[Haskell] Monadification as a refactoring

[S.J.Thompson]

Dear Haskellers,

We are interested in finding out from the Haskell community what sort of monadic
programs people write, and in particular how they arrive at those programs. We
have (Deling and Martin) developed a particular algorithm and tool for automatic
monadification of Haskell programs, and (Claus, Huiqing and Simon) have built
the HaRe tool for Haskell refactoring. We want to offer a monadification option
within HaRe. There are various different, and apparently incompatible, styles of
monadification, and we are therefore keen to find out what would be most
effective for Haskell practitioners.

As we have said, there are many different styles of monadification.
On the web page

      http://www.cs.kent.ac.uk/projects/refactor-fp/Monadification.html

you can find a small program together with five example monadifications. We have
included these as examples, but we make no claim to have covered the whole space
of possibilities. We are interested in finding out

- which of the examples represent the monadification styles that are the most
  useful for your purposes,

- which other styles of monadification you are using, and

- which monadification(s) you would like to see implemented in a refactoring tool.

We are happy to receive simple choices as well as more detailed responses. Could
you please send your comments, thoughts, code examples and so on to

	refactor-fp at kent.ac.uk

We'll post a summary of this to the Haskell mailing list. We'd also be happy to
discuss these issues on Haskell Cafe, too,


Claus Reinke, Deling Ren, Huiqing Li, Martin Erwig and Simon Thompson