I'll take a crack at this one. (Somehow I doubt we will change your
thinking about anything, though :-)
The entire string, in this case "Roustabouts (Musical group)", *is* the
name according to a set of rules that has been used in the construction
of hundreds of millions of records. *Neither* portion of it is necessarily
exactly what appears on the item in hand. It is a constructed "controlled
form of name" designed to uniquely identify that entity wherever it
occurs, serving both to collocate those items that are produced under
meaningless variations of the "real" name and to differentiate it from
the same "real" name used by other entities.
As Rebecca pointed out, if you do not operate in an Anglo-American
Cataloging Rules environment, you do not have to construct your names this
way. MODS does not care. AACR2 formulates them like this. Other rule sets
will formulate them differently, or in most cases, not at all.
I wouldn't get hung up on this example, if I were you. Each community
will determine what constitutes a legitimate name according to its own rules.
Botanists have their conventions, STM has its conventions, libraries have
theirs. Of course, they might all be wrong... ;-)
--Robin
Robin Wendler ........................ work (617) 495-3724
Office for Information Systems ....... fax (617) 495-0491
Harvard University Library ........... [log in to unmask]
Cambridge, MA, USA 02138 .............
On Sun, 4 May 2003, Bruce D'Arcus wrote:
> I just realized I never received the below reply of Rebecca's, which I
> just found in the list archive.
>
> Anyway, this issue is, of course, a pre-cursor of the more recent
> discussion, so I'm not sure if thinking on this has changed as well.
> Mine hasn't though :-)
>
> I realize there are these rules, but I just think they're wrong. By
> this logic (in other words, if we want to be consistent), shouldn't we
> also have the following as valid, even though it's highly suspect and
> presents all kinds of processing problems?
>
> <name type="personal">
> <namePart>Doe, John (person)</namePart>
> </name>
>
> Or even worse:
>
> <name type="personal">
> <namePart type="family">Doe (person)</namePart>
> <namePart type="given">John (person)</namePart>
> </name>
>
> Is the solution you came up with for the GMD data, Rebecca -- in which
> you map the additional metadata where it better belongs so that it can
> be reconstructed -- applicable for names as well?
>
> Bruce
>
> > This is not a holdover from MARC, but is the way you would establish
> > this
> > name according to Anglo-American Cataloging Rules. It is the controlled
> > form that would be found in the national name authority file; one could
> > also include authority="naf" if encoding this way to be more explicit
> > about this. You could certainly include as you suggest below without
> > the
> > (Musical group) and consider it an uncontrolled form of name.
> >
> > Rebecca
> >
> > On Thu, 17 Apr 2003, Bruce D'Arcus wrote:
> >
> > > Looking through the example records, I see this:
> > >
> > > <name type="corporate">
> > > <namePart>Roustabouts (Musical group)</namePart>
> > > <role>
> > > <code>prf</code>
> > > </role>
> > > </name>
> > >
> > > If I understand right including the "(Musical group)" bit is a
> > holdover
> > > from MARC tradition. This makes me cringe a bit (musical group is,
> > > after all, additional metadata about the name; not the name itself),
> > > and am just wondering if this would work as well?
> > >
> > > <name type="corporate">
> > > <namePart>Roustabouts</namePart>
> > > <role>
> > > <code>prf</code>
> > > </role>
> > > <description>Musical group</description>
> > > </name>
> > >
> > > Bruce
> > >
>
--
