Personal Identity January 14, 2007Posted by Andre Vellino in Digital library, Personal identity, Social networks.
I am a relatively recent convert to blogging. One of my initial objections was: why bother? Aren’t plain old web sites good enough? Another was: there seems to be very little “raw material” on blogs – most of the content appears to be links to other content; the actual “meat” is on traditional web sites (not entirely true, but close.)
But if everything is fundamentally defined by one’s relations to other things (to go back to Peter’s attribute / relation thesis,) then the “raw materials” of individual identity are the links, the blog-rolls, the RSS-feeds and, especially, the sequential order of entries. So, on a blog, perhaps there is no need for substantive content (recipes, pictures, ideas.) The medium i.e. the links-of-the-day, links to other blogs and, especially, the historical sequence of the entries are the very things that define your personal identity on the Internet.
This matters because I think that personal identity is key for good quality information retrieval in a digital library. Both the identity of the user seeking content and the identity of the authors of the content. We can glean a lot of “identity” information about information seekers from browsing behaviour, search queries and consumption patterns. And, in addition to citation information, there is also a lot more identity information (in the broad sense of identity defined by social networks) that can be gleaned from the documents themselves, such as inferences form the meta-data about the authors, such as their professional affiliations, their relationships to one another and their identities as information consumers themselves.
For example, some things which are no doubt relevant (besides citation information) for determining the “quality index” of an article might be: Which journal was the article published in? How does this journal relate to other journals (e.g. how much “harder” are the peer-reviewers in “Communications of the ACM” vs. “Journal of Artificial Intelligence Research”?); Who is the “senior author” of a paper with multiple authors? (and who is the graduate student who did all the work :-)); What is the work-history / education of the author(s)? (That Brian Cantwell Smith was a senior researcher at Xerox PARC might lend some credibility to some of his more unconventional points of view about Information Science – similarly, David Bohm’s relationship with Einstein might give you some hints about how to read his interpretation of Quantum Mechanics.)
Then there are questions, not about the published literature, but about gray literature. Here’s where information about who the authors are and who the reader is could be especially useful, either for narrowing or broadening the scope of a search. I think that meta-information about the authors’ / institutions / journals / citations etc. in one class of documents (e.g. peer-reviewed information) might shed some light on how to rank the gray literature. And the reverse may be true as well.