June 28, 2006

Feed Subscription with Contact Details

I was chatting today with Jon Rosewell, a departmental colleague, about a departmental research meeting we had last week and in particular the use of academic publishing for reputation building.

At the time, this meeting threw up several recurrent themes, not least that whilst department members don't tend to engage in RAE related research activity (the only thing we are allowed to call research...) we are very involved in scholarly research effort. Another takeaway conclusion from the meeting, introduced by David Gorham (and the subject of an interesting discussion with him tonight...are you blogging it, David? Or perhaps you'll make a comment?;-) was that 'research' activity is as nothing if it isn't disseminated in some way. The outcomes of the meeting are being distilled in (yet another?) departmental mission statement, and the one that is emerging looks quite promising, not least because of the philosophy - and vision - it captures. When the final statement is decided, I may well blog it with a bit of commentary, not least because it looks like it could frame some of the reasons I have for maintaining this blog.

Anyway - back to the reputation building conversation with Jon, and one of the things that arose was the potential use of blogging for building academic reputation.

In certain extreme cases this may result in the creation of notable personal brands, that may often reflect well on, and raise the profile of, the reputation of the institutions that employ (and tolerate?) the individuals concerned. (At this point I have to ground this with a mention someone like Robert Scoble, of course, but I think in edublog circles at least Stephen Downes would also rate here.)

The perception that audiences will increasingly make appointments with personalities may also have a bearing here, as anticipated for example by the owners of PodTech.net who just recruited Scoble and tech blogger Om Malik, who recently left Business 2.0 to set up his own GigaOM .

Academic Reputation

Estimating the extent to which an academic's reputation may be enhanced (or harmed) by blogging is a fraught affair, and one that I haven't really considered, save for the occasional (ok, daily;-) look at my Feedburner stats. But that only captures traffic from the people who subscribe to my OUseful Feedburner feed and not those subscribing to the feed that comes directly off the blog (move over, chaps - there's more content (in the form of my top daily OUseful delicious links) on the Feedburner feed...).

Just measuring the number of subscribers is not necessarily useful for measuring academic reputation. Traditionally, such reputation is developed (in research terms) through publishing in high quality peer reviewed journals. I guess that high quality publications also improve the influence(?) of the academics who publish there, e.g. in the sense that those journals have a large and knoweldgeable audience who may well be influenced by the things the read. (Influence in two ways: awareness of the publication, and citable influence, where the paper makes such an impact on a reader that they go on to cite it in their own work).

Here, the key items are peer review (the gatekeeper on getting content into a publication), readership (or audience) - who subscribes to, and/or regulalry reads or trachks the contents of the journal, and citation - who picks up on the work strongly enough to want to refer to it.

Considering each of these in the context of using blogs to build academic reputation, two features would seem to be trivially (relatively!) mappable: readership (as measured by (academic) subscribers; and citation: links from other academic bloggers (with a reasonable reputation).

So I guess I'm proposing something like the following: academic blogger reputation is some function of the size and reputation of the academic readership, and the size and reputation of the number of citations the blogger has received.

Something like ExpertRank, perhaps?

Our ExpertRank algorithm goes beyond mere link popularity (which ranks pages based on the sheer volume of links pointing to a particular page) to determine popularity among pages considered to be experts on the topic of your search. This is known as subject-specific popularity. Identifying topics (also known as "clusters"), the experts on those topics, and the popularity of millions of pages amongst those experts -- at the exact moment your search query is conducted -- requires many additional calculations that other search engines do not perform.

So how might this work?

