>> As far as I know the beta team (incl Alison) were still testing as late as
>> two weeks ago.
>> How do you know that Sierra has deliberately delayed the release of the fix?
>I know because the major problem created by the 1.1 patch (slomo / new sync
>method) was known to have been fixed very soon after it's release. Yes, I've
>been leaked a couple of the fixes and the problem is not only fixed but online
>play is much improved over 1.0. GPL became practically unusable for many with
>1.1.
>My point here is that there is no good excuse for Sierra not to release that fix
>as an unsupported beta, IF, they really wanted GPL customers to be happy with
>the product. They can dink around with the final official fix as long as they
>want to, fixing other relative minor bugs and adding whistles and bells.
>Good software companies who really care about customer satisfaction don't
>hesitate to get temporary fixes out when the problem is as serious as this one
>is.
>So why is Sierra doing this? I'm willing to give them the benefit of the doubt
>that it isn't just stupidity...partly stupidity but not completely. My guess is
>the truth of the matter lies with the effect on their sales of their major cash
>cow, N3. I've heard N3 still isn't available in Europe.
Hanlon's Razor. It's stupidity, not malice.
There are two schools of thought on patches. The correct one, and the
overly cautious one. The correct one knows about Linus's Law (Debugging
is Parallelizable), and tries to get the latest code out to as many
people as possible as soon as possible, clearly labeled as beta. The
overly cautious one fears that the code may break something, and so holds
off releasing until several layers of internal and trusted testers have
blessed the code for a few weeks or months.
The problem is that it's easy for 20 people to miss a bug that would have
been caught within a few hours if 1000 people had a chance to test it, so
the overly cautious types end up with a buggier final product than if
they'd released early and often. (It's only possible to devote a limited
number of programmer-months to supporting a program, and the more of that
time is wasted waiting for the bug reports, the less is spent fixing the
bugs.) Note that the absolute worst bug in any released games, the
uninstaller bugs in Thief: the Dark Project and a couple of demos that
would wipe your whole disk if you installed to a non-default directory
off the root directory, weren't caught until after release. Which shows
that the internal and closed beta testers not only can't catch all the
little bugs, but they don't even catch all the bankrupt-your-company bugs.
The guys at Papyrus are way too smart not to know this, but they answer
to the suits at Sierra, some of whom apparently have a bit too much of
a CYA attitude and a bit too little commitment to true quality.
--
spamgard(tm): To email me, put "geek" in your Subject line.
