Comments

Currently watching the video and it's nice so far.

Also, when Bart used the word "aspect" in his explanation of what a monad is, I was reminded of the following paper: On the relation of aspects and monads.

Nice video. The stuff about ActionBlock and dataflow programming reminded me of coroutines in Python.

Question to Bill Buxton: Any good reading material/pointers on the bimanual asymmetric action stuff?

The Tweet-link seems to be broken. Can anyone confirm?

I'm not too familiar with it, but how do H98's data constraints compare to GHC's GADTs? The GHC docs argue that the behaviour of GADTs are more useful than H98's data constraints.

(But since I don't really have experience in this particular area of Haskell, I can't tell how strong this argument is.)

Currently watching and great so far...

But on slide 20,

data Expr x => Neg x = Neg x

you don't really need the constraint in the data declaration, right?

Actually, they are even considered to be a misfeature by some. (Post found via this Haskell wiki page)

Well, yeah, they might be a bit esoteric... but imho there's nothing wrong with that.

Nice lecture.

For those who're interested, Ralf Lämmel has some older great lectures online here. Of these, I enjoyed the following lectures:

• Continuation Style
• Fixed Point Theory
• Object Encoding in Haskell

 They are great for exercising your brain.

Now this is just a wild guess from someone who's extremely novice in Category Theory, but since the functions from P to {X} and from P to {Y} were treated specially, perhaps the functor from slice category Set/{X} to slice category Set/{Y} might be interesting to study?

> PS. Sum could be implemented such that it still returns an IMonad<T>

As discussed in the video, IO<T>.Sum() does return an IO<T>. Now for IE<T>, the problem is backwards compatibility. Changing IE<T>.Sum() to return an IE<T> instead of a T breaks existing code.