Re: [Gnash-dev] Re: no AVM2 for a while?

[Benjamin Wolsey]

> If it isn't working usefully today, do you think it's realistic that
> we plan to cut the next release in September?  AVM1 still works,
> so a release wouldn't be entirely pointless.
> 

Yes, there's been progress on AVM1, so it is worth a new release. 

I do not think there is any hope of AVM2 functioning usefully by
September. There are too many problems that are made very difficult by
Gnash's current design, and working on them without improving Gnash core
just stores up more problems for later.

Hopefully the Gnash VM will be reentrant and refactored by September,
which would bring many benefits, not least being able to implement new
things more quickly. 

> But it would be hard if all the AVM2 work that's been going on
> couldn't be stabilized by the next release.  No project prospers by
> carrying huge amounts of code that's slowly getting moldy because only
> three guys can usefully run it.

Gnash's AVM2 code was originally designed so that it would replace or
use AVM1 code. Although that was misguided in some ways, lots of it is
used already when AVM2 is disabled.

Besides, there hasn't been that much AVM2-only work over the summer.
Much of it applied to AVM1, and the work on AS3 implementation means
that, when Gnash's machine starts running more correctly (how much work
it needs isn't really clear), fixing the actionscript will be then
easier.

Even now AVM2 can be tested and improved by using the working
as3compile.all testsuite, but largely isn't because it uses the
non-working classes.all testsuite, which also contains bugs that no one
appears to want to fix.

bwy


--
Free Flash, use Gnash
http://www.gnu.org/software/gnash/

Benjamin Wolsey, Software Developer - http://benjaminwolsey.de

signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil