This is the mail archive of the
mailing list for the glibc project.
Re: [libc-alpha] Re: Wish for 2002 ...
- From: Shawn Starr <spstarr at sh0n dot net>
- To: Kaz Kylheku <kaz at ashi dot footprints dot net>
- Cc: libc-alpha at sources dot redhat dot com
- Date: Sun, 13 Jan 2002 17:01:13 -0500 (EST)
- Subject: Re: [libc-alpha] Re: Wish for 2002 ...
By Common user i mean the casual C programmer/developer. Not regular non
On Sun, 13 Jan 2002, Kaz Kylheku wrote:
> On 13 Jan 2002, Shawn Starr wrote:
> > > You are confusing Unix with some kind of ideology, or ``happy family''
> > > or something. Get over your stupid nostalgic emotions and start thinking
> > > rationally.
> > No, I'm concentrating on API functionality. Get off the 'zelotishness'.
> > You seem to concentrate on keeping the glibc library wrapped in politics
> > and forget about the COMMON USERS who write programs in C.
> Don't you realize that it is political to stand on a soapbox and
> pretend to speak on behalf of a mass of common users? Where is the
> demographic data to back up any of these claims? If you don't have it,
> at least have the courage to speak for yourself only and say that you
> want some functions in the library for your own personal satisfaction,
> and not out of any genuine concern for anyone else.
> Peple who write programs in C are not ``common users''. The GNU C
> library has lots of users who don't write programs in C, or in anything.
> They need to depend on this library to be stable and upgradable.
> I do know know a little bit about how naive programmers write in C,
> from years of answering questions in comp.lang.c. I can't recall the
> last time strlcpy or strlcat have come up in any questions, or appeared
> in anyone's program.
> An all-time Google search over comp.lang.c for strlcpy or strlcat reveals
> a grand total of seven articles (not counting followups to these which
> quote the metion). They were all posted within the past two years, and
> all of these articles were written by expert programmers, who mention
> the functions in passing or recommend them. Four of these seven were
> written by a regular named Tor Rustad who appears to advocate their use
> (specifically, borrowing the implemenations from OpenBSD, not blindly
> assuming that they are just available!)
> Also consider this: an all-time search for strlcpy or strlcat in the
> newsgroup comp.os.linux.development.apps reveals but one article!
> The article is not about strlcpy. The author quoted quoted the
> setproctitle() function from sendmail in response to someone's question,
> and that function happens to use strlcpy().
> There is not one shred of discussion about the function itself.
> If there was a significant demand for these functions by common
> programmers, don't you think that it would have manifested itself in at
> least a few questions on Usenet?
> Okay, let's widen the search to any newsgroups that contain .linux.
> or .gnu. as a component in their name. Alas, at last, here is someone
> asking in comp.os.linux.misc. But when we look at the article, it comes
> to light that it's some severely confused individual who apparently
> thinks that C library functions are shell commands. See Message ID
> Again, no actual discussion of the functions to be found, just mentions
> of these identifiers in make outputs, snippets of code and the like.
> So where is this mythical mass of programmers who are clamoring for
> strlcpy and strlcat on GNU Linux?