Sunday, June 13, 2010
.NET people tend to tell all those stories about languages that are ported to the .NET framework. Like Iron Python, Iron Ruby... Last week I examined how it looks like for Haskell and Scheme. In the Channel9 movies I heard few times that there is something like Iron Scheme and Iron Haskell, I wanted to compare VS against Emacs editor in terms of fast coding. I found out that there is no such thing as Iron Haskell, but there is a visual Haskell, and it only works for VS2003 and VS2005. It is just an interface via COM, between VS and GHC. This is also the reason why it won't work in VS2008 and VS2010. I've been thinking, what's the point of mentioning that there is a Haskell version for VS if not many people use VS2005 or VS2003 anymore. It is like saying that POSIX compatible application, written on Linux system will compile under MS Windows. Sure it will compile, but only on a Windows platform that was POSIX compliant (has a POSIX personality), and the last version that did that was Windows NT:) In other words telling me that there is an Implementation of Haskell running under VS and on top of .NET is like telling me to get a version of a Windows NT, and still not telling the truth, cause as far as I can see this Haskell does not behave as C# or VB.NET in terms of compilation to IL and running in a .NET framework.
Iron Scheme is a different story. It can be executed in VS2008, too bad that I want to use VS2010. It doesn't give me a power of a .NET framework and ability to use a scheme language. For me writing an application in this hybrid is like a functional-imperative disco polo kind of programming. A good subject for a separate post.