Alexander Johannesen wrote:
> > OK, I'm full well prepared to jump the XML bandwagon if only I saw the
> > full-scale, fully operative applications that demonstrate its vast
> > superiority over the legacy systems we have. Where are they?
>
> They are being used in non-library places. :) To put it in technical
> terms, we need to have / improve a) meta data control, b) structure
> control, and c) flow control. All three of these areas can be improved
> by using XML and Schemas out of the box by putting various validation
> at different part of our systems to trap errors as early as possible.
> I'll probably write more on this tomorrow, as now I need to get kids
> into bed, do the washing up, and do some other life maintenance.
The thing is, format doesn't really matter anymore. Anybody can have any format they want locally. When we want to exchange, it's another issue. So long as we are exchanging only with library communities who are using the same MARC structure, there is little problem, but if we want to exchange outside our community, we suddenly find that MARC is very little used. XML can be used as a universal exchange format.
I doubt if there are many genuine MARC databases around today, other than some of the CDS-ISIS databases. Especially after the creation of open-source MySQL and Postgres, there is almost no need at all for the traditional MARC format. Almost all information is stored in relational database structures, and the MARC format we see is reconstituted from the database. One of the main reasons we had such trouble with global changes in the old MARC databases was that all of the numbers at the beginning of the individual records had to change. You couldn't just go into an individual MARC record and change "The old man and th sea" to "The old man and the sea," because the record length changed, along with the directory information, etc. With XML, I can open the record up in Notepad and change it in a moment, since it's a completely different structure.
As far as the numbers vs. the written-out forms go, that is another matter. 245 $a has a very clear and precise definition and those who don't know are forced to look it up to discover the meaning, which is actually quite complex. <title> may mean different things to different people, and they may *think* they know the definition when they do not. Title may mean the title proper of a book, dependent on ISBD and chief sources of information, etc., or someone may think that it refers to someone's title, "Sir" or "Misses" or "Esquire." A legal person may think more in terms of ownership and who has title of something.
Ultimately though, it doesn't matter what is used since these are just computer codes and can display however we want them to. Each person could even set their own preferences.
Jim Weinheimer
Received on Mon Aug 27 2007 - 08:02:50 EDT