RE: [xml-dev] XML-DEV list - prior art

["Len Bullard" <cbullard@hiwaay.net>]

From: sterling [mailto:sstouden@thelinks.com] 

Mr. Bullard:
I am particularly interested in your observation that 
if stealth had been based on citation, it would have 
taken a lot longer to develop."

Seems to me just the opposite might have been the case?
Please explain, why?     Thanks. 

sterling


Hi Sterling:

Because citations weren't the source for the resource.  It was an engineer's
memory of an obscure Russian math paper, not a search engine that unlocked
the knowledge.  Further, had a PageRank search engine been used, only by
digging deep and knowing exactly which questions to ask would the obscure
paper been uncovered.  This gets to the heart of a problem of working in an
emergent field:  if you are relying on ranked citations, they don't exist at
the top, and if they do exist, they may be obscured by the popular trends in
to areas that don't return innovative ideas.  In other words, the top of the
Long Tail tends to be the slowest to see a new idea, and when they talk
about the 'velocity of ideas' they are talking about their own desperation
to be on the receiving end of those ideas so they can stay on top.   The
tipping point is not about maintaining position; it is about the
unpredictability of forces.   

You can predict the next new thing; being it's owner or benefactor is what
is difficult to predict.   The butterfly effect cuts both ways.

Some nasty ironies as an aside:

O  The source of the paper was a Russian mathematician.  The first systems
to suffer the full effects of the application of the work were the Iraqi
air-defense systems which were Russian in origin (really, the Panamanians
but that wasn't a test of air defense).  There is a lesson about being a
prophet in one's own land. An apocryphal comment made the next day was that
Soviet generals were crying in their vodka because they knew that stealth
left their elaborate air defense almost helpless.   You can be sure that for
some time after that, the obscure paper was reviewed by quite a few Russian
engineers.

O  Sometimes it almost goes that way but superstition intervenes.  The A4
rocket engines and airframes that rained down on London as V2s during WWII
had their origins in the work of American Robert Goddard.  Designed as a
wonder weapon, Hitler's constant meddling believing that his forces did not
need them nor would he need long range bombers left their development in
Germany until too late to change the outcome of the war although it could
have been a near run thing for the Western forces if the V2s could have
carried a more powerful bomb.  Instead, at the end of the war the two forces
that benefited the most from German weapons research were the Bolsheviks
that Hitler hated so and the Americans that he felt so superior to.  The
Americans did recognize the contributions of Goddard but they used him to
attempt rocket-assisted launch of heavy aircraft.   Hitler had forced the
German and European nuclear scientists out of the Europe calling it "Jewish
science" thus giving away the major and most powerful bomb ever invented.

No matter how good the citation, stupid defeats innovation almost every time
unless one gets lucky, and even then, stupid is more powerful than luck most
of the time.

Lesson learned or not:

O  It is a bad idea to get rid of the abstract thinkers in a corporation as
a means of squeezing out numbers for short term gains and bonuses for the
executives.  As a very excellent CEO once told me shortly before he retired
and turned the company over to the Ivy bean counters, "that's eating the
seed corn".  Beware of any charming manager who tells you that you work for
the stockholders.  That is a charming locust.  Beware of the CTO evangelist
who's pronouncements are only about technology that you already know about.


O  Math is abstract for a reason.  Numbers that reveal patterns in the
numbers themselves can be applied to any thing countable.  They don't always
produce useful results but they can enable innovation.  In fact, most of the
time innovation comes from taking ideas in separate fields and applying them
in unusual combinations.  It can take a few generations to get right, and
that is why I say the next new thing isn't invented, it is bred.  The
synthesists learn to look for patterns that are similar so they read broadly
first, then deeply.  

O  You can either share ideas to help develop them for common opportunity or
you can as a company owner protect your sources for resources.  It is my
experience that the companies who do well do a bit of both.  Listening is
everything and timing is everything else.  You may not act on but listen to
the crazy guy talking about PZE or some other insane idea if they have an
idea for applying it to your market.  When we were first selling
markup-based hypertext systems, a US Navy representative referred to us as
the 'left wing lunatic fringe of SGML'.  When I first presented the idea of
a world wide hypermedia based system for distributed management, engineering
and workflow to a local GE manager in the 1980s, his comment was, "That will
f**k up the business".  Get ready to fight for the invention, the credit,
and the success if you really want it.  Otherwise, share with your mates.

There are clues here in the best way to apply Google and other engines, but
be very wary of eigen-values that lock or second order systems (humans or
bot use of the numbers to derive the numbers) that distort the returns every
time.   There is gold buried in the long tail but you have to dig for it.

len