Quantified Self software I use

Blog
17 Jul 2016, 11:22 p.m.

This entry is part of a loose series of posts about (mostly) software tools I use to make my life better. Quantified Self is a slightly odd fit in that regard: a lot of the data I collect on myself is not currently analysed in any useful way, and I'm not certain that some of it even can be. Content note: exercise, weight, eating habits.

The areas of my life of which I currently keep records range from weight and physical activity to the films I watch and where I see them. I have vague hopes that at some future point I'll run everything through some R scripts and something will come out that's not entirely unlike Stephen Wolfram's personal analytics. That's not an absolute goal, though. Some things I track because I want to improve them, like step counting. With others, like the books I read, I write things down to remember them better later ... and, honestly, because I have a small horror of information being lost. Reading about Oak Island as a child gave me a shaky, panicked feeling at the thought of something lost at the bottom of a pit, something that could never again be known. I still get that feeling when I read of a writer's heirs burning their work, even at the wishes of the deceased. Books read and films watched are of much smaller consequence, but it bothers me when I forget these things.

In this sense, the proliferation of options for Quantified Self tracking is a danger. The more things I can keep track of, the more things I feel should be tracked, and the worse I'll feel when I lapse. So far, trying to be discriminating in what I pick up, and bearing in mind the effects of such projects on my mental bandwidth, seem to be enough to stop me going over the top.

Culture

Goodreads

I love Goodreads. It's a site you can use in any number of ways — as a community, as a book recommendation engine, as a personal reading log, or something in between. The web interface is easy to use, especially for the depth of information it lets you record, and the mobile app has equally low friction (that's not to say that either is totally intuitive, of course). Goodreads is owned by Amazon, which has troubling implications for Amazon's dreams of monopoly in the publishing and bookselling realms, but does mean one can access it directly from one's Kindle. It offers a nice set of statistics directly from the account page. In fact, so many of the metrics I wanted to see are now automatically available that I've had to dismantle my complicated original shelving system.

I held out on making a Goodreads account for years because I was worried it would be a trigger for obsession as I described above, but on the whole it's been fine. Tagging and rearranging my books there is, on balance, more soothing than it is a source of stress. Mind you, I'm currently a few books behind in my records and the knowledge nags at me unpleasantly.

Last.FM

Last.FM lets you store every audio track you listen to, and it has four years' worth of data from me. I had an account there during university, too, but deleted it in a rush of paranoia about being detectable on the internet while I was unemployed. Now I find that a shame; it would have made a nice graph, my transition from a well-rounded musical diet to listening to the same three atmospheric black metal albums on repeat all day. (One day I'll write a post about atmospheric black metal and how great it is. For now I'll just point out that I also listen endlessly to the Girl with the Dragon Tattoo soundtrack.)

Letterboxd

Letterboxd is the equivalent of Goodreads for films. I've only been a member since the beginning of this year. Apparently I've so far watched eight films all the way through, which is less than I'd like and more than I would have expected.

Food and drink

Untappd

Taking pictures of my drinks like some kind of hipster, I apologetically refer to Untappd as 'my ridiculous beer-based social network,' and David rolls his eyes as if he thinks I do apologise too much. It does have utility, though. I was persuaded to sign up for this site in a beer garden in Budapest, a city I had ripped to shreds in my hunt for a beer I vaguely remembered drinking there in 2003, remembering only that it smelled like sunflowers was named after the Hungarian word for railway. When I next make it back, presumably in 2029, I won't have any trouble remembering that I liked Soproni Fekete Démon.

YouFood

Like MyFitnessPal, which I tried but found too prescriptive and too time-consuming to use properly, I heard about YouFood from my dear coworker Michelle. The idea behind it is to use the smartphone app to save a picture of each meal or snack you eat during the day. That's theoretically it, though of course YouFood also wants to be a social network and a hub for all your thoughts on things comestible. I unsubscribed from all that and made my account private, since I only want a visual record of what I eat and drink. I should probably think about consistent captioning, to make the data more tractable in future. I have another friend whose Flickr page contains years' and years' worth of daily food photographs with hardly any metadata, and sometimes thinking about that makes me itch.

The app is promoted in the app store with a '7 Day Challenge', in which you log everything for a week and find yourself eating better at the end of it, presumably by magic. It doesn't surprise me that this sort of logging would result in a change of eating habits, both because of the idea that all your followers will see what you eat and because pulling out your phone to enter the data into it adds extra friction to the act, which is worth overcoming for a meal but perhaps not for half a dozen M&Ms.

Two days in, it just occurred to me that there's no obvious way to export data from YouFood, which would make it useless in the long run. I have written to ask about this and will update when they get back to me.

Physical health and activity

Clue

Clue is the menstrual tracking app some people search for for years, and I feel grateful that my friend Alex mentioned it before I even started. There is no pink in its layout and the text is gender-neutral; its recommendations are scientifically backed (with citations!) and there is no automatic assumption that you're tracking your periods because you want to get pregnant, although of course you can use it for that. You can track something like 40 different aspects of your life that might be related to your cycle, with full customisation.

I knew that my periods had improved after my surgery this January, but using Clue it's easy to confirm that with data. If I have to go to the doctor with menstrual problems again, I'll appreciate having that kind of backup.

Withings

This is probably the most traditionally Quantified Self entry in this list. Last year, a coworker offered me his old Withings Puls Ox step counter (isn't that a ridiculous name?) so I could join in with the current office craze, competing for the greatest number of steps taken per week. I liked it so much that, when it finally bit the dust, I bought a new one for me and one for David as well.

As a side note: these are not particularly cheap. You could get nearly the same results with a simple £10 step counter and a spreadsheet, just without the granularity of data and the fancy app. Since I'd already been sucked in by the nice UI and could currently afford it, I went the branded way, but it's not necessary!


A screenshot from the Withings Healthmate website, showing my sleep and walking results for this week.

I count my steps per day, measure my sleep overnight when I remember, and also record my weight and any other exercise I do. I have a vague goal of getting fitter and more active. That doesn't mean losing weight; I only record that because the app makes it easy and I thought it might be interesting, especially given my barely-there thyroid gland and artificial metabolism. So far it seems to be mostly influenced by my menstrual cycle. I should cross-reference this with the data from Clue!

Apart from observing the spikes in my step count caused by conferences and Pokémon Go, I'm not really doing much with this data so far. I still want to keep collecting it, though, mostly as a reference in case of any future health problems or my sudden transformation into an Olympic weightlifter. This will probably occur around the time I learn enough R to write the scripts I mentioned at the start.

Other

I've already written so much about Beeminder that I won't add more now, except to say that it's a de facto QS system.

I tried Reporter for a while and really loved the idea of it, but found it too distracting to answer the quizzes at random times each day.

A friend of mine takes the PHQ-9 regularly and tracks the results to monitor their depression, something I have considered doing as well but not yet started. In fact, the two people I know who have the most dedication to record-keeping of the Quantified Self manner both have complex health issues, in combination with statistics skills that make me weep in admiration. I've never managed to stick with pen-and-paper recording long enough to draw any useful conclusions, but apps and website interfaces are making it a lot smoother and perhaps, one day, I'll figure out what I want to do with all these datapoints.