Changing MARC

I haven't been involved much at all in the discussions of RDA (Resource Discovery and Access earlier called AACR3), but I do have a few ideas about what should change in MARC-21.

1. Reduce redundancy
2. More explicit coding
3. More authority control

Redundancy: In terms of reducing redundancy, the first thing I'd get rid of is superfluous punctuation, especially ISBD.  Don't try to mark-up information in two different ways, just use MARC tagging.  My understanding is that RDA discussions have concluded that it is 'too late to change'.  I disagree, and certainly hope that there is going to be more cataloging done in the future than in the past.

I remember looking at ISBD documents back in the 1970's and being intrigued (light-weight mark-up for citations), but puzzled as to where it would be used.  I'm still puzzled.  The inclusion of ISBD punctuation in MARC-21 is a major obstacle in not only getting the rest of the library world to adopt it, but a continuing problem in terms of interoperability with other formats inside and outside libraries.

Explicit coding:  A lot of information in MARC-21 is semi-free text, such as author dates.  Again, this makes interoperating with other formats more difficult.  There is a fine line between over and under tagging, but there are places where MARC-21 is under tagged.

Authority control:  For our FRBR and VIAF projects, the first thing we do after getting a WorldCat snapshot is to control as many headings as possible.  Authority control is one of the basic things that libraries are doing in cataloging.  We should do more of it and share more of it.

--Th

Comments

[Ed Jones]

Thom

You're right that it should be possible to automatically generate ISBD punctuation from MARC 21. It would require only a few changes (such as separate subfields for parallel titles and subtitles [which currently share 245$b]). Of course, this was also true back in 1981 when AACR2 was implemented, in what was then USMARC. Perversely, automatic generation of ISBD punctuation was a function of UKMARC that the British gave up when they joined the MARC 21 community. I've heard many arguments over the years as to why this is not feasible, none very convincing, but apparently effective.

[arkham]

I would say you're absolutely right about redundancy. It seems absurd to need to put the date in the fixed fields, the 300 subfield c, AND in the call number. It (or they) should only have to be entered once.

As for punctuation, it's absurd for catalogers to have to input nit-picky punctuation in an online environment. There should simply be an entry, sans punctuation, and the ILS-OPAC display should handle the punctuation. Talk about a waste of time and effort that could be easily remedied.

Essentially, I think an XML-type coding would, if properly executed, work much better than our current methods of MARC entry, allowing the better granularity of tagging that MARC currently doesn't support (like your example of author dates).