr class="header"><th></th><th>OM</th><th>TEM</th><th>SEM</th></tr></thead><tbody><tr class="odd"><td>Magnification</td><td>2000x</td><td>5,000,000x</td><td>500,000x</td></tr><tr class="even"><td>Resolution</td><td>200nm</td><td>0.2nm</td><td>20nm</td></tr><tr class="odd"><td>Depth of field</td><td>1 unit</td><td></td><td>300 unit</td></tr><tr class="even"><td>Microanalysis</td><td></td><td>crystal structures, specimen orientations and chemical compositions of phases, precipitates and contaminants through diffraction pattern, X-ray and electron-energy analysis.</td><td>chemical composition, crystallographic, magnetic and electrical characteristics</td></tr></tbody></table><h2 id="microscopy-basics">Microscopy Basics</h2><ul><li>Optical<ul><li><p>transmitted light</p></li><li><p><strong>contrasting methods</strong></p><ul><li><em>phase contrast</em></li></ul><p><img src="https://upload.wikimedia.org/wikipedia/commons/2/23/Working_principle_of_phase_contrast_microscopy.gif"></p></li><li><p><em>dark-field</em></p><ul><li><p><em>polarisation</em></p></li><li><p><em>differential interference contrast (DIC)microscopy</em></p><figure><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/4/46/DIC_Light_Path.png/1920px-DIC_Light_Path.png" alt=""><figcaption>The route of light through a DIC microscope. The two light beams should be parallel between condenser and objective</figcaption></figure></li></ul></li><li><p><em>fluorescence</em></p></li><li><p><em>laser microscopy</em>=Confocal microscopy</p></li><li><p>Digital microscopes</p></li></ul></li><li>Electron<ul><li><p>transmitted electron microscopy (TEM)</p></li><li><p>scanning electron microscopy (SEM)</p></li><li><p>environmental scanning electron (ESEM)</p></li><li><p>field emision-scanning electron microscopy (FE-SEM)</p></li><li><p>field emision transmission electron microscopy (FE-TEM)</p></li><li><p><em>Micro-probe EM</em></p></li><li><p>Energy dispersive spectroscopy EDS (or Energy Dispersive X-ray spectroscopy EDX)</p></li><li><p>Wavelength Dispersive Spectroscopy (WDS) (or wavelength dispersive X-ray spectroscopy)</p></li><li><p><em>SEM-CL:cathodoluminescence</em></p></li><li><p>Serial-section Scanning Electron Microscopy</p></li><li><p><em>Focused Ion Beam Microscopy (FIB)</em></p></li></ul></li></ul><h2 id="microscopy-and-measurement">Microscopy and Measurement</h2><ul><li><strong><em>Field Number (FN)</em></strong></li><li>more accurate estimate<ul><li>stage micrometers</li><li>ocular reticules</li></ul></li><li>calibration</li></ul><h2 id="photomicrography">Photomicrography</h2><p><em>The objective on the microscope is the lens for the camera. This means the aperture is that of the objective. As a result the camera must be set to aperture priority mode.</em></p><ul><li>Pixels</li><li>Data Processing</li><li><em>Binning Data</em></li></ul><h2 id="image-management">Image Management</h2><ul><li>Key Considerations In Modern Digital Systems<ul><li>Whether the image is colour or monochrome</li><li>Size of the chip</li><li>Pixel size</li><li>Quantum efficiency</li><li>Readout noise</li><li>Dynamic range in which the image is stored</li><li>Bit depth</li><li>Frame rate at data collection</li><li>Pixel number</li><li>Cooling</li><li>Gain or binning</li><li>Whether the camera is designed for a microscope or consumer use</li><li>Monitor quality</li></ul></li><li>Reconstruction<ul><li>2D to 3D: <em>Filtered Backprojection</em></li><li>3D to Visual Render: voxels (VOlume (X) ELements - similar to a pixel)</li></ul></li></ul><h1 id="scanning-electron-microscopy">Scanning Electron Microscopy</h1><h2 id="can-cant">Can & can't</h2><p>can:</p><ul><li>Image morphology of samples (e.g. view bulk material, coatings, sectioned material).</li><li>Image compositional and some bonding differences through contrast and by using backscattered electrons.</li><li>Image molecular probes by using metal and fluorescent probes in biological samples.</li><li>Undertake micro and nano-lithography (remove material from samples; cut pieces out or remove progressive slices from samples [e.g. using a focussed ion beam]).</li><li>Heat or cool samples while viewing them (this requires a specific type of stage).</li><li>Wet and dry samples while viewing them (only in an Environmental SEM)</li><li>View frozen material using a cryostage</li><li>Analyse X-rays from samples for microanalysis (requires a EDS or WDS detector)</li><li>Study optoelectronic behaviour of semiconductors (requires a cathodoluminescence [CL] detector)</li><li>View/map grain orientation/crystallographic orientation and study related information like heterogeneity and microstrain in flat samples (requires an EBSD detector).</li></ul><p>can’t</p><ul><li><p>Image wet samples</p></li><li><p>Image non-conductive samples</p></li><li><p>Colour images</p></li><li><p>Accurate height measurement</p></li><li><p>Sub-surface imaging</p></li><li><p>Imaging through fluid</p></li><li><p>Atomic imaging</p></li><li><p><strong><em>Elemental analysis below micrometre scale</em></strong></p><p><em>Elemental analysis of areas less than 1 micrometre can be very difficult in an SEM. This is due to the interaction volume between the electron beam and the sample which is often in the micrometre range. The interaction volume can be reduced by reducing the electron beam accelerating voltage. However, the corresponding reduction in signal can make it difficult to acquire useful data.