Collin Vs. Blog

Last week, when I was young and impetuous, I made a couple of suggestion re the WPA Technology Outcomes Statement. In the interests of positivity, I thought I might revisit one of the suggestions, namely

How about developing an awareness of the variety of search strategies made possible by the combination of physical and electronic information sources?

One of the things that I think is hugely important here is that too many think "search" is simply a single process, delimited by pluses, minuses and quotes. "Strategies" too often refers to the ability to narrow your Google results down from 1 million to 50,000 or so. Problem is that this vision of search is not unlike bobbing for apples. It's a little flip of me to say so, I suppose, but it's true. And at one point, that's what web tech and search tools permitted.

With Google, the tools improved. The strategies? Not so much. For all of the good that Google did, driving out some truly horrid site design and searches so slow you could almost hear the hamster running in the wheel, PageRank was a huge advance that made us pretty darn lazy. There's still plenty of value to be had from Google, but it's still an information retrieval service. I use it on a daily basis, but I use it in a very specific, constrained way that takes advantage of speed and the particular needs I have.

What's happening now is something that Ebrahim Ezzy calls "Search 2.0" over at Richard MacManus's Read/Write Web:

Third-generation search technologies are designed to combine the scalability of existing internet search engines with new and improved relevancy models; they bring into the equation user preferences, collaboration, collective intelligence, a rich user experience, and many other specialized capabilities that make information more productive.

Yeah, yeah, third-gen but 2.0. Where this essay (part 1 of 2) potentially misleads is in the implication of 2.0 as something that will replace 1.0, when it's more likely that they will work in concert. One of the things that these new search engines have in common is their potential use specifically for Long Tail information. (I'll get to a review of that book one of these weeks.) The more specific the object of your search is, the fewer people there are who are likely to want it, and thus the less accessible it will be via Google, barring some arcane system of +this and -that. I'm not going to find a new book in my field via Google, unless it's to look up the web address of a publisher.

One of the things that Search 2.0 is about is not finding X, but finding Y who knows X. Here's an example: a friend asked me how I find new music. Usually, it's a combination of three things: what my friends (with similar tastes) are listening to, what selected reviewers (with similar tastes) are recommending, and what iTunes tells me are the proximate bands, songs, and playlists.

Let me talk about that last for a second, because it can sound counter-intuitive if you think about it. I'm talking about "searching" for something that I already know/have, in order to see what else is in its neighborhood. We do it in the library sometimes, right? If we know one book is useful, we might see what else is on the shelf around it. Now think about a service like del.icio.us as a site where thousands of users are building dynamic shelves or neighborhoods. Bookmark an article that you know you want to work with, and you can use del.icio.us to find all the other users who are bookmarking it, all the other things they've bookmarked beside it, and the terms they've used to categorize it.

Goodness knows, these systems and engines are far from perfect. Nick Carr, for example, observes that sites like Amazon are limited in their usefulness:

Fallows makes one observation that hits home with me. He describes how underwhelming he finds all the automated product recommendations that are always being thrown at you on the web. "In nearly a decade with Amazon," he writes, "I've yet to experience the moment of perfect serendipity when it discovers a book I really like that I wouldn't otherwise have known about." I, too, have been waiting years to experience that moment - with Amazon, with Netflix, with iTunes, with all of 'em.

These are Long Tail companies with Short Head motives, and so this isn't that surprising to me--their market share is built on factors other than specialist searches (like convenience, e.g.). I find Amazon almost completely useless for automated recommendations. But they also enable more specialized, community recommendations, and these I've found useful. If I've liked several of the books on a Listmania list, there's a good chance that I'll add one or more unread ones from it onto my Wish List. And there are a few situations where I've found autorecs useful, like when I'm returning to a particular genre after a hiatus, for instance.

As some of us non-science disciplines start to catch on to the benefits of these kinds of networks (e.g., AnthroSource, MediaCommons, et al.), the tools will exist that allow alot of people to bypass Google altogether. Not that the occasional apple bob isn't necessary, mind you, but it shouldn't be the first and last step.

The other little piece of optimism that I have is this: once we start acknowledging the value of this kind of searching and KM, then perhaps more broad-based support will exist for the construction of resources that take advantage of it. Imagine the iTunes interface, for instance, as a gateway to the scholarship in a discipline, where professors can put together "playlists" of texts in the same way that users now publish mixes, only these playlists are the syllabi for the graduate courses taught in the discipline or the bibliographies of our articles and books.

What I come bumping up against again and again is that we're too accustomed to treating our expertise as disposable. How many hundreds of times a year do we put together carefully thematic and specifically paced syllabi/playlists, only to bury them in desktop folders and wastepaper baskets? We don't collect that information in any systematic fashion because our vision of the web lacks an understanding of the motivation behind links and networks. Google was successful precisely because PageRank algorithmizes that motivation, and if these Search 2.0 tools succeed, it will be by leveraging the networks even further.

I hope we're not still bobbing for apples.

That's about it from me. There's some spots above that probably need filling out, but it's getting late. This is part of what I mean when I say that we need to understand search better, both for ourselves and for our students.

(Almost forgot: Dean)