Does YDL use unreleased/buggy redhat-versions of KDE and XFree86?

[Stefan Bruda]

So here is the situation: I am a YDL user exclusively, except in our
departmental labs where we have Intel boxes and our sysadmin chose to
go with RedHat (8.0 at the time of writing).  My opinion abour RH 8.0?
Disappointing (RH 8.0 that is, not my opinion).

At 18:36 -0700 on 2003-3-23 Dan Burcaw wrote:
 >
 > Well there's a lot of mis-information here.  Red Hat's KDE 3.1 has no 
 > patches related to Red Hat's Bluecurve. All of the theme code is handled 
 > via the theme engine. So, if you change themes, the Bluecurve code no 
 > longer runs.  

I use Gnome, so I would not know about KDE, but why should one go
through hoops to change the window manager?  I don't like metacity, I
like sawfish, why should I go to `killall metacity; sawfish' in order
to accomodate my (however subjective) preference?

I certainly hope it is not that hard to change the KDE theme.

Some students of mine in the C++ course discovered the hard way that
the terminal on our RH boxes refuses to print anything following the
last newline in the output, flushing or not.  Of course, not putting a
newline at the end of output is pointless, but these are students
taking a 200-level course, sooner or later they will all do pointless
things (that's how one learns) and whenever their terminals do not do
what they expect from them they get even more confused.

 > Red Hat does a good job, and we do use there stuff... and I 
 > don't feel Red Hat is damaging free software projects at all.

This is not necessarily RH's fault, I would appreciate somebody
confirming or infirming it: So I had the other day this student with a
program segfaulting on him without any apparent reason.  We stayed
together a pretty long time to figure out where the problem is, but
nobody looked into his makefile (after all, make did its stuff without
any problem, eh).  What did my student do?  He forgot to mention the
-lpthread in his linking target, and so the program segfaulted upon
the first encounter of pthread_create!

Now my good ol' GCC 2.9x generates, as I believe to be normal, a
linking error.  So why does GCC 3.2 just links the damn thing, come
hell, high water, or missing references?  This made me loose a couple
of hours, and my student loose some marks (he did not have time to
complete the assignment in time).

In all, I have been (and still am) _very_ reluctant to upgrade my main
machine to YDL 3.0 since I heard it is RH 8 based.  I am simply afraid
that it will look too much as our lab machines.  Yes, you could
consider all of these minor problems, and I will by all means agree
(though I consider the absence of a linking error in GCC 3.2 pretty
serious).  But then I do use my machine, not play with it.  I have
little time (and even less energy) to go around annoying "features,"
no matter how minor they are.

I was thus wondering whether anybody on this list can confirm (or
better yet, infirm ;-) ) that I will get the same problems should I
decide to upgrade.  I really appreciate your opinion on the matter.

Cheers,
Stefan

-- 
If it was so, it might be; and if it were so, it would be; but as
it isn't, it ain't. That's logic.
    --Lewis Carroll, Through the Looking-Glass