Re: Making Emacs more friendly to newcomers

[조성빈]

Po Lu <address@hidden> 작성:

> 조성빈 <address@hidden> writes:
>
> > I can’t understand why you’re so opposed to changing defaults. Defaults
> > matter, and I think that users who are already accustomed to Emacs will
> > have an easier time changing options. (Like adding `(vanilla-mode 1)`
> > to init.el.)
>
> I oppose changing defaults, because I value my tools remaining stable.
> Having to frequently adjust my user-emacs-directory to encompass the
> latest fancy new idea is an unacceptable maintenance burden to me.

Then you can turn on ‘vanilla-mode’ on your init.el and call it a day.
‘vanilla-mode’ will be the current situation of Emacs - it will try to
be stable.

> Not
> to mention in practice, those changes are going to be far larger than a
> hypothetical `vanilla-mode'.

Why? If you’re fine with the current level of changes in Emacs, that would
be the level of changes in ‘vanilla-mode’.

> > Adding a variable that makes Emacs warn if it’s version number is
> > different from its value might work.
>
> I'm not sure what you're talking about; Variables don't persist across
> Emacs sessions.

I’m saying that one can add a variable like ‘expected-emacs-version’ and
when Emacs is loading init.el and ‘expected-emacs-version’ is different
from the current version, Emacs can warn you.

> > The features that are best discoverable are the ones that are default.
>
> And since changing the defaults radically end up hurting existing users,

No, it doesn’t. Better defaults help users and allow them to lighten their
configuration.

> and also existing packages and potentially the existing ecosystem.  You
> have to find a balance between that, and changing the defaults is not
> the balance you want.

The balance I’m suggesting (actually ndame) is to add a ‘vanilla-mode’ for
people who value stability over features.

> > Defaults have to change for experienced users to adjust. And for them,
> > adjusting is an easy task, not something that takes a decade.
>
> "For existing users to adjust" is not an excuse -- that puts an extra
> burden on users.

No, it doesn’t - see previous comments.