[Ocaml-biz] The strategic future of OCaml for 2..4 years

[Olivier Grisel]

Tony Edgin a écrit :
> Now I'm going to preach to the choir.  Brandon has a good point, but maybe his 
> point is too sharp.  For Ocaml hobby programmers to be able to take Ocaml to 
> work, project managers need to be convinced its the right tool for their job.  
> The biggest thing PMs are worried about is mitigating risk and an unproven 
> language is a very big risk.  To prove Ocaml is industrial strength, a few 
> substantial projects need to be built and heavily publicized.  To aid in 
> building these projects, a cohesive tool chest of easy to use utilities is 
> needed.  I personally think an IDE should be one of them.  (Unix junkies, 
> which I am one, need to be reminded that emacs is an IDE.)   

Yep, we need integration, but integration doesn't mean arbitrary
choices. For instance, the IDE could be the promising eclipse plugin
recently advertised on the caml-list, with some integration with some
cross-plateform building tool such as OMake and findlib and unit testing
framework such as OUnit. OMake propably need better integration with
findlib, und so weiter ...

But people who prefer working with cameleon, emacs, or _vim_ (just guess
the one I prefer ... :o) can still go with it.

Contributing patches to the maintainers of those projects is the best
way to naturally make standards emerge out of the fuzzy OCaml bazaar.

> The Ocaml-biz group's goal should be to first come up with a list of tools; 
> second choose a market where the projects could be made; third, research the 
> Ocaml bazaar to find best fits for the tools; fourth encourage/help the tool 
> owners to bring them up to professional standards; fourth,  integrate the 
> tools into an easy to install package; and fifth publicize the hell out of it 
> in the market we've chosen.

I totally agree! GODI is a very good cross-platform tool to help
integrate projects together. I really think is the candidate tool that
will boostrap such an emergence process. However it definetly need
contributions from the community to reach the critical mass. GODIVA can
make it really easier to contribute packages out of projects source
tarballs. Just check it here:
http://projects.phauna.org/GODIVA/

> What are people's thoughts on this strategy?  We should also put a time table 
> on this discussion.  How about after the end of next week someone summarizes 
> the discussion into a strategy and we start implementing it?

Yep. The summary goes on COCAN wiki.

cheers,

--
Olivier

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 256 bytes
Desc: OpenPGP digital signature
Url : /pipermail/ocaml-biz/attachments/20040908/ce041c4b/attachment.pgp