[rak-list] Re: [RDA-L] Justification of added entries

[Armin Stephan]

Ach ja, wenn die lieben Anglo-Amerikanischen KollegInnen es doch endlich 
begreifen würden, dass sie immer noch der EDV-technischen Steinzeit 
verhaftet sind .... Seufz!

Kaum irgendwo gilt der Vorteil der späten Geburt mehr als im Bereich der 
EDV. Vielleicht hatten wir hierzulande also einfach nur Glück, dass 
unsere Systeme more sophisticated sind. Aber egal, sie sind nunmal der 
lebende und sichtbare Beweis, dass es besser geht als es die breite 
Masse Anglo-Amerikanischer Bibliotheken praktiziert.

Am 30.08.2011 23:04, schrieb Heidrun Wiesenmueller:
> On 25/08/2011, James Weinheimer wrote (on the question of whether relator
> terms should be added manually in legacy data):
>
>> This is an example of the old thinking, as I mentioned. Before the web,
>> the solution you mention was pretty much the only answer but today,
>> instead of rolling up our shirtsleeves and putting our noses to the
>> grindstone, there are other options. We are*not*  alone and we should
>> utilize that possibility. We should try as hard as possible to
>> *cooperate*  with related metadata creators. IMDB is only one such project
>> that could provide a lot of help and we should not ignore it.
> It might be worthwhile taking a look at cataloguing conventions used
> outside the Anglo-American world: According to the German "Rules for
> alphabetical cataloguing", we've added relator terms for persons such as
> "editor" or "translator" for some 30 odd years - only we call this
> information "function designators", and the list of possible designators
> is much shorter than those in RDA. I suppose it would be possible to
> extract this information from German catalogues and add it to the
> corresponding records in Anglo-American databases. This might be
> a starting point for enriching AACR2 legacy data with this kind of
> information.
>
> And on 23/08/2011, James Weinheimer wrote (on authority control):
>
>> One thing: I believe that people would love good, solid personal name
>> authority control, but they would have to learn what that means since the
>> concept has been forgotten by the majority of people because most use
>> keyword searches.
> I don't think people have forgotten the concept. I'd argue they never knew
> what it was in the first place, and I also believe that it would not be
> fair to demand that users of catalogues struggle with that kind of thing -
> this is*our*  job. It's also*our*  job to design systems for authority
> control which work in keyword searches. In German catalogues, this is no
> big deal: It simply doesn't matter whether the preferred form or a variant
> form of a name is entered as a keyword - you always get the same title
> list as a result. And, of course, this also works for subject headings.
>
> The reason why German catalogues don't have a problem here is due to the
> different data model used in our systems: Instead of typing in authorized
> forms as a text string, we create links between the title record and the
> authority record for the person, corporate body, or subject heading in
> question (using the ID number of the authority record). This means that
> not only the preferred form is available for indexing, but the variants as
> well.
>
> Actually, it's been puzzling me for some time why American librarians seem
> to be simply putting up with the fact that an essential tool of our trade
> does not work with keyword searching in their systems. Shouldn't there be
> crowds of librarians demonstrating in front of the offices of ILS
> suppliers,
> demanding that a technical solution be found for this problem?
>
> Heidrun Wiesenmueller
> Stuttgart Media University
>
>
> -- --------------------- Prof. Heidrun Wiesenmueller M.A. Stuttgart 
> Media University Faculty of Information and Communication 
> Wolframstrasse 32, 70191 Stuttgart, Germany www.hdm-stuttgart.de/bi

-- 

Mit freundlichen Gruessen
Armin Stephan
Jefe de Biblioteca
Augustana-Hochschule / Bibliothek
D-91564 Neuendettelsau
Tel. 09874/509-300
  |
  |      ,__o
  |    _-\_<,
  |   (*)/'(*)