RE: [External] : Re: NonGNU ELPA work

[Drew Adams]

> > I don't use GIT to manage my code.  I just post
> > updates to EmacsWiki.  Nearly all of my code
> > used to be on MELPA, but they stopped pulling
> > from Emacs Wiki (my pages on the wiki are
> > locked to all but the wiki admins, FWIW).
> 
> MELPA banned any packages from EmacsWiki on MELPA because it is
> insecure to have automated installation of packages directly from a wiki.

Stopped supporting, not "banned".

> https://github.com/melpa/melpa/issues/2342

https://www.reddit.com/r/emacs/comments/7vocqa/update_on_melpa_removing_emacswiki_packages_they/dtuhzmt/

> > As for overlap or redefinition of vanilla
> > constructs: yes, that's no doubt a problem if
> > someone expects to somehow integrate it with
> > Emacs.  Loading some libraries redefines
> > vanilla code.  Users of the code need to know
> > that.
> 
> This is easy to fix by adding a prefix to your definitions.

No.  The point is to redefine a vanilla thing,
so that any code that uses that thing uses the
replacement instead.  That's a choice.  It's
the point of my redefinitions.  Like it or not.

Often, of course, either advice or a minor mode
can be used to get similar opt-in behavior,
which can allow turning it off.  And I do
provide minor modes etc.  And I do use library
prefixes.  But not when the point is to replace
a predefined thing.

Using a library is opt-in.  So is loading it.
Doc for a library should of course make clear
what's involved.  There should be no surprises.

Your "that's easy to fix" is simplistic.  Life
is complicated.  One person's "broken" can be
another person's "by design".  The history of
art, science, and everything else is full of
breaking on purpose (with good and bad results).