Dave Winer, in response to blogs by Mark Pilgrim, made possible the harvesting of connections between weblogs. Now referred to as 'neighborhoods' among ru's webloggers.

See Marc Barrot's neighborhoods (e.g.,http://radio.weblogs.com/0104487/misc/weblogNeighborhood.html) as examples)

Yesterday Dave Winer (link), Jon Udell (link) and Marc Barrott, among others celebrated the ability to 'harvest' blogrolls. What is a harvest? It is a tracing of your subscriptons, of your subscription's subscriptions, and your subscriptions subscriptions' subscriptions. The results of your harvest is a list of links to subscription sources (weblogs, rss feeds of various sources) and the number of times the link has come up in the harvesting. This listing of links and numbers has been called a 'Neighborhood'.

All of this reminds me of a mental experiment I conducted in the mid 80's. It had remarkable overlap with this endeavor. Its essentials are as follows: analysis would be in references of academic writings (in books and journals); start anywhere and go to the reference section; pick the three authors most referred to other than the writer of the starting document; go to those documents and do the same thing; continue for, say, three levels;diagram the results. This list of names and references would be a map of an intellectual neighborhood, i.e, those who contend with each other inthe present tense (reference) and depend on the work completed in recent and not so recent past tense (again, references). It's a neighborhood in the sense that there are references within it that link writers to each other. Some are bilateral reference processes - those disputing an issue in multiple writings, for example. Others are unilateral references in which one writer acknowledges another but who is not acknowledged in return. (A lack of acknowledgement is not necessarily negative-- it is not shunning or ignoring). Enough said, I suspect, to credential my 'Aha!' when following the construction of 'Neighborhoods' of weblogs and such via a process of 'Harvesting'.

As the shifted librarian noted--[http://www.theshiftedlibrarian.com/2002/6/3.html#a2130], I'm paraphrasing here, it would be great if we could harvest neighborhoods of substance as well as of connectedness. The fact that a group tends to mutually reference implies joint working of material (collaborative or argumentative won't be obvious from the fact of the link itself) but does not unpack or sort the material by substantive categories. With her I believe that this would be immensely useful.

My experience with ad hoc categorizations has been pleasant and successful in qualitative research. But those categories, derived for small groups with unique experiences and little concern for universal categories at the time of their request for support, would be less useful if one's intent was to put an idea out there for literally anyone to search for and find, those local designations would be less than successful.

When it comes to anticipating world-wide (potential) access based on subtantive flags... I would like to have some common category scheme as a starting point. I don't think labelling one's personal categories of blogging will be an exhausting effort since we tend to hold to a smallish number of categories of interest.

Having used a universal category scheme for our weblog entries and/or entry categories, I believe that searches that are universal-category-based , would at least at the start of one's blogging, provide a better harvest than names of authors, certainly, and even one's more arbitrary categorizations.

In short, I think accessing by universal category designation (in addition to the personal category designations, the names, and the neighborhood spaces) will expand our ability to relate to both the thinkers AND to interesting material.