"FP is a real competitive edge for a company" - The commercial potential of OCaml/FP [PDF] | Hacker News Reader

"FP is a real competitive edge for a company" - The commercial potential of OCaml/FP [PDF]

janestcapital.com

8 pointspaulgb14 comments

OT: Note to anyone who knows the Scribd guys, please have them create a feature that lets me do something like this:

http://www.janestcapital.com.pdf.scribd.com/yaron_minsky-cuf...

That way I don't have to download/upload or worry about copyrights but I can still view it on scribd automagically. I promise it will make them more popular.

For the perhaps less hardcore, and less Blub averse, F# is a caml implementation/extension? on .net. Less/no need to write homegrown, production quality c/c++ interfaces, since it integrates with the rest of .net. It will compile some/most ocaml code without too much change. I can't help but think using it, or at least being aware of it, can benefit some software startups.

The problem with f# is that its basically one guy working on it. If he were to get hit by a car, the language would die. I stopped using it until it is no longer a MS research project.

F# does look cool, and .net integration is definitely a killer feature. I haven't tried it yet, but I plan to once I am a bit better at OCaml.

I work with .NET by day and after spending quite a bit of time learning Lisp, I can tell you that even with a killer feature like that, it's not even close.

It isn't close to what?

Lisp.

F# might not be as conceptually cool as lisp (I don't know enough of either to comment yet), but I would bet that enterprise users, especially those who already have a .net codebase, would rather switch to F# than lisp. Not to mention that in the case of the former, it wouldn't have to be a switch; F# is meant to be used alongside C# instead of replacing it.

You're right about the probability of .net coders using F# over lisp. That said, F# is still a very big conceptual leap even if it does compile down to IL and run alongside all the other CLR compliant languages. The leap is having to shift your thinking from object-oriented programming to value-oriented programming.

The number of language users is not a function of how good that language is (ask PG or any lisp advocate!).

I will almost certainly learn F#, for some of the reasons above, but I don't think it'll ever match the elegance and succinctness you get with Lisp.

I agree that the number of users is not a function of how good the language is. And you are right about the conceptual difference between C# and F#. OCaml from the beginning prioritized practicality instead of elegance, so you are surely right that F# will never match the elegance of lisp. That isn't the goal though.

Here is an longer article from the same company on their experience with OCaml: http://www.haskell.org/sitewiki/images/0/03/TMR-Issue7.pdf

I notice I haven't read TMR since it became a pdf this year. Source text of the article seems to be available for those who prefer such: http://sneezy.cs.nott.ac.uk/darcs/TMR/Issue7/JaneSt.tex

I have to read about the company later, but let me at this point add a link to more such stories at the Commercial Users of Functional Programming web site: http://cufp.galois.com/

As mentioned in the presentation, using any LFSP provides a huge advantage when hiring: most of the people who aren't smart enough have already been weeded out.

Jane St. is also doing the OCaml Summer Project: http://osp2007.janestcapital.com/

Reply on news.ycombinator.com