</em></p></li><li><p>Image charged molecules</p></li></ul><h2 id="components">Components</h2><p><img src="/2019/04/13/MicroscopyBasics/figure_12.svg"></p><h3 id="electron-gun"><a href="https://myscope.training/index.html#/SEMlevel_3_5" target="_blank" rel="noopener">Electron gun</a></h3><p><img src="/2019/04/13/MicroscopyBasics/figure_13.svg"></p><p><img src="/2019/04/13/MicroscopyBasics/figure_15.svg"></p><h3 id="vacuum-system">Vacuum system</h3><ul><li><h4 id="types-of-pumps"><a href="https://myscope.training/index.html#/SEMlevel_3_8" target="_blank" rel="noopener">Types of pumps</a></h4><ul><li><p>Rotary pumps</p></li><li><p><em>Diffusion pumps</em></p></li><li><p><em>Scroll pumps</em></p></li><li><p><em>Turbo-molecular pumps</em></p></li><li><p><em>Ion getter pumps</em></p></li></ul></li><li><h4 id="vacuum-requirements">Vacuum requirements</h4><ul><li>high-vacuum mode: SE mode & secondary electrons</li><li>low-vacuum mode: backscattered electrons & characteristic X-rays</li></ul></li></ul><h3 id="water-chilling-system">Water chilling system</h3><h3 id="structure-of-the-column">Structure of the column</h3><ul><li><p>condenser lens</p></li><li><p>objective lens</p></li><li><p>scanning coils (two pairs of electromagnetic deflection coils)</p></li></ul><h3 id="specimen-chamber">Specimen chamber</h3><ul><li>stage</li><li>detectors</li></ul><h3 id="imaging-system">Imaging system</h3><ul><li>SE images(surface of sample)<ul><li>increase the yield of SE <= coat thin layer(~10 nm) heavy metals (gold/platinum)<ul><li>non-conductive coated to reduce surface charging</li><li>low atomic number (Z) specimens (e.g. biological samples) coated to provide a surface layer</li></ul></li></ul></li><li>The contribution of BSE to images collected with the SE detector<ul><li>higher kV => sub-surface information due to various backscattered effects (elastic scattering)<ul><li>SE1: interaction between primary electron beam and sample</li><li>SE2: generate by BSE</li><li>SE3: interaction of beam with sample chamber, pole piece etc</li></ul></li></ul></li></ul><h2 id="beamspecimen-interactions">Beam/specimen interactions</h2><p><img src="/2019/04/13/MicroscopyBasics/figure_26.gif"></p><ul><li><p>regions</p><ul><li>SE (top 15nm)</li><li>BSE (top 40%)</li><li>X rays (entire region)</li></ul></li><li><p>Electron-matter interactions</p><ul><li>Elastic scattering: BSE</li><li>Inelastic scattering<ul><li>phonon excitation (heating)</li><li>cathodoluminescence (visible light fluorescence)</li><li><em>continuum radiation (bremsstrahlung)</em></li><li>characteristic X-ray radiation</li><li><em>plasmon production (secondary electrons)</em></li><li><em>Auger electrons (ejection of outer shell electrons)</em></li></ul></li></ul></li><li><p>Modelling interactions</p><ul><li><em>Casino (Monte Carlo Simulation of electroN trajectory in sOlids)</em></li></ul></li></ul><h2 id="a-basic-guide-to-using-an-sem"><a href="https://myscope.training/index.html#/SEMlevel_2_11" target="_blank" rel="noopener">A basic guide to using an SEM</a></h2><h3 id="specimen-preparation">Specimen preparation</h3><p><img src="/2019/04/13/MicroscopyBasics/下载.png"></p><h3 id="accelerating-voltage">Accelerating voltage</h3><figure><img src="/2019/04/13/MicroscopyBasics/figure_34.svg" alt=""><figcaption>figure_34</figcaption></figure><h3 id="apertures">Apertures</h3><p><img src="/2019/04/13/MicroscopyBasics/figure_35-5469450.svg"></p><p><img src="/2019/04/13/MicroscopyBasics/figure_36-5469460.jpg"></p><p><img src="/2019/04/13/MicroscopyBasics/figure_37.svg"></p><h3 id="spot-size">Spot size</h3><p><img src="/2019/04/13/MicroscopyBasics/figure_27.svg"></p><ul><li>working distance smaller => smaller spot size</li><li>condenser lens setting higher current => smaller spot size</li><li>objective lens aperture smaller => smaller spot size</li></ul><h3 id="working-distance">Working distance</h3><p><img src="/2019/04/13/MicroscopyBasics/figure_41.svg"></p><h3 id="constrast-and-brightness">Constrast and brightness</h3><ul><li>Perfecting an image - Signal processing</li><li>Tilting to increase SE contrast</li></ul><h3 id="magnification-calibration">Magnification & Calibration</h3><h3 id="scan-rate-signal-to-noise">Scan rate & signal to noise</h3><ul><li>slower scan rate => more electrons at each point => better quality</li><li>quality limited<ul><li>spot size</li><li>signal(S)/noise(N)<ul><li>Noise <= beam brightness, spot size, SE detector sensitivity => salt-and-pepper, grainy</li><li>high-resolution => low S/N ratio & grainy.<br></li><li>increase electrons per point => improve S/N ratio</li><li>eg. Tungsten (W) filaments<ul><li>high resolution => small spot sizes => low quantity of electrons => low brightness => need high current detecter => low S/N ratio</li></ul></li></ul></li></ul></li></ul><h3 id="image-artefacts-and-trouble-shooting">Image artefacts and trouble-shooting</h3><ul><li><em>Astigmatism</em>: <strong>stigmator</strong></li><li>Lack of detail of surface structures: lower kVs (5-10kV)</li><li>Edge effects: lower kVs</li><li>Charging<ul><li>reduce electron energy(kV)</li><li>reduce number of electrons<ul><li>beam current</li><li>emission level of gun</li><li>spot size</li><li>apertures between gun & specimen</li></ul></li><li>recoat the sample with a thicker layer of platinum</li></ul></li><li>Specimen damage<ul><li>lower beam energy</li><li>increase WD => larger spot size on the sample <strong>but</strong> reduce resolution</li></ul></li><li>Beam-related contamination<ul><li>take micrographs at low magnification</li><li>ensure sample clean</li></ul></li></ul><h3 id="specialised-sem-techniques">Specialised SEM techniques</h3><ul><li><p><em>CL</em></p></li><li><p>ESEM:</p></li><li><p>Cryo-SEM-Cold stage</p><p>hydrated (wet) samples, delicate biological samples, hydrogels, food, biofilms, foams, fats and waxes, suspensions, pharmaceuticals and nanoparticles</p></li><li><p>FIB</p></li><li><p>EDS</p></li><li><p>EBSD</p></li><li><p><strong><em>EBL</em></strong></p><p><em>Electron beam lithography (EBL) is a maskless lithography technique used for patterning of computer generated layout structures on photoresists on silicon wafers.