Rephrase --recording on CLI Page

[ann0see]

@gilgongo this needs to be rephrased. I think it's not well written. What it does is basically making recording available but it needs to be enabled via JSONRPC or SIGUSR

Originally posted by @ann0see in #747 (comment)

[pljones]

Above is incorrect.

By default, recording will start whenever there's something to record, if a recording directory is given.

To prevent that behaviour, --norecord must be supplied on the command line.

[ann0see]

As I said, it needs to be re-phrased. At least it's a bit confusing. If -R and --norecord is set, the server can record, but doesn't. If -R is set, the server records by default, I think?

[pljones]

sets directory to contain recorded jams is insufficient. It's never just done that. It's always also turned on the "record when a session is active" state. The text that's there now isn't what I proposed.

disables recording (when enabled by default by -R) is also misleading as it doesn't prevent recording (the "disabled" wording may also be in the GUI with this wrong meaning). What it does is set the default state to not record when a session is active.

The GUI allows that state to be toggled. It also also allows triggering a new recording. Both those actions are also available through SIGUSR1/2 (on *nix) and JSON-RPC.

So let's see...

-R, --recording       set server recording directory (server will record when a session is active by default)
    --norecord        set server not to record by default when recording is configured

How's that?

[ann0see]

Yes. I'd stress the fact that by default recording is enabled a bit more (not in parentheses)

Maybe -R could in future take two parameters: -R (recording directory, default recording state[on,off]). Probably requires the CLI parsing to be adapted

[pljones]

Probably requires the CLI parsing to be adapted

We could have an optional alternative syntax of -R [0|1];[path], so if it starts 0; or 1; it's the flag for off/on by default, otherwise it's the existing on by default behaviour. That's backwards compatible and removes the need for --norecord (which would still be supported as we love being backward compatible).

[ann0see]

Yes. Sounds good.

[ann0see]

@pljones since the translation hasn't started yet for the website, would you mind changing this and open a PR on the website with your proposal: #756 (comment) ?

[ann0see]

Actually this is probably the text which I thought was already changed in the app.

[pljones]

Mm, I don't think this has had an issue in the jamulussoftware/jamulus project raised to fix the app yet, no.