This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Testsuite ad RedHat 7


In article <200010101603.SAA04655@quatramaran.ens.fr> I wrote:
>I think the Steering Committee made it pretty clear that shipping a 2.96
>e.g., *development* snapshot, with an official `release' is a very stupid
>decision.

Just to clarify, the Steering Committee only made it clear that 2.96 never was
guaranteed to be compatible with anything. The inference that shipping it
is a stupid decision is mine.

I do happen to think that shipping such a compiler with an actual release of
a product is an idiotic move.  Especially with a wide-spread product such as
RedHat. Because it will further the impression that gcc and C++ are unstable
stuff.  Also because people will need to update again in a few months, when
gcc 3.0 is out. And also because people will get burnt over this, as they 
don't really expect things not to be binary incompatible.

In the public eye, this will appear to be yet another gcc/glibc/whatever
screwup or something. In an ideal world, the blame would fall squarely on
the party responsible. In the real world, this looks like a PR disaster 
waiting to happen for gcc and/or C++...

The explanations Jeff gave for the decision are quite valid, from a certain
point of view. They don't change the bottom-line that some users---the ones
that install new linux versions all the time, grab binaries off the net, and
compile from source once in a blue moon, will see loads of problems apparently
coming from gcc, and will arrive to the wrong conclusion.


There's also that old saw of mine that I definitely disagree with shipping 
products with development snapshots used as `production' environment.
Why ? because there is a need for actual timely releases of *reliable* tools.
These snapshots are a placebo for the actual releases. They appear to work,
but they're not that reliable.  This does make things look as if tools are 
amateurish.  This also reduces the pressure to churn out actual releases to
the point that, for some projects, no new release gets done over an insanely
long period.  In turn, this drives other vendors to build their own in a
corner. And things slowly get out of joint: some patches don't get contributed
back. Some very important fixes don't get made for the release, and some bugs
are fixed ten times instead of once.  All the bleak points of bazaar 
development, because of cathedral-like releases...

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]