Search Solutions 2011: Highlights and Reflections by Tony Russell-Rose.
Of particular interest:
Probably the main one for me was Ricardo Baeza-Yates presentation “Beyond the Ten Blue Links”, which discussed Yahoo’s ongoing quest to satisfy the implicit and explicit needs of web search users, presented as a set of seven “challenges”. Some of these challenges you might have expected, such as Query Assistance (e.g. suggestions, related searches, and so on) and Universal Search (i.e. dealing with mixed media results). But other challenges were more unprecedented, e.g. “Post Search User Experience” and “Application Integration”. Both of these suggest a wider re-framing of the search problem, in which findability is just one (small) part of the overall search experience. In this context, the focus is no longer on low-level activities such as selecting relevant documents, but on recognising and providing support for the completion of higher-level tasks. This is interesting in its own right, but it also underlines Search Solutions policy of bringing together the web and enterprise search communities: in this instance, we clearly can learn a lot from each other.
The entire post merits you attention (the proceedings are online by the way) but I think Tony’s point about findability illustrates a weakness in at least how I have approached topic maps from time to time.
That is to approach topic maps as an excellent solution to authoring, finding, or maintaining information about subjects, without stopping to ask why we want to author, find or maintain the information about subjects?
However interesting or clever I find search, string comparison, networks/graphs, graph algorithms, etc., they are unlikely to be of interest to mainstream users.
Or at best, such concerns are a means to an end and not considered interesting enough to bother learning the names of the algorithms that others (including me) think are so bloody important.
Not that I think SC 34/WG 3 needs to expand its brief to include “higher-level tasks” but that in promoting topic maps, we should try to find “higher-level” tasks where topic maps can offer a substantial advantage.
Comments?