iHack, therefore iBlog

In my opinion, BitC is the most innovative take on systems programming we’ve seen since the invention of C.  While sad that it failed, I am deeply impressed by the thoughtful post-mortem by Jonathan S. Shapiro.   Here are links to the various threads of his analysis (and the equally thoughtful responses):

• [bitc-dev] Retrospective Thoughts on BitC
• [bitc-dev] Retrospective: The Issues with Type Classes
• [bitc-dev] Retrospective Thoughts on BitC [David Jeske]
• [bitc-dev] Retrospective: separate compilation and dynamic linking needs programmer knowledge?
• [bitc-dev] Instance coherence: the shape of a solution
• [bitc-dev] Retrospective: shape types
• [bitc-dev] BitC Lessons For Other Language Developers – Simple vs. Too Simple
• [bitc-dev] An unusual design pattern
• [bitc-dev] Type Clases, Overloading and Genericity
• [bitc-dev] Subtyping/subclassing
• [bitc-dev] Wrong notion of const-ness