Editor’s note: Here’s another post from friend of the ‘Lab and colleague, John Cartan. When John reached out, offering a review of the Narrative Clip (neé Memento), I jumped at the opportunity to read and publish his thoughts, and not just because I value his insights.
When Noel (@noelportugal) and I were in the Netherlands for the awesome event hosted by AMIS in March, we ran into Sten Vesterli (@stenvesterli), Ace Director and OAUX Advocate, who was sporting the very same Narrative Clip. We both quizzed Sten about it and were intrigued to explore future uses and cool demos for the little life-logging camera.
Anyway, John’s review reminded me, and now we have more anecdotal usage on which to draw if/when we get to building for the Narrative Clip.
For several weeks now I’ve been wearing a small orange gadget clipped to my shirt – a “lifelogging” camera called the “Narrative Clip”. We thought we might be able to use it for ethnographic field studies (following users around to see how they do their job), or maybe for recording whiteboards during brainstorming meetings. But I was especially curious to see how other people would react to it.
The device itself is small (about the size of a Triscuit) and easy to use: just clip it onto your shirt or collar and forget it. It takes a photo once every 30 seconds without flashing lights or any visible indication. At the end of the day you hook it to a Mac or PC with a 3-inch USB cable to both upload the day’s photos and recharge the device.
The camera can be temporarily deactivated by putting it face down on a table or in a purse or pocket. In practice I found that my pocket wasn’t dark enough so I made a small carrying case out a box of mints.
Once the photos are transferred (which takes only a minute or two) you can either leave them on your hard disk, upload them to a cloud server, or both. The server upload and processing takes anywhere from ten minutes to six hours or more. Once uploaded, the images are straightened, cropped, sorted to remove blurry photos, organized into groups, and made available to a free iPhone or Android browser app.
The cloud storage is effortless and requires no local storage but sometimes over-crops (it once chopped the heads off all the people in a meeting I monitored) and provides only limited access to the photos (you have to mail yourself reduced photos from the phone app one at a time).
So I think that for full control you have to enable the local storage option. This works fine, but creates more work. You can easily generate over a thousand photos a day, which all have to be sorted and rotated. The photos consume a gig or more each day, which may eventually overwhelm your local hard drive; for long-term usage I would recommend a dedicated external drive.
Each raw photo is 2592 x 1944 (5 megapixels). The quality is acceptable in full light, grainy in low light (there is no flash). But because the photos are taken mindlessly while clipped to a shirt that may bounce or sag, the results are generally poor: mostly shots of the ceiling or someone’s elbow. There is no way to check the images as they are taken, so if the lens is blocked by a droopy collar you may not discover this until the end of the day (as happened to me once). And the camera generally won’t be pointed in the direction you are looking unless you glue it to your forehead or wear it on a hat. You can force a photo by double-tapping, but this doesn’t work well.
For all these reasons the Narrative Clip is not a replacement for a normal camera. But the random nature of the photo stream does have some redeeming qualities: it notices things you do not (a passing expression on someone’s face, an interesting artifact in an odd corner of someone’s cube, etc.) and it creates a record of small moments during the course of a day which would otherwise be quickly forgotten. Even if most of the photos are unusable, they do tend to jog your memory about the actual sequence of events. And because the photos are un-posed they can sometimes capture more authentic moments than a more obvious camera usually would.
The key to designing a great user experience for enterprise software is to first understand your user: what her job is, how she does it, what challenges she has to overcome each day, etc. One way of doing this is an “ethnographic field study” – the researcher follows the user around and documents a typical day.
Our original idea was that the Narrative Clip could enhance ethnographic field studies. Either the researcher could wear it while following a user, or you could ask the user to wear it for a day and then meet later to review the photos.
I think both of these ideas are worth trying. The Narrative Clip would not replace a normal camera; it’s main value would be to jog the memory when writing up reports at the end of the day. Similarly, if the user wears the clip herself, the researcher should schedule time the next day to step through the photos together and answer questions (“What were you doing here? Who is that? It looks like you stepped briefly onto the shop floor after lunch – how often do you that?”).
There are other applications as well. I set up the camera in a meeting room to take a photo of the whiteboard every 30 seconds. This could be a quick and easy way to capture drawings during the course of a brainstorming session. Placing the camera far enough back to capture the entire board meant the writing was hard to discern; it might work with good lighting and strong marking pens.
Setting the clip on a table during an interview allowed me to collect a collage of un-posed portraits which, in total, gave a more accurate reflection of the subject’s personality than any single posed photo could provide.
Another possible application is using the camera to take photos from the dashboard of a moving car. For optimal results the camera needs to be placed near the windshield and high enough to avoid photographing the hood of the car. I achieved a stable mount by clipping the camera to a placard holder (from an office supply store) and placing that on a dashboard sticky pad (from an auto supply store).
As we enter the age of wearable sensors and the Internet of Things, we are starting to ask a new question during our design sessions: “is that creepy?” As technologists we are naturally excited by the new applications and the bounty of data made available. But as we think about the user experience of our customers, it is important to consider what it’s like being on the other end of the camera. Wearing the Narrative Clip was a great way to explore personal reactions to this brave new world.
I found that in general people didn’t notice (or were to polite to ask) about the device unless I brought it up. But once they realized it was a camera, some people were uncomfortable (at first). Most people didn’t seem to mind too much once they understood how it worked, but some people were definitely shy about having their photos taken. Some changed positions so as not to be in my normal field of vision. One person requested that I destroy any photos it might take of her. It helps to explain what you’re doing and ask permission first.
Here is what one acquaintance of mine confessed:
“What I think is that I value one-to-one time that is ephemeral. Not recorded. Felt in the heart. I feel threatened when recorded without permission. Sigh. I know. That sounds dumb. I mean, with cell phones everywhere, I don’t even have privacy in the gym locker room. Then the flip side of my brain starts blabbing: “What are you worrying about? Who would want to see your body or record your thoughts anyway?” Am I just prejudiced? I would not want to hire someone I interviewed if they wore one. I would leave the dinner table if a date wore one.”
I feel that it is very important to respect attitudes like this. If people are uncomfortable with a new technology, they will find ways to bypass or subvert it. Sensor-based enterprise applications will only succeed if we strike the right balance between convenience and privacy, are upfront about exactly what data we are collecting and how it will be used, and show respect by asking permission and letting people opt in as much as possible.
The Narrative Clip is a solid, easy to use device that could be helpful for tasks like ethnographic fieldwork, but culling through the flood of random images requires time and effort. Further experimentation is needed to determine if the trade-off would be worthwhile.
Recording entire days – and being recorded by others – was an illuminating experience. Sensor-based technologies can provide treasure troves of data, but it’s always worth asking what it would be like to be on the other end of the camera. A reasonable balance can be struck if we are transparent about what we are doing and show respect by asking permission.