</em></p></li><li><p>Backscatter</p><ul><li><p><strong><em>electron channelling contrast imaging (ECCI)</em></strong></p><p><em>The change in diffraction of the backscattered electrons as they interact with a dislocation in the material results in a higher backscattering coefficient than for the matrix; so individual dislocations appear as bright lines in a darker matrix.</em></p></li><li><p>Coating</p><ul><li>uncoated for compositional information</li><li>non-conductive coated with carbon with compositional detail</li></ul></li><li><p>Topography and BSE</p><ul><li>flat => best compositional information</li><li>BSE image (greyscale differences <= atomic number contrast)<ul><li>phases recognition & classification</li><li>differences in elemental composition or concentrations</li></ul></li></ul></li></ul></li></ul><h2 id="sem-simulator"><a href="https://myscope.training/SEM_simulator.html" target="_blank" rel="noopener">SEM simulator</a></h2><h1 id="transmission-electron-microscopy">Transmission Electron Microscopy</h1><h2 id="can-cant-1">Can & can't</h2><p>can:</p><ul><li>Image morphology of samples, e.g. view sections of material, fine powders suspended on a thin film, small whole organisms such as viruses or bacteria, and frozen solutions.</li><li>Tilt a sample and collect a series of images to construct a 3-dimensional image.</li><li>Analyse the composition and some bonding differences (through contrast and by using spectroscopy techniques: microanalysis and electron energy loss).</li><li>Physically manipulate samples while viewing them, such as indent or compress them to measure mechanical properties (only when holders specialised for these techniques are available).</li><li>View frozen material (in a TEM with a cryostage).</li><li>Generate characteristic X-rays from samples for microanalysis.</li><li>Acquire electron diffraction patterns (using the physics of Bragg Diffraction).</li><li>Perform electron energy loss spectroscopy of the beam passing through a sample to determine sample composition or the bonding states of atoms in the sample.</li></ul><p>can't:</p><ul><li>TEM cannot take colour images. Colour is sometimes added artificially to TEM images.</li><li>TEM cannot image through thick samples: the usual sample thickness is around 100-200nm. Electrons cannot readily penetrate sections much thicker than 200nm.</li><li>A standard TEM cannot image surface information.</li><li>The TEM cannot reliably image charged molecules that are mobile in a matrix. For example, some species (e.g. Na+) are volatile under the electron beam because the negative electron beam exerts a force on charged material.</li></ul><h2 id="components-1">Components</h2><figure><img src="/2019/04/13/MicroscopyBasics/schematic.png" alt=""><figcaption>A generalised cut-away diagram of the internal structure of a transmission electron microscope alongside an example of a modern instrument.</figcaption></figure><h2 id="beamspecimen-interactions-1">Beam/specimen interactions</h2><p><img src="/2019/04/13/MicroscopyBasics/tem_interaction.png"></p><h2 id="transmission-electron-microscopy-in-practice">Transmission electron microscopy in practice</h2><h3 id="sample-preparation">Sample preparation</h3><ul><li><p>Organic sample</p><p><img src="/2019/04/13/MicroscopyBasics/sample-biological.png"></p></li><li><p>Inorganic sample</p><p><img src="/2019/04/13/MicroscopyBasics/sample-nonorganic.png"></p></li></ul><h3 id="step-by-step-procedures-for-using-tems">Step-by-step procedures for using TEMs</h3><ol type="1"><li>Insert the specimen</li><li>Choose and set voltage</li><li>Turn on the gun and filament</li><li>Ensure the illumination system is aligned and operating as needed</li><li>Insert apertures as needed (e.g. objective aperture in if it is out)</li><li>Adjust the sample height to the eucentric position</li><li>Select the correct spot size</li><li>Select the correct aperture size for the apertures needed</li><li>Move the sample to an area of interest and view</li><li>Focus and correct astigmatism</li><li>Select magnification of choice</li><li>Photograph or collect digital recording of image</li><li>Exchange sample or undertake shutdown procedure</li></ol><h3 id="machine-settings">Machine settings</h3><table><thead><tr class="header"><th><strong>Thin Sections/Biological Samples</strong></th><th><strong>High Resolution</strong></th></tr></thead><tbody><tr class="odd"><td>80-120 kV</td><td>200-300 kV</td></tr><tr class="even"><td>Condenser aperture: 1 (biggest)</td><td>Condenser aperture: 2 or 3 (small)</td></tr><tr class="odd"><td>Spot size: 1 (unless sample is “drifting”, 5 or 6)</td><td>Spot size: 6 to 8</td></tr><tr class="even"><td>Objective aperture: 3 (smallest)</td><td>Objective aperture: 1 (biggest)</td></tr></tbody></table><p><strong>An increase in voltage</strong></p><p>= increased brightness = increased resolution = thicker sections can be used = reduced contrast</p><p>The condenser aperture