[Ocaml-biz] The tactical future of OCaml in 1 year's time

[Brandon J. Van Every]

Tony Edgin wrote:
>
> As for the time table, I must admit that I'm a email list
> newbie.  What is a
> reasonable time drop dead time for an initial strategy
> discussion?  I do know
> that if one is not set, the discussion will finally peter out with no
> resolution.

>From previous experience, going on about business sub-issues in the
Python crowd, 1 month.  If we cannot resolve basic questions of policy
and will in 1 month, then there is no such will.  I can point people at
the material evidence of Python archives if people want tangible
evidence of that.

> I may not have been clear.  I just want to decide on what
> needs to be done, not do it.

People in Python-land, with far more actual business credentials and
clout, went round and round about various things with the Python
Software Foundation and never got anywhere.  Leading to epithets from
one party, "I'm not here to save Python from itself."  And to me calling
Guido a Dilbert, after having been publically provoked by him.  I'm
amused that someone toned down my version of events on the COCAN wiki.
I accept the revision, that "the cooperation of the PSF was not
secured."  But that's like saying the cooperation of Microsoft was not
secured.  It's not my personality or 'street cred' that determines these
things.  It's a question of political will.  Who has it, who doesn't.

Based on PSF experience, I worry about INRIA's role.  But, they don't
have a role right now, and shouldn't.  We need to come up with an effort
that can sustain *itself*.  Then appoint the best diplomat to sweet-talk
INRIA into a positive and mutually beneficial relationship.  The problem
with the PSF was they held all the cards.  The business leaders did not
have the will to build their own independent effort.  They just folded
up their tents and went back to doing business when the PSF proved
obstructive.  I guess they were already doing enough Python business
that they didn't see their core businesses as being in any kind of
jeopardy.

> Modified proposed preliminary game plan:
>
> 1)  In  parallel,
> 	a) Determine market
> 	b) Bring COCAN people on board if they aren't already a
>        part of this mailing list.
> 2)  Determine useful set of tools for an Ocaml toolbox for
>     this market.
> 3)  Find existing best fit tools.

For action items, I think we should cut it off here.  All these things
are going to heavily determine what we decide as (3).  We aren't going
to come to consensus on (3).  There are going to be camps, and at some
point we'll have to make a decision about what strategy to back.
Otherwise we're talking about people picking up their toys and going
their separate ways, as I experienced in Python bizland.

Also, I have no reason to believe that (1)(b) is easy, yet it is
crucial.  I have said COCAN needs its own mailing list, but I don't see
anyone biting upon that idea, let alone championing it.  COCAN is
currently in a fairly terse, non-communicative mode.  'Somebody' is
worried about the wiki being more of a finished product for business
presentation, rather than a hotbed of discussion on what course of
action to take.  I see their use of a wiki more as "unrestricted version
control for a website" rather than a wiki culture of communication.  It
may be that the COCAN progenitors don't *want* a lot of discussion about
anything, and don't see the sense of it.

Looking farther down the road:

> 	b) Develop INRIA-approved marketing materials such as
> logos and mascots.

No.  Develop logos and mascots *first*, according to some sane strategy
that takes projected INRIA concerns into account.  (i.e. Please no OCaml
naked ninja ladies).  Leave INRIA out of the loop.  Get those logos and
mascots out into the field, get them used by the community.  Secure the
*community's* buy-in for something.  Use the field test to revise the
offering.  Hope that the 2nd shipment is a worthy faits accompli, that
the community is hooked.  Then handle the situation diplomatically with
INRIA.  Get them to bear the standard, building upon *other* business
partnerships that hopefully have been built by that time.

Hope that INRIA has no legal eagles concerned with trademark
infringement while these events are unfolding.  PSF had them and were
vigilant.

The big mistake with the PSF was doing the work, then asking for
permission.  They said no, and they were going to say no to *anyone*
submitting *anything*.  Why?  Because they had no buy-in to the effort,
and they weren't going to offer any channel by which anyone could secure
their buy-in.  At some ultimate root, it probably traces back to some
weird issues regarding Guido's personal employment history.  Issues I
unwittingly stumbled into and probably put me on his bad side.  If you
want to guess what the 'trademark turf wars' might be about, check out
http://www.python.com  I was (rightfully) concerned that the likely
website people would go to, showed a naked lady.


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

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