On Thu, 7 Mar 2013 14:15:15 -0800, Tennant,Roy <[log in to unmask]> wrote:
>What happened in the past can inform, but it should not be allowed to
limit,
>our vision for the future. That isn't to say that knowing how the data has
>been used (or not) in current systems wouldn't be interesting, but neither
>is it a requirement for creating a new future. Knowing what data currently
>exists is.
But sometimes useful data has never been recorded, or only indirectly: one
element I can think of is the language context of different forms of a
personal or corporate name. If we're to have context-sensitive systems (so
that a Spanish catalogue user can search under "Iņigo" as an English user
searches under "Ignatius", for the founder of the Jesuits) systems can't
reliably use that approach without the data (language of name) to support
it. VIAF goes some way but can't (yet), I think, provide the level of
language-form interchangeability that would enable us to guarantee results
under the user's language selection.
When the data hasn't been recorded, there's no use to appeal to.
One of the theoretical weakness of the FRBR analysis of bibliographic data
elements is that it starts from cataloguing practice at the time it was
formulated, and seldom goes beyond. Barbara Tillett's "access control"
(which he hoped would replace the one-form-for-all mindset of authority
control) can't go ahead without such new elements and broader approaches.
Hal Cain
Melbourne, Australia
[log in to unmask]
|
