Comments for Types, Programming, etc.

Comment on Error Handling in Haskell by jdgallag

jdgallag — Thu, 18 Dec 2008 13:59:26 +0000

yes Augustss, I have never understood why fail is in the monad class, and have never seen this point discussed. Anyone have a suggestioin why?

Comment on How to write programs in two easy steps by gilbertmana

gilbertmana — Thu, 29 May 2008 09:50:04 +0000

Intern program for more details check out... Intern Program

Comment on Correct-by-construction Concurrency by Neil Mitchell

Neil Mitchell — Thu, 03 Apr 2008 12:04:49 +0000

Very cool, I especially like the example, and the comparative lack of typing judgements!

Comment on Idris, a language with dependent types by Winiq

Winiq — Mon, 24 Mar 2008 02:48:15 +0000

nice work, brother

Comment on Idris, a language with dependent types by edwinb

edwinb — Tue, 11 Mar 2008 00:03:52 +0000

I’m certainly not dismissing Coq, and it’s a pity if what I said comes across that way. It’s an excellent mature system, can do lots of clever things, and I do use it from time to time. Unfortunately though, from my point of view, it’s really hard work to do any serious dependently typed programming in it – dependent pattern matching in particular is really hard. Things are getting better though, thanks to Russell.

I do a fair bit of dependently typed programming, and of course I know as a result that sooner or later you can’t avoid proving things. Fortunately, since Idris is actually a thin later on top of a proof assistant, such a thing is not far away.

I don’t think we are yet at a stage where we should standardise on one system for dependently typed programming. There are too many questions still to answer. A bit of diversity, with the freedom to experiment in various ways, is therefore a good thing. If I were to work with Coq (or indeed Agda) I wouldn’t have the same freedom. I look forward to playing…

Comment on Idris, a language with dependent types by Eelis

Eelis — Mon, 10 Mar 2008 21:18:27 +0000

I can’t help but feel that this is a really misguided effort.

When you start programming with dependent types, as your types get more descriptive you will inevitably end up needing to prove things. Hence, ultimately you will want the system to fulfill the role of a proof assistant, anyway.

It makes no sense, then, to dismiss Coq on the grounds that it is a proof assistant. Yes, it could use some facilities to make it more practical for actual programs. But it is a very mature system, based on a nice rich calculus, with lots of nice infrastructure, a growing standard library, lots of documentation and publications, etc. It makes much more sense to find out what Coq is missing and add it, rather than starting from scratch.

Comment on New Year Resolutions by Karolyn Leanne

Karolyn Leanne — Sun, 21 Oct 2007 19:36:41 +0000

that’s why it will never wor. Karolyn Leanne.

Comment on New Year Resolutions by Essa Morty

Essa Morty — Wed, 19 Sep 2007 17:08:34 +0000

that’s why it will never wor. Essa Morty.

Comment on New Year Resolutions by Silence, and adding things up « Types, Programming, etc.

Silence, and adding things up « Types, Programming, etc. — Fri, 11 May 2007 23:36:05 +0000

[...] by edwinb on May 11th, 2007 Well, so much for new year resolutions . I’d like to claim I haven’t posted anything for a while because I’ve been [...]

Comment on New Year Resolutions by Alex

Alex — Wed, 25 Apr 2007 13:17:44 +0000

Thank You