RE: New member, couple of questions :)

"David Griswold" <[email protected]> Tue, 15 May 2007 13:16:43 +0200
Newsgroups gmane.comp.lang.smalltalk.strongtalk
Message-ID <[email protected]>


> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of J J
> Sent: Sunday, May 13, 2007 5:51 PM
> To: [email protected]
> Subject: RE: New member, couple of questions :)
>
>
>
> Am I getting blocked or something?  I am offering to help here with what
> little time I have and what people I can collect from my work
> with Smalltalk
> interest, but I would like to know what to work on.  I know the
> page said to
> just look on the issue list, but I think there should be some discussion
> about who is doing what and what the goals are.
>
> For example the "port Squeak" task.  What does that mean?  The
> whole image?
> Make an importer for the image or for Squeak style file-outs?
>
> And the GUI task, for sure that one has to be discussed.  Or
> should we just
> do something and if everyone like it it stays?

Please see my post about project direction from 5/1.  Activity on the
Strongtalk image is minimal; effort will be focused on VM stability, OS
porting, and a Squeak port.  The problem is that a core group of experienced
people to work on Strongtalk does not exist, unlike other open source
projects.  What little time I have to work on Strongtalk will be dedicated
to the above tasks, and even that work will be slow, since I only have a few
hours a week to devote to it; there simply isn't anyone who has time to
coordinate large scale work on the system, or to educate newbies.  I'm sorry
for the situation, but you must understand that Strongtalk is a system that
was buried long ago and was only recently resurrected, and all of the
original VM coders are long gone.

So do whatever you want with the system; hopefully we will eventually get to
a point where there are enough resources to support a full development
effort, but we simply aren't there now.

> >I saw that it was desired to port Squeak to Strongtalk.  I don't quite
> >understand this.  What would be considered success here?
> Determining what
> >the core is and putting what ever core classes that are missing in?
> >Morphic?

Your guess is as good as mine.  I and a couple of other people will be
starting to work on a Squeak port in a few weeks, but the only goal defined
so far is to get enough of Squeak running to motivate the Squeak community
to help; hopefully by that point some more VM documentation will exist, and
the VM will be a bit more stable, so that there is a better base for a
bigger development effort to get going.

> >I also saw "portable GUI" project in the issues list.  I don't know what
> >you
> >had in mind, but personally I like the Dolphin MVP (a more modern MVC
> >pattern) framework for doing the GUI work.  Obviously Strongtalk
> would want
> >to have something like wxWidgets as the base views instead of
> only windows.
> >So what is everyone looking for here?  Do you prefer to go with more of a
> >Morphic framework (which I don't know much about, except that it
> seems to
> >be
> >a mess in Squeak :)?

Whatever you want to do.  If people want to get the Strongtalk IDE running
on other OS's, something will be needed.  There is no work currently planned
on this.

> >And lastly, I saw that someone requested native thread handling in the
> >image.  Personally I feel this is not necessary.  In my opinion
> the way to
> >go is stay green threads, make them preemptive somehow and make
> a nice way
> >to do message passing.  After a question came up in the Squeak
> mail list I
> >gave my overview of concurrency [1] options today.  If
> Strongtalk goes with
> >in-image native threads then you have two options to deal with
> them:  fine
> >grained locking (a la Java) or Software transactional memory.
> >
> >As I stated in that message, fine grained locking isn't going to be any
> >more
> >tenable in the future then manual memory management is now.  STM may turn
> >out ok, but I don't have high hopes for it in the long term.
> Personally I
> >think a pure message passing strategy like Erlang is the superior model
> >here.

Native thread handling is already there, but it isn't preemptive and the VM
isn't multithreaded.  Nothing is likely to happen on this front soon.

-Dave


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Strongtalk-general" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/strongtalk-general?hl=en
-~----------~----~----~----~------~----~------~--~---