The Big Pink Elephant of DataPortability

Big part of my new job is to make recommendations regarding which community initiatives Yahoo! should consider getting involved with. Given the recent attention received by the yet-to-be-defined DataPortability organization, it was only natural for it to come up in multiple conversations.

A few months ago I explained why I personally decided to leave the DataPortability group. Now I would like to explain why I am recommending Yahoo! to stay away as well. Obviously, I only mean to stay away from the organization, not the actual principal of openness and data sharing which are at the core of Y!OS.

I agree with many of the smart individuals who are still listed as members of the DataPortability group, that it had at least accomplished creating a common term and started a significant discussion around the topic. Regardless if the literal meaning of the term is misdirected, it is now being used as an umbrella term for all things open in the social space. This is not a small accomplishment and I am happy to give credit when credit is due.

In many public and private conversations, members of the OAuth, OpenID, and Microformats communities tried to convince the primary leaders of the DataPortability group to focus only on marketing and public relations. Everything else must be out of scope, from making technical recommendations to writing specifications and defining best practices. Obviously this advice fell on deaf ears, although there seems to be a pretty scary difference between positive responses we've received in person, to the rhetoric made in written public responses.

The DataPortability group does not speak on behalf of any of the standards they decided to promote, a process which by itself has never been properly explained. The only exception being APML, which was suspiciously included in the list of well established community and industry standards, as it has been written and endorsed by a leading member of the DataPortability group. The topic of the selection criteria for endorsement has been discussed numerous times on the public mailing lists, but remains unresolved.

At the end of the day, success in this space comes from leverage - from having something to offer the big players whose support is so highly desired. Key members of the OAuth, OpenID, Microformats, and other communities have the real power of being able to facilitate both discussions and changes to standards in order to accomplish adoption by the major players in the space. While these are still community-driven efforts, the most significant work is done behind the scenes.

Using OAuth as an example I am intimately familiar with, everything was done in the open and the community always had the final say, but much of the prep work and negotiation were (and still are) done using backchannels where corporate participants are more comfortable discussing adoption and key issues. It is also the only viable arena for complex discussions of Intellectual Property Rights and other legal issues which are part of any standard work. Politics never works in a public mailing list.

Being able to meet with companies as a knowledgeable member of the community and propose changes needed in order to allow them to adopt the standard has been critical in getting traction. Please note I am not suggesting any actual ownership or official role within the community, just that my involvement allows to me facilitate this kind of interaction.

The work done by these key members is rarely public until it is ready to be presented to the community, and OpenID would be another perfect example for this kind of work. It involves very little credit or acknowledgment - only the actual accomplishment is celebrated as it should be. This work has produced many successful results and initiatives than the DataPortability group will ever be able to claim.

The DataPortability group doesn't have the power to negotiate changes to these standards that will both enhance compatibility, but more importantly allow greater adoption. The common practice when someone has an opinion in support of other people's standards and ideas is to blog about it, not to start a bureaucratic organization just to amplify these views and call it a movement.

The obvious reply from the DataPortability folks to any complaint made is to suggest joining the group and help answer these questions. But when anyone actually follow this advice, the answer is a discussion-ending suggestion to form yet another workgroup where any complaint will be buried away from public eyes. Seriously, how many workgroup, committees, and sub-communities are going to be created before someone realizes that the DataPortability group isn't the federal government.

The DataPortability members are making demands but offering little new. They pick a choose standards, mix and match them, and then expect others to abide by their so called technical recommendations. Yahoo! knows how to call Google, Facebook, AOL, MySpace, Microsoft, Plaxo, and others. In fact, people don't realize just how much these companies talk to each other on a daily basis. Big players don't need the DataPortability group to help them facilitate a conversation.

Cool stuff is coming from Yahoo! and I am excited to be part of it. It will be open, portable, and deliver the promise of putting the users in control of their data. I invite the members of the DataPortability group to continue discussing ideas, promote open standards, give us feedback, and to continue doing the one thing they actually do very well - talk about data portability.