Squackers, Tenth-Birthday-of-Squeak edition

Hi all--

     For October, Squackers[1] will celebrate Squeak's tenth birthday at
Dan Ingalls's place near Santa Cruz, CA, USA, on Saturday the
seventh[2]. We'll start by hanging out at the beach in the afternoon,
then head up to Dan's for a barbeque and further merriment (and we might
be found on the Squeak IRC channel around 7pm GMT-7, see [3]).

     If you think you'll be able to make it, please let me know! And
please forward this message to others you think might be interested.


     take care,

-C

[1]  Squackers is a group of Squeak hackers in the San Francisco bay
     area who get together once a month to have a good time and show off
     what they're up to with Squeak (including learning it).

[2]  ...the October Saturday nearest the traditional birthday of 1
     October, the date Dan announced Squeak to comp.lang.smalltalk (see
     http://tinyurl.com/gqdhv for the original message).

[3]  http://users.squeak.org/irc

--
Craig Latta
http://netjam.org/resume

Craig wrote...

Folks -

I think it might be nice to spend a bit of time strategizing about Strongtalk and Squeak (if we can wait until then).  Also I might talk a bit about my whacky project to do a self-supporting kernel in JavaScript, and if Ian can make it he might talk about his new VM that should run both JavaScript and Squeak, and what's cool about that.

If you are coming, please drop me a line so I can explain about parking and other silly details.

        - Dan

Dan

do you have plan to release your java squeak virtual machine?

Stef
On 15 sept. 06, at 07:29, Dan Ingalls wrote:

>Dan
>
>do you have plan to release your java squeak virtual machine?
>
>Stef

Absolutely.  It's only my distaste for red tape that has delayed this.  Keep pestering me mercilessly, (but don't burden the list).  I will take the first step by Wednesday and report a time estimate by Friday.  I'm sure that it will clear, and am thus happy to send the code confidentially  to anyone who cares in the meantime.

Further discussion off-list, OK?

        - Dan

Folks -

I just found out that there is a Fireworks extravaganza right in
front of my house the evening of October 14 so we're moving the
(let's be frank) party to that Saturday:

        http://www.monte-foundation.com/images/MontePoster2006.jpg

The rest of the plan is the same -- come any time to the beach (see
above link for beach activities that day), I'll start the grill at my
house around 6, and the fireworks will be around 9-9:30.

Parking will be an issue.  Try to car-pool and please let me know
that you are coming (off list!) so I can instruct you about parking
and provide roughly the right amount of food/drink.

        - Dan

Hi Dan,

    You wrote:

> I think it might be nice to spend a bit of time strategizing about
Strongtalk
> and Squeak (if we can wait until then).

    You also had written:

    More conservatively, how hard would it be to port Pavel's
    KernelImage to StrongTalk and turn it into S3 (tm)
    (Strong Small Squeak ;-)

> 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.
>

    and once there is a strong and small Squeak with an even better JIT,
    supports for various Squeak's GUIs would have a better chance of
    becoming a reality.

    Isn't it a concensus, by now, that it is much more easier
    to understand and work with a smaller and robust code base.


> 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,

    Some mechanism to support Squeak' named primitives ?

    It is so excited to think of S3 with Assember, Compiler..., and of
    course, Interpreter, ObjectMemory and Primitive plugins ;-).


> and it may take some archaeology to get it all to compile with the
> latest tools.
>

    I think VS6 can accept VS5 project file, from there it can get
    straight to VS8 (aka VS 2005).

    The C++ codes can be compiled as managed code and, for now,
    the object files from the Turbo assembly code can be bundled into
    one classic dll and used through platform invoke mechanism
    (or the machine code can be coded into some Forth words ;-).

> That said, I think there would be a tremendous reward for doing the
> work.

    Definitely !

    And a clean slate for licensing issue as a bonus ?

    There is also news that David Griswold will port StrongTalk VM
    to run Squeak byte codes. This would be great.

    It doesn't hurt to have some fresh air and healthy competitions ;-)

    The children will have more than one Squeak to choose from.

    It is horrible to think of a mess of half cooked, rotten,
    prototyped bits and bites being thrusted down the throats
    of the kids !!!

> The ironman engineering of Strongtalk seems a perfect match

> for Squeak's cheerful insouciance.
>

    Hmm, what was the expression ?

