[ANSI-Smalltalk] squeakers

[Jecel Assumpcao Jr]

Bruce Badger wrote:
> > 1) eToys (and Scratch) users
> [possible interest]

Agreed, specially if nicer paths from the eToys to the Smalltalk-80
levels are created. There are several efforts related to this.

> > 2) free Smalltalk-80 camp
> 
> Getting people engaged in standards work is likely to be hard even if
> the people *ought* to be interested.  This applies just as much to the
> commercial Smalltalk crowd as the "free Smalltalk-80 camp" of
> Squeakers.  I think the key here is to keep the process open and let
> everyone know what is happening (or not happening, even) so that
> people can pick up the thread and get involved when their current
> interests overlap with something that needs doing for the standard.

This is important, which is why I won't be too sad if this effort ends
up moving away from ANSI. Though I was just as interested the last time
around, I was only able to see the (almost) final version.
Unfortunately, even very open but costly processes are something I can't
participate in given my current financial situation. But I can see how
real live meetings might be important - email and wiki aren't a full
replacement for that.

> IMO, one of most important groups we need to involve are those working
> on in-house projects.  The projects, and thus the companies, would
> surely benefit from having a richer Smalltalk ecosystem.  Getting
> these people involved will also be tricky.

Yeah, there is the danger that their boss will find out they are using
Smalltalk and make them switch to Java ;-)

> > 3) the "inventing the future" guys
> 
> I too hope that we'll see something beyond Smalltalk one day.  I mean,
> this can't be it.  We can't possibly have arrived at the perfect
> language already.

Like Lisp and Forth, it might represent a local maximum. These languages
tend to "devour their young", as Alan Kay would say.

> Having a stable Smalltalk platform should not inhibit this group,
> rather I would expect it to be a boon, much as it would be for the
> EToys group.
> 
> I agree with your point that we need to make clear that a standard is
> not a straitjacket, it's just a region of consistency that can be used
> as a baseline if that is useful - and it will be useful to the
> *Smalltalk* community IMO and may be useful to the *beyond Smalltalk*
> community too.

One problem is that it takes time and effort to come up with simple
solutions. So early efforts tend to be both complicated and
incompatible. If you keep going long enough, you can reach a stage where
your system does much more with less and probably can take existing
systems into account rather than throwing them away. My "SuperLogo" from
1983, for example, had lots of new features including stuff that would
make people moving to it from regular Logos have to rewrite their
programs. My "NeoLogo" from 1994, on the other hand, was far more
radical and though it was actually much smaller than normal Logos it
could execute their code just fine. It takes me 11 years or more to
figure these things out, but I hope other people can do it faster.

A standard might get in the way of quick and dirty experiments, but
shouldn't harm Smalltalk's logn term future in my opinion.

> For all the groups you have identified, I think the best way to work
> with them is to get on with work on the standard and while doing that
> invite (and listen to!) comment from the broader Smalltalk world.

Exactly. Which is what you were already doing - good job!

-- Jecel