Thursday, December 18, 2003

Archival Insights for Nonrivalrous Resources Like Intellectual Capital

Knowledge Management is a big topic. According to the pundits, we need to use our intellectual resources and extract the "intellectual capital" from our organizations. I'm a bit suspicious about rendering knowledge (or even information) so it morphs into a line item on the balance sheet. Perhaps Thomas Jefferson-in an 1813 letter to Isaac McPherson-provided the best description of my concerns:

"He who receives an idea from me, receives instruction himself without lessening mine; as he who lites his taper at mine, receives light without darkening me." (quoted in Lessig, 2001 p. 94)

Jefferson indicates that certain resources-like information or knowledge-don't become consumed through usage. Unfortunately, our accounting standards have been developed specifically for tangible goods (barring the pesky "Good Will"). Stanford University professor Lawrence Lessig distinguishes between resources that can be accounted for and controlled and those that can't:

" 1. If the resource is rivalrous, then a system of control is needed to assure that the resource is not depleted-which means the system must assure the resource is both produced and not overused.
2. If the resource is nonrivalrous, then a system of control is needed simply to assure the resource is created-a provisioning problem, as Professor Elinor Ostrom describes it. Once it is created, there is no danger that the resources will be depleted. By definition, a nonrivalrous resource cannot be used up." (Lessig, 2001 p. 95)

It may be difficult to "use up" a nonrivalrous resource such as knowledge, but the resource can be lost. How do we control these intellectual resources? How do we ensure that they can be reclaimed and utilized?

It seems ironic that in order to control the "shrinkage losses" inherent in open information sources, we have to turn to control mechanisms. Instead of legal mechanisms, however, perhaps we should incorporate the best practices of archivists. Richard Cox (1994) provides a set of principles for an "Archival Documentation Strategy" that may throw some light on ways to prevent the shrinkage of nonrivalrous resources:

  • Principle One: All recorded information has some continuing value to the records creators and to society.
  • Principle Two: The immense quantity of recorded information is an impediment to the information's continuing value, leading to the need for the reduction of this quantity.
  • Principle Three: This reduction of documentary sources may occur through accident and natural events, resulting in a random or, at the least, partial aggregation of documentation that may harm the records creators and society.
  • Principle Four: Even a faulty archival appraisal decision process is better than records surviving haphazardly or not surviving at all.
  • Principle Five: Because of the immensity of this documentation and the importance of recorded information to its creators and society, a well-developed set of universal or common archival appraisal criteria is one of the most important elements for appraisal.
  • Principle Six: The criteria that provide the basis for archival appraisal decisions are independent of records creators and their institutions and are generic to recorded information.
  • Principle Seven: The most fundamental aspect of appraisal is the consideration of records as part of an organic whole related to institutional purpose and function.
  • Principle Eight: The quantity of recorded information should be reduced in a planned manner, based upon carefully determined and tested selection criteria.
  • Principle Nine: The archival appraisal selection criteria should rest not on unpredictable future research practices and trends but upon the more predictable sense of determining what are the salient and important features of contemporary institutions and society.
  • Principle Ten: Archival appraisal is an incomplete process if it is done without consideration of the information found in non-textual records that archivists often do not take responsibility for in their work.

Cox may have provided more. A copy of his paper came to me in an accidental manner and has experienced extensive "accidental reduction." At least I'm doing my part to stop the shrinkage of this particular resource.


Cox, R. J. (Fall 1994). The Documentation Strategy and Archival Appraisal Principles: A Different Perspective. Archivaria 38: 11-36.
Lessig, L. (2001). The Future of Ideas : The Fate of the Commons in a Connected World. Vintage Books: New York.

[UPDATE]- I'd like to thank a number of people who emailed and informed me that Lessig is at Stanford and not Yale. I know this, but for some reason my typing betrayed me. The error did have a good side as a number of people contacted me directly and provided feedback. It seems that the error acted like a type of easter egg. BTW- here's a link to my favorite easter egg in Excel 2000.

Wednesday, December 17, 2003

Social Networks and Email

I recently completed some work on Social Network Analysis (SNA) and Email. Of particular interest to me is the combination of a standard corpus with the social network behind its creation. Towards these ends, I proposed combining standard Information Retrieval (IR) techniques and SNA. I even went so far as to do some numerical modelling to study the possibilities (and work on my coding skills!).

Since I completed the project, a number of new resources have come to my attention. If I had known about them a few weeks ago, I could have save a lot of time!

Social Network Fragments is a very interesting project created by danah boyd. It demonstrates the possibilities of SNA and email archives.

It seems that Valdis Krebs has been doing this sort of work for some time.

The topic has also been covered in the popular science community:

Emerging Technology :: Pass your e-mail through some new software and the answer will become obvious
By Steven Johnson
DISCOVER Vol. 24 No. 04--April 2003

E-mail reveals real leaders :: Network analysis maps companies' informal structure
20 March 2003
By Philip Ball
NATURE--20 March 2003

Email traffic patterns can reveal ringleaders
NEW SCIENTIST--27 March 2003

There is also some pertinent academic literature that had escaped my notice (damned preprint servers!):

Email as Spectroscopy: Automated Discovery of Community Structure within Organizations
Authors: Joshua R. Tyler, Dennis M. Wilkinson, Bernardo A. Huberman

Featuring Web Communities based on Word Co-occurrence Structure of Communications
Authors: Yukio Ohsawa, Hirotaka Soma, Yutaka Matsuo, Naohiro Matsumura, Masaki Usui

Self-similar community structure in organisations
Authors: R. Guimera, L. Danon, A. Diaz-Guilera, F. Giralt, A. Arenas

One of the major problems I had was determining a way to visualize the networks. I had no intention of creating a custom spring embedded app to represent nodes and I couldn't find a publicly available one... until now:

HP's Zoomgraph

If anybody is interested in this sort of work, please feel free to contact me [email link].