2012/12/1 Marcus Denker <[email protected]>: >> IMO, the secret of Smalltalk is a delicate combination of elements, take >> some out and the magic is gone. > Another aspect is that what got me hooked to Squeak back then was the idea to > go further than Smalltalk. > To take what humanity learned since 1978, to take the hardware advances, and > to build a system that > is Smalltalk by *Spirit*, yet just *much* *much* more so.
Can you share what you have/had in mind...? I find it very interesting, and could be inspiring for others... > How can you do that if your main concern is "how do I get it to run on the > JVM?". Well... you... don't... :-D I think it's not what the aim of Redline is. They want the language, some of its semantics (modifying classes of the fly should be supported, the author told me) and it should run on the JVM. It's a... pragmatic approach, if you want. I don't know what you guys do for a living - I used to program Java for the last 10 years, now I do Ruby at my current workplace; better, but still not Smalltalk. In order to extend and integrate with the current codebase I would even be grateful if there was, say, a Smalltalk-to-Ruby compiler. It would be nowhere the same as using Pharo, but it would be a start. But even then I don't know if my boss would approve (there's always the fear: "what happens once this dev is gone? who maintains his code?" - which is also sad, being Smalltalk so easy to pick up). > And why does it matter? In 2000, Java was hype. Now Java is the new Cobol. Do > I want to spend the rest of my > live to get a dead language to run on a boring platform? What do you mean exactly by "dead"? (the "boring" I have no problem understanding ;-) As to mattering, well... it matters to the guys wanting to code Smalltalk and forced to do projects on a JVM basis... ;-) I think I agree with you that no great amount of innovation will come out of the JVM pretty much soon... maybe that's what you meant by "dead". It's just not the VM for innovative stuff. I, for example, am delighted about Cog, NativeBoost, Helvetia, distributed computation (like Croquet did?) etc. Very hard to do (or impossible) in a VM like the JVM... > I, personally, decided that there are more interesting > things to do. I also decided to give it a shot. And that's why when I see that one of the (if not *the*) goals of Pharo is to make possible to use Smalltalk professionally, it gave me hope. If I could program Smalltalk for a living, then... I would reach career nirvana. But for now I have to pay my bills :-( > Just imagine what you could build instead! I'm trying, and it looks good! :-) Cheers, Sebastian
