RE: [h-e-w] EmacsW32, gnuserv, pathes in .emacs

[Drew Adams]

    >     The philosophy is that the default Emacs should
    >     work well enough for the new user not to be
    >     bothered by customization for quite some time.
    >
    > It's not a question of bother. Setting preferences in most UI
    > applications is not something users do only exceptionally and
    > only because they are bothered. This is all the more true of Emacs
    > customization. Customization is a part of normal Emacs use even
    > more than it is part of the use of other apps. Emacs *is*
    > customization, in oh so many ways. Do you want to
    > customize? Get Emacs. It doesn't matter what - you can
    > customize anything with Emacs.

    My point is that customization is a sufficiently advanced issue, so
    new users should get used to Emacs and read the docs for some time
    before they embark on the never-ending customize adventure.
    Meanwhile, they have the Options menu to start with.

Yes and no.

The Options menu is certainly a good place to start. The customization
mini-tutorial should start with the Options menu. Thanks for reminding me of
it. And, since those same options also live in Customize Land, the Options
guided tour leads naturally to an optional, short day trip yonder.

Customization can be as complex as you like. My initial suggestion was to
have a tutorial lead you by the hand on a very simple, short trip through
Customize Land - not to just dump you there naked, in the dark, and run,
laughing.

Users should be made to feel at home with Customize, and learn its basics
(even the basics can be confusing, because the Customize UI is not yet all
it should be). They should learn a simple, useful interaction: changing a
value, saving a value, resetting a changed (and possibly saved) value to its
default setting. I suggested having them change some innocuous setting that
affects appearance (so they can see the effect immediately). Have them open
.emacs to see the saved setting (but *DO NOT TOUCH* - that's an important
lesson, BTW).

This simple exercise would give users a feeling of confidence with
Customize, a bit of courage to confront it later with more important changes
in mind, and a certain familiarity with the basics. They will learn soon
enough (alas, all too soon) about its pitfalls, *regardless* of whether we
first lead them on a safe journey through this wild\(er\)?ness. Better to
know one safe route through the jungle than to know nothing about it. Then,
when you have to head out on a cold, black night to customize some beast
you're unsure about, it's better to at least recognize some of the
territory.

Customize, in spite of the fact that it doesn't yet live up to its promise,
was designed with newbies and non-lispers in mind. We repeatedly advise
users that this is the way to customize Emacs, as opposed to just fiddling
with .emacs. If we can't even risk taking them around the Customize block by
the hand once in the daylight, without getting beat up, then we're really in
bad shape. If Customize is not at all for the faint at heart, then we
shouldn't be advertising it.

Again, I'm not saying that customization should be lesson #1 of the
tutorial, but it should be available as an optional tunnel to explore with a
flashlight and sexy Ms. Tutorial as guide.

And, as I said, it's also important to offer a guided tour of 1) Help and 2)
using the manual. All of that is part of the Emacs Experience. When you take
your first trip to Disneyland, you at least want to see a little of each
significant part of the place. And if you're a naked newbie, then you want
someone to hold onto you as you take your first ride.