Mats Nygren wrote:
> 
> Henrik Gedenryd <Henrik.Gedenryd at lucs.lu.se> wrote:
> > ..
> > If your MOP interest would mean rewriting the Squeak MOP to something more
> > friendly, that would be more than welcome :)
> > ..
> 
> I'm considering implementing a simple CLOS-like language. Primarily to get
> some experience in this, following "The art of the MOP".

> And then try it out with some metaobjects for objects with history.

This sounds interesting to me!

> Objects
> with history can be used as a basis for editors with indefinite history and
> also for backtracking search, together with unification that will provide a
> Prolog-system within the MOP. I dont know in advance how feasible this is
> it seems within reach. With further (extensive) work this can be connected
> to the Smalltalk change-set-machinery and also CVS-like things.

I'm currently 'playing' around with CVS at SourceForge and would
appreciate to do this in Squeak instead...

And version control is an issue (as often stated before) in Squeak, too.

Greetings,

Stephan

> Production quality efficiency will not be a goal at first.
> 
> This could lead to some ideas about how to put Squeak into this kind of organization,
> but only later. At the present I don't even have detailed insights into the current
> Behavior-related machinery.
> 
> /Mats

-- 
Stephan Rudlof (sr at evolgo.de)
   "Genius doesn't work on an assembly line basis.
    You can't simply say, 'Today I will be brilliant.'"
    -- Kirk, "The Ultimate Computer", stardate 4731.3