His annual reports with visualisations of his personal data made Nicholas Felton famous. At the Infographics Congres the zen master of the quantified self presents three ways of collecting personal data.
the study of human activity in the past, primarily through the recovery and analysis if the material culture and environmental data that they have left behind – Wikipedia
“The first annual report I made to wrap up the year – to look back at 2005 in a way. I used different datasets: the books I read; restaurant I went to; trips I took; photo’s I took; and data about the music I listened to from last.fm.” Since Felton first thought of an annual report at the end of the year, there was no other option than to use archeology to get data. He didn’t recorded his data in order to make a report – he studied his own activities trough recovery and analysis of data left behind.
He did the exact same thing again in 2010. After his father past away, he used the artifacts to encapsulate his fathers life. “I used old diaries, postcards to figure out where he had lived; and photo’s some with some without context. Though I sometimes could figure out the context by looking closely to them.”
But working like an archeologist isn’t ideal: it makes you reliant on often incomplete source material. Besides that it’s hard to extend – costs are high and constant. Or as Felton states: “You can’t turn this into a product. It’s too labor-intensive: by no means could I have made a report like this for others.”
the excessive acquisition of and inability or unwillingness to discard large quantities of objects – Wikipedia
“In 2006 I started collecting data by tracking everything: what I drank, which animals I ate, mapping the places I visited.” It was also the first year Felton printed his report. “The year after that I was trying to get all the things I missed. I tracked all the streets I walked in in New York City. And mind you, this was pre-smartphone or GPS. I just wrote it down…” Adding new data every year, in 2008 he tracked his miles traveled: 38.524 miles, including 1.036 miles traveld within videogame Grand Theft Auto…
Hoarding data makes your datasets more complete, but the costs are high. “And”,Felton notes “ it may be addictive. It’s hard to stop tracking once you started.” Besides that, not everything is suitable for hoarding: “Somethings are just too imprecise or difficult, think about datasets about what you read, watch on tv, talk about, or how you feel.”
the selection of a subset of individuals from within a statistical population to estimate characteristics of the whole
“The first time I used sampling was in 2009. Since selfreporting my mood didn’t feel good, I outsourced it by asking people i met to fill in an online questionair about my mood. I figured that the external expression of my mood – how other thought I’d feel – maybe even more important than my mood itself.” Not everyone asked will answer the questions, which creates the selection needed for sampling.
The people invited through mechanical turk to decide wheter used description of Nicholas’ mood were either sad or happy and intro- or extravert, decided that ‘swell” is 78.9% happy.
“In 2013 I reported on communication, focussing on the difference between the actual data and the metadata. Metadata has a real power: it told me who I texted with and when I receive and send messages. Based on that metadata only the NSA knows who my best friends are. But metadata is in a way worse than data: it’s fuzzier and harder to interpret.”
For the 2012 annual report I commissioned an iphone app. I would get notifications to answer a survey while the phone would at the same time get metadata – including number of photo’s taken since last time and location.” The app Nicholas used himself formed the fundaments for his recently presented Reporter app, which helps you to keep track of your own data.