On Wednesday, I had the great fortune to attend the closing keynote at the annual CSCW conference given by Mike Krieger, co-founder of Instagram, a photo-sharing site now owned by Facebook, but still operating largely independently, at least from the user’s perspective. In case you’ve been living under a rock, Instagram now has 400 million active users (75% outside the US) sharing 80 million photos and videos daily. Those are some serious numbers folks. And while they require considerable technical chops, I am glad Mike spent his time at CSCW talking about the design elements and human-computer interaction aspects. I share some nuggets below. (I failed to take notes so I’m skipping all sorts of info, sadly, and welcome corrections/additions in the comments.)
As old-timers here may recall, I am a big photo enthusiast and was a huge Flickr fan for quite some time. More recently, however, I have started getting into Instagram and now use it daily. Having thought about how these services differ and how I ultimately ended up using Instagram so much more these days than Flickr, it was a real treat to hear the brains behind the service share many of the conversations and decisions that went into making it what it is today. It was genuinely interesting to learn about the many aspects that he and his collaborators discussed and continue to ponder as they enhance the app.
In the first few minutes Mike shared some of his background, including his failure to get a paper into CSCW during his early days. I mention that as a reminder that people should not take the occasional setback too seriously.
Mike and his co-founder Kevin Systrom had worked on an earlier app called Burbn. The Atlantic has a few notes on this. This was the era of check-in apps so they focused on check-ins, but the app barely took off (we’re talking no more than about a thousand users). The aspect of the app that seemed to appeal to folks most was its photo-sharing capability. So they set out to focus on that primarily.
One of the conscious decisions they made up front was to downplay the importance of reciprocal following. That is, unlike Facebook where you have to agree to be friends with another person to share content (except for something like Pages or having public posts followed), on Instagram, not only can you follow someone who does not necessarily follow you back (like on Twitter), but the system in no way makes it obvious to know if someone you are following follows you (this is noted on Twitter when you check someone’s account while logged into your own). Generally speaking, information about followers is not easy to come by. You can see on an account how many others follow it and how many others it follows, but you cannot click through to get details. So instead of feeling pressure to follow people for social purposes, you can feel comfortable following just those accounts that are truly of interest.
Another explicit decision was to allow “fake names”, that is, an account can have the user’s real name, but this is in no way required. This has allowed the emergence of all sorts of accounts that go far beyond individual identities. It lets multiple people use one account or facilitates one person having multiple accounts. It has also led to the emergence of imaginaries such as Karl, the San Francisco Fog.
A very recent feature add concerns the ability to toggle between accounts without having to log out of them. Someone responsible for a corporate Instagram account may also have a personal account. It’s nice not to have to log out of each every time the user wants to switch between them. This ease of maintainting a second account (and excitement over Kinder Surprise Star Wars figurines) prompted me to start an account called THEFORCE365 where I’m chronicling the adventures of R2D2 across the globe.
And of course, there is the square image. Devices at the time of Instagram’s launch would mainly offer rectangular image options. The squares worked well for Instagram as they allow for a very easy way to set up a grid of photos. Since then, devices have square image options built-in.
Mike also discussed the creation of add-ons such as Layout from Instagram, which was in response to suboptimal third-party apps helping people create collages of their photos. It’s smart to look to users to see what should be your next steps with a product. After all, that is how Instagram developed from Burbn. It is also how Flickr developed into a photo-sharing site despite its founders at first focused on social gaming, and how Friendster became a general social network site (waaaaaaaaaay back when) although it initially started out focusing on dating.
On the whole, Mike Krieger’s talk at CSCW was a fantastic lesson in just how many design decisions go into the human experience even with what seems like an extremely simple app. (It’s hardly simple with several hundred people working on it these days, but in the grand scheme of things, it’s still a relatively simple service, at least from the user’s perspective.) Larry Lessig already made this case very well in Code and Other Laws of Cyberspace over 15 years ago, but it still seems like a less-than-obvious and not-much-appreciated point about our experiences with technologies. Engineers and hopefully user experience researchers are behind-the-scenes of services we use regularly making decisions that allow for certain interactions with the system while closing off other avenues. Mike’s talk made it clear that such decisions can make or break a service so should not be taken lightly.
Oh, and obligatory link to my Instagram account:, esztergram.