A few steps backwards

[Nick Arnett]

I had a small epiphany at last week's distributed searching and indexing
workshop sponsored by the W3 Consortium.  It dawned on me that the
technologists who dominate the Web (and I count myself among them) are
asking librarians to compromise their standards to take advantage of the
Web.  And while I'll apologize for whatever feelings may be injured by that
statement, I'll also explain why this might not be such a bad thing... or
inevitable, perhaps.

A fundamental problem is that library science has been in existence far
longer and is far more developed than the Web.  A simple way to explain the
need for librarians to yield is that if you compromise what little there is
that makes up the Web, not much would be left.  The Web is so simple and
unsophisticated that it doesn't allow much room for accomodation!

The context of these thoughts last week was a debate, of sorts, between
Web-heads and librarians who were calling for full-on Z39.50 implementation
as the basis for distributed search and indexing.  The Web-heads were
mostly silent, shaking their heads, or explaining why that was too much
trouble, thanks-very-much-anyway.  (This is not to say that none of us are
working on Z39.50; we are.  It's a question of how much, when and for what
purposes.)

I think there's a historical precedent.  When cheap paper arrived, thanks
to the Italian paper mill nerds in the 15th century, as well as cheap
reproduction technology, thanks to Gutenberg and his fellow nerds,
publishers had to compromise a great deal to take advantage of the new,
inexpensive distribution channel.  Compared to the illustrated manuscripts
that preceded it, the Gutenberg Bible was ugly, indeed. I have no doubt
that librarians thought it an abomination.  To me, the Internet is cheap
paper; Web servers are printing presses.

Last week in Cambridge, when the idea of "Z39.50 lite" was tossed out, one
of the objections was "Don't bother, you'll just have to implement the
whole thing later, anyway."  I suspect that there were those who decided to
wait until the printing press could handle colors before they'd use it, but
they didn't see it happen in their lifetimes.

This is not to say that outside of the
Web-as-cheap-distribution-to-the-people that Z39.50 shouldn't charge right
along, doing much more than it may in the popular context.  But I suspect
that we'll all do much better if we acknowledge that the Web is still based
on very simple technology and needs to take small steps, not big ones.
Unfortunately, this means that the Johnny-come-lately nerds won't
compromise much, while far more experienced people compromise more.  But as
a technologist, I try to always bear in mind that everything we know about
Gutenberg comes from his bankruptcy records; his contemporary, Aldus
Manutius, who formed alliances with librarians, scholars and printers,
succeeded in business.

Nick