[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnash-dev] Stable branch of Gnash

From: Rob Savoye
Subject: Re: [Gnash-dev] Stable branch of Gnash
Date: Fri, 21 May 2010 14:31:12 -0500
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv: Gecko/20100423 Thunderbird/3.0.4

On 05/21/2010 01:47 PM, strk wrote:

> C'mon rob, I'm not working on the codebase right now but I've heard
> this argument from you so many times that my blood pressure raises
> as well here.

  Look, as I've said repeatedly, we have this bad tendency to talk past
each other, than actually discussing anything concrete. You guys make
these general accusations, often without anything but personal opinion
to back them up, and then get upset if I disagree. Considering I've been
coding professionally since before you were born, I see things differently.

> If your commits badly break things you should take immediate action
> to fix, or revert if you don't have time.

  Do *not* keep reverting my changes, as it causes me much lost work.
You wind up doing the very thing you are complaining about, but seem to
think it's ok when you do it. And what makes you think I'm not going to
fix things ? You seem to assume I don't care about the project I
started, and have maintained and funded for many years.

  You also have checked in code that breaks both Gnash or the testsuite,
and I've never jumped on your case for it. "Badly breaking" things is a
bit of a joke, as Gnash wasn't badly broken with my checkin. I mean,
come on, a few failing tests is not the end of the world. Look at all
the expected failure in the testsuite we have now...

  Considering I'm the only one working on Gnash full time still (for
free, btw...) I can't afford to waste more time than I already am. We
have very different coding styles. I prefer to work very publically,
checking code in frequently, and then fixing it over the next few
checkins. This is the way most all free software projects I've been
involved in for 20 years operate.

> Yes, but I'd add "small" and most important "not breaking the testsuite".

  Um, the testsuite was already broken...  I checked in a fix a little
while ago that has nothing to do with ExternalInterface. Plus I'm far
from the only one that breaks the testsuite. And when it really comes
down to it, temporary breakage of a test case is a pretty minor issue.
Why do you feel like blowing this out of all proportion ?

  I not only run the testsuite, I also surf the web for hours with a
branch before committing. Remember, I wrote the majority of our
testsuite, so I think it's fair to say I consider testing important.
Course I also wrote the manual, do all the releases, maintain the
configuration and build infrastructure, but nobody seems to care about
that at all.

  The problem with the "stable branch" concept is that stable in this
case means stagnant, with no real development of major new features. In
order for Gnash to be relevant, we *must* continue to push forward with
the types of features our end users want. ExternalInterface, my original
RTMP work (which you broke repeatedly), avm2 compatibility are what we
should be focusing on. Instead I get endless rewrites of existing code,
all aimed towards "code quality". This does not advance Gnash at all,
which is why our funding evaporated.

  I'm for pushing Gnash forward, not backwards, and not just spinning
our tires in the mud endlessly.

        - rob -

reply via email to

[Prev in Thread] Current Thread [Next in Thread]