    Ahh,  is it "straight from the Horse's Mouth" ;-)


    Cheers,

    SmallSqueak

----- Original Message -----
From: "Dan Ingalls" <[hidden email]>
Newsgroups: gmane.comp.lang.smalltalk.squeak.general
Sent: Friday, September 15, 2006 1:29 AM
Subject: Re: Squackers, Tenth-Birthday-of-Squeak edition

Strongtalk and Squeak (if we can wait until then).  Also I might talk a bit
about my whacky project to do a self-supporting kernel in JavaScript, and if
Ian can make it he might talk about his new VM that should run both
JavaScript and Squeak, and what's cool about that.
>
> If you are coming, please drop me a line so I can explain about parking
and other silly details.

I wrote a while ago...
> > The ironman engineering of Strongtalk seems a perfect match
> > for Squeak's cheerful insouciance.

and SmallSqueak responded (on the Squackers thread)...
>    Hmm, what was the expression ?
>    Ahh,  is it "straight from the Horse's Mouth" ;-)

...and hopefully not the other end, right? ;-).

It has been interesting to me to read the discussion here.

Most heartening is the warm support for Exupery, not only because it is from our community, but also because it is in the Squeak tradition of keeping things as metacircular (ie in Squeak itself) as possible.  I agree that it would be a mistake to undermine Exupery by whatever we did with Strongtalk.  It's my hope that we can continue to incorporate any improvements for performance in a Squeak-style framework as Bryce is doing with Exupery.

At the very least, Strongtalk can be taken as a living textbook on bold steps with Smalltalk, from which we can learn a great deal.  Just a few that come to mind:

        An optional type system
        I don't like having to declare types when I'm hacking, but I wouldn't
        mind a button in the browser that revealed them when I clicked on it.
        Plus people here have pointed out the extra leverage for refactoring.
        And in a novice scripting world, types and type inference (unseen)
        can provide extra leverage for the system to suggest valid options.

        Elimination of most global static state
        We keep talking about it but never doing it.

        A kick-butt interpreter
        I don't know how it compares with Exupery, but I think it's nearly
        three times as fast as Squeak.  I think there are tricks to learn from the
        way they assemble native code snippets to build the interpreter.

        A mature Mixin system which is useful to compare with Traits.

        And, of course, the type-feedback inlining compiler
        This is what puts Strongtalk in the "ironman" category.
        Again there is much to learn here.  it would be great to see
        this done in Squeak itself, as we have been mumbling for several years.

        Special optimization for floats
        This translates into the ability to do serious graphics without
        needing as much code to be in primitives.
        Their float benchmark is extraordinary.

The passion behind my earlier post was simply in the space of Smalltalk community synergy.  I look on this as a sort of celestial event -- I thought Strongtalk was the coolest thing ever done with Smalltalk (that and Self), I though it was dead with the added agony of having been buried alive, and suddenly the coffin lid has been raised and it is breathing again!  All of this at a time when dynamic languages are experiencing somewhat of a revival in popularity.

And I'm thinking, what if there's some kind of opportunity for another run at the fence -- some sort of scheme that opened the door for Pearl, Python, Lisp, Ruby, JavaScript and Smalltalk all to vie for being, say, the core dynamic language around which most Internet software would be written...

...Wouldn't it be cool if we could point not only to decades worth of cool media support, tutorials, games and development tools but also, a deployment engine that rivals the best in the industry.

It appears that there is enough of a community around Strongtalk to at least get it running again regardless of what we do.  But I felt (and thus the quote from before) that there was might be some low-hanging fruit worth reaching for, and that the result, independent of technical impact, would be to build the synergy of these two beautifully complementary communities, and thus strengthen the backbone of Smalltalk itself going forward.

The simplest project I had in mind (if I weren't already swamped (I know; we all are...)) was just to port Squeak's BitBlt to Strongtalk, fileIn all of Squeak (yes, this will take 206 tries ;-) make each native Strongtalk window into a Morphic project, and see how it all works.  Ideally this could be done carefully and a "bridge" procedure developed so that when the next Strongtalk or the next Squeak is released, someone can fairly easily prepare a corresponding StrongSqueak release.

Then there would be lots of fun stuff to do -- get speech and music going well, see how well Croquet runs with the fast float support.  See how many primitives can be thrown away in the faster system, etc., etc.

How hard would this be?  I don't know.  My guess is that this could all be done without even touching the VM, just by simulating a Morphic canvas in the Strongtalk world.  A lot more fun than the Sunday puzzle page, at least.

        - Dan

yes I really hope that one of the people working on that will deliver  
soon :)

> Elimination of most global static state
> We keep talking about it but never doing it.

Can you explain a bit more dan?

