I know it's a bit late for comments on this, but ...
After digesting the proposed search points for a while I have one
comment. In my opinion, it would make sense to have the "role" as a
modifier to the "name" search point. It could just be me, but the
proposed author search syntax seems unworkable. Since both "name" and
"role" are top level search points, one has to contort the meaning of
the search to link the role and name in some way. Also, searching
based on role as a primary search point doesn't seem to provide much
utility -- would you search for all materials that have an
illustrator? I guess you could, but that could be done by matching
any name with a role of 'illustrator'.
My proposition would be to add a role modifier to the name index, in
addition to type, part and authority, such that queries would be
expressed in this way:
mods.name/type=personal/part=family/authority=lcauth/role=author
Other than that minor objection, which I plan to add as a local
extension, I think it is a nearly perfect expression of the useful
search points in MODS.
Thanks for all the hard work on this!
On 2/14/06, Ray Denenberg, Library of Congress <[log in to unmask]> wrote:
> Recall the recent discussion here about MODS CQL indexes for searching via
> SRU. Thank you all for your input.
>
> The SRU Editorial Board has now developed a proposal for discussion at the
> upcoming SRU meeting, March 1-2.
>
> The proposal is in three parts and covers three context sets (1) mods, (2)
> marc, (3) openurl. The mods proposal is at:
>
> http://www.loc.gov/standards/sru/march06-meeting/bib-indexes.html#mods
>
> The MODS set is intended for bibliographic searching. The proposed MODS
> indexes are based on the MODS schema - they do not presume that the data
> being searched is MODS, but MODS is used for reference semantics (in a
> manner somwhat analogous to Z39.50 bib-1 and MARC - the semantics of the
> bib-1 core Use attributes are defined in terms of MARC).
>
> The MARC set is for searching on specific MARC fields, subfields, and
> substrings; it is intended for users who are familiar and more comfortable
> with the MARC format and prefer to formulate queries using MARC vocabulary.
>
> The OpenURL CQL Context Set consists of a set of indexes corresponding to
> OpenURL keys. These indexes are primarily intended for the use of
> resolvers, which upon receiving an openURL, wish to locate the desired item
> via SRU. These are not intended for general bibliographic searching.
>
> We welcome comments and discussion on this proposal prior to the SRU
> meeting, March 1-2.
>
> --Ray Denenberg
>
--
Mike Rylander
[log in to unmask]
GPLS -- PINES Development
Database Developer
http://open-ils.org
|