Hiatus Concluded

It’s been awhile, dear blogland friends!  And while the last three months have been no slough, I don’t have a valid excuse for abandoning you…all three of you, that is.  I’m back in school again and need to finish the portfolio for which I set up this blog.  Eloise and I devised a metadata schema last semester that will look nice in the technology section, I think.   I learned XML for the occasion, and I’m glad I did.  With such extensibility and wide use, there’s no excuse not to. 

Besides the portfolio, I’ve been reading Everything is Miscellaneous by David Weinberger, which has been on my reading list forever.  As someone who’s interested in the details of search, classification and usability, the book has been a nice step back, a big picture kind of read.   I do take issue with Weinberger’s polarized view of folksonomy v. taxonomy and his opinion is the former is inherently better in a digitized world.   I will have to revisit this issue, though, because I’m only half way through the book.

Finding

Last week was a busy one for me in the world of information science.  At work, I began a project that involves designing a SharePoint library for the university’s policy documents.  I’ve been reading up on the search and taxonomy capabilities of SharePoint, while trying to get my site management sea legs.  SharePoint offers a lot of customizable views and search filters that should appeal to a lot of our users.

Also, I finally got Peter Morville’s book, Ambient Findability, from the library last week.  Ever since I discovered his blog called Findability.org, I’ve been meaning to see what he has to say about web navigation in the world of information overload.  Morville’s ideas are bounded by Moore’s Law, which acurately predicted an exponential explosion in electronic processing power, as well as Mooer’s law, which states that “an information retrieval system will tend not to be used whenever it is more painful and troublesome for a customer to have information than for him not to have it.”  Given the reality of both laws, Morville describes the various ways that we navigate the online world (which he also explains is becoming increasingly intermingled–or, as he says “intertwingled”–with the physical world) and find the information that we need.  So far, I haven’t encountered much in the way of concrete design or navigation solutions, but then again, I have a couple chapters left to go.

I also decided to check out the lastest developments in the Library of Congress’ upcoming adoption of RDA (stands for Resource Description and Access, the new cataloging rules that are slated to replace AACR2).  Apparently, LOC, the National Library of Medicine and the National Agricultural Library are recommending that RDA be more extensively reviewed and studied before it is adopted.  Hmmm.  I must admit that I haven’t had a chance to look over the proposed RDA rules yet.  Another something, something to add to my list.

Tag, You’re It

Shoe, heel, snake, eve, ahh! Photo by maintenancepic. Obtained via Flickr Creative Commons on 8/20/08

My friend Naomi just showed me a research project that Google is conducting, and it comes in the form of a simple, yet addictive little game. They’re calling it Google Image Labeler, and here’s how it works: You are paired with someone else in Googleland. You are both shown a series of images. You and your partner tag the images with whatever words come to mind. When you both assign the same word to an image, you move onto the next image and get points. You do this over and over (or in my case over and over and over and over) and the peeps with the most cumulative points are listed on the screen like top scorers in a video game. The images are often of poor quality and cover everything from portraiture to ads to nudie shots, so your brain’s got to be nimble.

Google states that the purpose of the labeler is to help “improve the relevance of image search for users.” Of course, this leaves me to wonder how the company plans on using the user generated tags to improve relevance. Research on query mapping, perhaps?  Or folksonomy?  Maybe Google is having its users generate metadata for them.  Maybe a matched set of tags later becomes descriptor for the image.  If so, this is genius.  Why hire temps to do something that people will do for free? *sigh* I wish I was privy to Google’s search secrets. For now, I’ll just be content to tag their images.

The Rebound

Photo by Jamesfishcer. Obtained via Flickr Creative Commons, 8/8/08

I was off the grid for the last few days, due to a migraine that left me bedbound and screaming alternately for morphine and a lobotomy. It didn’t help that the Blue Angels were doing their annual Sea Fair fly-by all weekend, breaking the sound barrier over my head like so many anvils.  But I’m happy to say that I’m back, both mentally and physically, and feeling the need to catch up on all things library! 

In the fall, I will be taking a class on metadata, for which I am very excited.  In library science, metadata can refer to so many things, from the theoretical concepts of the semantic web to specific metadata schemas, intended to add value to datasets (and to the web at large) and to allow for controlled search and description.  I’m hoping this class gives me some practical experience with creating metadata and inspires ideas for implementing it at my work library.  

My professor e-mailed the class over the weekend to tell us of a book called Semantic Web for the Working Ontologist: Effective Modeling in RDFS and OWL by Dean Allemang and James Hendler that will apparently give us a leg up on the the practical and conceptual stuff before class begins.  Though the book is retailing for way more than I want to spend, I’m thinking about buying it anyway.  The only area library that seems to have it is UW, and somebody’s got it checked out until October.  What can I say, my curiosity is piqued.

Photo by Parksy1964. Obtained via Flickr Creative Commons on 8/08/08

I’ve never worked with the ontologies mentioned in the book’s title, but I took a look at the OWL Ontology Language Guide to see what it was all about.  As an ontology, OWL attempts to capture and describe the various ways that words are related to each other, thus allowing searchers to form queries that stretch beyond the bounds of simple Boolean logic.  Here is an excerpt from the guide that describes the specifics:

OWL is a component of the Semantic Web activity. This effort aims to make Web resources more readily accessible to automated processes by adding information about the resources that describe or provide Web content. As the Semantic Web is inherently distributed, OWL must allow for information to be gathered from distributed sources. This is partly done by allowing ontologies to be related, including explicitly importing information from other ontologies.

In addition, OWL makes an open world assumption. That is, descriptions of resources are not confined to a single file or scope. While class C1may be defined originally in ontology O1, it can be extended in other ontologies. The consequences of these additional propositions about C1 are monotonic. New information cannot retract previous information. New information can be contradictory, but facts and entailments can only be added, never deleted.

The possibility of such contradictions is something the designer of an ontology needs to take into consideration. It is expected that tool support will help detect such cases.

Do I fully understand this? No.  I’m going to spend some quality time with OWL today, though, and see about checking out some materials about the semantic web from work.  Maybe I’ll post more about OWL once I get a handle on it.