On 14 October 2017 at 09:00, Eli Zaretskii <eliz@gnu.org> wrote:
> From: Reuben Thomas <rrt@sc3d.org>
> Date: Sat, 14 Oct 2017 08:10:44 +0100
> Cc: João Távora <joaotavora@gmail.com>,
>       Sami Kerola <kerolasa@iki.fi>, emacs-devel@gnu.org, Alan Mackenzie <acm@muc.de>,
>       Eli Zaretskii <eliz@gnu.org>, Noam Postavsky <npostavs@users.sourceforge.net>, sdl.web@gmail.com,
>       Stefan Monnier <monnier@iro.umontreal.ca>
>
> These days, it seems much better to use Flycheck than Flymake (that's certainly what I do). See
> https://github.com/flycheck/flycheck
>
> It would be a pity for Flymake to become yet another part of Emacs that developers spend time updating and
> users largely ignore; better to spin it off into ELPA, and if people still want to work on it there, fine. Meanwhile,
> why not use Flycheck by default (in the same way as we've "in-sourced" Org and other packages)?

I don't understand: Flycheck is an external package; why should we
prefer it to Flymake, assuming that the latter will become supported
well by the built-in major modes?


And I suggested precisely bundling Flycheck with Emacs.​

IOW, what I see here is a serious effort to make Flymake a
sophisticated and flexible syntax-checking tool bundled with Emacs.  I
don't see why should we object to such an effort, when one of our
major goals is to provide a modern program development environment.

​Because with Flycheck this is already accomplished. Why not work instead on things that Emacs lacks?​ There are already far too many duplicated packages, leading to duplicated maintenance effort.

--