Tuesday, July 07, 2009

Thing 13, Tagging

Yes, I use Flickr, and yes, I use tagging on my photos. I also add tags to other people's photos to aid in their being found by other users. I have mixed feelings about allowing user tags into the catalog. I guess I'd be okay with them if they were indexed separately from LCSH and could be excluded from searches at the user's discretion. Those that want to find "something, anything" will be happy, and more serious researchers will appreciate being able to tune out the "noise".

Slightly OT, but I recently offered some Metadata consultation for our Special Collections, who host various Digital Image collections. They typically use the Thesaurus for Graphical Materials (TGM?) controlled vocabulary but expressed some dissatisfaction with it when it came to one of their newer collections. I showed them what traditional LCSH had to offer via LC's ClassWeb, and they were impressed with the expanded range of descriptors that they could put to good use.

I'm also learning more about "tag clouds" and how to interpret them, but I confess I still don't really use them all that much.

I still like the old OPAC default display and I hate it when libraries install new "Discovery tools" that either bury or remove the default OPAC. I hate OPACs and Discovery Tools that don't index subject string searches to make them list browsable but only generate and re-generate such searches as keywords; I don't care if that's how "most" users prefer or if most users "don't care", *I do*.

For sites like Flickr, tagging is admittedly fun to play around with. I think LibraryThing sort of gets into that as well, but I haven't played around with that site. The advantage of tagging is that the terminology is up-to-date, while LCSH always lags behind a little, of necessity. I've also floated the idea in the past of Amazon marketing their SIPs and CAPs Metadata to libraries, or making it available for free in exchange for an Amazon link; I suppose a clever programmer could just mine the data from Amazon.com outright, but I'd rather get their permission and acknowledgment first. SIPs = Statistically improbable phrases and CAPs = Capitalized phrases/words. These give you good "snapshot" metadata that convey more of the "aboutness" of a book. This metadata is much more valuable than the anemic, vague "subject cataloging" provided by Amazon.com itself. Amazon.com also allows user tagging, besides the SIPs and CAPs generated from the actual text of the book proper (provided to Amazon by the publishers). Amazon also allows "user images" of their products, which is especially useful for non-book items like toys and tools. User interactivity is definitely a sign of the times. Some libraries also link in Amazon.com book reviews straight into the display for bib records in their OPAC. With enough clever programming, this is easily done, though it is frequently a "hack" rather than a vendor-supplied feature. The only vendors flexible enough to permit this are the Open Source pioneers like Evergreen and LibLime (Koha). Before you "hack the OPAC", you do definitely need to know what you are doing, because you could stand to really royally screw things up, too.

Tagging has its place, but it is no replacement for professionally chosen Controlled vocabulary; it is an augmentation of existing Metadata schemes.

1 comment:

Amanda said...

Hi John,
I agree with most of what you say, except for this: "Yes, I use Flickr, and yes, I use tagging on my photos. I also add tags to other people's photos to aid in their being found by other users."

I DON'T like it when other people tag my Flickr photos to aid others in finding them - if I wanted others to find them (which usually then also means they use them without my authorization), I'd tag them myself!