Re: [Caml-list] Re: GODI is shutting down

[Fabrice Le Fessant <Fabrice.Le_fessant@inria.fr>]

[-- Attachment #1: Type: text/plain, Size: 2815 bytes --]

Hi Sylvain,

I am sorry you feel that way.

For me, it's actually the opposite: OCamlPro tries to improve the whole
OCaml development infrastructure, adding features and optimizations to the
compiler, writing new generic libraries and development tools, and taking
the burden of maintaining them on the long term, so that all of us, OCaml
developers and companies, can focus on hacking our favorite application in
OCaml, without having first to rewrite all the tools you usually find with
other languages.

Also, as a service-provider, we mostly only work on tools and libraries
that are funded by our customers, because they think that these tools are
important to their business. Our current workforce on this topic is not so
big, we could do much more to improve the environment, but for that, we
need to find more customers (hint :-) ) so that we can design the best
solutions tailored to their needs.

--Fabrice
Scientific advisor for OCamlPro

On Thu, Jul 25, 2013 at 6:11 PM, Sylvain Le Gall <sylvain@le-gall.net>wrote:

> This whole discussion has drifted into various technical topics but I
> would like to give my POV on the general topic of OCamlPro vs some
> members of the community.


> I think OPAM is a great piece of work and I wished we can have two
> package manager, including GODI for which I have contributed a little.
>
> OCamlPro has the worforce to do things that most of the individuals
> cannot do: have a team of engineers working on an OCaml topic. That's
> great expect that when you know your software is on the list of topics.
> When you are a team of 1 and OCamlPro has decided to work on something
> close to what you are doing, you know that you should stop working on it
> because you cannot compete (lack of time, lack of workforce).
>
> This kind of feeling makes me extremly uncomfortable about the future of
> all the libraries I maintain. This kind of FUD is extremly
> counter-productive for any OCaml dev.
>
> To give you a quick taste of that, you can read:
> http://hal.inria.fr/docs/00/66/59/62/PDF/paper_10.pdf (fr)
>
> where you can see that ocp-build is supposed to replace OASIS
> (explanation about why OASIS doesn't fit, p2 at the beginning) and that
> it should replace findlib's META file (p12, Conclusion).
>
> Some people will see a good news in this document and some people will
> feel uncomfortable (like me).
>
> I don't think making some people feel uncomfortable is great.
>
> Given OCamlPro position, I would love to see them achieve great things
> and solve long standing issue like:
> https://github.com/lucasaiu/ocaml (a reentrant runtime which should be
> the stepping stone for parallelism in OCaml)
>
> I think it would be a lot more productive than to replace every OCaml
> tools by something starting with ocp-...
>
> My 0.02 cents
> Sylvain
>

[-- Attachment #2: Type: text/html, Size: 3757 bytes --]