Tweet Research: Computers and Writing 2009

I'll be presenting at the 2009 Computers and Writing Conference at UC Davis this week, and I've included the slides for my presentation below. If you would like a transcript of the talk, please let me know--I'm happy to provide it.

One of the most rewarding aspects of preparing this presentation was my interaction with Peter West, a management consultant and principle of Continuous Innovation in London, Ontario. Peter is one of my most important Twitter contacts, and he was gracious enough to grant me an interview which detailed how he uses Twitter within a broader suite of knowledge management tools and ecologies. My work with Peter is so interesting that I hope to work with him on a more substantial project in the very near future, as much of the information that he provided must necessarily be truncated for a short conference presentation.

Regardless, I wanted to provide a preview of Peter's workflow now, in anticipation of a more detailed discussion on Friday:

Peter has developed what he calls an “environmental scanning system” predicated upon a complex ecology of information sources ranging from RSS feeds of peer-reviewed journals to tools such as Twitter. He also relies heavily on both Google and Windows Desktops. He says that “each day, I manually scan the new citation releases from over 10,000 journals. Experience has proven that this is the most effective way to get a general sense of the literature that is being produced and to comprehensively capture relevant material.”

To get a better sense of what this type of knowledge work entails, Peter graciously tracked his activity during the month of May, yielding some tremendous insights into how he scans and disseminates research:

He processed 23,970 total citations, and spent 89.1 total hours doing so.

The highest number of citations processed in one day was 1,630 and the longest one-day processing time was 6.42 hours. Alternatively, the lowest number of citations processed in one day was 108, and the shortest single-day processing time was 1.1 hours.

He averaged 773 citations per day for an average processing time of 2.87 hours.

The Total number of journals included in these statistics was 4,298, however, he scanned another 5,834 journals during the period whose articles were not posted in Twitter, primarily because they do not provide open access.

Peter began sharing citations publicly via a blog called “SynapShots” back in 2002. One of his objectives was to foster reciprocity and conversation, an ethos achieved more readily through Twitter.

Regardless of your perspective on Twitter, it's clear that Peter is a very unique user and participant in this evolving social space. For my own work, his diligence is invaluable. I'm very much looking forward to talking more about Twitter at #CW09!

The University of Texas at El Paso approval for this study is filed under IRB protocol ID #117301-1--"Tweet Research: Aggregating and Disseminating Organizational Knowledge Work through Twitter"