Sep 172011
 
TumblrShare

I have only just stumbled across Paul Clarke’s brilliant blog. Over the next few weeks, I’m aiming to rectify that omission.  It’s like when, as a kid, you came across a new writer in your local library with a whole shelf of offerings – and whose first book you then read and ended up loving to bits.  Do you remember how that was?  Well, I certainly do.

Paul has a beautifully fashioned thesis on what we might term Twitter’s automated unfollow “feature”.  In his own words:

You find out one day that you’re not following someone you know you used to follow. And you’re dead sure you didn’t do it yourself. Either you or they have spotted the omission on a list, or they’ve tried to send a DM and failed. They might let you know about it. They might not. The relationship gets reinstated. Or it doesn’t. Life goes on.

So is this cock-up or conspiracy? A bug in the system that lets people slip through the cracks like this?

I don’t think it’s a bug at all, but a feature. A piece of very clever social design. Here’s why.

And thus it is that he goes on to explain.  And if you want to find out, then I urge you to read the piece in full.

Now I don’t know Paul – in fact, we’ve only just befriended each other via Twitter itself.  But it seems to me that the explanation he gives for exactly why Twitter has this allegedly intentional and “buggy” implementation indicates the presence of a far kindlier soul than you might – for instance – find in myself.  My immediate thought – and it’s located in a nexus of thoughts which in different ways I’ve had before in relation to Facebook’s own “deficiencies” – is that “buggy” implementations of both social media software as well as search engine algorithms allow for a perfect excuse to break down growing nodes of interaction you might prefer to impede – a kind of virtual censorship, in fact, with no ownership required.  No one need know if three or four users, or a dozen or a hundred, who together might do big, useful but traditionally deconstructing things together, never actually get to know of each other’s presence.  Or if they do, never get to consummate their relationship because the software breaks the relationship down – and the disconnection is interpreted on both sides to be mutually deliberate.

Paul’s thesis is certainly seductive – and even convincing; and I suppose the fact that, in start-ups, we still have an environment hidden away from the prying eyes of state-run security organisations would suggest that software engineers in such a context still might have the liberty to play the kind of games he describes.  What’s more, the reality of such engineering might mean his idea, whilst seductive, is actually unnecessary to explain the circumstance under discussion.  As a commenter to his piece pointed out:

As interesting as this is as a theory. The developer in me just can’t buy the idea that Twitter are doing this deliberately. I’ve followed the Twitter API for as long as it’s been around – they’ve got their hands full just keeping it stable and adding important features without introducing something as discreetly clever as this

Meanwhile, even as I continue to love his idea, I’m still inclined to run with my alternative explanation: the ease with which my thesis could be implemented, the reliance people have on the first ten searches which appear on Google and the self-evident probability that these large American companies will do constructive things for large American governments are all things which make me believe my interpretation, whilst clearly compatible with his, in the long run will tend to take over in any network of communication on the scale we are talking about – and eventually become the dominant structure behind what is allowed to happen amongst its users.

Unless, of course, that network were to be technically visible and transparent – that is to say, open-sourced – to anyone who cared to look.

But then that is a matter for a quite separate post.


TumblrShare

 Leave a Reply

(required)

(required)

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>