Re: Way kewl new bug reportin thingy for 31337 XEmacs h4x0r d00dz
On Mon, 6 Aug 2001, Charles G. Waldman gibbered:
In fact, I believe that since .emacs can contain any valid Lisp
expression, the problem of determining what is and is not a password
is equivalent to solving the Halting problem....
It's worse than that. Because the user may have passwords in there which
will be used by packages that aren't installed yet (I have a couple like
that for things I haven't got around to installing) the bug-reporting
package would also need to be either telepathic, or clairvoyant, or of
human-level intelligence, or all three.
Also, it's no use for complex .emacs's: Mine has, just after the start,
(require 'init-prog-modes) ; Initialize programming-language-related stuff.
(require 'init-message-modes) ; Initialize email/news-related stuff.
and a bunch of others. Working out which `require' is needed could be
tricky.
(Plus, there's >2000 lines of sitewide config. You want that mailed as
well? And all locally installed and patched site-packages? This starts
to sound like a DoS attack...)
The replicability problem with (X)Emacs bugs on sites with complex
configuration will always remain, I think; the good part is that sites
with complex configuration probably have a local (X)Emacs guru who can
be trusted to do most of the debugging work on his own...
--- btw, my line_number_cache corruption patch from last month is still
quasi-unreviewed...
--
`It's all about bossing computers around. Users have to say "please".
Programmers get to say "do what I want NOW or the hard disk gets it".'
-- Richard Heathfield on the nature of programming