[Haskell-cafe] Data Flow Programming in FP

[David Barbour]

On Tue, Jun 21, 2011 at 3:18 AM, Richard Senington <sc06r2s at leeds.ac.uk>wrote:
>
> I have been looking through the papers by Conal Elliott and Paul Hudak,
>
Hudak's book "The Haskell School of Expression" (chapters 13,15,17) and
>
the latest version of the Reactive library on Hackage.
>


In the past I have looked at Arrows, but I think I should have another

look now, thanks for the suggestion.


Hudak's 'Yampa' is a reactive arrows model, which I believe started with
some of his older work on Frob (functional reactive robotics). There are a
few documents out there of the form "Arrows, Robots, and Functional Reactive
Programming" [1][2]. Ross Patterson's early work on arrows also provides
example of using of Arrows to model synchronous circuits as a form of
dataflow programming [3].

Conal's 'Reactive' is an interesting study, but I don't like the model - it
implicitly uses unsafePerformIO and futures under the hood to capture
real-time behaviors, but I feel that 'real-world' integration should be
captured more explicitly by our models.

[1] http://people.cs.uu.nl/johanj/afp/afp4/hudak.ppt
[2]
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.5.2886&rep=rep1&type=pdf
[3] http://www.soi.city.ac.uk/~ross/papers/notation.html (section 6)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.haskell.org/pipermail/haskell-cafe/attachments/20110621/9fa6f4aa/attachment.htm>