-Standard/Mainstream protocols, libraries, and infrastructures. We need to provide a bridge to all important pieces of software (e.g. OpenGL).
With respect to bridging to other languages in particular, we need to be careful not to impose a Smalltalk POV on things that are not Smalltalk. For instance, the community of the "foreign" language targeted by an FFI is not going to be impressed by our efforts if they see their language misrepresented or misunderstood. This is only fair, if some other language wrote an FFI into Smalltalk, we wouldn't like this other FFI to misrepresent Smalltalk either.
People like computer scientists and modelers who are trying to solve fundamental problems, get the data structures right, get the algorithms right etc. are far more likely to look past the gloss and glitz and appreciate the power and capabilities of a language like Smalltalk.
Yes, and in some areas I think we could be doing a better job at promoting Smalltalk to those people.
What saddened me was that the projects had great looking UI's and most were available as both desktop and web applications, but the fundamental functionality, integrity, sophistication of systems and thought in design was actually at a lower standard than we had a decade ago.
I hope we manage to stay as far away from that as possible, for otherwise we would not be leveraging Smalltalk's advantages when dealing with difficult problems. And it's hard to ignore people who consistently deal with difficult problems successfully... Andres.