Squeak's "general acceptance"

Doug Way dway at mailcan.com
Wed Jul 6 17:50:19 UTC 2005


On Tue, 05 Jul 2005 21:12:31 -0700, "Blake" <blake at kingdomrpg.com> said:
> On Tue, 05 Jul 2005 19:49:58 -0700, Gary Fisher <gafisher at sprynet.com>  
> wrote:
> 
> > Why saddle a leading-edge development project like Squeak with the  
> > mundane trappings of "just about any" other tool?  That seems like  
> > adding a luggage rack and air conditioning to a Formula One car or  
> > putting a wet bar and a
> > stewardess on the Space Shuttle just to make them seem more familiar.
> 
> That's one perspective. Another perspective is that it's like adding
> doors  
> or tires.
> 
> Just because something is prosaic doesn't mean it has no value.

IMHO Gary's perspective is a bit extreme here, and I personally disagree
with it & agree more with Tim's statement.  I think if someone actually
came up with a mostly-finished bundled set of general-purpose UI
"widgets" (entry field, grid, etc), the Squeak maintainers would be
happy to include it in the official Full/kitchensink Squeak distribution
along with all the multimedia and other stuff.  (Helping the
distribution live up to its "kitchensink" nickname. ;) )

There have been a few attempts at this (BobsUI, Prefab) which never
quite got finished.  wxSqueak is very cool but that's a somewhat
different animal as it relies on an external toolkit.  Anything included
in official Squeak would be fully internal (using Squeak graphics).

Gary has a point in that this type of stuff shouldn't be the
over-arching goal of Squeak... i.e. www.squeak.org will probably never
be a totally business-app oriented site.  But it can be part of the mix.
 Such business apps could take advantage of the power of the Smalltalk
language & the Squeak development environment, in the same way the
multimedia stuff takes advantage of it.

> > Certainly if anyone really wants to add functions to Squeak they can do  
> > so, and if those functions prove useful they could become part of the  
> > Squeak
> > base, but why tack on a bunch of bric-a-brac just because it's already  
> > been done elsewhere?
> 
> I don't think anyone was talking about sullying the pure release with  
> practical tools. Just pointing out that for an environment that used to  
> pride itself on being easy to use, there are plenty of areas where it's  
> really not very easy to use.
> 
> I'm not sure on what basis you can call Squeak a "leading-edge
> development  
> project" but perhaps we have different definitions.

I'd consider it leading-edge in a few (perhaps narrow) areas:

1. IDE's, specifically code/object manipulation/browsing tools.  Some
things like Squeak's Method Finder (
http://minnow.cc.gatech.edu/squeak/1916 ) are not available in any other
environment that I know of.  The original Refactoring Browser was done
in Smalltalk (in VisualWorks, ported to Squeak), and now there are
similar tools being added to Java environments.  Other code browser
experiments such as StarBrowser and Whisker Browser haven't really been
done elsewhere.  (I could have written the Whisker Browser in Java, but
it would have been a pain in the a**, probably taking three times as
long with three times as much source code.)  Also, Morphic has some
issues, but its direct-manipulation capability helps a lot with UI
development.

2. Some language research stuff, such as Traits, invented in Squeak and
(a variation on them) currently being added to Perl.

3. Educational/development tools for kids, see Squeakland, EToys, etc.

4. "Multimedia", mainly Croquet at this point.

I left out Seaside, Tweak & probably other things.  Of course, Squeak is
definitely not leading edge in some areas such as business-app oriented
UI widgets, but that could easily improve.  I mentioned #1 and #2 above
because they could be helpful with building business apps.

- Doug



More information about the Squeak-dev mailing list