I would rather have pointed to Smalltalk, and 40+ years ago...
Off topic, apart from it contributing on message-based goodness, the problem is Web was not scaled from early demos, or anywhere to be seen at that time. Neither did it have an OO idea baked into it, in fact JS made it harder to write classes but it catered for OO purists too. WPF/WCF and dynamic CLR is not testing those paths for no reason I believe, with attributes, mark-up or other. But it also shows in massive runtime penalty.
I don't quite follow - what are you suggesting? That VBScript and expando properties have been failures, and that open classes are the same?
I tried to hint those were properties and instance related (considered less of a sin?), not class-level 'extension'.
DDK: Device Driver Kit? Unit-mania?
Correct, it has a different name now, and nothing to do with context of runtime assisted duck-typing, with some compile-time checking or not. As for interpreted languages, I believe an approach (see for yourself) taken on by MS for much of core OS testing. It worked to some extent I guess, which is why a hammer always tests a nail, instead of compiler testing the programmer, or is it the other way around. But it was testing, not production was it? My point being it is the compiler that caught the original 'static class' interpretation, and it likely means different things to different languages (ECMA clarifies that too), different environments, different people and more, whether we would all like for everything to be 'dynamic' or not.
Ruby is certainly not the first
ack Brad, and wont ask which one really is most powerful, naturally :-) =================================== This list is hosted by DevelopMentor® http://www.develop.com View archives and manage your subscription(s) at http://discuss.develop.com
