One might file this under “posts I wrote while waiting for a huge task to finish”.

One might.

I find myself in a funny situation these days. Exactly a year ago I was doing fieldwork on Quantified Self, tracking my steps, sleep, and virtually everything else in my life for which I could find a quantitative metric. From that work came an anthropological ethnography of which I’m extremely proud. I strongly feel that it provided some insight into the cultural landscape we broadly describe as “Quantified Self”, and it offered some theoretical explanations for why and how self quantification appeals to so many (especially Westerners).

While I made an effort to analyze my own data in meaningful ways, it became clear to me - as it did to those who offer QS tools and systems - that the value in self quantification is actually in the quantification of the world; the trends of people at large scale might offer real insight that can help people. Countless papers in CS seek to answer questions about society from a perspective informed, ultimately, by statistical analysis of massive numbers of people. Businesses are predicated upon the notion that this approach is sound.

I’m not saying it’s not true; on the contrary, if this is the paradigm which generally describes the culture of QS, then I’d be remiss not to explore it, but how can I explore the “Quantified Other” when the IRB at UC Irvine was so reluctant to let me study the Quantified Self? So I focused on the self, exploring my own data and carefully avoiding dealing with anyone else’s data per se.

Flash forward to a year later, and I’m waiting on a script to parse through several hundred gigabytes of Quantified Self data made available by Azumio, a Bay Area company that has offered the infolab a massive dump of QS data.

A year ago I was very much in the trenches, seeing QS at a personal, individual level. I was interpreting my own data and wondering whether others were seeing things the way I was. Now I’m 10,000 feet up, only seeing my own data when I go through several layers of winnowing to find exactly my data point, and this time I know this isn’t how anyone else sees their data.

The questions I’m asking are fundamentally the same - how can I make life better using quantified self data? What can I learn at large? What can I learn about my own data?

This post got off my radar, and I plan on writing more formally about this for CSCW, so I’ll leave this alone for the moment. Sorry if it seemed to end abruptly.

Posts I still plan to write up:

• making Python useful for compute servers with lots of cores (in spite of the GIL :P)
• Online(?) worker cooperatives (did I write about this before? I need to write more on it)
• ???