Invisible wrote:

> Ah yes, Haskell's legendary documentation. It's scattered all over the 
> place, half of it is published as academic papers that you and I can't 
> access, much of it is written for compiler designers and academics 
> rather than actual *programmers* who just want to know how to use the 
> thing, and lots of stuff isn't documented _at all_.
> 
> For example, laziness can be a bad thing. This is highly non-obvious, 
> and it ISN'T EXPLAINED ANYWHERE. If you pop into the IRC channel and ask 
> about it, somebody will explain it. But then it's *still* not written 
> down for the next newbie who wants to know. It's sort of an oral 
> tradition; there are many "well known" techniques, but nobody has 
> written any of this stuff down anywhere, making it infuriatingly hard to 
> learn about this stuff.
> 


> 
> ....which neatly demonstrates my point. Programs compiled with GHC 
> contain C debugging symbols by default - even though these are 
> completely useless in a *Haskell* program. By using a special 
> command-line switch, you can "strip" the symbols, making a typical 
> program about 200KB smaller. But this isn't written down anywhere; I 
> found out about it by chatting to people on IRC.
> 
> And yes, Haskell is designed for writing complex algorithms easily. If 
> you want *speed*, it's very non-obvious how to do this. There is some 
> scattered documentation about it, but currently the most expedient 
> method is to post the source code and ask Don to look at it for you. ;-)


> It's the website for a book that just got released and is meant to 
> explain all this stuff - not just the language syntax, but how to work 
> the compiler, what the best coding strategies are, etc.
> 
> I must say, I was very excited about this book, but having actually read 
> it now it's out, I'm dissapointed. Still, it's a single *large*, 
> coherant body of information, all in one place. That's a step in the 
> right direction...
> 

It sounds like you are getting / are very involved with Haskell and know 
  at least more than me about its inner workings.

FWIW Why not start writing some of the things that you just mentioned 
down and make them available.

You could provide a web site with simple articles explaining some of 
these things.

You could turn it into a <gasp> published book </gasp>.


Tom

Post a reply to this message