Landing : Athabascau University

Excellent news: Twitter Makes A Bet On Protocols Over Platforms.

Well this is good news! Of course, the road to Hell is paved with good intentions and there is much that could go wrong in between plan and execution, but it seems that Twitter is recommitting itself to openness, standards, and the use of protocols for a federated social Web (see also and for the announcements by Twitter's founders). It is a bit worrying that Twitter wants to help invent a new protocol when there are plenty of established ones that already exist (ActivityPub, OpenSocial, FOAF, XMPP, OStatus, OpenID, OAuth, PubSubHubbub, Zot, Diaspora, etc, etc). Also, there is already a pretty serviceable Twitter competitor in the form of Mastodon, that does most of what they seem to want to do. However, the fact that they are thinking about protocols rather than platforms at all is very heartening. The world needs much much more of this.

Twitter, as it evolved in its first couple of years, was brilliant. What made it great was that it could act as a highly efficient social bookmarking system *plus* commentary *plus* folksonomy, *plus* instant messaging, *plus* social networking, all through one incredibly simple, flexible, open field.  It was, in part, a descendant of social bookmarking systems that people like me developed in the 90s, but there were no predetermined fields for URLs (you could have more than one, or none at all); there were no predetermined categories; the tags (#hashtags) were trivially easy to include, without separate fields (this is what makes it highly supportive of social sets, in which the topic matters more than the person); and it had the lowest threshold social networking (especially through @mentions), again without the need for separate fields. It was a single small text box that did everything, and that could be used to share more or less anything with more or less anyone but, thanks to its size,  was primarily used to connect to other things. Part of what made this so cool is that #hashtags and @mentions were not designed into Twitter at the start, but emerged memetically from practice: the system evolved (at first) through a collective design process. Twitter's implementation of such things in software ingeniously used automation to make the overall system even softer and more flexible than it was before. It was generous in what it shared, too, so a flourishing ecosystem grew around it, at least for the first few years. You could use pretty much any Twitter data to which you had access in any way you liked. It was a very simple, very powerful component, a tool rather than an environment or platform. In retrospect I wish we had used Twitter as a model when developing the Landing, rather than the kitchen sink approach that we settled for.

Twitter is widely viewed as a competitor to Facebook - increasingly even by the company itself - though it was (and still is, to an extent) a very different animal. Facebook has tried to emulate all of Twitter's features as a subset of its own horrible evil mess, but completely misses the point. The strength of Twitter is that it (still) does one simple thing very well: it is primarily a hub that makes the rest of the Web more connected, rather than (like Facebook) sucking everything into it. However, that one simple thing is as soft and open to countless, unprestatable uses as an elastic band, a screwdriver, or good old fashioned email.  Jack Dorsey's announcement of the new move itself is a classic example of this, creating a long-form announcement from short tweets. Beyond simply connecting stuff, people have used it to write novels, coordinate social protests, conduct personal conversations, influence elections, and thousands of other things. It is a very soft, very human-driven tool.

For a few years it was very open, and it seemed to be getting more so, but it lost its way after that and became much more the self-contained platform we see today, pulling a lot of features into its core, closing off many ways of connecting with and using it, and increasingly hardening things that should have stayed soft, notably in its algorithmic placing and sorting of tweets. Though its old character limit was frustrating at times, it was actually a very good idea to set such severe boundaries because it ensured that Twitter remained as a connecting hub, rather than a self-contained site. The new higher character limit is still somewhat constraining, but it makes longer-form conversations increasingly possible - especially when combined with the easy upload of video, files, images, etc - thus drawing people to stay more at the hub, rather than to visit the things that it connects. It has become more and more a social media platform, increasingly isolated, increasingly its own bubble, increasingly driven by the popularity contests and narcissism amplifiers that seldom end well. Twitter's announcement, I hope, marks a reversal of this pattern. I hope (though don't expect) that they get the Mastodon gang on board. I will watch with great interest, whatever happens.