> A kick-butt interpreter
> I don't know how it compares with Exupery, but I think it's nearly
> three times as fast as Squeak.  I think there are tricks to learn  
> from the
> way they assemble native code snippets to build the interpreter.
>
> A mature Mixin system which is useful to compare with Traits.

:)
Dan read the paper of TOPLAS on my web page. Traits are cooler :)
But we really got influenced by Strongtalk. We met Gilad before  
Strongtalk before it was public
and I was the nasty that launched this public gathering of signature  
to have strongtalk open-sourced ;)
>
> And, of course, the type-feedback inlining compiler
> This is what puts Strongtalk in the "ironman" category.
> Again there is much to learn here.  it would be great to see
> this done in Squeak itself, as we have been mumbling for several  
> years.

:)

But this is.
It would be great to see how we could run squeak on the strongtalk VM  
too.
Even if I do not know the impact of their mixin aware VM. May be  
traits could gain from that.

Gilad told me that they were doing also clever copy down (only  
accessors would access state in strongtalk if
I understood correctly) so that copied down these methods on the  
mixin using classes ( since mixin has state and they could be
composed arbitrary and we could not garantee a linear order of ivs).
where should we sign for you to jump into it.
I think that it would be good to get some young blood there too. And  
we should really propose a Summertalk around that topic
next year (assuming that we will find a mentor willing to emails with  
a young smart guy).

>
> It appears that there is enough of a community around Strongtalk to  
> at least get it running again regardless of what we do.  But I felt  
> (and thus the quote from before) that there was might be some low-
> hanging fruit worth reaching for, and that the result, independent  
> of technical impact, would be to build the synergy of these two  
> beautifully complementary communities, and thus strengthen the  
> backbone of Smalltalk itself going forward.

I agree!

:)

>
> - Dan
>

Hi, Stef -

>> Elimination of most global static state
>> We keep talking about it but never doing it.

Getting rid of Pools, and dropping metaclass instVars

>> A mature Mixin system which is useful to compare with Traits.
>
>:)
>Dan read the paper of TOPLAS on my web page. Traits are cooler :)
>But we really got influenced by Strongtalk. We met Gilad before Strongtalk before it was public
>and I was the nasty that launched this public gathering of signature to have strongtalk open-sourced ;)

Thanks for your part, then.  One or two other factors were critical as well ;-).

>It would be great to see how we could run squeak on the strongtalk VM too.
>Even if I do not know the impact of their mixin aware VM. May be traits could gain from that.
>
>Gilad told me that they were doing also clever copy down (only accessors would access state in strongtalk if
>I understood correctly) so that copied down these methods on the mixin using classes ( since mixin has state and they could be
>composed arbitrary and we could not garantee a linear order of ivs).

I don't know if the VM is aware of the mixins.  What I do know is that all access to inst vars is by messages, and this frees the system from worrying about how inst vars are ordered in a particular concrete class.  In an ordinary VM this would add a substantial cost, but the inliner converts these to native access, so there is no extra overhead.

>>And I'm thinking, what if there's some kind of opportunity for another run at the fence -- some sort of scheme that opened the door for Pearl, Python, Lisp, Ruby, JavaScript and Smalltalk all to vie for being, say, the core dynamic language around which most Internet software would be written...
>>
>>...Wouldn't it be cool if we could point not only to decades worth of cool media support, tutorials, games and development tools but also, a deployment engine that rivals the best in the industry.
>
>where should we sign for you to jump into it.

I'm (over-) committed right now.

>I think that it would be good to get some young blood there too. And we should really propose a Summertalk around that topic
>next year (assuming that we will find a mentor willing to emails with a young smart guy).

Next summer? !!

Tell ya what...
Here's $5000 that says "some smart guy" will do it before the end of 2006.
<lays money on table with a bottle of champagne on top>
        ;-)

Dan,

I am glad to see you and so many others excited about the possibilities
that the Strongtalk release brings. The ideas you propose are very good
too. So I hate to be the "been there, done that" pessimist and hope that
what I write below will just encourage reflection but not cause anybody
to give up.

In my experience projects tend to have personalities just as people do
and tend to attract some kinds of people more than others. So while
there is no technical reason why somebody couldn't write eToys in Python
I will be very surprised if it happens, just like I don't really expect
to see Squeak become a great Unix scripting environment or embrace
native GUI widgets. These things are all possible but they would depend
on having people working on such projects who are naturally attracted to
different kinds of projects.

