The Intermission

I read an interesting article today. The Creative Director at frog design, Robert Fabricant, critiqued the iPhone for already showing it’s age. He picked out five “mistakes” in the iPhone, I think primarily pointing out that Apple wasn’t able to foresee some user interface and technical developments that have occurred since they started designing the iPhone who knows how long ago.

Some of his points don’t interest me too much, like when he says the TV icon for YouTube and the camera shutter animation are outdated. I would tend to think those were conscious design decisions that Apple made to give the device a warmer, more human feeling. But one of Robert’s points really hit home with me. He pointed out that the iPhone forces you to view your contacts as a feature of the phone tool. It would have made more sense, he argues, to let users see a list of contacts, and then choose any of the available methods to communicate with that person. For example, you would click on the name of a friend, and then choose to call, text, or email that person.

I’ve been thinking along the same lines for a new web application. In my head, I’ve been calling this application “Google Social” but it’s important that I point out that this is in no way affiliated with Google. I’m using their name as a conceit, because when I dream of web apps I dream in Google (and sometimes 37 Signals).

So what is Google Social? It’s a web tool that shows you a list of all your friends who are on the web. You click on a friend to view the social networks that they belong to (e.g. Facebook, LinkedIn, YouTube, etc.), and then you see a list of that person’s recent activity on that social network. It’s nothing terribly new — you can probably do most of this today through RSS and email notifications.

The important thing to me is the change in mindset. I’m no longer thinking “I’d better check LinkedIn to see what my contacts are doing.” and “Now I need to log in to Xanga to check my friends’ blogs.” and “Now I need to go to my friends’ Flickr accounts to see if they’ve posted new photos.” Instead, I go to Google Social and think “I wonder what Chris has been doing.” It’s a human-centered approach to social networks, instead of a platform-centered approach.

This web application interests me so much that I couldn’t help but design it in wireframes. So I invite you to look at Google Social, as presented in beta clickable wireframes.

These wireframes aren’t great, but I hope they communicate what I was thinking. You’ll notice some red dots around the pages. Not everything is clickable yet, so you’re mostly limited to what has a red dot by it. It’s like a Choose Your Own Adventure wireframe.

There are two ways to add friends in Google Social. First, you can search for people who have already created Google Social accounts. When you find someone, you can invite them to be your friend (very similar to LinkedIn). If they accept your invitation, then that person gets added to your friends list. Alternately, you can create a friend from scratch. This involves entering a fair bit of information. For example, if you have a friend who is a user on YouTube, you would need to enter her YouTube user name. Then Google Social monitors YouTube for any activity by that user and pulls it in for you to see. I’m hoping a lot of this kind of activity can be handled by RSS feeds that social networks are already putting out.

I know some things are still missing. I haven’t designed the sign up process yet, but it will be very important. It’s also missing a notification center for viewing invitations you have sent and received. It’s probably also missing a lot of other important stuff. But, hey, it’s only beta!

Please let me know what you think. Is this a compelling way to keep track of your friends online? What would make it better? Do you see any interaction problems?

Additional research keeps killing off my complaints about the state of social networking. It turns out that there is an open standard for describing people online, and it’s called OpenID (more on Wikipedia). It was created by someone at LiveJournal, and integration of OpenID is planned for Firefox 3 and Widows Vista. So this standard seems to have some momentum. What it needs is larger adoption from the white label social networking platforms I mentioned earlier.

Fortunately, I’ve found at least one platform that seems to be supporting it. Everyone needs to take a look at PeopleAggregator. They have a fantastic feature set, a great stance on open standards, and what appears to be a robust set of APIs. I’m going to show this to a colleague at work and see if it really is as good as it looks.

Here’s hoping!

Earlier, I pointed out three problems I see with social networking. My first two problems are, I think, still valid. But I’ve done a bit more research and found out that there is probably no need for another startup to create a set of social networking APIs. They would be entering an arena with at least 40 other competitors. Apparently, in my earlier searches, I wasn’t using the correct terminology. These companies that make APIs for social networking are called “white label social networking platforms”. “White label” meaning anyone can rebrand them.

I haven’t had the time (and who would?!) to look at all 40 of these platforms, but I can already share some preliminary thoughts…

First, the current arena of social networking platforms looks like many other emerging categories have looked before. There are way too many competitors for the marketplace. In the next few years, these competitors will die off, merge, or get bought. Ultimately, I’m guessing the major players will look like they do in other software categories. Microsoft will have SharePoint; Google will have its offering (probably a revamped version of a startup they’ve bought); maybe one more name brand offering (like Cisco or Oracle); plus one or two others that develop organically out of the mess of current competitors.

Second, I’m immediately drawn to a couple of the platforms I see on Jeremiah Owyang’s list. Ning looks awfully interesting, in part because of their feature set and in part because I just learned that they’re one of Adaptive Path’s clients. Ning gives me almost everything I’m looking for: a flexible tool for deploying user profiles, blogs, and groups, the ability to search for users within my social network, the ability to host my social network on my own domain, and the ability to include (or not include) advertising. However, they don’t make the cut simply because they don’t allow me to make my social network as independent as I’d like. What I mean is, your social network is always going to be a Ning social network. You can’t integrate a Ning social network with an existing registration system. And when a user registers to use your Ning social network, they’re also registered to use every other Ning social network. This is a major shortcoming for business applications. For a company that wants to create a social network, half the point is to get a bunch of users’ names and email addresses into your database so you can start sending them emails and coupons. Until Ning makes that possible, I don’t see it being used for many corporate-run sites.