is the top one on the column</p><p><strong>A smaller condenser aperture</strong></p><p>= increased resolution = decreased beam damage to sample = decreased brightness</p><p>The objective aperture is the middle one (just below the specimen holder)</p><p><strong>A smaller objective aperture</strong></p><p>= increased contrast = decreased brightness</p><p>Spot size on the JEM-1010 gets smaller with increasing number</p><p><strong>A smaller spot size</strong></p><p>= increased resolution (for some imaging modes) = decreased brightness = decreased beam damage to the sample = decreased sample “drift” (stability improved)</p><h3 id="alignment">Alignment</h3><ol type="1"><li>Illumination system: beam tilt; beam shift</li><li>Condenser aperture alignment</li><li>Condenser astigmatism correction</li><li>Alignment of illumination with respect to the objective lens</li><li>Objective aperture centering</li><li>Objective astigmatism correction</li><li>Alignment of intermediate lens</li></ol><h2 id="virtual-tem">Virtual TEM</h2><ul><li><a href="https://myscope.training/legacy/tem/practice/virtualtem/basic.php" target="_blank" rel="noopener">Basic Imaging</a></li><li><a href="https://myscope.training/legacy/tem/practice/virtualtem/advanced.php" target="_blank" rel="noopener">Diffraction and advanced imaging</a></li></ul><h3 id="additions">Additions</h3><h4 id="image-formation-basics">Image formation basics</h4><p><img src="/2019/04/13/MicroscopyBasics/tem_beampath_stage_aandb.png"></p><ul><li><p><strong>Stage A</strong> is the scattering of an incident electron beam by a specimen. This scattered radiation passes through an objective lens, which focuses it to form the primary image.</p></li><li><p><strong>Stage B</strong> uses the primary image obtained in stage A and magnifies this image using additional lenses to form a highly magnified final image.</p></li></ul><h4 id="imaging-modes">Imaging modes</h4><p><img src="/2019/04/13/MicroscopyBasics/modes_new.png"></p><h4 id="convergent-beam-electron-diffraction-cbed">Convergent Beam Electron Diffraction (CBED)</h4><p><img src="/2019/04/13/MicroscopyBasics/convergence.png"></p><h1 id="microanalysis">Microanalysis</h1><figure><img src="/2019/04/13/MicroscopyBasics/detection%20limit&%20spatial%20resolution.png" alt=""><figcaption>detection limit& spatial resolution</figcaption></figure><p>Energy Dispersive X-ray Spectroscopy (EDS, EDX or XEDS)</p><table><colgroup><col style="width: 17%"><col style="width: 57%"><col style="width: 25%"></colgroup><thead><tr class="header"><th>EDS</th><th>SEM</th><th>TEM</th></tr></thead><tbody><tr class="odd"><td>spatial resolution</td><td><img src="/2019/04/13/MicroscopyBasics/generation-scattering.png" alt="generation-scattering"><strong>Figure</strong>: The typical spatial resolution of different signals, Secondary electrons, Backscattered electrons and X-rays, in the scanning electron microscope.</td><td>order of sample thickness.</td></tr><tr class="even"><td>detection limit</td><td>~0.1-0.5 wt%</td><td>~0.01-0.1 wt%</td></tr><tr class="odd"><td></td><td></td><td></td></tr></tbody></table><h2 id="x-rays-generation">X-rays Generation</h2><h3 id="bremsstrahlung-x-rays">Bremsstrahlung X-rays</h3><figure><img src="/2019/04/13/MicroscopyBasics/generation-continuum.png" alt=""><figcaption><strong>Figure</strong>: The difference between generated and observed Bremsstrahlung X-ray spectra. Although many low-energy X-rays are generated most of them are absorbed so the observed spectrum records a decrease in X-ray intensity at low energy.</figcaption></figure><h4 id="kramers-law">Kramer's law</h4><p><span class="math display">\[I=i_p\cdot Z\frac{E_0-E}{E}\]</span></p><p><span class="math inline">\(i_p\)</span>(electron probe current),<span class="math inline">\(Z\)</span>(mean atomic number),<span class="math inline">\(E_0\)</span>(energy of the electrons in the primary beam)</p><h3 id="characteristic-x-rays">Characteristic X-rays</h3><figure><img src="/2019/04/13/MicroscopyBasics/characteristic-production.png" alt=""><figcaption><strong>Figure</strong>: An electron from the primary beam dislodges an electron from the K shell of a Si atom in the sample. An electron from the L shell fills the vacancy and a Si Kα X-ray is generated. The energy of the X-ray is equal to the ionization energy of the K shell minus the ionization energy of the L shell.</figcaption></figure><h4 id="nomenclature">Nomenclature</h4><ul><li>Siegbahn: element name + electron shell produce the X-ray + relative intensity of the line</li></ul><figure><img src="/2019/04/13/MicroscopyBasics/characteristic-specimenatom.png" alt=""><figcaption><strong>Figure</strong>: The electron transitions involved in generating the Kα, Kβ and Lα X-ray photons.</figcaption></figure><ul><li>IUPAC</li></ul><h4 id="the-x-ray-spectrum">The X-ray spectrum</h4><figure><img src="/2019/04/13/MicroscopyBasics/characteristic-both-5052765.png" alt=""><figcaption><strong>Figure</strong>: The X-ray spectrum displayed by the Energy Dispersive system comprises Characteristic X-ray peaks superimposed on the continuum (Microanalysis/characteristic-both-5052765.png).