Data displays for self-awareness
Can our Kindly Contributors think about data, information, and displays for life tracking?
Diaries, calendars, blogs, tweets, income tax returns and their back-up data, Google’s retention and display of searches, ISP records,
credit card bills, and email history all generate large amounts of data that might be relevant for making life-tracking displays–and,
more importantly, for understanding and possibly changing one’s life or one’s allocation of time.
What kinds of effective displays are used by people in planning, recording, and assessing their daily activities? Since so much
information is compiled, it is important to identify what kinds of data are relevant to understanding and changing one’s activities, to
go beyond score-keeping. So we’re looking for data that would help see what we’re doing, how much is habit, and how one can be
more effectively self-aware. The issues are less about display methods and more about choosing the key data worth looking at.
It would be interesting to see what I suspect are endlessly repeated looping patterns in internet use (in my case, for example,
NYTimes to weather to email to flight status to ESPN, and then over and over during the day), with the only variation being whether
the loops are executed on my desktop, laptop, or iPhone. Thus one might generate one’s own ISP logs or get the records from your
ISP.
Here is a start:
There was an article in Nature that examined this that may be of interest.
Gonzalez MC, Hidalgo Ca, Barabasi A. Understanding individual human mobility patterns.
Nature. 2008;453(7196):779-82. Available at: http://www.ncbi.nlm.nih.gov/pubmed/18528393.
Abstract
Despite their importance for urban planning, traffic forecasting and the spread of
biological and mobile viruses, our understanding of the basic laws governing human motion
remains limited owing to the lack of tools to monitor the time-resolved location of
individuals. Here we study the trajectory of 100,000 anonymized mobile phone users whose
position is tracked for a six-month period. We find that, in contrast with the random
trajectories predicted by the prevailing Levy flight and random walk models, human
trajectories show a high degree of temporal and spatial regularity, each individual being
characterized by a time-independent characteristic travel distance and a significant
probability to return to a few highly frequented locations. After correcting for differences
in travel distances and the inherent anisotropy of each trajectory, the individual travel
patterns collapse into a single spatial probability distribution, indicating that, despite
the diversity of their travel history, humans follow simple reproducible patterns. This
inherent similarity in travel patterns could impact all phenomena driven by human mobility,
from epidemic prevention to emergency response, urban planning and agent-based modelling.
I keep track of my exercise for triathlon training. More than just a running log, I also have categories for biking and swimming. For each I track distance (miles and kilometers) and time, which allows me to judge whether I’m actually seeing endurance benefits. With this basic data I can calculate pacing as well. I’ve started measuring my weight every two weeks to see how discernible weight change is. The most informative displays of this data (after collecting one year of data) showed where I fell out of the exercise rhythm, for various reasons. Vacations, family visits, colds, running injuries, and a bicycle accident all conspired to reduce the frequency of exercise. For me, it was useful to visualize this data as miles completed (by all means) per week, as well as the number of activities per day (usually just one, but sometimes two and three in preparation for the actual triathlon).
As an intern at a hospital in the US I am obliged to log the number of hours I am in the hospital. The weekly limit is 80, averaged over 4 weeks. The data is compiled on http://newinnov.com and the individual and aggregate results are compiled for the residency program directors who report them to the ACGME.
As a surgical intern I am obliged to log my surgical cases on acgme.org.
As a medical student I was obliged to provide similar data through similar online services.
Unlike Gmail, where I can recover all my messages back to my own computer, in these systems I get little or no access the data for introspection.
I have some introspective plugins for Firefox. Boomtango allows me to inspect my browsing history in fairly sophisticated ways. Zotero, del.icio.us, and instapaper allow me to review what I’ve read with bibliographic information, tagging, and bookmarking for later reading — I can examine my past plans. TimeTracker gives me a simple clock to log how many hours I spend in the browser.
Each year I get an annual report from one of my credit card companies (Capital One, I think) showing all of my spending over the prior year, sorted into categories (food, gas, clothing, etc.) It’s very straightforward but very illuminating. You could of course do much more with the data by looking at when expenditures occurred (i.e., what time of day, what time of month, what time of year), what the mix of very small vs. very large expenditures was, etc.
A few semi-random thoughts, without actual answers, sorry:
Don’t forget the phone as a data source, too. Not just records of phone calls and txts, but if you have a location-aware smart phone, your physical location is also easily tracked. If it has an acceleromoter, your general level of physical activity can be tracked over time (noting that phone-on-the-desk is usually easily distinguisshable from phone-in-pocket-while-you-are-sitting-still). Proximity of certain other items is alsio detectable: is the bluetooth car hands-free kit nearby? Then it’s a fairly safe bet that I’m driving. Similarly, am I near my bluetooth-enabled computer?
To further combine the data at another level up, if your friends are also having their location tracked, proximity to them can be derived. Other behaviourial correlations can also be derived – e.g. whenever mother leaves the house, child starts using playstation.
