The interesting problem with Periscope and Meerkat
Periscope and Meerkat are mobile livestreaming apps. Lots of tech people are talking about these apps, and using them feels like exploring a new frontier. Livestreaming isn’t new, but broadcasting video from a smartphone to your Twitter feed is, and maybe that experiences will capture lots of people’s attention and become huge. (Or not.) Stranger things have happened.
This week, I played around with Periscope and Meerkat to see how much data they require when you broadcast. I had assumed that these new smartphone broadcasting services would be doomed because they would kill people’s data plans. Periscope co-founder Kayvon Beykpour cautioned us that the data usage rates are variable, not just based on the speed of one’s cellular network, but what’s being filmed as well. Videos with lots of movement will require more data than if you leave the camera trained on a houseplant.
- Bernie Sanders and Some Democrats Get Ready to Lick Elon’s Boots and Practice the Politics of the Past
- Nancy Mace Is an Irredeemable Garbage Person Who Loves Bullying Vulnerable People and Yet the Media Still Believes Her
- NBC Seems to Suggest a Children's Video Game is to Blame for UnitedHealthcare CEO's Killing
But, in any case, neither app is a huge data hog, at least according to the cursory test I ran. I hit 20 megabytes used after 3 minutes on Periscope and 4 minutes on Meerkat (using the tracking app My Data Manager). That’s a lot, but it’s not outrageous for the occasional user; it’s comparable to streaming music from Spotify. One would have to broadcast for hours before chewing through a big chunk of a data plan.
But while running a test on Periscope and Meerkat’s data consumption, I discovered a different, but significant, problem that may hamper these apps’ growth: namely, the difficulty of keeping up with a flood of real-time comments.
While I was testing Periscope, something like 80 people began to watch me. And as the people started piling into the Periscope room and the comments started flying around (most were about my shirt, my hat, and my ears, among hellos and shoutouts), I realized that I could not maintain my line of thought. Anyone who wanted to actually talk about what I was doing was drowned out by randos—and as soon as I tried to respond to one comment, there were three more pushing it off the screen. Suddenly, the 12 sentences that I was trying to deliver turned into a mush of random, disjointed responses to individuals, while I tried in vain to circle back to my original intent.
Imagine standing at the front of a room with everyone shouting questions and comments simultaneously at you. That’s Periscope and Meerkat right now.
It’s not just me: I have watched anybody whose broadcast starts to get attention stumble as people start saying talking at them. Kids, media pros, TV broadcasters: No human brain is meant to read a stream of commentary while talking. Cognitive psychologists have found that even a task as simple as tapping a finger or walking can make it harder for people to talk.
My point is this: The one-to-many broadcast model on the livestreaming apps works fine. But the one-to-many communication model just does not scale. I’m sure anyone who broadcasts a lot would get much better at dealing with the inputs and outputs, but I think there are real limits to how good anyone can get at this kind of task.
Beykpour, the Periscope co-founder, agrees that this is a difficult problem. They’ve actually decided they think a 100:1 ratio is tenable right now. “We handle this in a crude way right now, by cutting off commentary after 100 viewers,” he told me. Beyond that point, only the first 100 viewers and people the broadcaster follows can comment.
But he doesn’t see that as a hard limit. Beykpour insists that there are ways they can execute the user interface that would allow more than 100 people to communicate with a broadcaster. “I’m super fascinated by the idea of how UX and product mechanics can help push that number even higher,” he said.
I’m fascinated, too, but skeptical, too, based on watching people perform on the app. What if a single person’s mind simply can’t do it?
As Periscope and Meerkat evolve—and they will, of course, to differentiate from each other and in response to user behavior and product vision—maybe the social norm will be to ignore the chatter. Maybe only the people you follow will be allowed to comment. Maybe you’ll be able to dynamically toggle comments off and on. Or maybe the real action will run through Periscope’s “Private Broadcast” functions.
Funnily enough, porn sites (who were doing one-to-many livestreaming with chat rooms a decade ago) seem to have landed on a solution to this problem: moderation! On most popular camgirl/camboy sites, the person on camera doesn’t do most of the communicating with the people who are watching. That’s handled by some other person or persons, whose job it is to block trolls, threaten haters, and generally try to keep things under control.
Perhaps, as in most Internet things, porn will lead the way.