Of/About, part 2: Getting to Work

To summarize a number of articles (LOC, Neugebauer, Jacobs, Enser and Brighton), here are some specific hints to keep in mind when assigning your subjects:

1. Remember your audience!  Know their searches, and you know how to analyze your images to answer their searches.
2. Who, What, Where, When, Why?  They’re questions.  They’re facets of knowledge.  You have the answers to varying levels of specificity (“when” could be “May 13, 1992, 9:33 a.m.”.  It could also be “morning” or “daytime”).
3. You can’t index everything.  Try to find what makes this image worthy of being in your collection.  Is something in it significant to your audience?  Is something shown in a way that is different than the rest of your collection?  Basically, if this image shows up in a search, is the searcher informed/happy, or did you just waste the searcher’s time?
4. Is this part of a series of images?  Does the series provide a context to this image?
5. Aboutness is powerful, and audience awareness becomes even more important in addressing it.  There isn’t a distinct divide between ofness and aboutness, and a consistent policy will help control your indexers, letting them know when it’s okay to stop extracting.  In fact, as we’ll see, it may be better suited for descriptions than subject headings.

However, I’d like to throw in some other ideas, especially considering readings from earlier weeks.

1. Subjects are not the time for guessing.  I have four pictures here.  They are very similar, but your task is to assign their location.  Here are your choices:
   • Seventh Arrondisement, Paris, Île-de-France, France, Europe
   • Orlando, Orange County, Florida, United States
   • Las Vegas, Clark County, Nevada, United States
   • Minato, Tōkyō metropolis, Kanto, Japan

   I could’ve cropped the pictures to really up the ante, but no, full context.    You can probably do this, but your student assistant?  They’d have to know of these other towers before even attempting to give subjects to these.  And so…

2. Be ready to cite your sources for subjects.  And maybe your description field is the place to do so.  It’s (more) okay to be incorrect if you can say WHY you’re incorrect.
3. Unless you have an artificial-intelligence in your indexing system (spoiler: no), then you’ll need to add broader as well as narrower terms.  The system won’t know to add your image to the list of dogs if your tag is “dachshund”.
4. Be as specific as possible. But wait!  What if you only have three dog pictures in your collection.  Do you really have to add the breed tag as well?  What if that’s not the focus of this collection.  Well, here’s the thing…
5. All that stuff about “know your audience”?  All of those interoperability readings should have told you THAT was a fairytale.  Assume that, sooner or later, something will grab all of your data, cram it into Dublin Core, and throw it out into the world for everyone.  Your audience is every expert in your field, and every kid writing a report, and what was 1 of 3 dog pictures is now one of a million.  If your average dog picture is a great dachshund picture, let everyone know now, because sooner or later, everyone will want to know.
6. Balance time and resources with results.  Don’t have time to look up the breed?  Okay.  Keep your documentation ready so that, when the time comes for more indexing, you can do so.