From: Nat Pryce Date: 2001-10-08T06:54:30+09:00 Subject: [ruby-talk:22213] Re: Let's work on Windows support > Personally, my concern is not with there being multiple > implementations of Ruby (though I'm stil intrigued by/not clear on > exactly what that would mean), but rather with the possibility of .NET > versions of languages becoming de facto standards, in a situation > where those .NET versions are *not* drop-in replacements for the real > versions of the languages. Your scenario, in which there's no > difference, would be fine, but I can't imagine that's why Micros**t is > throwing its .NET over the world of non-MS programming languages. Microsoft is encouraging the port of many languages to .NET because it wants to create the most widely used VM. To compete against Java, which is already very entrenched, it will have to make it *much* easier to do certain tasks in .NET than in Java. That requires different languages because languages are better suited for some tasks than others. E.g. Perl and Ruby great for doing text processing, Haskell is great for doing tree transformations, and so on. If different languages are used in the same app, there must be a unification layer. This is what the .NET CLR provides. It also by definition requires that languages fit with the common type system, runtime semantics and security framework. That's why some languages have had to be changed. There's nothing nefarious about such changes, such as Microsoft trying to control development of all languages. After all, that would dissuade language implementods from supporting .NET, and so dissuade programmers for writing .NET apps. The evolution of languages that run on .NET is still in the hands of their implementors or standards bodies. Cheers, Nat.