Learning to code, or, My life with angle brackets

I think we can all agree that not all metadata librarians need to learn programming. But, depending on your library's setup some of you will. I fall into this latter camp. Our digital library development is on the geeky side and I've had to learn XQuery programming. It started painfully, but has turned into a joyful experience. I'm hooked.

So, you may wonder why. Well it started out that I'd be using XQuery to develop an in-house tool kit for my everyday metadata work: for things like quality control. Also, we're using DOIs and I'm extracting metadata to create an XML document to submit to CrossRef, our DOI provider.

But the plot thickens: I'm working with my colleague (who's much better at programming than I am) to develop some XQuery code to use the new WorldCat Search API for one of our digital project. I may also be involved with our digital library user interface.

At some point I'll weigh in on why I think it's helpful to understand computer programming. For now I just wanted to share a little piece of my metadata life with XML: the world of angle brackets.

Comments

[Bryan]

Christine,

I am thinking about learning some basic programming myself if only to better understand the language that other library-programmer types are speaking these days on the lists, whether be AUTOCAT, NGC4LIB, CODE4LIB, etc. I want to be able to evaluate if what they say make sense to me. I hate being ignorant of these things, and I don't want to be left behind. I also have sense that some of my work (our work) could be made easier if I just knew more about programming, database building, data modeling, validation, etc.

I am working on a small project at work to retrospectively add links, abstracts, and series information for some technical reports we produced
locally (http://www.worldcat.org/oclc/28194844?tab=details#tabs). We have scanned most of the the reports, and the .pdfs are serving as the source for the abstracts, thank goodness. I can not imagine keying all that data.

I know there must be some ways of speeding up some of the process. Perhaps a program could automatically validate that the series numbers are the same, that all the elements that I need are there, or even recommend elements that could go there besides the additional few that I am adding. I can imagine Connexion telling me "Those who have cataloged reports in this series in the past have also added these other elements. Do you want to add them to the record(s)?" Maybe Connexion could sport a recommender system (a la Amazon) for catalogers. The list seems endless.

Learning programming might also give me some insight into what in our current practice could easily shake out or be dealt with differently.

Anyway, good luck to you. I hope to hear about your progress. Thanks for the inspiration.

Bryan

[Christine Schwartz]

Hi Bryan, thanks for comment.

You are spot on that understanding and using programming can make our cataloging/metadata work easier.

For example, using XQuery programming and regular expressions I'm able to extract parts of text strings that I need for the XML document that I'm creating for CrossRef. Another example: my colleague and I wrote code to remove Dublin Core elements globally across our whole digital collection. So, with programming we are able to do significance changes and improvements to our metadata.

Your abstract looks great. Can you imagine keying in all that information?

I would definitely encourage you to explore and learn more about programming.