</figcaption></figure><h2 id="x-ray-intensity">X-ray intensity</h2><ul><li><p>Concentration</p><ul><li>major elements (> 10 wt%)</li><li>minor elements (1-10 wt%)</li><li>trace elements (<1 wt%)</li></ul></li><li><p>Beam current, probe current or spot size</p></li><li><p>Accelerating voltage or overvoltage ratio</p><p><span class="math inline">\(\text{overvoltage ratio }U=\frac{\text{electrons energy of primary beam }E_0}{\text{critical ionization energy }E_c}\)</span></p><p><span class="math inline">\(U\ge2\)</span> & <span class="math inline">\(U_{opt}\sim2.7\)</span></p></li><li><p>Fluorescence Yield</p><p><span class="math inline">\(\text{fluorescence yield}=\frac{\text{X-rays}}{\text{Auger electrons}}\)</span></p><figure><img src="/2019/04/13/MicroscopyBasics/auger_production.png" alt=""><figcaption><strong>Figure</strong>: After inner shell ionization, the atom may relax by emitting a Characteristic X-ray or an Auger electron. The energy of the Auger electron is related to the electronic configuration of the atom that was ionized by the primary electron beam.</figcaption></figure></li><li><p>X-ray Absorption</p><p>photo-electric effect</p></li></ul><h2 id="x-ray-detection-by-eds"><a href="https://myscope.training/legacy/analysis/eds/xraydetection/" target="_blank" rel="noopener">X-ray detection by EDS</a></h2><h3 id="three-basic-parts">Three basic parts</h3><ul><li><p>detector (<em>SDD=silicon-drift detector</em>)</p></li><li><p><u>pulse processor</u></p><p><em>Process time, dead time</em></p></li><li><p>multi-channel analyser or computer display</p></li></ul><h3 id="care-and-calibration">Care and calibration</h3>]]></content>
<summary type="html">
<p>MyScope was developed by Microscopy Australia to provide an online learning environment for those who want to learn about microscopy. The platform provides insights into the fundamental science behind different microscopes, explores what can and cannot be measured by different systems and provides a realistic operating experience on high end microscopes.</p>
</summary>
</entry>
<entry>
<title>Machine Learning Foundations & Techniques</title>
<link href="https://github-9233.github.io/2019/01/17/Machine%20Learning-Hsuan-Tien%20Lin/"/>
<id>https://github-9233.github.io/2019/01/17/Machine Learning-Hsuan-Tien Lin/</id>
<published>2019-01-16T16:00:00.000Z</published>
<updated>2019-04-11T02:38:17.634Z</updated>
<content type="html"><![CDATA[<p>Machine learning is the study that allows computers to adaptively improve their performance with experience accumulated from the data observed. Our two sister courses teach the most fundamental algorithmic, theoretical and practical tools that any user of machine learning needs to know. This first course of the two would focus more on mathematical tools, and the other course would focus more on algorithmic tools.</p><a id="more"></a><h1 id="machine-learning-foundations"><a href="https://www.csie.ntu.edu.tw/~htlin/course/" target="_blank" rel="noopener">Machine Learning Foundations</a></h1><h2 id="topic-1-when-can-machines-learn">Topic 1: When can machines learn?</h2><p>###Lecture 1: The Learning Problem</p><p><em>A</em> takes <em>D</em> and <em>H</em> to get <em>g</em></p><ul><li><p>Course Introduction</p><p>foundation oriented and story-like</p></li><li><p>What is Machine Learning</p><p>use data to approximate target</p></li><li><p>Applications of Machine Learning</p><p>almost everywhere</p></li><li><p>Components of Machine Learning</p><p><em>A</em> takes <em>D</em> and <em>H</em> to get <em>g</em></p></li><li><p>Machine Learning and Other Fields</p><p>related to Date Mining, Artificial Intelligence and Statstics</p></li></ul><h3 id="lecture-2-learning-to-answer-yesno">Lecture 2: Learning to Answer Yes/No</h3><p>PLA <em>A</em> takes linear separable <em>D</em> and perceptrons <em>H</em> to get hypothesis <em>g</em></p><ul><li><p>Perceptron Hypothesis Set</p><p>hyperplanes/linear classifiers in Rd</p></li><li><p>Perceptron Learning Algorithm (PLA)</p><p>correct mistakes and improve iteratively</p></li><li><p>Guarantee of PLA</p><p>no mistake eventually if linear separable</p></li><li><p>Non-Separable Data</p><p>hold somewhat ‘best’ weights in pocket</p></li></ul><h3 id="lecture-3-types-of-learning">Lecture 3: Types of Learning</h3><p>focus: binary classification or regression from a batch of supervised data with concrete features</p><ul><li><p>Learning with Different Output Space Y</p><p>[classification], [regression], structured</p></li><li><p>Learning with Different Data Label yn</p><p>[supervised], un/semi-supervised, reinforcement</p></li><li><p>Learning with Different Protocol f ⇒ (xn, yn)</p><p>[batch], online, active</p></li><li><p>Learning with Different Input Space X</p><p>[concrete], raw, abstract</p></li></ul><h3 id="lecture-4-feasibility-of-learning">Lecture 4: feasibility of learning</h3><p>learning is PAC-possible if enough statistical data and finite |H|</p><ul><li><p>Learning is Impossible?</p><p>absolutely no free lunch outside D</p></li><li><p>Probability to the Rescue</p><p>probably approximately correct outside D</p></li><li><p>Connection to Learning</p><p>verification possible if Ein(h) small for fixed h</p></li><li><p>Connection to Real Learning</p><p>learning possible if |H| finite and Ein(g) small</p></li></ul><h2 id="topic-2-why-can-machines-learn">Topic 2: Why Can Machines Learn?