Organizing Web Informat

[Roxanne Missingham]

What a stimulating debate!!

I am delighted to hear George Porter suggest that authors are not 
the most appropriate document indexers!!  This is not to denigrate the 
knowledge of authors about their subject but to acknowledge that both 
motivation and knowledge of the most relevant indexing thesauri (eg MESH
is not suitable for indexing of ecology material) is a specialist task.

I find automatic web indexing frustrating because of the limitations through
cultural use of language (for example US use wildfire where we in Australia 
would use bush fire), reliance upon self reporting, vaguaries of the search
options
and difficulty to determin from the 20,000+ hits factors such as level of
information.

With automated tools we need solutions which are convenient for authors 
and searchers, centralised cataloguing may not be timely or cost effective.  

Do we then need librarians/records managers/information managers to take 
responsibility co-operatively for subject areas of organisations?  Yes, I think 
that this approach which should be collaborative would take us further than 
a Fordist division of labour.

Does every index need to be down to document level?  (I think not - the 
comparision between a catalogue which lists journal titles and an index 
which lists journal articles is relevant - often I want just to go to an
organisation
on the web, and other times library clients want to find everything
published on a topic)

Does it matter which interface a clinet uses to get the information?  I
would like to
see easy use of indexes of internet resources through library catalogues as
well 
as search engines - the cutting and pasting of a search from one to the other 
should just be a temporary solution to interconnectivity

Metadata will be an excellent way of providing core information for 
automated indexes, but it will be a service when:

        * the search software is flexible (how can you search 50 million 
          records well when "near" is within 100 words in one seaarch engine?)

        * the basic information (meta data) is truely accurate and consistent
                (could spamming really diappear?)

        * document are really available on the Internet and are preserved for
          use beyond the immediate future (of a list of Internet ecology
journals &
          texts I compiled  January almost a third had moved or no longer
existed in April!)

Just my 2cents worth

Regards

Roxanne Missingham     

At 07:10 AM 16/07/1996 -0700, you wrote:
>Subject:RE>>Organizing Web Information (was: Something...  Date:7/16/96
>Time:9:12 AM
>
>Dr. Heinrich Kuhn suggests:
>
>>With now probably some 50 million documents on the web (and the 
>>number still rising fast) it is obvious that it will be impossible
>>to catalogue them all.
 ======================================================
 Roxanne Missingham		phone:  (06) 242 1659
 Divisional Librarian			fax:  (06) 241 3491
 Division of Wildlife and Ecology	email:  r.missingham at dwe.csiro.au
 CSIRO
 PO Box 84
 LYNEHAM  ACT 2602
 ======================================================