[Ocaml-biz] Ocaml & Eclipse

[Brandon J. Van Every]

Marcin Wudarczyk wrote:
>
> Hello,  I  was  reading  the  archives  of ocaml-biz but now I see the
> traffic is almoust zero. What happend?

What happened is, nobody had financial stakes in OCaml business
anything.  It was all 'nice ideas'.  Until someone shows up who has
money on the table, *and* thinks that discussing things in a public
forum is a value add to them, nobody's actually going to get anything
done.  Generally speaking, open source developers don't tend to get
certain things done, because they usually have no financial stakes to
focus their attention.  Given this, I believe it will be 5 years before
OCaml is as popular as, say, Python, assuming the community slooooooowly
does a few things in the interim.

I've moved on.  After 1 year of talking to open source techies about
'biz stuff', I finally, belatedly realized it's a complete waste of
time.  In Python-land we actually had CEOs of large companies whose
buisness models relied on Python.  They had a motive to get marketing
things done... and we still couldn't get anything done.  Guido and the
Python Software Foundation applied 'stop energy' to the problem.
http://www.userland.com/whatIsStopEnergy  What I can say positively
about OCaml, is at least it's so early in marketing-days for it, that
nobody's paying attention.  Maybe someone will show up and get something
done when nobody's looking, but it won't be me.

If OCaml had turned out to be my ideal technology platform, I might have
continued with it.  But I've never liked the C FFI, and I've figured out
that low-level coding is rather important to me.  Currently I'm looking
at HLLs with better C FFIs.  Bigloo Scheme is the current guinea pig,
but the Windows toolchain for it isn't polished enough to make me
immediately happy and productive with it.  So I've been working on AI
designs rather than slogging it out with so-so toolchains.

> Anyway, I'd like to say that I cleary recognize the necessity of Ocaml
> IDE,  especially  with  integrated  debugger. The available plugin has
> very limited functionality, and it needs much work.

OCaml Eclipse support is in a rather early stage.  It's not compelling,
particularly to a Windows developer.  One could work on the Eclipse
support for that; everyone on this list has known that for 2 months,
roughly.  But speaking for me personally, I've sworn off doing basic
support for languages and tools "just to get started."  That's not
productivity.  There's no value in some "super language" if you have to
implement all these very basic things just to get a good development
environment going.  I'd sooner go find some other "super language" that
has better tools.

Haven't found it yet.  Bigloo Scheme, Clean, and Mercury all have their
various tradeoffs.  As I work more on my AI designs, maybe I'll decide
that one of these languages has clear advantages.

> Currenly  my  small  company  located in Polad is running a project of
> creation  of  Eclipse  plugin  for  programming  DSPs  (Digital Signal
> Processors) - http://www.nglogic.com:9081/nglogic/Projects/BoxView for
> our  client  from USA. Here at NGLogic we are enthusiasts of Ocaml and
> are  willing  to use it in business implementations. We are also closy
> linked  to  Faculty of Computer Science at Warsaw University, which is
> extensively using Ocaml in teaching and research.

Congrats, you're the 1st developer on this list with some semblance of a
financial stake!  At least that has spoken up, that is.  I hope you find
some willing and capable partners, but take all the foregoing as major
caveats.

> I   am   willing   to   help   in  Ocaml  plugin  implementation,  but
> unfortunately, my company is too small for investing in development of
> such  plugin  as  the  benefits for us will not be immediate and large
> enough at the beginning.

Same boat everyone's in.  Everyone wants to follow, no one wants to
lead.  Then, when someone *does* try to lead, people say they want to do
something different.  Only money will focus people.  If you don't have
the money, you're DOA.


Cheers,                     www.indiegamedesign.com
Brandon Van Every           Seattle, WA

When no one else sells courage, supply and demand take hold.