This is an "inofficial release" of the version of Yampa that uses GADT-enabled
dynamic optimization as described in the ICFP'05 paper "Dynamic Optimization
for Functional Reactive Programming using Generalized Algebraic Data Types".

The code is being made available only because a number of people have asked
for it. It is not supposed to be useful in any way as it stands. If it happens
to be useful to you, great, it's your lucky day!

The code is very messy at least at two levels:

  * Work in progress: lots of code fragments from trying out various
    approaches around; the code for the most successful approaches never got
    tidied up; the GADT optimizations have not been implemented throughout;
    lots of notes and comments around, some of which may still be relevant,
    others which are not, and others which only ever were of interest to me;
    etc.

  * The fundamental approach taken does not really scale that well
    as one gets a (small) combinatorial explosion of cases to
    consider for optimization. Thus there is way more code here than
    one would hope, some of it quite repetitive.

Use at your own risk! ;-)