"Stephen J. Turnbull" <turnbull(a)sk.tsukuba.ac.jp> writes:
Yoshiki> Why? It might become a little bit slower for large
Yoshiki> English man page because of regexp search, but other
Yoshiki> than that it does no harm. I can't imagine a situation
Yoshiki> where
Speed is no issue here, man pages just can't get that big.
Speed is definitely an issue with large English man pages. First,
nroff takes a while to process them, and Emacs takes time to clean
them. This is less dire on newer machines, but man pages really do
get that big (try `man zshall' on a slower computer.)
Unfortunately, I have no useful input for Japanese stuff, I just
wanted to point out that the speed problem for man pages is real.
Yoshiki> I'm reluctant to do the same thing for Chinese
and
Yoshiki> Korean characters as I don't know how they localize man
Yoshiki> pages.
...why not give these other guys a chance of fixing the same problem
_without_ needing to read the code? With the defcustom, _they_ can
decide for themselves even if they don't grok elisp.
Ideally, XEmacs should just do the right thing, either by hardcoding
the stuff or by guessing it. We should generally be careful about
introducing new user-options.