I find many multidisciplinary books to be especially fascinating, and Ambient Findability by Peter Morville was no exception. Morville posits that in the age of the search engine, one of the first considerations for data or a service is how easily it can be found and not necessarily how easy it is to use (though the usability bit certainly comes into play after it has been found).
…design of a useful information system requires a deep understanding of users and their social context. We cannot assume people will want our information, even if we know they need our information. Behind most failed web sites, intranets, and interactive products lie misguided models of users and their infomration-seeking behavior. Users are complex. Users are social. And so is information.
Of course, there is a great deal of information out there to be found and not all of it is terribly relevant. It should also be obvious that the sheer amount of data ‘out there’ is growing. In particular, the amount of telemetry information from GPS, RFID tags and sensors of all sorts is adding more unstructured or semi-structured data to the mix. Morville covers this ground deftly while touching on a very approachable discussion of the nature of information, how wayfinding in the physical world might provide clues to wayfinding in the data-space, and push versus pull models of data acquisition and filtering.
Chapter 6 introduces semantic web technology and approaches and its goals in dealing with data and metadata in a structured, machinable way. These approaches are contrasted with the folksonomy approach taken by the roll-your-own-tags way of doing things such as del.icio.us .
It would seem that unless you take charge of your own findability, someone else will do it for you and not necessarily to your benefit. Typo-squatters were the unfortunate pioneers of this space. More subtle variations on this are competitors who create site to disparage your product and play up their own under the guise of some ‘neutral’ third party opinion.
What you don’t find can be just as bad. I identified a malady back in the early 90s in the context of the then emergent graphic user interfaces that mirrors the contemporary ‘Google mindset’. At the time, everyone was extolling the virtues of What-You-See-Is-What-You-Get. The problem that I saw was that the vast majority of people will assume that What-You-See-Is-All-You-Get and not venture beyond the eye candy that is most immediately in front of their noses to discover richer functionality and interactions. There are folks who truly believe that ‘if it is not in Google, then it doesn’t exist’ and that what they find in Google is the truth. This can be extremely dangerous, as it has been demonstrated over and over how easy it is to manipulate search results (not only on Google but Digg, Technorati and others).
Another issue with current search technology is that it simply isn’t fuzzy enough. For example, when I travel, I seek out CD shops to browse the racks and find new music. Part of finding new music is also simply having an ear out for what is playing in the shop itself. I have found quite a bit of new music that I would never have known to search for but was introduced to by a chance listen or stumbling across it in a section of the shop that I was looking for an unrelated item in. I hold out a great deal of hope that application of semantic web technologies might help to search for and discover items in a manner similar to these physical, chance finds.