Data Wallability

Data Portability is the option to use your personal data between trusted applications and vendors. That’s not as clear as it should be, but it does imply that you own the data, the option to use it, and the decisions about trust.

So what options do vendors have? According to Mike Arrington, they can make announcements about Data Portability while fighting for control of your data, so that Data Portability is the new walled garden.

You can get a more concrete and interactive view of vendor options by playing the game hosted at ReadWriteWeb:

you can choose to play the role of any of 5 different players: Google, MySpace, Microsoft, the Data Portability Project, or Facebook. You can then predict what will happen, or voice your opinion about what should happen. Or both.

These players will no doubt make further moves next week. More then.

DataPortability Seeks New Logo

So the DataPortability project got itself a logo. Then it received a cease and desist letter from Red Hat, which claimed that the DataPortability logo was too similar to the Fedora logo.

I think that the logos are uncomfortably similar, and so wish that DataPortability had gone with something more distinctive. But I wish that Red Hat had let it go, or objected in a less legalistic way.

Anyway, DataPortability is running a contest to find a new logo. Cofounder Chris Saad describes the contest. Mike Arrington approves of it.

I also approve of the logo contest, and, in particular, the decision not to expend resources fighting Red Hat over this. I wish that there was a contest to replace the rather misleading name DataPortability with something better.

Data Property Rights

You have data in at least one web application. Given that you’re reading this, the chances are that you have data in many web applications. If you don’t like the way your data are being used by the application, you can always leave. So you want your data to be portable.

Nitin, posting at GigaOm today, argued that data portability isn’t as important as what he terms data property rights. He went further, to state that “data portability is a non-solution to a non-problem.”

Is the DataPortability WorkGroup, then, solving a non-problem? I don’t think so. I just wish that we could solve the problem of its misleading name.

I like Nitin’s post, in part because it highlights this problem. I also like it because of his application of the exit-voice framework to the issues. Portability relates to exit, to taking your data elsewhere. Data property rights relate to voice, to having a say in what is done with your data.

Exit and portability are important as threats. If you don’t like what an app provider is doing with your data, you should have a credible threat of exit. The existence of the threat is one of the reasons for app providers to behave themselves.

DataPortability’s first principle is that: We want sovereignty over the profiles, relationships, content and media we create and maintain. Portability itself is only one aspect of that sovereignty.

Portadentity: All Right Then, Data Portability

Your portadentity is your identity, portable across web services. You probably haven’t heard the term before, because it just occurred to me, and Googling it yields no hits. The concept may well be familiar, since it has received a lot of coverage in the last month or so. A recent example comes from yesterday’s Financial Times.

It is a frustrating fact of modern internet life. Users of websites such as Facebook and Google spend hours building up and maintaining friend lists and e-mail address books, but when it comes time to move such social information to another online service they frequently find it impossible to get their data back out. Instead, they must start re-entering their personal details from scratch.

Another statement of the problem is the Smashcut video. It provides part of an argument for “data portability.” I don’t find that term very useful, but suspect that we are stuck with it, and regard it as a price worth paying for the achievements of the Data Portability Working Group (DPW).

At the heart of data portability is your identity. The DPW, consistent with its policy of using existing standards, uses OpenID for digital identity. Your OpenID, together with the data attached to it, is what I’d call your portadentity.

Today’s big news about the DPW is that Microsoft has joined. Read/Write Marshall reacts as follows. Microsoft’s joining the group is an event of sufficiently complex historical meaning that I’m hesitant to try and interpret it here. I won’t try to interpret it either.

I’m more inclined to spend my interpreting energies on Matt’s statement yesterday about Automattic’s vision of a better web not just in blogging, but expanding our investment in anti-spam, identity, wikis, forums, and more. I added the emphasis on identity.

Here’s what I hope Automattic will do about portadentity… all right, data portability.

  • Appoint someone to the DPW.
  • Make a consumer of OpenIDs. It’s currently a producer only.
  • Make it clear where Gravatars might fit in to all this.

Why DataPortability: Video Explains

The term DataPortability refers to something important, but doesn’t refer to it very well. A new video does a good job of explaining the problem that the DataPortability Workgroup (DPW) exists to address.

The video is by Michael Pick of Smashcut Media. I found it via Read-Write Web. There are some interesting comments arising from Marshall Kirkpatrick’s post there: I now have more idea about what it means to join the DPW.

I think that the video is better as a statement of the problem being addressed than as a description of DataPortability. That’s less of a criticism of the video than a request for a sequel.

Data Portability: Worthy Aim, Misleading Term

Given that you’re reading this, the chances are that you’ve created profiles and network links across a bewildering variety of web services. You may feel that you shouldn’t have to recreate this data: you should be able to access existing copies of it, and otherwise have control of it.

That’s the problem that the Data Portability Workgroup exists to address. I applaud the aim of the DPW, while wishing it had a more appropriate name.

Data portability suggests to me that, if I have data in, say, Facebook, and I don’t like Facebook’s policies and practices around that data, I can take (port) it elsewhere. But that wouldn’t really help me if my friends’ data are still in the Facebook silo.

So I’m glad to say that the DPW’s aim goes beyond this. I think that the best characterization of what the DPW is up to comes from Duncan Riley.

The DataPortability Workgroup is actively working to create the ‘DataPortability Reference Design’ to document the best practices for integrating existing open standards and protocols for maximum interoperability (and here’s the key area) to allow users to access their friends and media across all the applications, social networking sites and widgets that implement the design into their systems.

So it’s particularly good news that there are people from Google (Brad Fitzpatrick) and Facebook (Benjamin Ling) on the DPW. It is, of course, a long way from people joining a workgroup to that workgroup developing a reference design to that design being implemented.