Organizing Web Information (was: Something Missing)

[Temple Hoff]

This has become quite a conversation.  One from which I have already learned a great 
deal.  A quick look at my sig. at the bottom of this message will tell you that I am not 
a Librarian.  I am more to the technical end of the spectrum.  

There seems to be great amount of work out there dedicated to the idea of indexing, 
cataloging, and organizing the Interent.  While I don't what to devalue these valient 
efforts, I would like to make one observation.

It won't work! (And I don't say this often in the realm of technology!)

I don't mean to say, stop what you're doing.  These reference lists and cataloging 
attempts are good bandaids for now.  But they will never be current enough, accurate 
enough, complete enough, or easy enough to provide the research power you and your 
patrons will want.

You must understand that you are applying 16th century, two dementional concepts of 
organization to a dynamic, evolving, uncontrolable wilderness of data.  The Internet will 
not fit on a 3x5 notecard.  The same is true for cataloging.  The Dewey Decimal System 
was a cumbersome inadiquate system at conception.  Infact no library information 
organization scheme from Dewey to LC, ISBN or MARC records has ever been really 
functional.

The problem is that at the root of each scheme is the idea that the data must be 
organized and controled.  The books must be on the shelf in a certain order. This is not 
the case on the Internet, or anywhere in computer science.  Rather than attempting to 
organize data on a disk when we know the computer will be constantly moving things around 
and changing things,  we allow the computer to haphazardly store bits of data everywhere 
and charge the computer with finding it when it is needed.  

This is what must eventually happen with the Internet.  We cannot mandate that tag fields 
be attached to everything, or that it be stored in an organized manner, or that you must 
notify the world before you move or change a web page.  We must instead have super search 
engines, intelligent search engines,  that will find what you want, whatever you what, 
not from a premade list, but by searching everywhere availible at the instant you hit the 
search button.  I mean search the entire text of each source, not the first few lines, or 
the subject headings. Also, instead typing a few keywords with some operators inbetween, 
you will type in (or speak) whole paragraphs or dialogs of explanation about what you 
want in plain English, or what ever language, for the super search enigine to digest and 
essentually research for you.


When it comes right down to it, noone really wants an index of sources to dig through to 
find the answer to a question.  They just want the answer, or maybe lots of answers.


This technology may seem far off, but then again we could wakeup tomorrow to find that 
some grad-student in Washington, or Texas has made this niffty new search engine.  After 
all it wasn't that long ago that the Web started just that way. 
-- 
Temple Hoff                  E-Mail:temhof at mohave.lib.az.us                 
Library Services Coordinator           Phone:(520) 692-5703  
Mohave County Library District           Fax:(520) 692-5788