[
Lists Home |
Date Index |
Thread Index
]
Michael Champion wrote:
> I'm reminded of the (possibly apocryphal) story
> that Tim Berners-Lee was ignored or scorned by
> the hypertext community of the late 80's /
> early 90's because his stuff was so trivial and
> didn't address the interesting problems.
The story isn't apocryphal. It is true. Tim's paper was, for
instance, turned down at the International Hypertext Conference (the
one in Pittsburgh if I remember right) for similar reasons. And, the
truth is that what Tim did *was* trivial and it *didn't* solve the
"interesting problems." What Tim did wasn't even new. The important
thing here isn't whether people correctly judged the technical merits
of Tim's work -- rather it is that people rarely realize that even
"trivial" solutions that don't solve "interesting problems" can be
massively and profoundly useful.
My first reaction to Tim's work was much the same as the rest
of the community at the time. Hearing that he'd worked at CERN, my
first thought was: "Oh! This is just some guy who has written a Unix
version of the VMS-based "Memex" system that I had installed at CERN
back in mid-80's." In fact, my hypertext system was even more capable
then Tim's was since it had images in its first version and was even
able to link to non-text objects like records in databases, files in
code management systems, etc. (This is why we called it
"hyperinformation" -- not "hypertext." We even supported creating
links to and from media in which you couldn't embed the links.
Nonetheless, Tim's work resulted in the massive revolution of the Web
while mine did not, nor did any of the other systems (many much more
capable than Tim's) that preceded his work. Tim's work didn't succeed
because it was technically brilliant. It succeeded because of timing,
being on the "right" platform, being open-source, solving a large
enough problems to be initially useful, etc. If you had been in the
Hypertext community during the late 80's, you probably would have
"ignored or scorned" his work as well since the reasons for its
success had little to do with actually solving what were considered
the problems of hypertext.
The "interesting problems" that Tim's work didn't solve are
*still* interesting and little progress has been made in solving them.
Read the old Vannevar Bush article "As We My Think" and you'll see
that we *still* don't have systems like what he described. We still
can't, for instance, trade "trails" of links with people and we still
don't have systems that even record or analyze such trails for private
use. We still have tremendous problems with cataloging what is in
hyperspace -- not even Google can solve this problem. And, we still
have really serious problems representing visually what a large
collection of links look like -- people still get "lost" in
hyperspace... We still only have one kind of link. (i.e. links don't
have types.)... These are only some of the "interesting" problems that
Tim's work didn't address. What surprised many people who knew about
this stuff was that you could build a useful system without solving
these problems.
Those who said that Tim's work was "trivial", or "not-new" or
"didn't solve interesting problems" were right. But those who said it
wasn't "important" or "useful" were wrong.
bob wyman
|
