Karen Coyle wrote:
>>
>
> It seems to me that we should be trying to increase the
> number of different routes that users can take to get to
> information, knowing that none will be complete and none
> will be entirely consistent and predictable.
>
What we should keep in mind here is that we have two or three
very different search types (user intentions):
1. Known item search
For this, consistency is of the essence: we dont want to
have to do several tries just to make sure what we are looking
for isn't there. (In the card catalog we knew: if it is not
in this place then we don't have it.)
Name authorities are very relevant for this purpose.
2. Collocation search
Expressions/manifestations need consistent indexing or they will
not be collocated as RDA (and AACR) would want them.
The relevant indexing for this may use identifiers for works
instead of textual strings for names and titles. If you have these
identifiers in the bib records, which is not the case in MARC.
3. Subject search
This is what the discussion here mostly focuses on!
And this is where consistency, however desirable, is inachievable.
B. Eversberg
Received on Tue May 22 2007 - 01:49:01 EDT