Learning through observation

I am enrolled in a usability research course this semester, and my first assignment for class was to conduct a site visit, to observe someone doing his/her job. This goes directly to usability research because, to perform usability testing, one must learn to observe users, must learn to pay attention to what users do--only to take in, not to assign value, criticize, or make judgment. Just to observe.

I greatly enjoyed the site visit, and I will post my write-up on here, if I get permission from the person I visited. I visited one of my former colleagues, to observe her as she created catalog records for materials; when we worked together, I never had time to do this--I was always simply a beneficiary of her labor, and I had no need to pay attention to how metadata in my system originated with this person. Now, I have learned the process that goes into creating a record--both creating a record for our own university's search engine, and a record with OCLC, that can be localized and adapted to other libraries' holdings. (I have lots of librarian friends who are welcome to post on the actual process that I watched; I'm just stating what I observed, in my own words. Please feel free to say, "You were watching this person do XYZ.")

What hit home even more to me was the goal of user observation. In understanding what people are doing, in recognizing how they make things happen--in whatever job they do, it matters not--could be making smoothies or making metadata--developing a habit of shutting one's critical brain off, that part of the brain that says, "This user did this badly, and should have done this instead, and that's how I would fix it because I know better,"--and being able to get into the habit of watching what people and respecting their work as people who use tools, teaches us how to perform usability research. Usability research is not at all about, "Here's what would fix this product." No. It's about, "Here is how this user uses the product. Here is how the product was designed," and then letting whomever hired the usability researcher know about user habits and design intents; it's not about making judgments and criticisms.

Iterative design--the constant re-evaluation of how users use a product to redesign that product--requires constant observation, constant taking in, constant understanding what people are doing with a product. Iterative design in software and websites is becoming increasingly important, particularly in terms of library/archive websites, which are constantly gaining new information, new data, and which aren't necessarily being redesigned to accommodate this new data. For instance, at ALA in San Diego this year--which I did not attend, merely heard about it--OCLC announced a new tool that harvests data from digital archives and allows their materials to be available on OCLC, which basically means that OCLC now recognizes digital materials (that have proper metadata and OAI-PMH harvest capability) as physical holdings at local libraries. This is huge--and now it is contingent upon local libraries to design their sites to be even more usable, and to constantly revisit their designs--iteratively design based on how users use their information--so that they can support the high traffic that will begin to appear.

It's an exciting time. Questions of quality metadata come heavily into play, and quality metadata will rely upon strong observation of how people search. Not observation with the intent of, "I can fix this whole situation because it's not currently usable," but observation with the intent of, "Here is how users are using this right now. This is what I'm learning and taking in."

Speaking of Discipline(s)

I am currently pursuing a PhD in Technical Communication and Rhetoric. What in the world does that mean?

Technical communication is the umbrella under which technical writing falls. Technical writing itself fills a very recent gap that opened during the mid- to late-1800s; this gap grew even larger with rapid technological innovations. These new technologies led to a need for people to be able to communicate how to work with them. Engineers aren't writers; writers aren't engineers. There needed to be some sort of bridge-people who could liaison between technology, engineers' information, and the regular person.

Technical communication became a full-blown field of study with the availability of graduate degrees, with the inclusion of critical theory into the discipline, and with the growing need for research rigor. (How, for instance, do I know that my documentation about the development of a server is functional without testing it on other people? That would require research.)

With the .net boom, technical communication has enjoyed a heyday--so many people who use computers and the Internet need to know how to do specific functions. So much information can be tracked about what people are doing as they interface with technology. These are things that a technical communicator can study, interpret, and convey.

Metadata. Where does metadata fit in?

At the moment, the field of technical communication conflates social tagging with metadata. I have read a lot about technical communicators' conception of metadata, and I've discussed it with other technical communicators, and everyone thinks that social tagging=metadata. They don't realize a few things.

Controlled vocabulary.
Authority.
The role of metadata in digital preservation.
Digital preservation. (Where will all this stuff we've written be in 10 years? 20 years?)

Right now, I plan to put these pieces together so that disciplines besides Library Science can understand how metadata, digital preservation, Web surfing, and social inter-function.

What does usability test?

Usability, for the purposes of my research interests, measures the functionality of software or a website for people to achieve the purposes for which it was intended and created. Usability tests how graphics and text interact on the image of the page; it tests whether the right words attach to the correct kind of link; it measures how effective specific text is.

Usability testing is not, however, interested in meta-information. I very deliberately choose to call this "meta-information" because I'm approaching the original idea of "meta," and I want to separate "metadata" and all its connotations from this, for right now. Meta-narrative, for instance, is the narrative outside a text that elucidates the text. Metaphysics deals with transcendence from science.

