Re: Proposal for an Emacs User Survey

[Thibaut Verron]

Le ven. 16 oct. 2020 à 09:24, Marcel Ventosa <mve1@runbox.com> a écrit :
>
> On Fri, 16 Oct 2020 08:52:49 +0200
> Thibaut Verron <thibaut.verron@gmail.com> wrote:
>
> > Le ven. 16 oct. 2020 à 08:03, Marcel Ventosa <mve1@runbox.com> a
> > écrit :
> > >
> > > On Thu, 15 Oct 2020 23:59:07 -0400
> > > Richard Stallman <rms@gnu.org> wrote:
> > >
> > > > I hope that only a minority of Emacs users know about MELPA, and
> > > > I'd rather not inform the rest about it.  But if something is
> > > > going to inform them anyway, it is better to do it with a
> > > > denunciation.
> > >
> > >
> > > I've been using Emacs (and MELPA) for the best part of a decade and
> > > knew nothing about this! I'm concerned to use only free software and
> > > actively avoid proprietary software, so this is a bit of a shock.
> >
> > As I understand it, Melpa packages cannot *be* or *install* non-free
> > software. But some will not work without such software, which can in
> > theory encourage users to install it.
> >
> > So unless you yourself installed non-free software, Melpa cannot have
> > made your Emacs configuration non-free by accident.
>
> I understand, thanks for the explanation. In that case, I think I'm
> well informed enough to have avoived the dangers. I wonder how many
> people are not.

I personally don't think many users install non-free software because
they saw it wrapped in a Melpa package.

Taking the example of emacs-lastpass given above, I don't see how
anyone would even find this package without searching for it with the
keyword "lastpass".

The audience, rather, is users who are currently using Lastpass in
their browsers but are interested in bringing some of their online
activities to Emacs, but rely on their password manager to do so.

In due time, with the new "taste of freedom", they might even switch
to Keepass or Bitwarden (note: I don't use a password manager in
Emacs, so I have no idea of the quality of each support package,
beyond their existence), but in any case, I believe that for those
users, the existence of transitional solutions is a good thing.

I absolutely support the fact that Melpa is not activated by default,
and that there should be a warning about the existence of those
packages everywhere possible. But I still consider that the value of
those packages outweigh their dangers, just like the win32 build of
Emacs.

> > ELPA means Emacs Lisp Package Archive, so both Melpa and GNU Elpa are
> > ELPA's. I think that commonly referring to GNU Elpa as simply Elpa
> > (which I am also guilty of) is a bigger source of confusion than
> > Melpa and GNU Elpa sharing the same suffix.
>
> As written on the MELPA Github about page:
>
> "MELPA is Milkypostman's ELPA or Milkypostman's Experimental Lisp
> Package Archive if you're not into the whole brevity thing."
>
> Perhaps the shared `E' in `Elpa' is purely coincidental?

I didn't know about the Experimental version, thanks. If anything, the
quoted sentence is self-contradicting for me.
Also, the title of the melpa.org page is "MELPA (Milkypostman’s Emacs
Lisp Package Archive)".

In any case, I don't know if an acronym with a different vowel but
still ending with LPA, such as Milpa, would be less confusing.