A Photo Every Minute: One Year Later

Hello, so I wear this camera phone around my neck and it takes a photo every minute, and I’ve been doing that since August of last year and I have a long presentation about the whys and hows that you can find on the Quantified Self website if you look for Rob Shields. So I thought I would just take a little bit of time to talk about what has changed, what’s new and things that have been working, and some things that haven’t been working, and share may be a little bit of data there that you are looking at.

But first, may be some quick hit stats. I’ve taken over the course from August 2012 to 2013. I have taken over a quarter of a million photos, which if you printed that out. It would be a stack more than seven story’s tall. That averages about 781 photos a day. If you slice that differently. That is 196 Gig per year’s worth of photos, or about half a gig a day which is not bad considering the photo resolutions are actually pretty large. That’s about what I ingested when I undertook this project last year, which is self-confidence building.

Some of the things that are improved is that I used to just have a strap that went through the back of the battery compartment of my phone, and my lovely wife made me this harness which has a pouch but also a safety strap over the top, which is important because I lost my first phone in a tragic Frisbee related accident. This is literally the last photo that phone took on its way down.

So other things that I’ve improved is the amount of time that I can go between charges. I used to be able to go about half a day between charging, but now I can go two days, which makes it more convenient to carry around. And I thought. Because of all of this photo taking was the problem, but it was actually the phone trying to be a phone that was the problem; trying to talk to the cellular network.

I used to have to plug my phone into my laptop, every time I wanted to download photos, and now actually it’s just whenever they are both on the same Wi-Fi and they move the photos around which makes it nice and easy for me. The small barriers becoming big problems later, but some things have become tedious. I started off tagging all my photos and kind use Picasa and Picasa basically does two types of tag, there are people tags and there are context tags.

So I was tagging them every day and it would take about 10 minutes, but it become more and more because it catches the TV that I watch and I started tagging all of the faces on the TV that I watched and that took a lot of time. But it is useful and valuable data, so I would actually like to get back in doing more of it.

Interestingly, something that became tedious was actually looking at all of the photos, which you might ask isn’t that kind of the point, and it is kind of the point. So one of the things that I have been fiddling with is a system where I can playback the photos for a particular day and added accelerated time alongside my GPS, which is associated with those photos and tags of who was in the photos and what I was doing, just as a way of being able to review them and more interesting, rather than just flicking through it.

Some things that haven’t happened, I originally thought of this project as a way I would start being able to search my life. I would have photos, and I would have extra data and I could go there and say, where was I when I was talking to so-and-so and have a search kind of result that would give me that information. And it hasn’t happened yet. I have done it a little bit, but the reason I haven’t done it is because it’s physically cumbersome and this stuff is all actually at my house so I have to be at my house. The interface in search in Picasa is not great.

But I am starting to explore the data, and with the help of our QS tech group in Portland, I was able to get the meta data out of Picasa and actually into MySQL so I could do some stuff with it. And one of those points of data that I have is faces, and I have seen more than 1000 faces in the year from August that ended up in my system; 266 just in June of this year.

This is a graph of just a count of people that I saw in the month of June, so you can see that over the course of any particular day I had so more or less people. And just to give you a sense of what was going on for these things, here are some of the activities that I was doing in Portland Road trip in the middle of all of this.

Then I realised that since I’m tracking faces, and also tracking TV, so I could do things like look at when I was watching what kind of TV. So he is a graph of when I was watching Game of Thrones in June, and interestingly as you look at this, you can see how many characters there were in each one, which got me to thinking well, one of the things that I could look at is how long was each character in the show. How many photos of that character did I get? So you can see the story art changing from one episode to another. So this is those same two episodes and you can see where people had a lot of screen time in one and less in another. There was a whole lot more people in one than the other, and some people just carry on through them. So you can see how did this narrative art change over time.

In case you think I’m a kind of a beheading and backstabbing kind of TV watcher, here is a different TV show that I also watched in June, with some people that work in that.

So, on top of all of this data about faces, there is also contextual data, which is data you can do in Picasa. It’s kind of free-form and you can put whatever you want, and I use it to kind of say, what am I doing, where am I at sort of thing. So here is a picture where I was driving on our summer road trip this year. You can take that same sort of approach to do what I did as opposed to faces. So this is just a count of different activities, different tags that were in my photos in the month of June. And you can see, sometimes there was a lot of things going on, sometimes where there wasn’t much at all going on and just to give you the sense of it.

Here is a breakdown and a kind of a spike, where we were just coming back from the end of our road trip. I was doing some driving and we stopped at restaurants along the way. We came home and we binged on TV for a while.

You can take the same data and then actually look at how many photos contain each tag and that’s what this chart is, and that gives you a sense of how long I was spending doing any particular activity in any given day. So here are some labels in the things that I was involved in, and you can see there was the road trip which had all of the photos that were involved in that road trip were tagged with that. You can see and I did a bunch of driving at the end, and you can see TV, the times that I was at home and the times I was at work and even at QS meet up time.

That allows you to kind of slice and dice this to see how much time I spent doing some things. So you can see how much of my waking photographs life, I was at home, on the road trip, at work, watching TV, driving, playing poker. You can see that I was hiking the same amount of time that when I was at the QS, which is way too much time hiking for me.

Interestingly, for some reason the values didn’t come through on this slide for how long I spent in Baskin and Robins and I’ll leave that to your imagination. Even how long I spent in the bathroom, which brings me to the question I get asked very frequently which is, do I leave this running when I’m in a public restroom. And the answer is yes, because I don’t know what happens in those restrooms that you visit, but this is the most exciting photo that I have ever gotten in a public restroom.

Thank you.