[OT] Sun has released the Strongtalk VM as Open Source.

> Wow! Excellent, thanks for the information. On a more regrettable
> note, they seem to have let go of David Ungar and Adam Spitz who were
> working on Self and Klein (which is now also released): http://
> tech.groups.yahoo.com/group/self-interest/message/1943
>
> On Sep 11, 2006, at 10:54 PM, Andrew Tween wrote:
>
>> I just caught this on the Strongtalk mailing list, and thought that
>> it may be of
>> interest to some of you. See  http://www.strongtalk.org  for more
>> info.
>> Cheers,
>> Andy
>>
>
> --
> -Brian
> http://tunes.org/~water/brice.vcf
>

> The Strongtalk VM is organized as a high-performance interpreter  
> (2-3 times Squeak speed, I believe), and an inlining JIT that  
> achieves roughly 6x Squeak speed.  Gilad reports the following on  
> his Intel Mac:
>
>         Squeak 3.8    345,712,356 bytecodes/sec;  7,855,215 sends/sec
>         Strongtalk  1,805,996,472 bytecodes/sec; 48,075,256 sends/sec
>
> My mind reels at these numbers.  Moreover Robert Griesemer had a  
> design for an even better JIT and, if this became an active  
> project, I bet he would help out.
>
> Strongtalk is set up to support native windows, and it probably  
> makes sense to keep it that way, but this would be a parting of the  
> ways from Squeak's run-anywhere agility. It would be nice to  
> introduce a layer in the UI with a separate bitblt-only  
> implementation to retain extreme portability.
>
> The VM is not simple -- it is a large body of C++ code.  However it  
> was written by smart people and is well-organized (I haven't looked  
> through it carefully).  It probably has some bugs, and it may take  
> some archaeology to get it all to compile with the latest tools.
>
> That said, I think there would be a tremendous reward for doing the  
> work.  The ironman engineering of Strongtalk seems a perfect match  
> for Squeak's cheerful insouciance.
>
>         - Dan
>

>> I just caught this on the Strongtalk mailing list, and thought  
>> that it may be of
>> interest to some of you. See  http://www.strongtalk.org  for more  
>> info.
>
> This could be the most important thing to happen in the Smalltalk  
> community in years.  The Strongtalk VM was faster than any other  
> when it was written, and I believe it is still comparable to the  
> VisualWorks VM (it would be fun to test).
>
> One could ignore the type system and simply port all of Squeak into  
> Strongtalk (of course there are parts of Strongtalk that are better  
> and should not be lost ;-).  Then, or in the process, if one did a  
> tasteful job of supporting the types optionally (ie a browser  
> switch to show them or not), it would be the first opportunity to  
> have the best of both worlds in Smalltalk -- or anywhere for that  
> matter.
>
> The Strongtalk VM is organized as a high-performance interpreter  
> (2-3 times Squeak speed, I believe), and an inlining JIT that  
> achieves roughly 6x Squeak speed.  Gilad reports the following on  
> his Intel Mac:
>
>         Squeak 3.8    345,712,356 bytecodes/sec;  7,855,215 sends/sec
>         Strongtalk  1,805,996,472 bytecodes/sec; 48,075,256 sends/sec
>
> My mind reels at these numbers.  Moreover Robert Griesemer had a  
> design for an even better JIT and, if this became an active  
> project, I bet he would help out.
>
> Strongtalk is set up to support native windows, and it probably  
> makes sense to keep it that way, but this would be a parting of the  
> ways from Squeak's run-anywhere agility. It would be nice to  
> introduce a layer in the UI with a separate bitblt-only  
> implementation to retain extreme portability.
>
> The VM is not simple -- it is a large body of C++ code.  However it  
> was written by smart people and is well-organized (I haven't looked  
> through it carefully).  It probably has some bugs, and it may take  
> some archaeology to get it all to compile with the latest tools.
>
> That said, I think there would be a tremendous reward for doing the  
> work.  The ironman engineering of Strongtalk seems a perfect match  
> for Squeak's cheerful insouciance.
>
>         - Dan
>