[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Please reconsider removing the `defadvice' doc (ref: bug #16491)
From: |
Drew Adams |
Subject: |
Please reconsider removing the `defadvice' doc (ref: bug #16491) |
Date: |
Sat, 9 Aug 2014 14:19:46 -0700 (PDT) |
You want to push your replacement for `defadvice'. OK, that's
presumably all well and good. No doubt it is the best thing
since sliced bread (ou le fil a couper le beurre).
But it is a real disservice to users to have removed all of
the information about using `defadvice' from the Elisp manual.
I *really* hope you will reconsider this decision.
Lots of users write or maintain code that works across multiple
Emacs versions, including code that uses `defadvice'. Caring
about Emacs users means not throwing obstacles at them
unnecessarily. Making them hold onto Emacs 24.3 just to have
access to the `defadvice' doc is uncalled-for annoyance.
Such archaeological digging by users should not be necessary.
I can imagine some newbie developer having such a mindset, but
it is disappointing - and yes, shameful - that the leadership
of GNU Emacs has come to this.
At least please consider making this doc available as a
separate Info manual, if you insist on not letting users see
it in the Elisp manual. Call it "Stupid-Old-Defadvice Manual"
if you like. Discourage using `defadvice' anyway you like,
short of depriving users of information about it.
And no, "I kept the 60KB of docs in the Commentary section
of advice.el" is *not* an answer. You might as well have
said that you tossed 600 kilos of old literature in the
dumpster. "60 KB"! What an attitude!
And no, this is not about me. I keep multiple Emacs versions
anyway. It is only a relatively minor annoyance to have to
bring up an older version to get to this doc.
Not everyone who uses or wants to understand `defadvice' in
existing code will keep multiple Emacs versions, or will know
what version to backtrack to, to find this doc. And even a
minor annoyance imposed gratuitously on lots of users is
better avoided, especially for something as important and
complex as advising functions.
Bug #16491, with the same request to restore this doc, was
marked "done" back in January. Is "done" the new "wont-fix"?
According to NEWS, `defadvice' is not even deprecated. And
even if it were to become deprecated, it would presumably
- it should - continue to be supported for quite some time.
And documented.
- Please reconsider removing the `defadvice' doc (ref: bug #16491),
Drew Adams <=