RE: Dublin Core as a replacement for MARC21

Posting to Autocat

On Fri, 10 Dec 2010 11:16:26 -0600, Joel Hahn wrote:
[Concerning my statement that MARCXML is just as incomprehensible as our normal formats]

>I’ve heard that argument before, but that very issue is true of many if not most XML implementations, across every industry. (Especially if you don’t speak English.) If you want to work with a data file, you need to have the file format’s documentation and the data format’s documentation handy.

>In the case of a programmer who wants to work with MARCXML, he doesn’t necessarily need a trained cataloger to translate it all for him (though that would make the process significantly easier); what he truly needs is access to <> and <>. Access to AACR2 & RDA would also help for figuring out the detailed foibles of the data formatting within the tags & subfields (and having both locked up and only available for a pretty stiff fee certainly makes matters worse), but since neither AACR2 or RDA is directly tied to the actual coding, the content definitions at <> alone should be sufficient to get a project going.

I agree with much of this and is why I mentioned that I have gone back and forth over what the future should be in this regard. Although MARCXML *can be* manipulated, I don’t think non-librarians will take the effort to do it. One example is exactly what you mention: looking at the MARC coding standards to try to understand what a field means is incredibly complicated. As an example, I want to point to a discussion I had on NGC4LIB with a programmer on exactly this issue, discussing something that is seemingly so simple: title. The very concept of a title is far more complex than a non-specialist would ever suspect, and the definitions are incomprehensible to a non-specialist.

Therefore, telling people to look things up in highly technical standards is just not a realistic answer, since only specialists can understand what is written in the standards. I can’t understand what is written in the standards for housing or medicine or electrical conductors or much of anything outside of the few specializations I have. Expecting outsiders to understand our MARC standards is simply asking too much of them. As an example from our own field, they got rid of the 440/490 structure for series even though it duplicated information and effort because (I guess) it was so complicated for people to understand the purpose of the 440 vs. 490. OK, I accept that, but this is child’s play compared to other parts of the bibliographic record.

I am not saying I know the solution, but it is clear to me that it will be up to us to fit our records into the new information world since the new information world will not exert itself for us. Why? They don’t feel they need to. We are already seen as terribly backwards by the rest of the information community, focusing our efforts on “pieces of dead trees” (I do not agree), but these are the people who will be in control of the new “information environments” and not us.

Whatever we do, we must not expect everyone to redo everything and spend loads of time reworking and understanding our records. We must fit into what they make. And that represents a profound change.