On 2011/03/27 22:31, Jecel Assumpcao Jr. wrote:
> Casey Ransberger wrote on Fri, 25 Mar 2011 15:28:16 -0700
>
>> Hey while we're thinking about a small clean modular Squeak, I
>> strongly recommend looking at what Juan has done with Cuis.
>> It's *really* nice, and it's small and it's clean. I really wish we
>> had something like Cuis after running #unloadAllKnown on Squeak.
>
> I am very much in favor os small and simple. Of course, sometimes you
> need more funcionality. The first time I tried to use Cuis for a
> presentation I ended up having to go back to a traditional Squeak image
> because I needed some of the things that had been eliminated.
>
> The ideal is a simple and tiny core that can be expanded as needed in a
> way that is as transparent as possible.
>
>> As far as modularity goes, have you checked out Environments?
>> I think if we decouple the OmniBrowser bits from it, we'll have
>> a namespacing mechanism that's actually real and works in Pharo
>> today. I got it to mostly work in Squeak without *that* much
>> hacking too, so I know it's doable. Not sure how much work is
>> baked into freeing it from OB though.
>
> I don't care that much about name spaces and would have been perfectly
> happy with Göran's solution, for example. On the other hand, I would
> like to see some more fundamental solution, like the "worlds" thing VPRI
> is currently working on.

Alessandro Warth's Worlds thing looks very, very interesting. 
(http://www.vpri.org/pdf/tr2008003_experimenting.pdf for those who 
haven't read it yet).

I found many parallels (if I recall correctly: I read these papers about 
a year ago) between Alex's work and Henry Baker's Worlds in Collision 
(http://home.pipeline.com/~hbaker1/TreeShadow.html).

frank