blog feed

Latest Articles in this Channel:

02/07/09--08:29: Moggi :: CT -> Hask (chan 1675890)

The monad found its way into Haskell from E. Moggi’s research in the late 1980s integrating Category Theory into the domain of programming language semantics. He began by observing that there are three common approaches to proving equivalence of programs:

by gar




02/09/09--07:56: Formal semantics for side effects (chan 1675890)

Got an idea for formal semantics for operations with side-effects like IO operations, random, etc. It's inspired by Haskell's use of Category Theory as a semantic device and by the Z specification language (the ISO Standard is downloadable; see the small print below the price list. An outdated but still useful manual is here.)

by gar




02/10/09--05:20: Programming Language Semiotics (chan 1675890)

In a recent posting to the haskell-cafe mailing list, Richard O'Keefe put his finger on a major problem that afflicts most if not all expositions of Haskell's evaluation semantics with respect to IO operations:

by gar




02/13/09--07:50: "Computation" considered harmful. "Value" not so hot either. (chan 1675890)

The term "computation" has at least six different senses, all of which are commonly used in technical literature:

by gar




02/13/09--09:26: Fixing Haskell IO (chan 1675890)

No, Haskell IO isn't broken; that's just a shameless bid for attention. But the metalanguage commonly used to describe Haskell IO is broken.

by gar




02/14/09--07:35: Pernicious Myth No. 1: The Container Fiction (chan 1675890)

Newcomers to Haskell quickly learn that lots of containers are involved. Lists and tuples are containers; a monad is like a container; constructed data contain the arguments used in construction, and so forth.

by gar




02/15/09--04:17: In Praise of Elitism (chan 1675890)

A respondent to one of my earlier posts complained that approaching Haskell by way of category theory is a "radical approach that would lead to elitism", since such an approach is likely to scare off all but the mathematical sophisticate.

by gar




02/15/09--08:06: Intergalactic Telefunctors and Quantum Entanglement (chan 1675890)

In a previous article I argued against the pernicious Container Fiction. Some respondents have quite reasonably responded that this seems a bit pedantic; it may be true that e.g. a list is not really a container, but it's such a useful metaphor that only a pedant would insist on banning it.

by gar




02/21/09--04:00: Performance Economics (chan 1675890)

Another cooked-up performance "controversy" is bubbling on the haskell-cafe mailng list: is gcc a "better" compiler than ghc? The question hinges on the performance (speed) characteristics of a relatively small, simple chunk of code when run through the respective compilers. It seems gcc makes faster code in this isolated case. I can't follow the technical details, but I do know that this kind of "performance" comparison is pointless and probably harmful.

by gar




03/01/09--08:50: Turing's Euclidean Stroke of Genius (chan 1675890)

Everybody knows Alan Turing was the father of the computer. What is less appreciated is that Turing stands right next to Euclid in the pantheon of creative geniuses. What is almost never mentioned is why Turing deserves so lofty a ranking.

by gar