At 05:47 PM 7/7/2003 -0700, Richard L. Hess wrote:
>At 05:06 PM 7/7/2003 -0400, Bradleys wrote:
>>Steven C. Barr's views are congruent with my own.
>>
>>Google is a good model to consider. How useful would Google be for most of
>>us most of the time if it only indexed important sites and the most
>>important parts of those sites on the internet? For me it would have very
>>limited usefulness. Because Google indexes everything and doesn't try to
>>make a decision as to what matters we can find things that are amazing.
>
>The expense is for conversion to Google-friendly records from
>paper/film/tape and Google does not YET index sound or image records for
>content. That raises the ante somewhat (understatement).
>
>Virage, DATE, and others do some of this.
What Google does is to compile information from the extant electronic
forms. It does not preserve the information (except in that some sites are
cached) and links to the sources.
If the documents under consideration were already in electronic catalogues,
then indexing would be quite worthwhile. The problem of collecting the
information from those documents is prohibitive, in some cases likely to be
more costly than simply preserving them.
Google does index some images and is developing some sound indexes (I keep
wanting to write indices, but of course that's another sort of index).
However, they are not based on content of the image or the sound but on the
associated electronic text at the site.
Mike
[log in to unmask]
http://www.mrichter.com/
|