Re: FC4 good new tech, bad legacy support

[Richard Kelsch <rich csst net>]

Steffen Kluge wrote:

On Wed, 2005-06-29 at 01:52 -0700, Richard Kelsch wrote:
  

Good luck on getting many non-rpmed Perl CPAN modules to work, even 
though they worked fine in FC3.  Not everyone is a C programming master 
with a PHD.  Trying to figure out why someone in their right minds would 
make a compiler not compile code it's previous versions compiled quite 
happily is just beyond all logic in my opinion; especially since no 
English readable error is generated, except something cryptic that only 
a hippie-haired college professor would decipher at a glance (and 
probably with a condescending tone too).
    

Let me throw in a shred of fact among all the hand waving and hyperbole:

I just did a "install Bundle::CPAN" on a freshly upgraded FC4 system,
along with all the related module updates this drags behind it. I also
installed a bunch of other modules (that I don't really need) from CPAN,
just for the heck of it. What can I say, it didn't even blink.
Especially the gcc compilations went through without a single complaint.

So, maybe rather than generalising FC4 compilation is broken from a few
modules that fail compilation, it might be more worthwhile giving those
modules good looking into.

Cheers
Steffen.

  

I whole heartedly agree.  I do feel it was technically the problem software's fault not the gcc4 in FC4.  However, it was more than Perl modules that was becoming a trend.  I also have some Sourceforge software and such not in the core, extras, nor freshrpms trees.  Basically your average every day open source software distributed as source not binaries.  Most of which were failing compiles when FC3 did not have a problem.  Fixing all of that C code just to satisfy a standards committee's idea of what is acceptable in C code sounds just too darn French like (the French are "anal" about their language and have a committee as well), and down right un-open-source like, in my opinion.  An OS with an open-source community can not be tied down at the whim of a group of bearded men with a clean coding fetish.  It runs totally counter to the idea of software from the community.  Tightwad standards only make sense with strict API's governed by non-disclosure agreements and intellectual property issues.

Rich

P.S.  For the record, Bundle::CPAN worked fine for me as well, but then again, it was not a concern I mentioned.