The good, the bad, and the ugly of Google Plus

Of all the Google social efforts, Plus has the best chance of making something great.

Google Plus(or: Why Johnny Can’t Friend)

Google Plus is, of course, Google’s second attempt at conquering the world of social networking (or third, or fourth, depending on which of Orkut, Buzz, Wave, et al. you count). Google has rightly been dinged in the past for being a over-cocky engineer-driven Johnny-come-lately, without the design sense of Apple or the agility of Facebook. Plus represents a significant change in how they approach and release product, so it’s worth stepping back a little to see how it stacks up.

The good

It’s instructive to compare and contrast Wave with Google Plus. Wave was a skunkworks run from Australia, they built a separate system to replace email/IM/Docs/etc., they launched with a bang at Google I/O, then discovered nobody could figure out how to drive it. Plus is run from Mountain View, they’ve built some new stuff but a lot of work has been integration with other teams in Google, and they’ve soft-launched (as softly as Google can launch anything) outside of a major event, and they’ve had a UX team working on it headed by Andy Herzfeld of Mac fame. They have learned from experience.

Of course, once you accepted that Wave Was Going To Replace Everything, skunkworks made perfect sense. You can develop a self-contained app or feature in a skunkworks. But every Google service will have a social dimension, so they couldn’t build Plus in isolation. The conception of Google Plus as the social infrastructure for every Google service means that it has very different development path, release, and success conditions.

Furthermore, they’re learning as they go. They’re releasing early (the Google Plus for Android app resets to the home screen every time you switch to another app and then back again; Plus is far from feature-complete; the Photo UI is undercooked) but they’re learning from feedback. They have a great feedback mechanism (taking a screenshot from within the browser, letting you highlight and black out regions, so your comments appear in context) and are getting a heap of it. Features are being added and code cut as a result of that feedback.

It’s obvious that the Google Plus team are not blindly copying Facebook or Twitter. They’ve set out to solve the problem that “friend” or “follower” doesn’t accurately represent our social relationships and doesn’t permit fine-grained sharing. They want to solve problems like: I’ll show my kid pictures to my family and friends, but everyone can see the pictures from my visit to the Computer History Museum. Circles and activity streams are how they’re tackling the problem: you tag your friends (“put them in a circle”) and your posts (“share this post with these circles and individuals”), then see the latest activity from people within a circle.

They were very clever to launch with Hangouts, group video chat. It undermined Skype, Facebook were left flatfooted (cf their rather lame launch of video chat), and well, they’re quite cool. Already there are success stories of family hangouts and business use of them.

And finally, they launched with a huge pile of PMs and developers on the product. This feels different: I can hear straight from the horse’s mouth what’s coming and what they acknowledge is broken. It’s more immediate and personal than the product blogs that Google has used in the past. It is, of course, a massive poaching tool for Facebook and other orgs hoping to recruit from Google. I hope that doesn’t stop it being used for user-to-developer contact the way it is now.

The bad

On the flip side, though, it’s another social network. Like all Google projects, Plus makes an attempt to Not Be Evil: you can export your profile, posts, and circles of contacts thanks to the Data Liberation front. But, at its heart, you’re telling Google who you know, what you listen to them about, and what you’re interested in. It’s a massive advertiser’s honeypot.

That is, of course, why Google and Facebook and Twitter are in this business: they make money from advertising. By using their services you’re participating in a giant Faustian bargain: you get free photo hosting and the ability to yap away to your mates, but in return they get to use all that information to advertise to you. And, of course, simply by collecting the information it’s now available to regular law enforcement, the secret police, lulzsec, and other opportunistic parasites feasting off the corporation’s data reserves. And if you make it public, it’ll be scraped by credit agencies, the secret police, and everyone else who wants to build a profile of every person on the planet.

