[This continues from the comments in my previous entry "Weblog spam and (my form of) retribution..."]
A couple of points, Guy:
1) You are right… the comment doesn’t exactly fit the definition of spam, which is more commonly defined as “Unsolicited Commercial E-Mail“. Comments are certainly solicited in our weblog, but — lack of TOS notwithstanding — it’s obvious that the comments we are soliciting should have at least a smidgen of relevance to the topic at hand — in this case, Globalisation. The comment posted was most definitely commercial, but it wasn’t e-mail. So, technically, it wasn’t spam, but imagine if you got this same message in your inbox at work: you’d call it spam then. Maybe we need a new term for what’s sure to be a growing problem… maybe velveeta or cheez-whiz or, just in case Kraft isn’t as easy about its trademarks as Hormel has been, cheese-food?
2) Certainly, expertise location software might have a bright future ahead of it, but… it’s got a long way to go yet. I’m still reading through Tacit’s literature and waiting for them to send me their white papers. (I don’t see that kind of depth represented in the LearningIdeas site.) What I do see in both of these products is something that resembles — dare I say the words — Artificial Intelligence. The algorithms behind the matching will require expertise at discovering expertise by burrowing through e-mail. (if u r like me, a prog burrowing thru my email wont b able 2 tell if im literate, let alone expert!)
3) What is the market for this software? My guess is that it will be useful only in large, multi-location companies or organizations where any one person is unlikely to know everyone else. But it is exactly in these large companies where the social effects of office politics will come into play — that is not to say that politics does not occur in smaller companies. In their very interesting and informative paper entitled “Just Talk to Me: A Field Study of Expertise Location” [pdf], David W. McDonald and Mark S. Ackerman quote research from T.J.Allen which stated;

“Information seekers weigh the psychological cost of asking, including loss of status, expected reciprocity (i.e., likelihood of returning the favor), and social equity (i.e., how well they know the person socially).”

During their own research, McDonald and Ackerman spent 5 months onsite at a 100-person software company, recording, surveying and observing the behavior of the employees in requesting expert help from other employees. One of their major observations was that the software company employees — purposely or not — divided their expertise-hunt into 2 tasks: expertise identification and expertise selection. Both of these tasks have deep social components, access to which might never be available to a computer. Three examples mentioned in the paper are the expert’s “attitude”, her language skills, and simple social protocols such as whether her door is open or closed. The political — and legal — aspects flow right from these deteminations: should such things as “uncooperative attitude” or “inability to put an English sentence together” be entered into a computer system as criteria for determining who is expert?
One of the paper’s conclusions is that expertise location systems might be most useful as “augmentative systems; these would help users identify and narrow (rather than completely select) potential candidates for expertise.” I can see this as a potential role for the software, but is it enough to justify the tens of thousands of dollars that will be spent on these systems? And will the publishers of the software be willing to admit such a minor role for such expensive installations?
LearningIdeas’ product is already over-promising, since the implementation of its “patent-pending Match and Route engine” claims to remove any requirement for choosing a resource from the person in need of assistance. Tacit, because of its concern for privacy, has more of a role for the individual in choosing whether or not to contact or be contacted.
I don’t think that “Expertise Location” has become big enough yet to be recognized by its initials, let alone suffer the spectacular flameouts of such terms as AI or KM. Still, a look at the list of papers presented at the ACM-sponsored Computer Supported Cooperative Work conference, CSCW2002, finds nary a mention of the word “expert” anywhere on the program. Debunker that I am, I know that an anecdotal observation doesn’t prove anything, but it does point to only a few possibilities: the conference organizers didn’t accept papers on that topic, there were no finished papers on the topic ready for the conference, or maybe, there’s not much work going on in that field.
Perhaps, too, the idea of “expert”, with its connotations of class, rank, hierarchy, due deference, etc., just doesn’t fit in our new collaborative, cooperative, team-based working world?