So if he borrowed from Python and Icon as well as from Java, what did Hejlsberg originate in C sharp? Is it just a mish-mash of other languages? (Maybe that's a good thing, I'm just asking.)

Hejlsberg has made sure that C-sharp has some great borowings from other languages, such as the concept of generator functions from Python and iterator functions from languages like Icon and Python.

Bluebottle certainly looks nice and lean... it's like AmigaDOS done with Oberon. No wonder the system calls are fast: they're not actually system calls, they're just procedure calls. And unlike, say, Java, they don't even set out to stop you from clobbering other objects in the shared address space - they just assume that you will do almost everything in type-safe Active Oberon, and will be very very careful if you don't. Bad luck if you want to support multiple people working on the same system.

But reading some of the Bluebottle sources, what is it with the ETH-style languages (Modula-2, Oberon) and the SHOUTING KEYWORDS? Do ETH people actually like having to hit the $#!? caps lock key every few words? Or do they use an editor which does the stupid capitalization for them? If so, what actual point is it? It's like the old Star Trek episodes where the computer had to yell in a grating monotone just to remind you that *it was a computer*.

Please, be serious and really do some research :
both DOT NET and J2EE are bloated everchanging copycats working over very old OS architectures dating from the seventies.

If you really want some efficient, lean, mean and SIMPLE architecture, really long-lasting, please visit ETHZ site's :
http://bluebottle.ethz.ch/
Those researchers have done a marvelous job. Their job. It's your job to use it. If you want. Test it.

Box's comments are very discouraging. He must realize that when you don't have a giant corp snatching up ideas and crushing the originators, then you end up with something quite astonishing: Choice. Hibernate, iBatis, JDO and the like exist because individuals and companies contribute - because they receive benefits. Case in point, Java has ant, quite happy with it, if not try Maven, if not...build something better. It's such a good idea, MS would like to steal it, and call it MsBuild (quite flattering, actually).
Now, the .Net people who wrote (n)ant are wringing their hands because MS is going to crush them, people are asking (should I even bother...). Constrast that with the various regexp projects before java supported regex, and now they are happily existing along side the official Java (ORO etc). Same for logging (commons, log4j java 1.4 etc).

To compare it with Vietnam is obsurd. Hmmm. Java developers are actually happily volunteering for this and deriving great value. Hell, if you're a Java programmer and you don't benefit from an o/r mapper or persistence layer, more power to you. Choice. We even have tools that make changing our minds less expensive too (springframework.org). Ah, and yes, there is a .net port underway. Does that make .Net developers the greatly underestimated lemmings of the Vietnam War? Please, MS spokespeople, continue trying to tell me why choice is bad.

what does "o/r" mean?

Sure .NET has a ways to go yet at the enterprise level but I doubt that this Microsoft Java envy will last much beyond 2006...after that everything may change

Praise from Hejlsberg is praise indeed. Makes a change from the barbs you usually hear from language gurus. Perhaps there's something to this co-opetition idea after all, with Java and C# as rivals rather than enemies.

I was there in Vancouver. The *really* cool moment was when Microsoft's "Indigo" maven, Box, conceded that he loved the fact that the JVM's footprint is so small.