Next big thing?
Messina says the next thing the DiSo project that he is part of ("an initiative to facilitate the creation of open, non-proprietary and interoperable building blocks for the decentralised social web") will work on after Activity Streams is groups and a peer-to-peer messaging system that enables you to send a message from one URL to another.
Salmon is one mechanism to achieve this kind of functionality. In fact, social networking tools Cliqset and Status.net recently announced the first live implementation of the open source Salmon protocol, which means users don't have to be on the same network to send messages to each other.
If others follow, the significance for the open, interoperable web could be huge. In the end, users will be able to send messages and share photos across networks, without any one company in the middle.
"I think this is a great step forward," Messina enthuses, "and a perfect demonstration of how we're advancing interoperability between networks. The Salmon spec isn't even finished yet but we're nearing a moment of rapid acceleration in distributed social web technologies."
Messina is also proposing an overhaul of OpenID (Google is by far the biggest OpenID provider, with 61 per cent of registered accounts). "OpenID has been greatly challenged by Facebook Connect," he concedes.
"On the OpenID website we've been running around like our hair's on fire. The problem is that OpenID historically took the approach of putting privacy at the forefront and we didn't account for people who actually want to share data with certain people. Because we didn't build this into the protocol to start with, we're only now getting to the point where we can create these interoperable technologies. I describe this as OpenID Connect.
"I use the word 'connect' very purposefully because Facebook did a great job in coming up with a verb that makes sense and has a different connotation than 'sign in' or 'register'. OpenID needs to be much more expressive, providing richer information and making it easier for people to sign in across the web and bring data with them in a way that's more secure."
Messina's aim is to combine OpenID with OAuth 2.0 to tie together various open technologies like Activity Streams and Portable Contacts, but he's not convinced by Facebook's recent decision to kill off the 'Facebook Connect' branding and instead implement OAuth 2.0 for user authentication as part of its new Open Graph platform.
"Regardless of what Facebook does, I think that the 'connect' verb is an important one that will only become more relevant as people need to provide long-lived access to their data. As an industry, we still need to educate people about the choice that they deserve in who provides their online identity.
While I'm all about making online identity easier to leverage on the web, taking the user out of the identity selection process may result in unexpected behaviour. That is, just because you're signed in to an account in one tab doesn't mean that it's the right identity for you to use in another tab."
In November and December, Messina also collaborated with Mozilla to envision how the browser could become more social. The concept he came up with already included OpenID Connect (see his blog entries).
"The idea was to think about the browser as a social agent," Messina explains, "somebody who helps you deal with all this information that you're producing and consuming on the web. It starts with indicating to your browser who you are. In the same way that I can take my SIM out of my phone and put it in another phone, I should be able to bring my identity into a browser and have all my preferences and friends be part of my experience."
So, while Apple increasingly counts on a closed system and Facebook seems intent on centralising user authentication and the storage of all identities, Google is much more about the open, interoperable web.
Now Messina has taken his campaign for openness inside the biggest internet company, and has its support in the battle to get more players on board. The opportunity is there, he says, to work together on these technologies while they are in their early stages and then compete on providing better services and user experiences. Surely, this can't be a bad thing?