Well, it's late, this post is way too long already, but here's some stuff for the mix:

  • Host the blog on an institutional server; (here's an earlier post on a related topic: Blogging to an Institutional Repository;
  • collect stats on the domains subscribers come from, scoring ones form .edu or .ac.uk as "academic related"; similarly for comments (which tend to have a self-identification element of the person posting the comment) and trackbacks (which similarly incorporate self-identification of the person making the Trackback);
  • find a mechanism by which subscribers could set their reader user agent so that it identifies the human reader...for example, use the reader agent to transmit the location of something like a FOAF file for the human reader. In this way, readers could announce to a publisher that they have subscribed to a feed, whioch would be good for network building. (This would be open to abuse of course...but I'm just flying kites here...);

This last point has traction I think - finding ways by which subscribers can identify themselves as individuals to feed publishers - and I need to think about it some more, so no more on this here for the moment... (except to say I'm thinking of something more elegant than a user getting a feed from a password protected account that they had to set up previously. What I want is a client side solution - the one-click user subscription model used for normal feeds would have to remain... ).

(I also wonder whether a client side solution could be used to implement paced feed delivery...? )

One further potential payoff of this identity sharing from the part of the subscriber is that it could be used to support personalised feeds....

Personalised Feeds

How this also relates to the delivery of personalised feeds (e.g. as in paced feed delivery) is something I need to think about. Certainly, there may be server side implications, as this post (from November '05) on personalised feeds suggests:

Personalized RSS feeds is one of the issues that hit me in the face recently. In the past few weeks, I've subscribed to a few RSS feeds that were personalized just for me. Specifically, when I subscribed, the URL that ended up in FeedDemon / NGOS (the aggregator that I use) had a unique identifier at the end. If I subscribed a second time (pretending I was a different person), I got a different unique identifier and ended up with two feeds. This is distinct from a feed that I've customized such as a delicious tag feed that is still a generic feed that presumably multiple people will subscribe to if they use the same parameters that I do.

(The post goes on to describe problems with this model - the need for the feed producer to generate multiple versions, effectively, of the same feed; the way this approach breaks sharing the feed via an OPML feedroll, and so on.)

The feeds could be further customised by the institution through the inclusion of limited forms of advertising etc.

Just one final related point - a debate I haven't been following at all (though I have doodled about related things before as Peer Review 2.0), but think I should try to catch up on: Nature Publishing Group's Peer Review Trial and Debate:

In Nature's peer review trial, lasting for three months, authors can choose to have their submissions posted on a preprint server for open comments, in parallel with the conventional peer review process. Anyone in the field may then post comments, provided they are prepared to identify themselves. Once the usual confidential peer review process is complete, the public 'open peer review' process will be closed. Nature will report on the results after the trial period is over.

Linking this back to the idea introduced above: what I'd like is an academic blogging system in which the feed reading client can reveal the (institutionally authenticated?) identity of the person subscribing to the feed to the publisher. A commenting client that similarly reveal the (institutionally authenticated?) identity of the person commenting on an academic post. A blog posting client that can send (institutionally authenticated?) identified trackbacks to other academic blogs. A blog publishing system which can log and report on academic, identity verified subscriptions, comments and trackbacks. The publishing system should also work with non-academic subscribers, commenters, trackbacks etc., but not necessarily use the related data for academic reputation logs.

Using IP addresses, or Shibboleth, or maybe even a blogezproxy equivalent to libezproxy, to validate the academic credentials of the subscriber/commenter/trackbacker may not be so difficult???

Maybe this could all be built into some weird combination of EduPress and an eprints server?

Posted by ajh59 at June 28, 2006 01:17 AM
Comments

I really like your ideas, but I'm not sure academic blogging service is the right angle. Surely what you're describing is a social networking service. Call it MyScienceSpace(tm).

Posted by: Richard Akerman at June 29, 2006 08:01 PM

Maybe...what really got me started on the post was though was the idea of being able to identify myself to a feed publisher whenever I subscribed to their feed (without any need for logging in via an a/c name and password combination to the publisher's site), so they could see who was reading them. Only then did I strart to rambleon about how this might be used to measure reputaion.

There is an academic myspace I think already, though I forget the name of it? The NPG open peer review trial is also interesting.

Posted by: Tony Hirst at July 1, 2006 06:43 PM