Aside from that, environmental factors are also relevant: is the weather hot, cold. raining, sunny? Are there any special events on which may affect your activities even if you are not directly part of them (e.g. first day of school affects traffic)?
Other patterns of interest I’d like to see: in what ways does my behaviour differ if I had poor sleep the night before? What about other people’s reactions to me – am I less sucessful in my social interactions? Are there people who avoid me? Are there correlations I had not noticed before – eg the above three things when I have a particular choice of clothing, or ride the bus vs. drive, or eat certain foods?
In case you all haven’t seen it, here is a writeup of Dan Meyer’s personal report foe 2009,
which he designed into a 2.5 minute infographic video:
http://infosthetics.com/archives/2010/02/an_animated_annual_report.html
One I forgot: mint.com provides ever improving daily analysis of your finances using the same sort of Bayesian analysis Gmail employs for identifying spam. Their president offers some insight into Mint’s security practices. This probably the most persuasive introspective tool I use. Mint has definitely changed our family spending more in 6 months than MSFT Money did in 7 years, and Mint is infinitely more user-friendly.
You may also be interested in the Quantified
Self blog that Kevin Kelley maintains. Additionally, he hosts meetings with other
people interested in collecting data.
There is a website and iPhone app called “Trixie Tracker” that allows users to enter data on when your infant has eaten, napped, needed a diaper change, etc. The app generates graphs to help identifying trends that might not otherwise be evident.
See http://www.trixietracker.com/
Sounds trivial and OCD, but anyone who has parented an infant knows that it’s very reassuring to know that there are patterns in your infant’s behavior rather than just the random chaos it seems.
I found it amusing that when I checked the “quantified self” blog, the post “Why I Stopped
Tracking”, by Alexandra Carmichael came up. Sometimes the quest for a way to do something
masks the reason to do it in the first place.
There’s a forthcoming New York Times Magazine story profiling a number of people who have done “personal data projects” — the collection and analysis of data on various aspects of their daily lives.
See http://www.nytimes.com/2010/05/02/magazine/02self-measurement-t.html?hp
Excerpt:
“In the cozy confines of personal life, we rarely used the power of numbers. The techniques of analysis that had proved so effective were left behind at the office at the end of the day and picked up again the next morning. The imposition, on oneself or one’s family, of a regime of objective record keeping seemed ridiculous. A journal was respectable. A spreadsheet was creepy.
“And yet, almost imperceptibly, numbers are infiltrating the last redoubts of the personal. Sleep, exercise, sex, food, mood, location, alertness, productivity, even spiritual well-being are being tracked and measured, shared and displayed.”
Thoughts:
The act of tracking as awareness – just the act of tracking whatever you want to be aware of (i.e. eating, exercise, internet use) calls attention to that activity, increasing mindfulness. This sort of tracking exercise can be limited in duration, making it seem more palatable (i.e. “I’m not going to have to do this forever”).
The act of tracking as obfuscation – using tracking activities as avoidance of dealing with the issue. Using the busy-ness of tracking as an excuse to reinforce your initial proclivity for not dealing with the problem.
Measurable goals: Not all issues have measurable goals. “How do you feel?” – there are no right or wrong answers to this question (although for the depressed, the issue may be “do you feel?”).
Data displays as external diagnosis tools: I often bring lists of questions and illustrated health histories to visits with doctors. This information becomes part of a dialog, often sharpening the focus on problems, and increasing my awareness of my body. While most self awareness tools are about self help it seems any analysis can be aided by outside opinions.
In Praise Of Vagueness – Jonah Lehrer, WIRED
http://www.wired.com/wiredscience/2011/07/in-praise-of-vagueness/
Describes a study that suggests that precise feedback may be less effective in motivating
change than fuzzy or vague feedback.
“Some subjects were given a precise readout of their HHI, which they could compare to the
ideal HHI. Others were given an HHI range that varied by 3 percent in either direction – it
was a deliberately vague form of feedback. What the researchers found was that the less
precise feedback led to far more weight loss, especially for those already overweight. For
instance, while those with an initial HHI of 75 gained, on average, one pound over the
course of the experiment after being given precise feedback, those given vague feedback lost
nearly four pounds. That’s because the vagueness provided an illusion of proximity to the
ideal, thus making the goal seem more achievable. The fuzziness of the facts kept them
motivated. ”
Dear ET & readers,
You may be interested in this piece in MIT Technology Review. For this service offering to
be successful the applications will need to be highly functional AND have user interfaces
that provide engaging data displays of a users personal data.
http://www.technologyreview.com/business/38485/?ref=rss
Matt
http://blog.stephenwolfram.com/2012/03/the-personal-analytics-of-my-life/
Stephen Wolfram recently published an analysis of personal data he has collected since
1989, on emails sent and received, keystrokes, calendar event times, phone call times and
durations, walking steps taken, and type of file modified, all versus hour of the day.
There is also a graph of book chapters modified over time, apparently with intensity of
color showing completeness of change.