Hacker News Reader

Comment by CJefferson | Hacker News Reader

I have only looked at the slides. This looks like an interesting overview of Haskell, but, like most functional programming talks I see seems to think the rest of us really want to be using functional programming.

Notice the slides that thinks the connection between C/Java and Haskell is that C/Java programmers envy Haskell, and take ideas from it. No envy or ideas ever flow the other way.

View on HN

While there is little envy of Java in Haskell land (is there much Java envy anywhere?), there is actually a healthy flow of ideas to and from .NET, particularly C# and (naturally) F#. (Simon Peyton-Jones works at Microsoft Research, after all.)

For example, I have heard talk of extending Haskell's already nice list comprehensions with some LINQ features like orderBy.

So while the slides--which are just a backdrop for the talk rather than its entirety--may imply Haskell is insular like that, the reality is that Haskell is always happy to take features from other languages.

Finally, I think the idea isn't that everybody wants to use functional programming languages, but rather that everybody should want to use functional programming languages ;).

GHC has a feature very similar to LINQ: http://www.haskell.org/ghc/docs/7.2.1/html/users_guide/synta...

SPJ actually discusses this a bit in his talk. He says that Haskell has tried to achieve the ease and intuitiveness of imperative programming through things like the "do" syntax sugar for monads. He agrees that there's still plenty of improvement to be made; on several occasions he jokes about how Haskell programmers envy other programmers' ability to get things done quickly.

He's also joked about making strict/eager (non-lazy) eval the default. I think the heart of the joke is that learning haskell (and all the GHC extensions) involves a lot of staring at error messages, type signatures and the rest of hte code and trying to make associations.

"No envy or ideas ever flow the other way."

I think that is a misread of the slides. I see why you thought that, but in the talk itself it was more clear.

>Notice the slides that thinks the connection between C/Java and Haskell is that C/Java programmers envy Haskell, and take ideas from it. No envy or ideas ever flow the other way.

If you had actually watched the talk, you would have noticed that that's not the case at all.

The envy slide doesn't make his intent particularly clear. He is talking about how Haskell envies the usefulness of the dangerous languages, and then contrasts that with how Haskell contributes more theoretical ideas (e.g. retry/orElse for STM) back to those languages.

You've been corrected already on your misinterpretation of the slide without watching the video (at 38:08), but I wanted to add that Simon has repeatedly made the point (including this presentation) that prior to IO Monads, the designers of Haskell were envious of other languages and the slide is repurposed from earlier presentations.

Reply on news.ycombinator.com