Many thanks for mentioning and quoting me in your post, I'm honored!

Most likely social media services will jump on the bandwagon and enable their activity streams to be used by/processed through Twitter clients & services. What service wouldn’t want to access this enormous ‘installed base’ of clients and ecosystem of 3rd party services to increase engagement with users?

If we would look back in the future to this period it could indeed have been the tipping point. However, currently, I would be more convinced when 1) I see the first one or two clients that are actually enabling users to converse and interact between different social media services/activity streams and 2) that there is a build up of users that actually like using it. My hunch is we will most likely be heading toward 2011 and beyond before we see this, but I hope I’m wrong!

Regarding the thoughtful comments of Sam Sethi (hello Sam!): I think hope is not lost for better APIs & protocols that are more suitable for a web of activity streams and its applications. I can imagine, in a kind of transition period, services and activity stream clients will support multiple APIs and protocols, simply because we find out what other useful things we can do with this emerging web and how we like to use it most.

IMHO supporting multiple protocols and APIs is doable as long as it’s confined to a few layers in the software and when it’s not affecting the stack/architecture as a whole, especially on the client side.

Over the holidays I will take some time to dive in to the details & references of your post and Sam’s comment. It would be interesting to have a better assessment of what would really be required for a successful web of activity streams and, subsequently, what the consequences would be for the future if services would massively start implementing Twitter ‘clone’ APIs.

Cheers,

Freddy

That being said I like everything you proposed in your comment and totally agree about the namespace issue. It is critical to allow people to establish personal ownership and control of their data.

Good post and agree we are close to a tipping point. I however do NOT think that the twitter api is the way forward. Yes it is RESTful but sadly it is an HTTP pull based publishing model which is therefore not realtime unless Twitter decided to support PubSubHubbub.

Instead I would prefer to see XML/Atom based ActivityStreams over XMPP, AtomPub or PuSH being the way we distribute realtime content openly between social endpoints like twitter, linkedin, facebook, flickr etc.

As for Tweetdeck, Seesmic and others (Google Reader), my guess is they will evolve to become ActivityStream aggregators/filters but they will never be my personal domain namespace. i.e sites like posterous, tumblr and WP.com are fast becoming personal domain namespaces where people “create, curate and control” their own content aka social objects which they distribute through autoflows to their social endpoints.

Personally in 2010 I hope to see Posterous or Tumblr support ActivityStreams, Portable Contacts and Salmon Protocol. e.g Cliqset does most of this already but sadly is not a customisable personal domain namespace.

But I do agree there is so much going on right now and 2010 will certainly be an exciting year for the Open Social Web. I think people will become more aware of data ownership and privacy.