The Problem with MARC and AACR: the World Doesn’t Disco Anymore

My undergraduate background is in computer science, and from that perspective I have a great deal of admiration for MARC and AACR as well as their creators and proponents: Henriette Avram and Michael Gorman. At their creation, MARC and AACR propelled library services to new heights of efficiency and usefulness. Here’s my problem, though: we no longer live in the 1970s, and the fundamental tools of our trade should not be based in nearly 40-year-old technology.

This post started out as a comment to an LISnews thread by Kathleen de la Peña McCook that pointed back to a posting with the title Michael Gorman, Fischerspooner, Amnesty International and Hamlet. The more I thought about it, though, the more the comment went beyond a criticism of my perception of Michael Gorman’s 1;

try {
var pageTracker = _gat._getTracker("UA-16323142-1");
pageTracker._setDomainName(".libraryjournal.com");

pageTracker._trackPageview();
} catch(err) {}

GS_googleAddAdSenseService("ca-pub-1872221973353849");
GS_googleEnableAllServices();

GA_googleAddSlot("ca-pub-1872221973353849", "LJ_160x160_B");

GA_googleFetchAds();

Articles">world view. I must admit that I was also influenced by the Evolution of Dance video that is making the rounds on the net (thanks, Thomas!) — hence the byline to this post. So I’ve edited the comment and posted it here (with comments and trackbacks open, I might add) for posterity.

I noted this statement in the 1;

try {
var pageTracker = _gat._getTracker("UA-16323142-1");
pageTracker._setDomainName(".libraryjournal.com");

pageTracker._trackPageview();
} catch(err) {}

GS_googleAddAdSenseService("ca-pub-1872221973353849");
GS_googleEnableAllServices();

GA_googleAddSlot("ca-pub-1872221973353849", "LJ_160x160_B");

GA_googleFetchAds();

Articles">announcement of Henriette Avram passing: “Though Avram was a systems analyst by training, not a librarian, her work … revolutionized access to library materials.” Right on the mark. With that background in computer science I’d like to think I have a somewhat unique (or at least minority) perspective on Avram’s and Gorman’s work. They did what they did in a world with the twin challenges of expensive computing cycles and scarce storage. MARC and AACR were created when punch cards and 9-track tapes ruled the world and computers took up rooms. (Wanted one for your desk? Hah!)

Fast forward to today. Computer cycles are so cheap that we stack CPUs in huge racks to collectively work on solving a problem. Storage is so cheap that we consider a verbose, ASCII-based markup language (a.k.a. “XML”) as the state of the art in the transmission of data. Compare the information density of 1000 characters of MARC versus 1000 characters of XML. And more to the point, let’s not forget AACR where every colon, period, and dash carries meaning versus the explicit description of attributes in XML. Example? Look at

300    |a1 v. (unpaged) : |bill. (some col.) ;|c26 cm.

versus

<datafield tag=”300″ ind1=” ” ind2=” “>
  <subfield code=”a”>1 v. (unpaged) :</subfield>
  <subfield code=”b”>ill. (some col.) ;</subfield>
  <subfield code=”c”>26 cm.</subfield>
</datafield>

See what I mean? The first is 51 characters, more or less, and the second is 185 characters. Now I’d argue that the second is not much more expressive than the first. (The second is MARCXML — a raw translation of the MARC record format to XML.) But take a look at this more mainstream XML format (from MODS):

<subject authority=”lcsh”>
  <geographic>United States</geographic>
  <topic>Politics and government</topic>
  <temporal>20th century.</temporal>
</subject>

A human can read and understand that as well as being parsable by a computer.

Now anyone that would have proposed the second or third of these examples in the 1960s or 1970s would have been laughed out of the machine room and told never to come back. Sure, you could write a computer program back then to read and write those XML-based formats, but it would have been so computationally- and storage-expensive that it would never have been taken seriously, much less actually contribute to the spread of machine-based cataloging tools.

So back to this decade, when storage is cheap and computer processing power even cheaper. Interoperability with other systems outside the library domain is more and more important. There is a computer on every desk…and one in your pocket. The user is empowered with a combination of better human/machine interfaces (we’ve exchanged punch cards for keyboards, mice and graphical user interfaces) and inexpensive communication mechanisms (that make machine-aided tagging and recommendation engines possible).

I’ve never met Ms. Avram, nor have I followed her work (and in fact I didn’t know of her connection to MARC until this announcement), but I wonder what she would think of the MARC/AACR combination now. Given her training in systems analysis, would she say we are making the best use of computing technology today?

Is this the death knell for the librarian? Not necessarily. If the profession continues to train and promote the librarian of the 1960s, 70s and 80s, then yes. If that librarian is one that recognizes the shift to user-empowered technology in the past decade, nebulously characterized as “Library 2.0″, then we have a valuable value-added role to play in the information-seeking and -use activities of citizens in this new world. Are our users disco dancing? If not, we’d better figure out how they are dancing now…

The text was modified to update a link from http://lisnews.org/article.pl?sid=06/05/13/1719246 to http://lisnews.org/node/26973 on January 13th, 2011.

Footnotes

  1. "https:" == document.location.protocol) ? "https://ssl." : "http://www.");
    document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E" [] []
(This post was updated on 22-Aug-2012.)