RE: Is Elisp really that slow?

[Drew Adams <drew.adams@oracle.com>]

> We have this discussions so often because we only get the developer's and
> old users opinion. But not a general public survey to measure global
> opinions.

The opinions of Emacs developers and experienced Emacs
user are the most helpful opinions, including in terms
of guiding where Emacs can and should go.

The opinions of non-users and non-developers of Emacs
are useful mainly in terms of discovering their
misconceptions and their current, often limited or
unfortunate experience with applications only somewhat
similar to Emacs (editors and other).

We have this discussion, especially about key bindings,
so often because there are _lots_ of non-users of Emacs
who discover Emacs, realize it's something powerful and
potentially even more powerful, but who have ingrained
key-binding etc. habits that they imagine are superior.

Eli nailed it when he gave his list of the kinds of
things Emacs really needs to improve its usefulness,
especially for programmers (e.g. IDE features).  It's
NOT at all about key bindings and such.

Perhaps most (nearly all?) newbies to Emacs initially
feel that the key bindings and interactions they were
used to before Emacs are superior or easier.  That's
been the case since Day One.

In general (and no, not always), they've been wrong
about that.  When they've been right they've just
customized their Emacs to do what they prefer.  And
in some cases Emacs has itself ended up changing
default behavior along the lines of what they've
implemented as, initially, personal customizations.

Yes, we took decades to turn on `transient-mark-mode'
by default.  And yes, we still haven't turned on
`delete-selection-mode' by default (something I'm in
favor of, FWIW).  So what?  It's so easy to turn
something on or redefine something in your init file.
The argument that Emacs should change to reflect,
out of the box, what most non-Emacs users are used to
is super-flawed, and hollow.