English version
Accueil     À propos     Téléchargement     Ressources     Contactez-nous    

Ce site est rarement mis à jour. Pour les informations les plus récentes, rendez-vous sur le nouveau site OCaml à l'adresse ocaml.org.

Browse thread
More cores
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: 2008-12-23 (11:50)
From: Jon Harrop <jon@f...>
Subject: Re: [Caml-list] More Caml
On Tuesday 23 December 2008 09:43:59 Oliver Bandel wrote:
> Hello Jon,
> where have you been for such a long time?

My time has been split between building products around F# (in preparation for 
its world domination in 2010 ;-) and my bouncing baby boy.

> It seems, your destructive ages of blaming OCaml-team
> instead of implementing things by your own, are now gone
> (or at least on decay).

I don't want to blame the OCaml team: they did a great job with OCaml and it 
was never their goal to make what I want. After all, I would not be here were 
it not for them (I mean on the caml-list, not in existence ;-).

But I do think it is time to face facts: the current OCaml implementation has 
some serious issues that are holding it back but they can be solved with 
enough effort. My personal belief is that the other significant FPLs on Linux 
and Mac OS X (i.e. Haskell, Scala and Erlang) have even more serious 
practical issues. So I perceive OCaml's stagnation as a catastrophic loss in 
language diversity that I would like to prevent if at all possible.

I cannot see a revenue stream in this new language implementation so I can 
only devote my spare time to it, which is usually extremely limited. If 
anything, this is a very long term investment with the hope that this new 
language implementation might eventually become the dominant open source FPL 
outside Windows in a few years time. If that happens then we would be able to 
earn money from it by writing books about it and selling software for it, but 
that is a long shot.

> To have the LLVM-project implementing this is fine.
> This makes the start easier.
> Seems to be interesting and mature enough to use it.

Yes. I have found a couple of limitations and a performance issue in LLVM but 
nothing serious and I would not hesitate to recommend it.

LLVM is really thriving because it is commercially viable and, consequently, 
has serious commercial backing. Indeed, LLVM is likely to become more popular 
than OCaml in the not-too-distant future:


I think it would be very productive to learn from this development model: an 
open source ML with industrial backing would have huge potential.

Dr Jon Harrop, Flying Frog Consultancy Ltd.