You’ll notice that none of the social networks have subscription options. Nobody says “pay me $100/yr and I’ll keep all your data private and you can have an ad-free experience.” My hypothesis is that this is because your data is worth more to Google, Facebook, and Twitter than you can justify paying for it: they don’t want $100 from you when they can earn $500 or $1,000 targeting advertising to you as you use their sites. They certainly don’t have a federation model.

Nobody’s thinking beyond a centralized profit model, either. AdSense made money for small website publishers, who previously didn’t have a way to commercialize what they did. Mac App Store has made it so easy to make money from software that people now sell rather than give away. There’s no vision in Google Plus to reinvent social networking in a similar platform fashion, creating more value than they capture.

Ok, I’m being melodramatic and ranting. The tl;dr is: Google Plus has not innovated at all in business model.

The ugly

What’s ugly about Google Plus? The UX. I was, like everyone else, mesmerized by the Circles UI. “Circles” seemed a great metaphor for my social network, I mean we already talk about “social circles.” I spent an OCD-happy morning dropping all my friends into the appropriate circles: New Zealand, tech, O’Reilly, etc.

The problems arose when I started to use the circles. If I post something to a circle (e.g., kid pictures to “Family”), someone can reshare that outside the original circle with two clicks. There has, of course, been considerable debate about whether this is a good thing (after all, some say, they could just copy and paste the picture anyway), but I come down firmly against it. If I’m using circles for privacy, I don’t want items to be reshared. Just being able to see my photo doesn’t make you the administrator of it.

That “share” button on other people’s posts makes for another problem: fragmentation of comments. Let’s say Tim posts, asking for places to visit when he’s in New Zealand. I see that and realize I know many more people who could contribute to that discussion, so I reshare it. Now people can comment either on Tim’s original post or on my reshare of it. This makes more work for Tim when he wants to participate in the comments discussion. Sharing blurs endorsement, amplification, and the invitation to comment. Compare with retweeting, for example, where I know that my audience sees it as coming from the original person via me, and replies go to the original poster.

More fundamentally, though circles are used by readers and by writers and they are not adequate to the task. Writers decide who will be interested in what they write: I’m finding that I have to model the mental state of all my friends — will they care about stories of my kids, pictures of gigs I go to, thoughts on technology, rants on NZ politics? And, of course, I can’t model this perfectly: many people in my NZ circle won’t care about my political rants. But they can’t unsubscribe from my NZ rants, they can only take me out of a circle. Unsubscribe isn’t fine-grained enough to be useful.

My classic use case for filters is my good friend James. James tweets about technology and his family, and I want to read those. But he also avidly supports a Canadian ice-hockey team and during games he tweets non-stop about stuff I don’t care about. I want to unsubscribe from his #habs tweets. I know friends who don’t follow me on Twitter because of the floods that come from me during conferences. Google Plus doesn’t deal with these common use cases.

In general, the UI makes it hard to find the stuff I care about. What do I care about? I want to see new things from my friends, I want to see replies to things I’ve written, I want to monitor comment threads I’m a part of, I want to see the stuff my friends like, and I don’t want to see the same stuff again and again. The Google Plus UI mushes all these into a few overlapping streams such that I see the same threads again and again yet can’t find the categories of things I do care about. I think they hope that machine learning will promote relevant items to the top, but the results so far do not make me confident that they can deliver a useful service on this approach. My experience is one of noisy irrelevance.

Currently, Facebook and Twitter both offer a more functional user interface to social activity.


Google is rolling out Plus in the right way: starting small and expanding slowly, learning as they go. I don’t think they’ve got the conceptual model and UI right yet, but I’m enjoying watching it change before my eyes. I still can’t see substantial benefits to consumers in switching to a different centralized social network: it looks like freedom but we’re just leaping out of the Facebook and into the Google. But I’ll continue to experiment because of all the Google social efforts so far, this one has the best chance of making something truly great.

(Google’s Joseph Smarr, a member of the Google+ team, will discuss the future of the social web at OSCON. Save 20% on registration with the code OS11RAD.)


tags: , , , , , ,