Thoughts on chi.mp

by Marty Alchin on June 23, 2009

After following some discussion about it, I signed up for chi.mp once it launched as a beta service, and it hasn’t really impressed me all that much. Recently, though, I was asked to take a survey about my experience, and I was disappointed to see that the survey didn’t really ask what I feel are the right questions. Since surveys only allow you to answer the questions that are asked, I’d like to take a few minutes to speak candidly instead, in hopes of better addressing the issues I have with the service.

To set the stage, here’s an excerpt from the first post on Own Your Identity, which introduces chi.mp in the context of the problem it’s trying to solve.

We need an Enlightenment online. An evolution of personal identity that says I’m free to throw sheep at people on Facebook or explore new bands on Myspace, but my content, my contacts, and my communications are in my control not theirs. I want people to be able to connect and interact with me through one unique identifier that doesn’t change, no matter where I choose to host my identity. I want to own my identity, I’m tired of being owned.

A group of us have started Chi.mp to try and jump-start this evolution of identity. We are building a Content Hub and Identity Management Platform that can be deployed on any domain and puts the individual in control of their own identity. People using Chi.mp will have identities that are importable, exportable, interoperable, portable and most importantly theirs. By deploying it on the domain of your choice you can move from Chi.mp to another identity provider without losing the unique signifier that represents you. Oh and if you don’t have your own domain we’ll give you one (like everything else) for free. We’re turning the social networks inside out and making the Internet the Platform again.

This sounded very intriguing to me, despite its obvious tendency for marketing-speak. The notion that a third-party service could allow me to control my identity, content and contacts on other social networks sounded—well, I’ll say it: it sounded pretty damn cool. I had just started to get into Twitter and Facebook and I had already noticed the burden required to maintain two separate sites with two audiences and two sets of content (after all, different audience should get different content).

But I’m a level-headed geek, so I had my reservations. After all, what they were promising would certainly require the involvement of the social networks themselves, in order to provide third-party access to that level of control. And that also meant those networks giving up some measure of control, which I doubted from the start would be possible. All the same, I followed the Own Your Identity blog, which promised to provide more detail on developments of chi.mp and its services as time went on.

More posts followed, and the chi.mp were eager to respond to comments, creating a friendly dialogue that did manage to convince me they had their heads on their shoulders. We discussed spam, domain ownership and attention management, and the conversation was genuinely collaborative, which I hadn’t really experienced before on a blog, much less a blog for a still-in-development new service. I felt like I was, in some small way, helping to shape the service by speaking up about what I thought it should do.

Then I got my invite. I excitedly signed up for gulopine.mp (Gulopine is what I use to communicate online), eager to see all the goodies that were waiting for me. Here are all the features I had, as far I could tell, anyway.

So, if I take out the OpenID endpoint, I’ve got FriendFeed. Except FriendFeed supports far more services, allows me to actually read content from the contacts it imports (chi.mp only aggregates my own content) and even lets me add feeds for people it wouldn’t otherwise know about. Curiously, the chi.mp team believes “there isn’t a whole lot of life in lifestreams,” considering them nothing more than “dumb rivers of updates.”

Tragically, chi.mp’s one and only advantage over FriendFeed—OpenID—proved to be unusable for me most of the time, though that seems to be isolated to just me. So I just signed up for MyOpenID, went back to FriendFeed and almost forgot about chi.mp entirely. As a side note, I’ve since stopped using FriendFeed as well. I guess I don’t really need all that aggregation anyway.

What’s that? I left out the most important feature that chi.mp has? Oh yeah, personas. I can create a different persona for each aspect of my life that features a different audience, then publish (or rather, aggregate) content to only those groups that should see it. My mother-in-law doesn’t need to read a programming discussion, but my programming colleagues don’t need to know my plans for Thanksgiving. Sounds good, right?

So I created a few personas and then realized the fatal flaw: there wasn’t much to differentiate. I can put in different profile information (mainly just contact info) for each persona, but that’s about it. All my content is aggregated from sites that don’t have personas, so it’s open for all my contacts—or possibly everybody—anyway. There just wasn’t any real-world value in personas.

Eventually (after I gave up on it), they also added the ability to write status updates, upload photos and write blog posts, all of which are tied in to personas, so I can limit who they go to. But each content update comes with checkboxes to publish it to external services, like Twitter, Flickr and Facebook. Again, the whole value of personas vanishes as soon as you publish to a site that doesn’t support it. Which brings me to the most important point of all.

chi.mp’s most prominent features are only valuable if you point people to your chi.mp site. Sure, they give you your choice of domain, whether on the .mp namespace or elsewhere, but you have to use their service for your public-facing identity. It’s a curiously strong lock-in all its own, even though the whole goal was supposedly to avoid lock-in. The only thing that prevents it from really being perceived as such is that it doesn’t really have much influence. People either aren’t using it or at least aren’t using it for much, so it doesn’t really seem like anyone has noticed.

But there’s a bigger problem at play here, and it’s something I hadn’t even noticed myself until fairly recently. chi.mp doesn’t just lock me into their service, it locks my audience into their service as well. If someone wants to know what I’m up to, they have to use chi.mp to do it. They don’t have to sign up or anything, but they can’t use whatever service they’re already more comfortable with. Of course, my updates are already available on that other site, but chi.mp seems to want to pretend they’re not (because then I’m not “owning my identity”).

I think what it boils down to is this: truly owning my identity invariably means controlling how others perceive me. That sounds good in a way, but I don’t mean controlling what aspects of my life they see and how it’s presented. I mean controlling the physical (or technological) manner in which they gain access to me and my information. If you want to look at me, you have to be wearing my preferred brand of sunglasses, so to speak.

In the programming world, I’ve experienced the audience side of this problem a lot recently. There’s a lot of debate going about with source control system is best. Each has its own unique features and advantages, which also means each one has a different way to use it. That’s all fine and dandy for people who are using those systems for development. Choose the one that works best for your needs and enjoy. But what about people who aren’t hacking on your code? What about someone who just wants to run it? Or maybe just read it?

Currently, if I want to get the latest version of someone’s code, I need to know how to use CVS, Subversion, Git, Mercurial and possibly some others that I’m not thinking of. Not because I care about how any of those systems work, but because different projects use different source control, and getting code for a project means using the system that project uses. Which means learning how that system works and sometimes downloading way more than I really need (some seem to work only with a full local history of the entire development tree).

Personally, I hate it. If I have to learn a new system just to get your code (or read your updates, or get in touch with you), I probably won’t bother, unless it looks really intriguing. And now that I see it from that perspective, I don’t think I can use chi.mp in good conscience. If it had stayed true to its promise (or rather, what I perceived its promise to be), I’d be able to control my identity on other services, so my audiences wouldn’t have to change their behaviors just to interact with me. Alas, no.

To be fair, I do realize the technological and political problems involved with what I really want chi.mp to do. It’s pretty much impossible to expect sites like Twitter, Facebook and Flickr to give up that much control to a third party, I know that. But until they do, I’m afraid the truest form of identity ownership is personally juggling the various sites out there yourself. Third-party services just can’t deliver the real features that are necessary to change the game.

Therefore, chi.mp, I wish you luck, and I hope someday you can realize your potential, but I’m not holding my breath.