Re: [Caml-list] Some comments on recent discussions

[Martin DeMello <martindemello@gmail.com>]

On Tue, Dec 13, 2011 at 12:51 AM, Alain Frisch <alain@frisch.fr> wrote:
>
> But in order to get really good support in the long term, which includes
> community tools (packaging, porting libraries, support for Windows API and
> .Net, documentation, etc), I think we need to find a way to "bootstrap" a
> larger community of OCaml hobbyists, who consider Windows as their main
> platform.  (It might be the case that "native Windows users" are culturally
> less inclined to participate to an open source project, but I don't believe
> this is the primary explanation for the lack of community work for Windows.
> We simply need more people on board.)

ruby had the same problem of being heavily unix-centric for a long
time. i think what tipped it over was first a small handful of really
dedicated people putting in the hard work to make a one-click
installer and ensure critical libraries worked, followed by the
web-dev community at large deciding ruby was a really good language to
get stuff done in and thus attracting a flood of people on windows who
found that there was at least a good installer, some decent
development environments and IDEs, and a helpful community to
encourage them to stay and work out any teething pains they might
have. if ocaml could provide windows users some compelling reason to
use over other languages that already have better windows integration,
it might take off.

i personally see the big advantage of ocaml is being able to write
native apps in a powerful and expressive language, but it seems
interest in "native" is declining rather than growing over time.
(quite frankly, i'd be happy to see any of ocaml, haskell or chicken
scheme get the sort of community support and enthusiasm that c++
currently has as a language for writing cross-platform desktop apps,
but it doesn't seem likely to happen any time soon.)

another big draw would be a popular open source project that is easy
to write small ocaml extensions for, to encourage people to get their
toes wet. something like a plugin-oriented irc bot, for instance,
would really play to ocaml's strengths, and might have gained a decent
amount of mindshare if it was around when eggdrop was (i got the sense
that a lot of people who didn't particularly like or even really
master tcl nonetheless picked up enough of it to hack on eggdrop
plugins and extensions). as it stands, it's not even just a windows
problem - i see very little, in the way of either tutorials or open
source projects, to engage the casual hobbyist.

martin