</h2><h3 id="lecture-5-training-versus-testing">Lecture 5: training versus testing</h3><p>effective price of choice in training: (wishfully) growth function mH(N) with a break point</p><ul><li><p>Recap and Preview</p><p>two questions: Eout(g) ≈ Ein(g), and Ein(g) ≈ 0</p></li><li><p>Effective Number of Lines</p><p>at most 14 through the eye of 4 inputs</p></li><li><p>Effective Number of Hypotheses</p><p>at most mH(N) through the eye of N inputs</p></li><li><p>Break Point</p><p>when mH(N) becomes ‘non-exponential’</p></li></ul><h3 id="lecture-6-theory-of-generalization">Lecture 6: theory of generalization</h3><p>Eout ≈ Ein possible if mH(N) breaks somewhere and N large enough</p><ul><li><p>Restriction of Break Point</p><p>break point ‘breaks’ consequent points</p></li><li><p>Bounding Function: Basic Cases</p><p>B(N,k) bounds mH(N) with break point k</p></li><li><p>Bounding Function: Inductive Cases</p><p>B(N,k) is poly(N)</p></li><li><p>A Pictorial Proof</p><p>mH(N) can replace M with a few changes</p></li></ul><h3 id="lecture-7-the-vc-dimension">Lecture 7: the VC dimension</h3><p>learning happens if finite dVC, large N, and low Ein</p><ul><li><p>Definition of VC Dimension</p><p>maximum non-break point</p></li><li><p>VC Dimension of Perceptrons</p><p>dVC(H) = d + 1</p></li><li><p>Physical Intuition of VC Dimension</p><p>dVC ≈ #free parameters</p></li><li><p>Interpreting VC Dimension</p><p>loosely: model complexity & sample complexity</p></li></ul><h3 id="lecture-8-noise-and-error">Lecture 8: noise and error</h3><p>learning can happen with target distribution P(y|x) and low Ein w.r.t. err</p><ul><li><p>Noise and Probabilistic Target</p><p>can replace f(x) by P(y|x)</p></li><li><p>Error Measure</p><p>affect ‘ideal’ target</p></li><li><p>Algorithmic Error Measure</p><p>user-dependent =⇒ plausible or friendly</p></li><li><p>Weighted Classification</p><p>easily done by virtual ‘example copying’</p></li></ul><h2 id="topic-3-how-can-machines-learn">Topic 3: How Can Machines Learn?</h2><h3 id="lecture-9-linear-regression">Lecture 9: linear regression</h3><p>analytic solution wLIN = X†y with linear regression hypotheses and squared error</p><ul><li><p>Linear Regression Problem</p><p>use hyperplanes to approximate real values</p></li><li><p>Linear Regression Algorithm</p><p>analytic solution with pseudo-inverse</p></li><li><p>Generalization Issue</p><p>Eout − Ein ≈ 2(d+1) on average N</p></li><li><p>Linear Regression for Binary Classification</p><p>0/1 error ≤ squared error</p></li></ul><h3 id="lecture-10-logistic-regression">Lecture 10: logistic regression</h3><p>gradient descent on cross-entropy error to get good logistic hypothesis</p><ul><li><p>Logistic Regression Problem</p><p>P(+1|x) as target and θ(wT x) as hypotheses</p></li><li><p>Logistic Regression Error</p><p>cross-entropy (negative log likelihood)</p></li><li><p>Gradient of Logistic Regression Error</p><p>θ-weighted sum of data vectors</p></li><li><p>Gradient Descent</p><p>roll downhill by −∇Ein(w)</p></li></ul><h3 id="lecture-11-linear-models-for-classification">Lecture 11: linear models for classification</h3><p>binary classification via (logistic) regression; multiclass via OVA/OVO decomposition</p><ul><li><p>Linear Models for Binary Classification</p><p>three models useful in different ways</p></li><li><p>Stochastic Gradient Descent</p><p>follow negative stochastic gradient</p></li><li><p>Multiclass via Logistic Regression</p><p>predict with maximum estimated P(k|x)</p></li><li><p>Multiclass via Binary Classification</p><p>predict the tournament champion</p></li></ul><h3 id="lecture-12-nonlinear-transformation">Lecture 12: nonlinear transformation</h3><p>nonlinear X via nonlinear feature transform Φ plus linear X with price of model complexity</p><ul><li><p>Quadratic Hypotheses</p><p>linear hypotheses on quadratic-transformed data</p></li><li><p>Nonlinear Transform</p><p>happy linear modeling after Z = Φ(X )</p></li><li><p>Price of Nonlinear Transform</p><p>computation/storage/[model complexity]</p></li><li><p>Structured Hypothesis Sets</p><p>linear/simpler model first</p></li></ul><h2 id="topic-4-how-can-machines-learn-better">Topic 4: How Can Machines Learn Better?</h2><h3 id="lecture-13-hazard-of-overfitting">Lecture 13: hazard of overfitting</h3><p>overfitting happens with excessive power, stochastic/deterministic noise, and limited data</p><ul><li><p>What is Overfitting?</p><p>lower Ein but higher Eout</p></li><li><p>The Role of Noise and Data Size</p><p>overfitting ‘easily’ happens!</p></li><li><p>Deterministic Noise</p><p>what H cannot capture acts like noise</p></li><li><p>Dealing with Overfitting</p><p>data cleaning/pruning/hinting, and more</p></li></ul><h3 id="lecture-14-regularization">Lecture 14: regularization</h3><p>minimizes augmented error, where the added regularizer effectively limits model complexity</p><ul><li><p>Regularized Hypothesis Set</p><p>original H + constraint</p></li><li><p>Weight Decay Regularization</p><p>add λ/N*wTw in Eaug</p></li><li><p>Regularization and VC Theory</p><p>regularization decreases dEFF</p></li><li><p>General Regularizers</p><p>target-dependent, [plausible], or [friendly]</p></li></ul><h3 id="lecture-15-validation">Lecture 15: validation</h3><p>(crossly) reserve validation data to simulate testing procedure for model selection</p><ul><li><p>Model Selection Problem</p><p>dangerous by Ein and dishonest by Etest</p></li><li><p>Validation</p><p>select with Eval(Am(Dtrain)) while returning Am∗ (D)</p></li><li><p>Leave-One-Out Cross Validation</p><p>huge computation for almost unbiased estimate</p></li><li><p>V -Fold Cross Validation</p><p>reasonable computation and performance</p></li></ul><h3 id="lecture-16-three-learning-principle">Lecture 16: three learning principle</h3><ul><li><p>Occam’s Razor</p><p>simple, simple, simple!