Re: isearch multiple buffers

[Dan Nicolaescu]

Eli Zaretskii <address@hidden> writes:

  > > Cc: Miles Bader <address@hidden>, address@hidden
  > > From: Dan Nicolaescu <address@hidden>
  > > Date: Fri, 26 Oct 2007 08:16:26 -0700
  > > 
  > > Eli Zaretskii <address@hidden> writes:
  > > 
  > >   > > Because the 8.3 name restriction has a negative effect for _all_
  > >   > > developers.
  > >   > 
  > >   > What negative effects? can you show them in the archives?  AFAIR, this
  > >   > is about the first such issue in years.
  > > 
  > > Here are quite a few exmaples: go to http://search.gmane.org/ search
  > > for 8.3 and use "gmane.emacs.devel" for "In group". It produces a good
  > > amount of hits. 
  > 
  > All I see, in addition to this thread is the following:
  > 
  >   . An issue with foo-+.texi in June 2004
  > 
  >   . A couple of messages about ChangeLog-unicode in February 2005
  > 
  >   . A discussion back in August 2005 when MH-E was added to Emacs, and
  >     a related discussion in September that year about image files.
  > 
  >   . A comment by Kim Storm in June this year about renaming a single
  >     file.
  > 
  >   . A few other hits that have nothing to do with conflicts in file
  >     names due to 8+3 restrictions, such as the discussion in May 2004
  >     about the doc string of convert-standard-filename
  > 
  > So there are, like, 4-5 instances since 2004---more than I thought,
  > but still very much negligible.

I wouldn't call it negligible, it is annoying. And that is just one
search, looking other keywords might find more stuff.  And again, the
feeling of a lot of people on this list is that this dealing with this
is unjustified, as the platform does not seem to be too much alive

  > > Another situation where the DOS port had to be taken into account:
  > > recently I wanted to clean up the code and get rid of the MULTI_KBOARD
  > > #ifdefs. You asked me not to because the MSDOS port might need it. I
  > > am 100% convinced that it won't be needed, but preferred to drop the
  > > issue rather that try to convince you of this. 
  > 
  > So you are saying that my stubbornness is also a problem related to
  > the MSDOS port?

I did not say, nor did I want to imply anything about "stubbornness",
I am sorry if that's what you read into this. I just felt that it
wasn't worth my time or yours to try to convince you about
this. (which is quite possible with the right technical arguments).

  > > Also, there are a LOT of DOS #ifdefs in the code that we could get rid
  > > of, that should simplify future maintenance. 
  > 
  > You can say this about many other supported platforms.  Look at
  > sysdep.c, for example, or at process.c.

Exactly, if there are many other systems that are dead, we should just
delete the corresponding cruft.

  > This thread was only about 8+3 file-name limitations.

I might have confused the thread, I thought the discussion was in
general about if it was about if keeping the MSDOS port altogether.