Home Original page

Neil Schemenauer's GC patch in python1.6?

Thomas Wouters thomas at xs4all.net
Sun Jul 16 05:23:48 EDT 2000 
On Sun, Jul 16, 2000 at 02:02:08AM +0100, Robin Becker wrote:

> In article <20000716012003.L7340 at xs4all.nl>, Thomas Wouters
> <thomas at xs4all.net> writes
> >On Sat, Jul 15, 2000 at 06:30:19PM +0100, Robin Becker wrote:
> >[ amk explains cycle-GC will be in 2.0, though possibly optionally ]
> >> If GC has finally/nearly made it, what about Chris Tismer's stackless?
> >
> >A different story altogether. I'm sure Christian can give a better
> >explanation that I can, but I believe the main showstopper is the fact it's
> >impossible to implement (currently) in JPython, not to mention future Python
> >compilers. There may be a lot more behind it, though.

> bit strange that features/defects in java should determine where python
> goes. I would certainly like generator style programming to be easier.

Not really. Unlike Perl<wink>, Python is more than its implementation.
JPython is a popular implementation on its own, and shouldn't be treated
like some backward kid ;-) I'm fairly certain this isn't the only problem
though: stackless code is pretty complex, incorporating it means a lot
changes in Python's innards, it means all developers that hack there have to
adjust to it. It also means totally new ways of programming, which may be
contra to what Guido wants with Python. And don't forget that Stackless
still has no 'killer app'; there isn't a package that requires stackless
that has everyone going 'add stackless!'

These arguments are pretty old, though, so I'm not sure if they still hold.
Probably not, in fact ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!
More information about the Python-list mailing list