</p></li><li><p>Sampling Bias</p><p>match test scenario as much as possible</p></li><li><p>Data Snooping</p><p>any use of data is ‘contamination’</p></li><li><p><strong>Power of Three</strong></p><p>relatives, bounds, models, tools, principles</p></li></ul><h1 id="machine-learning-techniques">Machine Learning Techniques</h1><h2 id="topic-1-embedding-numerous-features-kernel-models">Topic 1: Embedding Numerous Features: Kernel Models</h2><h3 id="lecture-1-linear-support-vector-machine">Lecture 1: Linear Support Vector Machine</h3><p>linear SVM: more robust and solvable with quadratic programming</p><ul><li><p>Course Introduction</p><p>from foundations to techniques</p></li><li><p>Large-Margin Separating Hyperplane</p><p>intuitively more robust against noise</p></li><li><p>Standard Large-Margin Problem</p><p>minimize ‘length of w’ at special separating scale</p></li><li><p>Support Vector Machine</p><p>‘easy’ via quadratic programming</p></li><li><p>Reasons behind Large-Margin Hyperplane</p><p>fewer dichotomies and better generalization</p></li></ul><h3 id="lecture-2-dual-support-vector-machine">Lecture 2: Dual Support Vector Machine</h3><p>dual SVM: another QP with valuable geometric messages and almost no dependence on d ̃</p><ul><li><p>Motivation of Dual SVM</p><p>want to remove dependence on d ̃</p></li><li><p>Lagrange Dual SVM</p><p>KKT conditions link primal/dual</p></li><li><p>Solving Dual SVM</p><p>another QP, better solved with special solver</p></li><li><p>Messages behind Dual SVM</p><p>SVs represent fattest hyperplane</p></li></ul><h3 id="lecture-3-kernel-support-vector-machine">Lecture 3: Kernel Support Vector Machine</h3><p>kernel as a shortcut to (transform + inner product) to remove dependence on d ̃: allowing a spectrum of simple (linear) models to infinite dimensional (Gaussian) ones with margin control</p><ul><li><p>Kernel Trick</p><p>kernel as shortcut of transform + inner product</p></li><li><p>Polynomial Kernel</p><p>embeds specially-scaled polynomial transform</p></li><li><p>Gaussian Kernel</p><p>embeds infinite dimensional transform</p></li><li><p>Comparison of Kernels</p><p>linear for efficiency or Gaussian for power</p></li></ul><h3 id="lecture-4-soft-margin-support-vector-machine">Lecture 4: Soft-Margin Support Vector Machine</h3><p>allow some margin violations ξn while penalizing them by C; equivalent to upper-bounding αn by C</p><ul><li><p>Motivation and Primal Problem</p><p>add margin violations ξn</p></li><li><p>Dual Problem</p><p>upper-bound αn by C</p></li><li><p>Messages behind Soft-Margin SVM</p><p>bounded/free SVs for data analysis</p></li><li><p>Model Selection</p><p>cross-validation, or approximately nSV</p></li></ul><h3 id="lecture-5-kernel-logistic-regression">Lecture 5: Kernel Logistic Regression</h3><p>two-level learning for SVM-like sparse model for soft classification, or using representer theorem with regularized logistic error for dense model</p><ul><li><p>Soft-Margin SVM as Regularized Model</p><p>L2-regularization with hinge error measure</p></li><li><p>SVM versus Logistic Regression</p><p>≈ L2-regularized logistic regression</p></li><li><p>SVM for Soft Binary Classification</p><p>common approach: two-level learning</p></li><li><p>Kernel Logistic Regression</p><p>representer theorem on L2-regularized LogReg</p></li></ul><h3 id="lecture-6-support-vector-regression">Lecture 6: Support Vector Regression</h3><p>kernel ridge regression (dense) via ridge regression + representer theorem; support vector regression (sparse) via regularized tube error + Lagrange dual</p><ul><li><p>Kernel Ridge Regression</p><p>representer theorem on ridge regression</p></li><li><p>Support Vector Regression Primal</p><p>minimize regularized tube errors</p></li><li><p>Support Vector Regression Dual</p><p>a QP similar to SVM dual</p></li><li><p>Summary of Kernel Models</p><p>with great power comes great responsibility</p></li></ul><h2 id="topic-2-combining-predictive-features-aggregation-models">Topic 2: Combining Predictive Features: Aggregation Models</h2><h3 id="lecture-7-blending-and-bagging">Lecture 7: Blending and Bagging</h3><p>blending known diverse hypotheses uniformly, linearly, or even non-linearly; obtaining diverse hypotheses from bootstrapped data</p><ul><li><p>Motivation of Aggregation</p><p>aggregated G strong and/or moderate</p></li><li><p>Uniform Blending</p><p>diverse hypotheses, ‘one vote, one value’</p></li><li><p>Linear and Any Blending</p><p>two-level learning with hypotheses as transform</p></li><li><p>Bagging (Bootstrap Aggregation)</p><p>bootstrapping for diverse hypotheses</p></li></ul><h3 id="lecture-8-adaptive-boosting">Lecture 8: Adaptive Boosting</h3><p>optimal re-weighting for diverse hypotheses and adaptive linear aggregation to boost ‘weak’ algorithms</p><ul><li><p>Motivation of Boosting</p><p>aggregate weak hypotheses for strength</p></li><li><p>Diversity by Re-weighting</p><p>scale up incorrect, scale down correct</p></li><li><p>Adaptive Boosting Algorithm</p><p>two heads are better than one, theoretically</p></li><li><p>Adaptive Boosting in Action</p><p>AdaBoost-Stump useful and efficient</p></li></ul><h3 id="lecture-9-decision-tree">Lecture 9: Decision Tree</h3><p>recursive branching (purification) for conditional aggregation of constant hypotheses</p><ul><li><p>Decision Tree Hypothesis</p><p>express path-conditional aggregation</p></li><li><p>Decision Tree Algorithm</p><p>recursive branching until termination to base</p></li><li><p>Decision Tree Heuristics in C&RT</p><p>pruning, categorical branching, surrogate</p></li><li><p>Decision Tree in Action</p><p>explainable and efficient</p></li></ul><h3 id="lecture-10-random-forest">Lecture 10: Random Forest</h3><p>bagging of randomized C&RT trees with automatic validation and feature selection</p><ul><li><p>Random Forest Algorithm</p><p>bag of trees on randomly projected subspaces</p></li><li><p>Out-Of-Bag Estimate</p><p>self-validation with OOB examples</p></li><li><p>Feature Selection</p><p>permutation test for feature importance</p></li><li><p>Random Forest in Action</p><p>‘smooth’ boundary with many trees</p></li></ul><h3 id="lecture-11-gradient-boosted-decision-tree">Lecture 11: Gradient Boosted Decision Tree</h3><p>aggregating trees from functional gradient and steepest descent subject to any error measure</p><ul><li><p>Adaptive Boosted Decision Tree</p><p>sampling and pruning for ‘weak’ trees</p></li><li><p>Optimization View of AdaBoost</p><p>functional gradient descent on exponential error</p></li><li><p>Gradient Boosting</p><p>iterative steepest residual fitting</p></li><li><p>Summary of Aggregation Models</p><p>some cure underfitting; some cure overfitting</p></li></ul><h2 id="topic-3-distilling-implicit-features-extraction-models">Topic 3: Distilling Implicit Features: Extraction Models</h2><h3 id="lecture-12-neural-network">Lecture 12: Neural Network</h3><p>automatic pattern feature extraction from layers of neurons with backprop for GD/SGD</p><ul><li><p>Motivation</p><p>multi-layer for power with biological inspirations</p></li><li><p>Neural Network Hypothesis</p><p>layered pattern extraction until linear hypothesis</p></li><li><p>Neural Network Learning</p><p>backprop to compute gradient efficiently</p></li><li><p>Optimization and Regularization</p><p>tricks on initialization, regularizer, early stopping</p></li></ul><h3 id="lecture-13-deep-learning">Lecture 13: Deep Learning</h3><p>pre-training with denoising autoencoder (non-linear PCA) and fine-tuning with backprop for NNet with many layers</p><ul><li><p>Deep Neural Network</p><p>difficult hierarchical feature extraction problem</p></li><li><p>Autoencoder</p><p>unsupervised NNet learning of representation</p></li><li><p>Denoising Autoencoder</p><p>using noise as hints for regularization</p></li><li><p>Principal Component Analysis</p><p>linear autoencoder variant for data processing</p></li></ul><h3 id="lecture-14-radial-basis-function-network">Lecture 14: Radial Basis Function Network</h3><p>linear aggregation of distance-based similarities using k-Means clustering for prototype finding</p><ul><li>RBF Network Hypothesis</li></ul><p>prototypes instead of neurons as transform</p><ul><li>RBF Network Learning</li></ul><p>linear aggregation of prototype ‘hypotheses’</p><ul><li>k-Means Algorithm</li></ul><p>clustering with alternating optimization</p><ul><li>k-Means and RBF Network in Action</li></ul><p>proper choice of # prototypes important</p><h3 id="lecture-15-matrix-factorization">Lecture 15: Matrix Factorization</h3><p>linear models of movies on extracted user features (or vice versa) jointly optimized with stochastic gradient descent</p><ul><li>Linear Network Hypothesis</li></ul><p>feature extraction from binary vector encoding</p><ul><li>Basic Matrix Factorization</li></ul><p>alternating least squares between user/movie</p><ul><li>Stochastic Gradient Descent</li></ul><p>efficient and easily modified for practical use</p><ul><li>Summary of Extraction Models</li></ul><p>powerful thus need careful use</p><h3 id="lecture-16-finale">Lecture 16: Finale</h3><ul><li><p>Feature Exploitation Techniques</p><p>kernel, aggregation, extraction, low-dimensional</p></li><li><p>Error Optimization Techniques</p><p>gradient, equivalence, stages</p></li><li><p>Overfitting Elimination Techniques</p><p>(lots of) regularization, validation</p></li><li><p>Machine Learning in Practice</p><p>welcome to the jungle</p></li></ul><h1 id="notes">Notes</h1><ul><li><p>Terms</p><ul><li>正则 岭回归</li><li>伪逆矩阵</li><li>离散函数</li><li>拉格朗日对偶法</li><li>拉格朗日最小化</li><li>拉格朗日乘法</li><li>空间转换</li><li>原始SVM</li><li>对偶SVM</li><li>Decision Stump</li></ul></li><li><p>L1-5 25</p></li><li><p>L2-19</p></li><li><p>L4-18 19</p></li><li><p>L5-4 5 9 15 17</p></li><li><p>L7基于以上的两个例子,我们得到了aggregation的两个优势:feature transform和regularization。我们之前在机器学习基石课程中就介绍过,feature transform和regularization是对立的,还把它们分别比作踩油门和踩刹车。如果进行feature transform,那么regularization的效果通常很差,反之亦然。也就是说,单一模型通常只能倾向于feature transform和regularization之一,在两者之间做个权衡。但是aggregation却能将feature transform和regularization各自的优势结合起来,好比把油门和刹车都控制得很好,从而得到不错的预测模型。</p></li><li><p>L07-20 bootstrap aggregation (BAGging): a simple meta algorithm on top of base algorithm A</p></li><li><p>L07-21 值得注意的是,只有当演算法对数据样本分布比较敏感的情况下,才有比较好的表现。</p></li><li><p>L08-20</p></li><li><p>L09-10</p></li><li><p>L10-2Bagging能减小variance,而Decision Tree能增大variance</p></li><li><p>L11-10</p></li><li><p>L12-6感知机模型每个节点的输入就对应神经元的树突dendrite,感知机每个节点的输出就对应神经元的轴突axon</p></li><li><p>L12-13 16GD 21</p></li><li><p>L14-5 9 10</p></li></ul>]]></content>
<summary type="html">
<p>Machine learning is the study that allows computers to adaptively improve their performance with experience accumulated from the data observed. Our two sister courses teach the most fundamental algorithmic, theoretical and practical tools that any user of machine learning needs to know. This first course of the two would focus more on mathematical tools, and the other course would focus more on algorithmic tools.</p>
</summary>
<category term="Computer science" scheme="https://github-9233.github.io/categories/Computer-science/"/>
<category term="Machine learning" scheme="https://github-9233.github.io/categories/Computer-science/Machine-learning/"/>
</entry>
</feed>