To take the points James makes further—  [hopefully] we’ll start seeing
data produced outside of the library context brought into our workflows
more and more, with less and less overhead.

One area where the content standards and carrier/models intersect is with
regard to how Authorized Access Points are prescribed to be formed to
avoid “collision”. In an open world workflow, to attempt to get everyone
in the world to agree to refer to a person/place/thing/etc with the same
label is not advisable, nor in a linked data context do we have to. We
just need need enough data about the entity so that we can reconcile the
URIs and to allow users to disambiguate one entity from another in users
interfaces. This data doesn’t have to be captured in an authorized access
point, but rather through the various properties permitted for used with
that entity type. 

A lot of what I’ve said above is theoretical, of course, because library
systems by and large do not make use of URIs in these ways.

Steven Folsom
Metadata Technologies Program Manager
Harvard Library

On 2/9/16, 4:30 AM, "Bibliographic Framework Transition Initiative Forum
on behalf of James Weinheimer" <[log in to unmask] on behalf of
[log in to unmask]> wrote:

>On 2/8/2016 8:03 PM, Karen Coyle wrote:
>> This brings to mind something that I have often heard Diane Hillmann
>> say: that catalogers do not talk in terms of AACR (or now, RDA) rules,
>> but instead they speak "MARC". This of course is a confusion of the
>> content and the carrier, something we should be more carefully aware
>> of. To consider yourself cataloging in BIBFRAME would be making the
>> same mistake. The cataloging rules are RDA or ISBD or DACS (etc.), and
>> that determines the content of the catalog entry. The carrier is MARC
>> or BIBFRAME or ISBD or any number of other possible data carriers. It
>> should be possible to transport our data in a variety of carriers, and
>> to define a variety of profiles for different uses.  We do need to be
>> able to catalog in all of the detail of our rules, even if many
>> applications use only a portion of that detail.
>I don't think I agree with this. When catalogers speak, they speak
>primarily in terms of rules and procedures, not so much in terms of
>MARC. A quick review of Autocat or RDA shows this. When catalogers do
>use MARC, it is more of a shorthand for the rules, e.g. they might use
>245$b in place of typing out "other title information".
>What catalogers do not talk about is rules other than their own.
>Occasionally, catalogers might bring up other, related cataloging rules
>that are still library-based, e.g. rare books, but they almost never
>discuss how non-library related sets of rules, e.g. Onix, handles a
>bibliographic concept. When the universe is widened even further, for
>instance in the case of author-created metadata, no discussions at all
>take place. In my opinion, this is highly unfortunate since library
>records will be encountered more and more often *outside* of the
>traditional library systems--after all, that is one of the main purposes
>of linked data--and catalogers should concern themselves with making
>their records coherent in those settings.
>A single example of what I mean is: how do we make a subject heading
>such as this:
>Dostoyevsky, Fyodor, 1821-1881--Homes and haunts--Russia
>which is both useful and coherent when it is within the library catalog,
>but when taken out (as here) becomes something very strange. Absolutely
>nobody (except for a library cataloger) would ever think to look up
>anything like this. How do we make this type of bibliographic oddity
>less bizarre in the linked data universe?
>James Weinheimer [log in to unmask]
>First Thus
>First Thus Facebook Page
>Personal Facebook Page
>Cooperative Cataloging Rules
>Cataloging Matters Podcasts
>The Library Herald
>[delay +30 days]