Is posted here.

Although it isn't a NFJS event, Bruce has done many sessions at NFJS. This is a pretty interesting talk because he blends a little bit of his "Beyond Java" material with some "Where is R/R going?"

A couple of choice quotes and ideas:

• Ruby is in a good place because there is no legacy code to muck it up. Bloat happens, but careful; Ruby is not necessarily immune to that
• Ruby isn't successful (just) because it is beautiful, it solves certain business problems well
• Where could Ruby go wrong? What if IBM owned R/R? What would Websphere/Rails look like?
• Bad places to go: COBRA, distributed transactions, concurrency.

Interesting that he says applications may not be the right place to handle concurrency. Perhaps it is best left to lower layers (i.e. C/C++). OO languages might not be the best way to handle concurrency at all and he mentions in passing that functional programming might be the best suited. He says he is "very afraid" of Brian Goetz's Java Concurrency in Practice book.

Finally, he suggests the best way to avoid bloat is to deprecate like crazy and never abandon beginning programmers; that this is why Java can't train young programmers without a deluge of docs and API binders.

The link is an embedded flash player. I couldn't figure out how to embed it here...