Wednesday, March 9, 2011

What do you mean, you don't have metadata experience?

(This has been in my drafts for a while. I finally decided to publish it. Maybe one of these days I'll publish all the other drafts I have sitting around.)

A few weeks ago, a cataloger who thought she might soon be unemployed emailed one of the listservs I subscribe to. One of the things she wrote really caught my attention: “The few cataloging jobs that I saw required metadata experience, which I don’t have…”

I could say who wrote this and when it was written, and the post can be found in this particular listserv’s archives, but since I don’t know if this person would be ok with her name also being in a blog post, I’m not including it here. The important thing is simply that she wrote this, and that this is not the first time this has come up.

Offlist, I emailed her about this statement, saying that, actually, she does have metadata experience. MARC is a metadata standard. In fact, it’s a very complicated and unintuitive metadata standard – lots of fields and subfields. The average person, looking at a list of MARC fields, would probably not be able to immediately equate, say, “245” with “Title statement,” and yet for many catalogers who regularly use MARC it actually becomes more comfortable to see fields and subfields rather than plain English labels.

If you know MARC, you have metadata experience. You may not have experience with Dublin Core or EAD or ONIX or XML or whatever else, but you do have metadata experience, and you can apply what you have learned from MARC to learning another metadata standard. (This, of course, takes an employer who is willing to train you or give you the time to get trained - which is a valid worry, what with all the employers who seem to want new employees who can be put to work with little or no training.)

Not everyone writes or talks about MARC and metadata as if they are two completely separate things, but it has still somehow been embedded in some people’s brains that they are separate things. On listservs and in blogs, I have read complaints from people that catalogers aren’t very good at recognizing transferable skills. The mental separation of “metadata” and “MARC” is, I feel, one of the main reasons why this is so, and it cripples catalogers and makes them afraid. Skilled catalogers don’t think they are qualified to be metadata librarians (or whatever else they are called). They think that what they do is becoming obsolete.

I do believe that, one day, catalogers will probably be using something other than MARC. However, I don’t live in fear of my future and my career(*)…because I believe there will always be a place for someone who can create and edit metadata. I can learn a new metadata scheme if I need to. After all, I learned MARC.
If you'd like to know more about MARC's place in the metadata world, check out "A Visualization of the Metadata Universe." Actually, this shows you the place of 105 metadata standards - it's awesome and kind of pretty.
* - I do worry that, one day, there won't be any jobs for people like me in libraries. I could work for a corporation if necessary, and almost ended up at one during the course of my post-grad school job search, but I'd prefer to work for a library. With all the outsourcing that's going on, however, that may not always be possible.

Wednesday, February 23, 2011

The blog hasn't died, really it hasn't

I know, I haven't updated this blog in ages. I figured I'd make my first update in such a long while an update about some of the projects I've been working on, either on my own or with Tracy:
  • Fixing up the diacritics problem in the Classical Music Library and Contemporary World Music MARC records. There are still no diacritics (I'm not sure if would be possible for our OPAC to display those diacritics, anyway), but at least the stuff that looked like gibberish and was unsearchable and unreadable by human beings has been dealt with. Tracy and I worked for an hour and a half on this problem ("Diacritics Hell"). Neither one of us had any idea it would take quite so long.
  • I'm still working on adding 505 fields (contents notes) to records. Right now, I'm concentrating on PS647-PS648 (in our General Stacks area). We've got records for anthologies in this area that don't say what titles or authors they include, meaning that the only way someone would know that, say, History Revisited has a story by Harry Turtledove is through some source that is not our catalog (the record for this book now has a contents note, by the way).
  • The records to which I'm adding 505s get other special treatments, which is why I only do a few per day, or there wouldn't be any time to do anything else. To list a few things I do: add our holdings to OCLC if they're not already there, fix up the title control numbers, and add genre/form headings (which are actually just LC subject headings used in 655 fields - for now, until there are more genre authority records and I'm actually able to load those authority records, this seems to be the best way to go).
  • I'm cleaning up 505 notes that already exist in our catalog. Some of them are enhanced when they shouldn't be, which creates false hits in our title searches. Some of them are unenhanced even though they should be enhanced, which means things that should come up in a title or author search don't. I can fix it globally, but the fix is imperfect, so I have to at least glance at the records before I reload them, slowing the process down a bit.
  • I'm fixing 245 fields (Title) that have subfields in an incorrect order. The incorrect order creates display weirdness that sometimes makes it incredibly difficult to figure out what the actual title is, or see that it's the audiobook version instead of the print version. So far, I've done our vinyl records and VHS tapes, which probably took care of the worst offenders.
  • I'm still loading authority records in order to authorize our headings and make it easier to keep them up-to-date. It's a slow process that I will never finish.
I think that's it, not including all the projects that have ended up on the back burner.

Friday, January 21, 2011

Something cool - Amazon to MARC and IMDb to MARC

The Amazon to MARC converter takes information from book records and turns it into MARC. I don't see myself using the MARC records this produces, because the records would take so much cleanup that it might actually be easier to start from scratch, but I still think it's pretty cool. Plus, some aspects could be useful for my work: I could copy and paste summary information from here and avoid (I'm pretty sure) having to hunt down quotation marks and apostrophes that Connexion doesn't like, and I could potentially use this as a starting place for call numbers and subject headings.

The IMDb to MARC converter (prototype) takes information from IMDb and turns it into MARC records. I think this converter's output is actually even more helpful than the Amazon to MARC converter's - video recording MARC records take a lot of work, because of all the name access points and various notes. This would take care of some of that work, although there'd still be a lot of fixing and fiddling to do. I love the "verify names" feature (also present in the Amazon to MARC converter). I could see this tool being especially popular with libraries that, in order to save time, have a policy of basing video recording cataloging on container information - this would probably help them save even more time. Again, as with the Amazon to MARC converter, I probably wouldn't use the MARC records produced by the IMDb to MARC converter, but there are still certain things I could copy and paste into the records I end up using in Connexion.

UPDATE: The Amazon to MARC converter doesn't just do book records - I just had it generate a record for a DVD, VHS, and CD. The "classify" information seems to be drawn from OCLC - too bad, I was hoping it could help Tracy and Trudy in those cases where they have trouble finding OCLC records that match the Contemporary World Music records they're assigning call numbers to.

Wednesday, January 5, 2011

Electronic theses and dissertations

I added 115 URLs and their corresponding e-resource item records to bibliographic records for theses and dissertations today. In theory, every thesis and dissertation for which we have a bibliographic record and that is available via Proquest should now be searchable as Type: Thesis/Dissertation, Location: Online Access. Yay!

While I was at it, I also cleaned up some stray issues in the records, and added abstracts to records I was editing that did not already have them.