Re: rant on the morality of confidentiality (fwd)

[dlv@bwalk.dm.com (Dr.Dimitri Vulis KOTM)]

Jim Choate <ravage@ssz.com> writes:
The Nuriweiller wrote:
> > if so, they are not SCIENTISTS. a key aspect of SCIENCE is publishing
> > results. science cannot advance without it. name me one scientist
> > who did not publish an important result, or is considered a good
> > scientists for doing so!
>
> I must agree here. If a technologist (ie one who studies science for
> profit, hence creating a technology) chooses not to publish their results
> that is fine. However, a scientist is one who studies nature and its
> interactions, profit is not and should not be a motive. Simplistic or not;
> in fact some things are better understood when simplified (ala the
> scientific principle). A scientist has an obligation to discuss and publish
> their results for other scientists (and even technologist) when they are
> reasonably sure their results will stand up to indipendant verification (a
> critical issue in science, not in technology however).
>
> Don't be confused by Timy's claim to be a scientist, he is a technologist at
> heart. Many of his views and beliefs are motivated by issues of control *not*
> curiosity.

That's a very interesting idea.

Consider Fischer Black, who passed away a couple of years ago. His most
important contribution to science was the basic Black-Sholes equation.
What were his direct economic rewards for having come up with it? Not much,
really. He was already a tenured full professor at MIT. However as the
result of his discovery he got hired away by Goldman Sachs as a VP, and
later became a full partner. He did quite a bit of work at GS; none of it
as spectacular as the Black-Scholes equation; almost none of it published
in the open literature. Can we say that the bulk of his $50 million was for
the research he did while at MIT and not at Goldman (and which benefited
everyone in the industry, not just Goldman)? Did Goldman bet that Black
would deliver results comparable in importance to the B-S equation, which
Goldman would keep proprietary? Did Goldman win this bet (meaning, we
wouldn't really know if they did)?

I know another guy whom I won't name because he's still alive. He too is a
tenured professor and has published numerous papers in refereed journals.
His results are used by many people in the financial industry to make money.
A few years ago he made an interesting discovery in statistics. Instead of
publishing it, he and his coauthor took it to some investors and showed them
how to make lots of money trading on these results. The investors then said,
basically: yes we signed a nondisclosure agreement, but now that we know
what this is about, we're going to use it and we won't pay you a penny and
you don't have the money to sue us. Which is precisely what happened; the
result is still not published, but is slowly circulating through the quant
investing community.

The same guy informed me later that he discovered a closed-form solution to
some very interesting problems 9related to Black-Scholes) which according to
the open literature either can't be done accurately at all, or require
incredible amounts of cpu time for monte carlo simulations. he doesn't
wish tio publish it (although it would make him quite a celebrity) and
now he's going crazy trying to figure out a way to sell it in such a way
that he can't get screwed again. How would crypto help if at all?

Oh and by the way I suspect that a few minor crypto ideas in my PhD thesis
were known to certain British cryptographers in the 30s but never published
in the open literature.

---

<a href="mailto:dlv@bwalk.dm.com">Dr.Dimitri Vulis KOTM</a>
Brighton Beach Boardwalk BBS, Forest Hills, N.Y.: +1-718-261-2013, 14.4Kbps