Of course you can pay somebody to work on something that they might not
seek out on their own, or if you are their thesis advisor you can also
get them to do it. My impression is that Craig Chambers and Urs Hölzle
were more C++ guys than Smalltalk ones even though they did such
fabulous work on the Self VM. This is based on things they said while
doing the work, the feeling I get while reading their code and also the
direction their carreers went after that project.

The source for the Self VM has been available since 1992. Perhaps the
timing was wrong (people used to complain about having to upgrade their
Sparcstations from 8MB to 24MB of RAM to run Self) and perhaps now
different people are looking into this and will make it work. But my
guess is that most people who enjoy Squeak will find the Strongtalk
sources tiring and will end up moving on to other projects. I will note
that Klein, the Self-in-Self, has also been open sourced and though I
haven't looked at it yet I expect it will be far friendlier to
Smalltalkers. Even Ian seems to have grown tired of the C world
otherwise he would have been my best bet for making something
interesting with Strongtalk.

For those Squeakers who do feel they can deal with the pain of a huge
C++ system given the rewards that await them I can just say "you have my
admiration and my thanks!"

-- Jecel

> >I think that it would be good to get some young blood there too. And we should really propose a Summertalk around that topic
>>next year (assuming that we will find a mentor willing to emails with a young smart guy).
>
>Next summer? !!
>
>Tell ya what...
>Here's $5000 that says "some smart guy" will do it before the end of 2006.
><lays money on table with a bottle of champagne on top>
> ;-)

In answer to a question from off-list, this is a *prize being offered*.

        - Dan

Dan,

> The simplest project I had in mind (if I weren't already swamped (I
> know; we all are...)) was just to port Squeak's BitBlt to Strongtalk,

How many times, all over your life, did you implemented BitBlt? ;-)

I guess, 10 times?

Cheers,

-- Diego

Resent with changed subject in case someone missed this.
Karl
Dan Ingalls skrev:

On 9/26/06, Jason Shoemaker <[hidden email]> wrote:

Benjamin Pollack puso en su mail :

So, no, this is not a bet, this is a bounty.

--Benjamin

If Dan add he travels to any place of world to drink champagne with who wins ... I could start a quest for smart guys in Argentina.

Edgar

     Edgar writes:

> If Dan add he travels to any place of world to drink champagne with
> who wins ... I could start a quest for smart guys in Argentina.

     It's not just a matter of being smart, but also somewhat
masochistic... there's a lot of C++ to deal with in Strongtalk. :)


-C

--
Craig Latta
http://netjam.org/resume

On Sep 27, 2006, at 1:21 , Craig Latta wrote:

>
>      Edgar writes:
>
>> If Dan add he travels to any place of world to drink champagne with
>> who wins ... I could start a quest for smart guys in Argentina.
>
>      It's not just a matter of being smart, but also somewhat
> masochistic... there's a lot of C++ to deal with in Strongtalk. :)

..and not just C++, but old C++.  Old C++ code and modern, standards-
compliant C++ compilers often don't get along too well...

Marcel

Dan,

Let me thank you for Smalltalk and Squeak. Your works are a source of
inspiration to me. I learn from your code. I keep re-reading "Design
Principles Behind Smalltalk", and watching your lectures on video. I
enjoy reading every message you send to this list.

I want to program like you. I want to write like you. I want to think
like you.

I first read about Smalltalk in 1984, in one of the very first computer
magazines I read. It was like scientifiction. I only found it again ten
years later at the university. By then, I had been programming for ten
years, and I was completely shocked by Smalltalk. In 1997 I knew about
Squeak, and I got my first job in Smalltalk. After that, I never took a
job on anything else. Your impact in my life hasn't diminished a bit
since then.

I would really love to meet you at Squeak's birthday. I'm sure lots of
us would but can't. I hope I'll be able to meet you and to thank you
personally.

Happy birthday, Squeak!

Cheers,
Juan Vuletich

Dan Ingalls wrote:

Hi Edgar,

----- Original Message -----
From: "Edgar J. De Cleene" <[hidden email]>
Newsgroups: gmane.comp.lang.smalltalk.squeak.general
Sent: Tuesday, September 26, 2006 4:09 PM
Subject: Re: [ANN]Port Squeak to Strongtalk VM for 5000$ Re: Strongtalk VM
for Squeak


    I am sure you will be able to find really smart guys with open mind
    in Argentina.

    I am looking forward to the celebration day, even if Dan cannot
    make it to Argentina, I will raise my glass over the web cam ;-)

    Cheers,

    SmallSqueak

    P.S: This posting was originally sent out on Sep. 27