Views from an Red Hat -> Ubuntu -> Fedora migrator

[Lennart Sorensen]

On Mon, Oct 22, 2012 at 03:48:23PM -0400, D. Hugh Redelmeier wrote:
> Thanks for the pointer.  I was conflating two different things 2.96
> and EGCC (which mostly became 2.95, I think).  But I don't think that
> they were unrelated events.

egcc took over after gcc 2.7.2.3 stalled forever.  I believe the first
egcc was released as 2.8, but I might remember that wrong.

> So object code from g++ 2.96 could not be used on a system with a
> different compiler.  Big deal.

And libstdc++ was incompatible too, so you couldn't run the binaries
anywhere else because no one else had a compatible C++ runtime.  That part
was a big deal.

> Red Hat did support 2.96.  Since they supplied a lot of the manpower for 
> GCC, glibc, etc., this hardly seems to be a stretch.

Sure, but no one else (distribution wise) did.

> That doesn't seem to be the case.  Non-RH GCC developers' time wasn't 
> wasted if they paid no attention to 2.96 users' bug reports (as I infer 
> from the page you posted).

They would have to filter them out first.

> I admit that this is old news.

That it is.  But to me it means redhat will always make decisions in
the interest of their needs, not their users needs.

I like the support they have for a lot of important projects, but they
have certainly showed some bad judgement when it comes to releasing
their distribution over the years.

-- 
Len Sorensen
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists