RE: [RDA-L] rdacontent terms – dataset

Bernhard Eversberg wrote:

15.02.2011 20:48, Weinheimer Jim:

> In my opinion (and not only mine), this is the world we must enter, whether we want to or not. How do you enter this world? By creating Web Services. In order just to start to do this, you must use XML, since this is the language. It is not ISO2709.

Now that is of course right. Only “you must use XML” does precisely *not* mean “use XML as your internal syntax”! It just means “be able to use XML in the production and use of services”. That, in fact, is posible on systems that use MARC internally, and even ISO-MARC.

The misunderstanding here is the same that led to the internal use of MARC in ILSs in the first place. That was never really necessary, nor intended by the creators of MARC, for MARC was meant to be a communication format. In modern parlance, a service format, only that it was offline bulk services (magnetic tapes) at that time.

Again, to make it clear: Internally, in the black box that is your system from the viewpoint of the world, you can do whatever you want to structure your data. You can even (although you should think twice) use ISO-MARC – only just let nobody see it. As long as you are able to answer requests in XML *and in other syntaxes that may be asked for*, in services that the world can use. XML is not a be-all or cure-all, and in 10 years’ time it may be obsolete – we have no control over that.

May we now put that matter of ISO to rest? I’ve never liked it myself, and it *ought* to be gotten rid of, but that’s actually off-topic in this forum.

This is correct. I never mentioned ISO2709 being used internally. The internal format also probably won’t be MARC, but some kind of relational structure, or an XML structure (as in Lucene), or a mixture of both (as in Koha). Each system internally can and probably will be, quite different, just as they are today. That is beside the point. The only catalog I know of that stores records internally in ISO2709 is CDS-ISIS, but there are probably others. All that matters ultimately however, is that the final product transfers its data according to a specified format.

That aside, the matter of ISO2709 *is* of incredible importance for the transfer of our records, because so long as we use it for transferring records, we remain locked into all its deficiencies, no matter how great our internal systems may become. It’s like having a dam within a drought-suffering populace that needs water. Your dam may be able to deliver 200,000 gallons of water a second, but if the pipes are old and can only deliver 100 gallons a second, the fact is, you can only deliver 100 gallons a second, and this remains the case even if you do more work and you can deliver 400,000 gallons. Although everyone wants your water, and you want to deliver it, the pipes must be upgraded if you are to help. And if we don’t upgrade those pipes, we cannot blame the populace for looking elsewhere for what they need.

So, perhaps we create these wonderful sites that internally have, e.g. 100 subfields in a field. Maybe we want fields beyond the 999 that we have now. None of this can be transferred using ISO2709. *If* we wanted to get rid of the single main entry, by making the 100 repeatable and everything associated with it, it would be a huge undertaking in ISO2709, if it could be done at all, but fairly simple in XML. There are lots of problems.

I don’t know if XML is the ultimate solution, but that doesn’t matter. It would certainly be a step forward; a step we could take now; the developers could start working with our records now and the public–perhaps–might even begin to appreciate them; and it wouldn’t cost nearly as much as instituting RDA (to bring the topic back).

But you are right. I really do not like saying bad things about RDA on this list. The reason I harp on this is to provide a concrete example how we could adopt changes that are much less disruptive to us than the adoption of RDA, far less expensive, and that would (or at least could) have far more profound effects on the world “out there”.