[ih] History of Online Databases

[Dave Crocker]

Rick,

Friday, January 17, 2003, 12:09:02 PM, you wrote:
R> Does anyone know (or can point me in the right
R> direction) when the first online databases appeared?

R> I'm particularly interested in the first web-enabled
R> databases -- how long after the launch of the WWW was
R> it before anyone wrote a web-specific program, and
R> what were the first databases used for? 

The web came at least 15 years after there were significant online
databases, including commercial ones.

For example, there were databases for lawyers by the mid-70s, I
believe.  And the New York Times database (online morgue of articles)
was around by the mid-70s.   These, obviously, were for the kind of
non-technical users that you cite.

The NYT database was interesting.  The public history is that it was
wildly successful as a public service.  In fact, however, it was a
complete failure, relative to its original goal.  Its original goal
was to replace the experts, down in the morgue, who did research on
behalf of reporters.  (Reporters would call down to the morgue, say
what they wanted, and the expert would forage around to find it.) That
is, the goal was to have reporters directly make queries to the
database and find the articles they needed.

The problem was that the user interface was far too difficult to use.
So the reporters still had to talk to an expert -- an expert at the
database user interface -- to specify what was needed. However the
result of using the online database was *vastly* better research. And
the ability to make the search service available publicly.

Along with such examples as the host names database -- which I tend to
agree was probably the first large-scale distributed data base --
there was also the database of RFCs.  I do not know when it became
accessible online, but it would have been pretty early.

The folks active in Library and Information Sciences probably also had
online databases quite early. I know that the best computer user
interface design work at UCLA in the mid-70s was in that department.
Not engineering or psychology.

As to the CCA Datacomputer, it is probably worth noting that very
early on they operated a network service. You would connect to a
special port and it would spit out a randomly chosen limerick and then
close the connection. Most of the limericks were filthy. Exceptionally
filthy. But that was not a requirement.

If you contributed 3 limericks that were not already in the database,
you got a copy of the whole thing. I think this was around '73 or '74.

d/
-- 
 Dave <mailto:dhc2 at dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850