Simon Peyton-Jones
I realise belatedly that my message might have sounded dismissive. My apologies; it wasn't intended to be. Good ideas are just that: good. Reinventing them is a sign of good taste.
As to documenting GHC, we try to do that by writing papers. That's easy to motivate because we get research brownie points for papers.
One of the irritating effects of this process is not that the reports are research papers, but that they are on-line sporadically and only very rarely html. The overhead of having to download a big file (or search through one's own copies) and fire up some other viewer (for .ps or .pdf) -- or worse find a printed copy or fork out £55.97 to read something online) is a significant obstacle when all one wants to do is to is check the syntax of something or look up a short bit of code. A hyperlink of the form <a href="http://.../long-research-paper.html#interesting-paragraph"> interesting bit</a> is far more useful than one of the form <a href="http://.../long-research-paper.pdf">look for section 49.7.3</a>. It may not seem significant, but when one is attempting to learn some new part of Haskell it's really off-putting. -- Jón Fairbairn Jon.Fairbairn@cl.cam.ac.uk