This is the mail archive of the cygwin@sources.redhat.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

The semi-open development model


Ben,

I am not quite certain that the cigwin list is appropriate for this thread
but I will answer here for now.  If clarification is requested, I may pursue
reply on another appropriate forum as advised.

I am aware there has been quite some discussions to the effect that most if
not all software development projects should benefit from the open source
model.  I have found such debates inspiring and useful.  However there are
some instances where economic realities may warrant to clarify, extend and
perhaps amend the open source model stance.

I was very unsettled at the news a handful of people became instant
millionaires marketing Linux software  The Linux community seemed to cheer
their good fortune on the grounds it validated their OS of favor.  The
unprecedented inequity in the monetization of Linux seemed to have escaped
all but a few.

The conscious decision to avoid valuation of the parts (contributions)
resulted in a tremendous inherent value of the all (the OS) which astute
investment bankers leveraged on the open market.  Ironically the rank and
file open source proponent unwittingly toiled away for the benefit of the
cathedral business model for the sheer sake of avoiding to define a bazaar
style business model.

We will never know how much useful code open source proponents would have
written and how much larger a movement it would have become if just a
portion of the "loot" had been re-distributed.

In order to avoid the duplication of such costly mistake, there ought to be
a clause some place that outlines some level of code ownership so that when
serious dollar derives from a piece of work those who contributed would
profit as well.

That those new millionaires did not voluntarily correct that inequity and
the willingness of the open source community to swallow the pill (i.e. not
re-evaluate their stance) is beyond my understanding.

In the case of mechanical money making algorithm the necessity for a
contribution arbitration can be very readily demonstrated in terms of
economics.

Let's hypothetically consider you are developing an algorithm that will
predict lottery numbers.  There is a point at which additional source code
exposure becomes counter productive in terms of the original aim: win
lottery prizes.

The available pool of prize money being finite, the more working copies in
use of that particular software, the more winners and the lesser the prize
share.

Financial markets being a zero sum game, profits necessarily equal
corresponding losses in somebody's account.

In both cases the success of the software does not depend on data, but on
critical algorithms.  Hence the revenue producing core logic and by
extension the code implementation may benefit from the proper amount of
obscurity.

It derives there must be an optimal number of developers that maximizes
profit distribution per contributor.

The semi-open model does not specify what should be open source and what
should not.  It simply offers a springboard between strictly open and
strictly closed models and leave it to the players to arbitrate the matter
themselves.

The players may decide that only those with significant contribution will
have full access to the code, full use of the software or may come up with
any other arrangements they wish in an open market place that seeks to
eliminate pre-conditions and arbitraries.

The semi-open model takes the view that most code is purely of supportive in
nature and that by segregating revenue critical business logic, the open
model can be extended to most sections of current proprietary commercial
software.

For a more detailed case supporting the application of market mechanisms to
the software domain see http://www.agorics.com/agoricpapers.html

"When also open to human society, computational market ecosystems will
enable diverse authors to create software entities and receive royalties for
the services they provide, and enable diverse users to mold the system to
their needs by exercising their market power as consumers. Computational
markets can be made continuous with the market ecosystem of human society."

http://www.agorics.com/agoricpapers/aos/aos.1.html

Thierry

> > One side interest of our project is to see if the bazaar style
> development
> > model can be extended to commercial semi-open source software and to
> > business building.  The reason it's semi-open is because there
> are no more
> > compelling reasons to make public a profitable trading system
> than placing
> > your private credit card information under the GPL.
>
> I don't understand what you mean here at all.  It sounds like your
> motivation for not making your source code 'fully' open is that would
> require users to copyleft their *credit card numbers*.
>
> If that is what you're worried about, have no fear.  The GPL strictly
> refers to the code, not its data.  Furthermore, consider that a
> copylefted public key encryption system exists, called gnuPGP.  If
> such a GPL'd program wasn't secure for the data put into it, people
> wouldn't use it for security.
>
> That said, I don't understand why you wouldn't want to GPL your
> trading system.  Sounds like an excellent case for bazaar development
> indeed - everyone who participates in the trading would be interested
> in seeing that the code was air-tight!
>
> ben taylor
>
> mithras@dhp.com / http://www.dhp.com/~mithras
> "If you buy a person a ticket, they'll see a movie that day.
>  But if you show a person how to sneak in, they'll see movies for
> a lifetime."
>  Mark Stanley, _Freefall_, http://www.purrsia.com/freefall/
>
>
> out quitting your day job.


--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]