Sam Kinsley

The launch of Google Buzz has prompted me to raise some things that have been lurking in the back of my mind for some time. These thoughts began when the discussion about the ‘walled garden’ nature of facebook et al. emerged a couple of years ago and lead to the initiation of tentative steps towards interconnection and (that horribly overused word) ‘openness’ in the guise of ‘friend connect‘ and ‘facebook connect‘. Twitter was already sort of ahead of the game with their API, as the glut of applications for ‘tweeting’ attests.

Lots of talk on the interweb’s various locations for commentary centred on the social web, real-time web etc. being based in discrete platforms. This remains somewhat true today. We can certainly connect these services together and form extraordinary information gathering tools in the form of what Howard Rheingold usefully describes as ‘personal information dashboards’, using services such as netvibes and pipes in concert with the various APIs for the platforms we all use. However, this all takes quite a bit of effort at the moment [but! for a good tutorial, please check out Howard's super videos: #1, #2, #3].

However, for the majority of internet users to usefully stick all of these various platforms and applications together there needs to be a much lower threshold of effort to achieve the desired results. Jyri Engstrom, co-founder of Jaiku and one of the big brains apparently behind ‘Buzz’, articulates the argument well here:

Most of the conversation over the last 24h has been centered around predicting if “Buzz will kill” this or that service. This debate starts from the assumption that Buzz and the rest of the social web are mutually exclusive. It’s arguably fair to assume so, considering all the social networks we’ve got so far are silos. To no longer assume everyone has to be using the same branded system to talk to each other is disruptive to the tech biz discourse, which is obsessed with turning everything into a war over which company is “the one”. So much so that the alternative is almost unthinkable. If the new standards succeed, in 2015 we’ll look back and shake our heads like we shake our heads today at the early days of proprietary phone networks and email systems. The thought that you couldn’t call, text or email people just because they happen to be on another phone operator or email client is laughable. Doubly so for the social Web. The reason many of the current commentators miss this point is that they are, in the immortal words of Walt Whitman, “demented with the mania of owning things.” (borrowing that quote from Doc Searls)

What are these ‘new standards’ then? Well, if we’re to take our cue from Google they consist of the development of the various existing data formats for syndication: extensions of Atom and RSS, such as activity streams and mediaRSS. There may well be families and hierarchies of such data formats and I’m sure hundreds, if not thousands, of developers are already working on creating these things. But I’m still left with this question: what if I don’t want my stuff (information, pictures, etc.) always held on servers owned by Google, facebook etc? What if I’m happy for such ’stuff’ to be transient? Which of course such companies don’t want because your ’stuff’ is incredibly valuable and they want to mine it for all its worth. Nevertheless, my half-formed thoughts are: what’s the equivalent to IMAP for social media?

To my mind, the missing ‘glue’ for the social networking ecosystem is the missing service architecture to allow all of us to host our own streams and tie together the various bits of our rapidly growing, perhaps increasingly ‘public’, ‘digital identity’. Social media could easily be distributed, just as blogs and ‘web 1.0′ are. What’s to stop a community creating something like wordpress or drupal for activity/social streams? If the standards suggested by Google really are that versatile then all that is necessary is to create a system that imports/exports using them. Search would be renewed in its importance, but companies/services like twitter could remain successful by facilitating that search functionality and helping users subscribe to one another’s feeds/streams.

A couple of years ago I thought about it in terms of a ‘meta-platform’ or ‘platform for platforms’, but we’ve kind of seen these, in the form of friendfeed and their ilk. Now I think, well, it could all still happen over port 80 with web traffic - it just needs an architecture to allow people to stick things up on their own servers and interconnect. If we take what Jyri says (above) seriously, it seems to me the logical step is to really set the social web ‘free’ and build the elements required to allow people to host their own activity streams. Maybe this is already happening. To build on Jyri’s theme of looking to a future and to paraphrase Alan Kay: “the best way to predict the future is to [build] it”. Go to it!!