bug#50331: 28.0.50; Propose to obsolete cwarn.el

[Zhiwei Chen <condy0919@gmail.com>]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Zhiwei Chen <condy0919@gmail.com>
>> Date: Thu, 02 Sep 2021 14:30:44 +0800
>> Cc: andlind@gmail.com
>> 
>> # conclusion
>> 
>> Since the first two cases are not existent, and the last one is false
>> positive (IMO), I propose to obsolete cwarn.
>
> I don't understand how you arrived at that conclusion.  cwarn provides
> the diagnostics while you write the code, without any need to invoke
> the compiler, and potentially long before you submit the code to the
> compiler.  Some people might prefer that, so why inconvenience them?

Why would I like to obsolete cwarn?

Actually, I'm making an introduction to Emacs builtin modes (in Chinese)
to see how powerful the vanilla Emacs is. But I found, Emacs is not so
orthogonal in functionality. e.g., `whitespace-mode' provides
`whitespace-cleanup' while simple.el has `delete-trailing-whitespace'.
Is it time to do subtraction for Emacs? There are too many packages
merged into Emacs, but a few are removed from Emacs (to elpa maybe).

The main reason for obsoleting cwarn is that its functionality can be
superseded by flycheck/flymake. The result of
https://grep.app/search?q=cwarn-mode shows that there are few users of
cwarn. Of course this result may be wrong, after all https://grep.app/
only searches the git repo and all data on the Internet. The safe way to
do this is to raise a poll and let the community decide if cwarn should
be removed. Since Emacs is such a monster of 40yrs, the expected result
is that most users never know there is a package named cwarn.

[1]: https://emacs-china.org/t/emacs-builtin-mode/11937

-- 
Zhiwei Chen