On Dec 30, 2007 6:24 PM, Joost Behrends
I've already browsed through the docomentation of all that. Sorry, but i will not use WASH. I like things to be direct, to write >> p { ... } or similar things instead of <p> ... </p> is worsening things for me.
Haskell is not a good "quick and dirty" templating language. See perl for 500 templating approaches. So, now that we're past that. One of the beautiful things I have noticed about Haskell is that there has been essentially nothing I have not been able to factor out. If I have any common repeated logic anywhere in my program, I have always been able to naturally factor it out so it appears in only one place. So how, prey tell, do you factor out an expression which includes <p>...</p>? It is not Haskell, Haskell has no power there. What the libraries you are looking at do is precisely to encode HTML/SQL into Haskell, so it may be manipulated by Haskell functions without resorting to string mangling. Haskell however does not look anything like HTML or like SQL, so there is a mapping you must learn. But that mapping is a refined version of what you might come up with if you were starting from scratch and diligently refactoring as much as possible. Surely learning that mapping is easier than building your own (which will doubtlessly be worse (no offense, that's the first law of library use)). And since you are a Haskell beginner, learning a library will teach you not only the library, but loads about common idioms and Haskell programming in general. As an example, it was only after using the Parsec library that I finally came to terms with monads; for whatever reason, I was incapable of grokking them studying only the standard built-ins. I dunno, it just seems odd to me to avoid "extra learning" when you're trying to learn the language in the first place. Luke