Project Direction

I'm sure some of you have noticed that I've been mostly off the list for the
last few months- my apologies.

I became fairly discouraged after it became apparent to me late last year
that there wasn't a critical mass of people with the right kind of
experience ready to pick up the Strongtalk system and run with it, at least
in terms of developing and supporting the system as a whole right now.  Not
that I ever thought there was that high a probability of that happening
immediately, but it was worth a shot to see if the Smalltalk community would
rise to the occasion.  Dan Ingalls even offered a $5,000 reward last year
for someone to port Squeak to the Strongtalk virtual machine, but there were
no takers. While the Smalltalk community as a whole did not rise to the
occasion, those of you on this list are the exceptions to that, and I thank
you for your interest, enthusiasm, and participation so far.


You have to understand that no one from the original Strongtalk project, me
included, has the time or desire to pick up that load.  In the years when
Strongtalk sat on a shelf at Sun, we all moved on to other,
non-Smalltalk-related things.  I got involved again because the overall
vision for the system was originally mine, so I probably care the most about
it, and I wanted to make sure that the design philosophy was carried on, and
that someone was available to answer questions. But this is still very much
a side project for me, although I absolutely still believe to this day, that
there is still no programming language and virtual machine as advanced as
Strongtalk.

The big problem at the moment is that the system is far from complete, and
there are large, highly complex parts of the system that there is no one to
work on, for example
  a.. The user interface needs major work to update it, finish it, and port
it.  I could do this, but I simply don't have the time.
  b.. The typechecker needs significant design changes that are probably not
going to happen soon.
  c.. A debugger needs to be written, and I don't have the time to do this,
either.
The bottom line is that in the near term I don't think we are going to be
able to really support Strongtalk as a complete system that is usable by
developers.  I considered giving up my involvement in the project, but after
a lot of thought, I've concluded that there is still a path forward.

Development Direction
  a.. So how can we get Strongtalk on the path to being a supportable,
complete system, or at the very least, something that survives and is of use
to people, not just as a demonstration?  For now, I think we need to defer
any attempt to support the entire system right now, and focus almost
entirely on the following three things:
    a.. Increasing VM reliability
    b.. Porting the VM to UNIX-like operating systems like Linux and MacOS
    c.. Porting Squeak to the Strongtalk VM
For Strongtalk to survive, we need a larger community to become enthusiastic
about it, and convinced of the need for it to survive.  I think the only
feasible way of accomplishing that right now is to make the Strongtalk
virtual machine a usable alternative to the Squeak virtual machine.  To do
that it will need to be grafted to Squeak, made more reliable, and made to
run on some major platforms that Squeak is used on.

Once a larger group of people see how fast the Strongtalk VM actually is,
running a code base that they are familiar with, I believe we would create a
lot more interest in the Strongtalk technology.  Also, in the short term, we
would get a finished, supported Smalltalk stack sitting on top of the VM.

Hopefully, this would help get a critical mass of developers interested in
working on the virtual machine.  Once that happens, many other things would
become easier, including spending more time on development of the Strongtalk
code base above the virtual machine.  But for that, time will tell.

At the moment, it looks like there is starting to be a reasonable group of
you folks out there working on how to port the VM, so that's a good start-
thanks again.  I'll start trying to consolidate your changes into the trunk
where that is appropriate.  Otherwise, I'm going to start focusing on a
Squeak port.  It looks like we have the possibility of getting a little
money to pay someone to help out with that from Summertalk, so hopefully we
can get going on that pretty quickly.

Cheers,
-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
-~----------~----~----~----~------~----~------~--~---