One of my interests is building highly scalable (web-based) software applications. The advantage in using Smalltalk is that it has the nice possibility of being able to get (1) the wonderfully powerful development environment, (2) the incredible flexibility of the language and (3) *very* high performance deployment.
In my opinion, Pharo is taking the right direction in the first two (getting proper closures finally into the Squak runtime is an example, as is the progress in refactoring and browsing, license cleaning, increased test coverage, etc.)... and adequate performance for many situations. For the very high performance need, Strongtalk looks like a very interesting platform to me, others may have their own preferences. For this to work there are two significant impediments: sharing code, and networking compatibility. Interestingly, Avi Bryant seems to have had his hand in addressing both (Monticello and Seaside compatibility modules). As Strongtalk seems to be a bit orphaned these days, I think a fruitful route would be to make the interfaces compatible with squeak/pharo, where possible. As for the overall question of compatibility, I think we should strive to have the lowest level infrastructure as compatible as possible, but lots of experimentation above that level. ../Dave _______________________________________________ Pharo-project mailing list [email protected] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
