Posting to RDA-L

On 10/27/2015 11:25 PM, Gene Fieg wrote:
For instance a 260 might read: New York, N.Y. : Mcgraw-Hill, c1960
That is data string. Now we have Data bits: New York, N.Y.
then McGraw-Hill,
then c1960,
But they are not connected
How will the bits be brought together so that the patron can acquire, obtain what he/she wants?

Not entirely correct. “New York, N.Y. : Mcgraw-Hill, c1960” is a data string in this email, but the moment it is put into a MARC database and coded correctly, it is: 260 $aNew York, N.Y. :$bMcgraw-Hill,$cc1960.

At that point we have “data bits” that we can decide to display however we want, and librarians have been able to do that from the beginning of MARC. For the sake of continuity–but more probably from the inertia of printing cards–it was displayed in traditional ISBD format for a long time, and our catalogs became online card catalogs. Eventually libraries decided to experiment with record display, although they could have done so from the very beginning, to e.g. display 260$c and then 260$a and 260$b with any text or other data bits they wanted to add.

Most of these things are nothing new at all. They’ve been happening in other databases for decades, but for some reason they were just never implemented in library catalogs.

With linked data, we would have essentially the same thing except each data bit can become a URI. So:
260 $a[URI for New York, N.Y.] :$b[URI for Mcgraw-Hill],$c[Probably just a text string here, but who knows? Somebody may want this to be some kind of URI for c1960].

The coding could change into XML or JSON or Turtle, but it wouldn’t necessarily have to and in any case, the function would be the same. The real–terribly difficult–problem would be to turn the information imported through these URIs into something coherent and useful for the public instead of overwhelming everybody with floods of “information.” Also, I think we can see definite similarities with authority control (adding the correct URI instead of the correct text string).

As a result, libraries would probably opt not for the URIs for publication information but keep the text strings. With URIs, ISBD may not be applicable, but with text strings, ISBD definitely would be applicable. It is hard to predict the future, however.