July 20, 2011

Google+ and cognitive overhead

Are you tired of hearing about it yet? I’m getting there. There’s more chatter about Google+ right now than there is on Google+, and most of what’s on Google+ is conversations about Google+.

I’m sure Google didn’t set out to build the world’s most narcissistic social network, but considering how it’s designed and the “problems” it’s designed to solve, it also seems inevitable that Google+ would turn out this way. In trying to make a slicker, friendlier Facebook they’ve built a machine that requires so much active maintenance that its users are bonding over how frustrating it is to use, like guests at a dinner party whose hosts have decided not to have plates, where the lack of plates becomes the best, easiest thing to talk about.

It’s clear that Google is closer than they ever have been to getting people excited about a social networking product, but right now I’d say people are more intrigued than anything else. Some friends and I were talking about it at brunch a few days ago, and someone who didn’t have Google+ yet asked how it’s supposed to be better than Facebook. “Well,” one said, “you’re supposed to be able to control who sees your posts better. There are these Circle things.”

Ah, yes. The Circle things. Kevin Cheng wrote an excellent post the other day on Google+ Circles; he points out that while Google’s UI for creating and sorting individuals into Circles is delightful, there are some serious flaws in the overall idea. He writes:

I think I could run an open card sort for myself and probably come up with some good categorizations for my friends. However, once I’ve created these fancy Circles, will I actually remember who will see a given post? From my experience organizing my Facebook and address book, I’ve found that I don’t remember the complex taxonomies I dream up. In fact, I don’t know that I can list every person that’s in my “Family” group in Flickr even though it’s less than twenty.

When compounded with the high overhead of maintenance and likely outdated groups over time, it’s even less likely that I’ll know who I’m actually sharing a post with.

One use case where recall isn’t a concern is in consumption. If you’ve created a “Celebrities” group to read their content, it doesn’t matter if you don’t remember every individual in the group.

In designing or evaluating a user experience, I think a lot about “cognitive overhead,” which is to say, how many logical connections or jumps your brain has to make in order to understand or contextualize the thing you’re looking at. It came up a lot in our last job because our Agile project management tool had a lot of options for categorizing or bucketing stories, but did a poor job of helping us understand where a story fit into the overall narrative, and it was important to understand so we could get work done efficiently and right. Cognitive overhead is incurred when you have to work or think harder in order to get what you want out of the software, and most people’s response to that is to simply ignore it.

User experiences are most successful when they protect users from unnecessary work, either by only including features that are easy to use (like Apple does in most of their consumer products), or by designing features in such a way that users can do something cool very easily and opt into harder/more powerful features when they’re ready for the extra work.

The problem Circles are meant to solve is talked about in the same context as privacy, but is actually more about audience targeting. The idea is that you may want to post something personal or compromising to your Google+ profile, because you want to share it with your friends, but you don’t want work contacts or the public to see it. Google+’s solution is for you to target each post to one or more Circles, but that means for every single post or photo upload, you have to think about the thing you’re posting and which of your various Circle contexts it would be appropriate for. Google seems to think that’s an easy solution, and it may be the best way for software to solve that problem.

But for many people who aren’t engineers or information architects, there are more natural solutions:

  • Post everything to everyone anyway, and deal with the embarrassment or (hypothetical) public shame and future diminished job prospects.
  • Don’t post anything that isn’t appropriate for your whole network.
  • Don’t post anything at all.

The only real difference between the first two is in what a person considers to be appropriate public information. In the last one, no information is public so it doesn’t get posted. Either way, the idea is that posting something to a social network means making it public. And in this context, ‘public’ just means ‘not compartmentalized’. Most things you post to Facebook will go only to the friends you’ve explicitly let into your network, but that’s just a slight difference in how the service defines ‘everyone’. Ultimately it’s far easier to not post than to have to think about your friendships in terms of categories, either every time or just once.

But even though its model is flawed, Google+ is very impressively crafted software, and you can’t beat the convenience factor of having it so tightly integrated into all things Google. Google+ notifications being visible from every Google app is nice; being able to reply to or ‘+1’ something from the notifications tray, or post something new, without leaving Gmail or Google Reader is just killer.

Google is very good at solving this kind of problem — at making it dead-stupid-easy to communicate and share information with people. I just wish they’d stopped there instead of trying to make the social graph more rational. People don’t want their social graphs to be more rational, except after doing something embarrassing when they wish their boss hadn’t seen them so drunk in those pictures. Software is great at giving us overly concrete ways to talk about our anxieties and regrets. It’s the same as what happens when a client or boss asks for Feature X from Competitor Y, but really is just saying they’re afraid Competitor Y will be more successful and they believe Feature X to be the magic talisman that makes it so.

Access controls may lead to safer online relationships, or use software more responsibly, but that’s not really what people want. What they want is contradictory, and hard to rationalize or represent in code. They want all the spontaneity online communication can offer — all of the freedom to share or over-share — but none of the regrets or dangers of over-sharing. That’s not a social problem, not a social software problem, and in trying to solve it Google may have hurt their chance of becoming the world’s way of chattering online.