Third, I’m scanning Jeremiah’s list to find a platform that is open source, modular, and purely utilitarian. I’m looking for the social networking platform equivalent of PHP or WordPress. Something easy to set up, without a lot of branding, lets me remix the features in a modular way (e.g. I want to install forums and blogs, but not groups), provides a registration engine, but also lets me integrate with an existing registration database. I haven’t seen one on the list that sounds like that yet, but I’ll keep reading. For my money, that’s would be the best of breed winner.

As an interaction designer, I try to keep my eye on what’s happening in the world of online communities. Also, as an interaction designer, I’m occasionally asked to help design online communities. This has got me thinking about the problems with online communities as they currently exist. So I’m going to examine some of the problems I see and recommend some solutions.

Problem #1

Most communities allow too much anonymity to provide a useful view of who’s reputable and who isn’t. Consider this example. In the real world, I might own a restaurant. However, I’m rude to my customers and the food is rancid. It won’t take very long before people stop coming to my store. If I tried to overcome my bad reputation by opening a different restaurant, people would take one look at me behind the counter and leave. I have ruined my reputation and now nobody trusts me. Rightfully so.

In contrast, in the online world, I might be a private seller on eBay. If I consistently make false claims about my products, ship them late, or ship them damaged, my reputation goes down. But I can always just create a new email address and start all over again.

This problem is compounded when you consider how it plays out across multiple communities. I might be an excellent community member at Slashdot, but when I open an account with eBay, none of that good karma carries over. Alternately, I might have proven myself to be an unreliable seller at Amazon, but I get a fresh start at eBay.

Solution #1

Somebody (I’m nominating Google, but any Internet startup could take a shot) needs to create an XML-like file to describe an online person. This online person must have a single match to a real life person. Your online person contains links to all of your online activities… your Facebook profile, videos you’ve uploaded to YouTube, photos you’ve uploaded to Flickr, product reviews you’ve written on Amazon, etc. It also contains a single “karma” or “reputation” points system, which reflects on you wherever you go.

Whoever eventually does this will be taking on a huge task. It involves getting numerous web services to all agree to one standard. But the practical upshot is that people will be just as accountable for their actions online as for their actions in real life. I can no longer troll on Slashdot and still expect to get good business on eBay.

Problem #2

I have a hard time keeping track of “where” my friends are online. An average friend online might have an account at Facebook, Xanga, Flickr, LinkedIn, Orkut, YouTube, Digg, Meetup, not to mention IM accounts with AOL and MSN, and email accounts at Gmail and Yahoo. How can I keep track of what they’re doing if all of their online activities are at different sites, likely with different user names? I want to keep track of my friends, not my friends’ online accounts.

Solution #2

After everyone has a single online persona, then I can (like in reality) become friends with a person, rather than make all my online accounts become friends with all their online accounts. When I log into my online persona at Google (or whoever develops it), I see a list of my friends, and each friend in my list is tied to all their online accounts. For example…

Welcome Jeff Stevenson!
Your reputation is currently 80/100.

Your Friends:

Sandy Phelps
View Sandy’s profile at YouTube.
- Sandy uploaded the video Max Jumping Through Hoops (27Aug2007, 3:47 PM EST)
- Sandy commented on the video Roller Nightmare (27Aug2007, 3:30 PM EST)
View more…

View Sandy’s profile on Facebook.
- Sandy added 5 photos to the gallery Girl’s Night Out (25Aug2007, 12:10 AM EST)
View more…

View Sandy’s profile on Digg.
- Sandy dugg 8 Most Successful Movie Franchises of All Time (22Aug2007, 1:13 PM EST)
- Sandy dugg How to Wash Your Clothes (22Aug2007, 1:10 PM EST)
View more…

…and so on. No more logging into 10 different accounts to see what my friends have been doing.

Problem #3

Every one and his mother wants to build an online community these days. I won’t name names, but some of the clients I work for are basically asking us to build MySpace inside of their site. This is a tremendous waste of resources for several reasons, even aside from the fact that they’re just throwing in features without any understanding of what they do (”give us a tag cloud, and blogs, and groups” which you might as well translate into “give us Web 2.0″ in terms of meaningful requests), yet they’re unwilling to pay for something as simple as a host (I’m using “host” in the same way Powazek uses it in Design for Community) to raise the quality of the community. But if you do end up creating an online community for a client, it costs an arm and a leg, because all the community features need to be developed from scratch.

Solution #3

Someone should create a decent set of freely available APIs for adding community features to a web site. (I’m nominating Facebook, since their work has already begun, or PublicSquare, since they seem to really understand what users and site owners need.) That way, next time one of my clients asks for community features, we don’t have to build an expensive, proprietary system. Instead, we’ll plug into the same architecture that everyone else is using, and we’ll share access to the same reputation points as all other online communities.

If Facebook or PublicSquare provides this toolbox of community features, then interaction designers can pick and mix the parts they need to create a community appropriate for their clients. It’s win-win-win!

Your thoughts?