"Write a new package" culture instead of patches?

[Stefan Kangas <stefankangas@gmail.com>]

Dmitry Gutov <dgutov@yandex.ru> writes:

> Another example is elegant-emacs, suggested in yet another thread by
> Nicolas P. Rougier. There's nothing stopping us from featuring it in GNU
> ELPA (right?), but we would get the most value if we really examine it
> and look for pieces to put into the vanilla Emacs by default.

Yes, this is the correct approach in many cases.

This reminds me of something else:

There's a general problem that when package <foo> lacks small feature
<bar>, some users don't see this as a chance to write a patch for <foo>.
Instead, they write a new library <foo>-<bar> to add this feature.

Sometimes, of course, this is the correct choice.  But I've seen some
very small packages just to basically patch this or that annoyance in a
package, or in core.  For example:

    https://github.com/Fuco1/eshell-bookmark/issues/1

(FWIW, I think we should have a policy to reject such packages on
technical grounds (and ideally MELPA would do the same).)

Now, this is an extreme example, but many more could be found.  Why are
the authors of "helpful.el" not helping us mainline some of their great
innovation, for example?

Has anyone else thought about this?  Is it correct to say that such a
"package first" culture has developed?  If yes, why has it developed,
and is there anything we could do about it?

Best regards,
Stefan Kangas