Re: Stability of core packages (was: Not easy at all to upgrade :core packages like Eglot)

[João Távora]

On Wed, Apr 19, 2023 at 7:32 PM Eli Zaretskii <eliz@gnu.org> wrote:

> No, it cannot, and this and the sibling discussions show why: the
> package maintainers are biased in favor of their packages.  That
> (completely understandable and expected) bias prevents them from
> seeing the overall picture objectively.

What is the part of the picture that I'm not seeing?  Who am I
ignoring, who am I missing? What are these people that I'm
(IYO unwillingly, fortunately) biased against.  What exactly are
do you conjecture or know they are doing right now in Emacs 28 that
my patch will prevent, affect, break, make difficult?  Just one
example, please.

> > Or make it a defcustom if you're really worried.
> That doesn't change the picture, unless the default for the defcustom
> will be nil.  Which I expect João to object to, because he wants Eglot
> to be updated by default and automatically.

And _only_ Eglot.  Because, that's what happened to it in Emacs 28.

> > > You assume that everyone will
> > > want Eglot and use-package automatically updated, but this assumption
> > > has no real basis.
> >
> > People don't call 'M-x package-install' automatically, nor do they put
> > those calls in their init files automatically.
>
> That's factually incorrect, AFAIU.  Moreover, the cases that bothered
> João (again, AFAIU) were exactly those which you say don't exist.

Yes, you're right Eli.  I pointed to numerous issues collected over
the years, where I found proof that users have this kind of form
int heir configurations:

(use-package eglot :ensure t :config() )

Even ChatGPT suggests this. Just ask it "How do I configure the Eglot
Emacs extension for the <Your favourite language> LSP server"?

João