Tuesday, 30 July 2013

Twitter - Information Network?

This blog post was originally published on the Convallis Software website
in 2012 but I think that it it is still just as relevant today.

I ended up being involved in a conversation which started after the remark of a top Twitter manager when he described Twitter as an Information Network. Many folk seemed to be disagreeing with that view and were
keen to call it a social network, and expressed concerns that Twitter management don't 'get it'.

I have to say I think calling it an Information Network is a far more accurate description.

Let's examine what Twitter does at its core. It takes data, which is passed into its systems by a client application which must have identified itself and its user to Twitter, and then delivers that data to all the other user accounts that follow the identified user. Incredibly, that data is in general only ever delivered to any given destination once, that's in spite of the hundreds of millions of messages that are generated every day. That's quite an impressive technical feat.

So Twitter is a huge and (hopefully) well engineered broadcast messaging system, i.e. it broadcasts the message to all those who have expressed an interest in following the user.

Why an Information Network? From a computer science point of view all those messages are being moved around a network, and each user could (I think) be considered to be a node on that network both generating and consuming content. But there's more to it than that.

Social Networking seems to be one of those things that raises a lot of passion in people, personally I don't pretend to understand why. But I think (and this is only an opinion) that Social Networking is just one of many Twitter use cases, albeit an important one.

There are others, a broadcast messaging system can be used to simply announce an event, or the publication of a new article or to inform a user that a particular event has occurred (a server backup for instance, although there are perhaps better ways of doing that). None of these use cases are 'social' (although the word seems to be redefined so often that it's difficult to know), especially if bots are used to send them - but that doesn't mean that people won't follow that account if they have sufficient interest in the subject matter.

I suspect that since it was first created, its users have created many use cases for it (probably even most of them) that weren't even imagined when Twitter was first designed.

So I think that the Twitter management view of Twitter as an Information Network is actually reassuring, because it means it's less likely to be pigeon holed into serving any one particular use case.

Author: Richard Isaac - originally published March 2012