Meta-DATA deals with the outside data, that yes, is descriptive to a website, but that is also representative of a website as a whole; meta-DATA is "beyond" the site; meta-DATA is NOT the site; and in fact, META-data is that which pulls users into a site, though they never actually SEE the means by which they are pulled (metadata).

I've talked to a lot of metadata librarians who keep telling me--whom I wouldn't blame of they get slightly annoyed at my questioning--"Metadata is just a tool." (And I say this fondly because I feel friendship and gratitude for all my metadata librarian friends. I've had such fascinating conversations with them, and their patience is something for which I also have undying gratitude.)

While I agree, metadata is a tool, I also agree that my car is only a tool, a means by which I drove from Lubbock to Austin this past week. As I drove, I stopped and took photographs of windmills between Fluvana and Sweetwater; I pulled off to look at cemeteries when they were marked; after arriving in Austin, I ventured out to Barton Springs one evening and sat on a boulder to watch the rain as it fell in the water.

Let's look at the verbs that the tool, my car, allowed me to use here: "to drive" (in a car), "to stop" (in a car); "to pull off," "to arrive," "to venture." That car is one heck of a tool. Albeit, it is a tool, but it is the tool by which I was able to explore a lot.

I assert that metadata is similarly powerful. Perhaps metadata is only a tool, but I argue that it is the highway by which we access information.* If usability tests structure, functionality, layout, even information architecture of a website, should it not also test usability of the meta-information attached to a website?** Also, while libraries and archives, in digitization, recognize that metadata is important and useful, industry has not started to recognize this--and I do wonder if units of digitization actually do usability testing on the metadata itself. I've heard of libraries running usability tests on their website, but I have only heard of libraries putting together focus groups for search engine ranking improvement--rather than use testing of metadata itself. Even more, metadata is also a set of decisions created by a specific person or persons to limn how the website should be located by/present itself users. It's a powerful and deliberate decision set that brings a person from Chechnya to a website in Texas--perhaps just a tool, but a strong tool nonetheless.

(I had a long talk today with Dr. Still, my dissertation director, and this was what we went over.)

*(Do not mix my car metaphor with the highway metaphor. It's a dangerous thing to get caught in Ana's metaphor land.)

**Writing an article on this right now. Hope to send it out for publication soon.

Metadata at the Core of Web 3.0

About a month ago, I met with my dissertation committee for my first year review in the Technical Communication and Rhetoric PhD program. When they asked me what topics of research interest me, I responded with two things, the first a "first" choice, and the second as a secondary research field.

The first: A study of how the capitalist economic system and Western legal systems will have to change to accommodate a Web 3.0 world. The committee expressed boredom and skepticism with this topic.

The second: A study of how metadata affects consumption of digital materials; more important, an examination into how metadata will not only be key to what shapes Web 3.0, but also about how different disciplines should combine to improve metadata creation practices. One of the committee members jumped out of her chair in excitement, proclaiming, "THAT'S web 3.0!"

So the second topic it is.

It's taken me a month to write about this because I've been thinking it over. I'm reading up on information architecture, and I'd like to know why metadata to a cataloger is so much more important than, say, metadata to a web designer, when, in reality, it should be of utmost importance to both.

And usability and metadata. My dissertation director has mentioned a few times now about how he'd like to conduct some sort of usability study on metadata. How should that be shaped? Where does usability fit into metadata, aside from the obvious, "I was able to find it, therefore it's usable."

But what makes me think of all this now is that I often hear people discuss how there are no metadata standards, or few metadata standards, or the standards aren't very good, or how there need to be grant-funded programs to create metadata standards. What exists right now? Why is it broken? What do we do to fix it? Who should write it to begin with--and who should be fixing what exists? How do we go about preserving it? These are important questions to ask because, if standards is a state- and nation-wide issue, then obviously, the metadata we're seeing right now isn't sufficient. What more must be done?

Thank you, Carol, for making me type some of these questions up here.

Carol's Blog, "Metadata, Cataloging, and Various Librarian-Like Stuff"

URL: http://carolslib.wordpress.com/2010/02/16/metadata-shemtadata/ (Linked in title.)

My friend Carol, who works at a library services company training folks to learn how to create and work with metadata, has started her own blog about metadata and cataloging. In addition Carol is my fiercest Scrabble competitor. She regularly beats me, in fact.

But, aside from Scrabble, Carol has some important ideas to get across about metadata. I've enjoyed reading through her blog so far.