I'm going to slide off on a tangent here. For me, the questions raised about blogs & communities and/or email v. RSS have gotten me to thinking about push & pull. And that, in turn, has connected for me with the discussion about citation that Alex, Seb, and David Brake have been holding.
When I think about push & pull, one of the first places that my brain travels to is this little section from Steven Johnson's first book, Interface Culture. Among a number of (I think) unappreciated ideas in that book, there's a spot where Steven tries to redescribe links as stylistic devices, using the example of Suck:
Suck's great rhetorical sleight of hand was this: whereas every other Web site conceived hypertext as a way of augmenting the reader experience, Suck saw it as an opportunity to withhold information, to keep the reader at bay (132).
Johnson labels the normal way of web linking (the click for additional info) as centrifugal, pushing readers to other sites, or other pages within the site. The links on Suck, on the other hand, encouraged readers to go to other pages but to return--other pages were used as a means of adding various dimensions to the page you happened to be reading. Compare this idea (of directional, or centripetal/centrifugal linking) with a lot of the other early hypertext theories and you'll find that for most writers, links are immaterial conduits. (There are some smart exceptions to this, of course.)
Okay. What does this have to do with importing academic citation index models into the blogosphere? Academic citation does a little bit of both push and pull. On the one hand, I select certain scholars and integrate their work into my own as a means of building my credibility, locating my work within a particular tradition, name-dropping, whatever. I pull their work into my own. But I also push my readers to these scholars--if I find them valuable enough to cite, then a reader who finds my work compelling may trace out my bibliographic network and read these other writers. Duh. Obvious enough.
Print bibliographies, however, blur these two different directions and various functions. In fact, they blur a lot of stuff. I've always thought it would be interesting to try and weight bibliographic entries according to how central they are to a given book, maybe just by messing with the alpha channel so that parenthetical mentions or footnotes are light grey while crucial texts get bold faced. But that's neither here nor there. For all of the imperfection of your run-of-the-mill bibliography, these different motives for citation all legitimately feed into the purpose of a citation index (unless they encounter widescale gaming, I suppose).
One of the wrenches that gets thrown into the mix with weblogs, though, is the fact that there is no generic "link." The links that I'm building into this entry are different from the links to the right in my blogroll, and those are different from the links when I visit Bloglines. Right now, I'm pushpulling with citation links, but I think of the blogroll as centrifugal and of Bloglines as centripetal. And that's to say nothing of comments or trackbacks. An "accurate" citation index would be able to weigh each of these appropriately, I suppose, but for me, one of the real advantages of that variety (contrasted with the flatness of the bibliography), is precisely that it doesn't lend itself to one-size-fits-all accounting. One example. More and more, I'm using BL as a filter for my roll, as a way of trying out sites and writers. If I stick with them, I move them to my roll, bc I understand that it's only there that they "show up." But I manage the roll by hand, so those changes tend to be rarer and slower.
And for me, those various weights attached to my links are important. They're not all equal for me. Mycology is more dynamic for the fact that I can decide how much pushing, pulling, and pushpulling to do. There is value in flattening out those categories, and treating them all simply as links--the interesting work that Alex is doing re Scholarati is evidence of that. And when Seb borrows Guédon's core-fringe metaphor to advocate for the margins, I can't argue. But I think David hits on it when he says, "Unfortunately, counting such links does not (usually) tell you anything about why the link was made (was it criticism? how significant is the linkee to the linker or vice versa?)"
The why of linking matters a heck of a lot less in academic citation, Erdos numbers nonwithstanding. But out here, as a variety of link types develop, the idea of an index is of limited usefulness, I suspect, and at worst, it would lead to even more gaming, and impose upon a dynamic, general economy of links the kind of scarcity that Guédon describes in relation to the ISI citation index.
