Role of librarians

[weibel at oclc.org]

In response to Tony Barry's notes about changing roles of net librarians:

> 1.  Scaling difference.
> 
>    The Web encourages small documents and publishing is so cheap that far
>     more will be.   

There will be more intellectual artifacts on the net, but NOT
necessarily more intellectual value.   The Web does not decrease the
huge majority of the cost of making available ideas... the cost of
someone's time to craft a reasoned position.  The Web DOES increase the
importance of so-called grey literature... lists such as this one, for
example.  The fact that we can so easily have this discussion supports
Tony's point.   

> 2.  Economies of indexing
>     You have the full text of web material available and computing gets
>     cheaper and cheaper.  Automated full text indexing looks more and more
>     attractive compared with manual indexing on a cost basis.

Indexing is valuable in inverse proportion to the size and diversity of
the collection.   Great for a book, terrific in a controlled vocabulary 
or limited domain (medical literature is the best example).   Automated
indexing can be made better, but will not replace cataloging and classification.

> 3.  Need for descriptive cataloguing
>     Descriptive cataloging is needed because catalogues _don't_ deliver
>     information.  You need to have a detailed description to work out if
>     you will justify the effort to get the book.  On the net this is not
>     needed.

What about description of encoding format of the resource, resources
necessary to view or execute? provenance of the object? terms and
conditions of use?  Number of bytes? Sounds like descriptive cataloging
to me.  May or may not be done by librarians.  Publishers have always 
had the opportunity to provide cataloging for thier publications, but
the incentives were (are) not their to do so.  

> 4.  Economics of publishing
>     One of the main reasons libraries exist is because people's
>     appetite to read greatly exceeds their capacity to pay.  With the net
>     substantially dropping the cost of publishing and access there will be
>     less need for libraries and their catalogues.

There is absolutely no decline in the cost of publishing on the net
versus paper at this time... this MAY change, though its not obvious
that it will.  The cost of supporting and testing high quality display
of information on a variety of hardware and software platforms dwarfs
the savings of paper, unless you limit yourself to lowest common
denominator formats or (ugh) formats such as Postscript.

The role of libraries in distributing such information may well, as
Tony suggests, change. 

> 5.  Difference in the content
>     Books are fixed artifcats.  What's on the net isn't.  It may be
>     dynamic and even the boundaries of a document become unclear.

Many resources will be dynamic, and the description of such resources
must account for this.  On the other hand, there must be means to fix
the record in a permanent way... scholarship cannot proceed without a
stable (and citable)  intellectual record.  My colleague had occasion
to cite Tim Berners-Lee's CERN papers on the Web in a journal article
he published.  Three months later, in checking the galleys, he found
that the URL no longer worked.  This is what comes of leaving the
functions of the library to the net.