Re: [NonGNU ELPA] New package: project-store

Lin Jian <[email protected]> writes:

> Richard Stallman <[email protected]> writes:
>
>> When you distribute this program via Nix, people who see it listed
>> there will know it relates to Nix.  But people who see it in ELPA may
>> have no idea of that.  I'm looking for a way to make that immediately
>> apparent when people come across the package in the ELPA list of
>> packages.
>
> In most cases, when people see ELPA packages, besides package names,
> they also see package brief descriptions.
>
> In my Emacs, I can see package brief descriptions even in
> `package-install', thanks to my completion config.

We should keep in mind that this is not generally the case.

> IMHO, project-store, together with its brief description, is very clear.

This is ultimately the reason why we don't want to insist on maintainers
changing the names of their packages if they have strong opinions to the
contrary.

But still: I asked a few friends who use Nix about what their
association with the word "store" is an if they would think of "Nix",
and nobody said they think it is intuitive.  Generally, "project-store"
seems to either be interpreted as "project-shop" (noun) or
"project-save" (verb).  I think the argument with Guix is also not that
strong, since the compatibility with Guix is coincidental, rather than
inherent.  It is really project support for "Nix stores" or anything
that behaves similarly.

A name is part of the public advertisement of a package, and (some of
us) appreciate having a name have a standalone meaning -- without the
help of a description.  While usually the issue is with project names
being too quirky, the other direction is also tricky.  I have witnessed
this myself with my "setup" package, where people have complained
numerous times that it is too generic, but that is also not the same
name as with your package, if we are being fair.

This is not to say that we have to change the name, but just to argue
the point that it doesn't seem like a _good_ name.  If we want to
continue this discussion, we ought to do so on emacs-tangents.