Modes of Conversation
Until recently, there were three typical approaches to holding a conversation on the web: instant messaging, email and newsgroups.
A year ago I used to draw the following diagram:
I argued that these different channels fitted different modes of conversation along a single axis of formality and that conversations only ever became more formal. IM’s are on-the spot-conversations I have because you’re online; when I want to cover my backside, I use email; and when I want to involve a larger community and contribute the conversation to a long-lived information base, I employed newsgroups. One never starts a serious conversation and downgrades it to be less formal (though less formal side-conversations may be spawned).
A sucker punch accompanies this argument: These channels are silo-ed and so conversations never do escalate; they remain in the channels to which they were born. Arbitrary protocols constrain the evolution of any given conversation. Probably most of us have had the following experiences:
• The need to transition an on-the spot conversation to one which is longer-living (e.g. questions that we need to research to answer, actions that can be signed off) and involves more people.
• The need to transition from a string (knot?) of emails to a structured discussion with outcomes and information that could be utilized in the future. It also explains emails dominance, despite its obvious weaknesses: Email was in the middle and therefore provided the best compromise approach. This property is something I shall keep returning to.
I reviewed this argument recently and in light of using Facebook and Twitter, I realised it was hopelessly simplistic; the sharp-minded among you will have detected the fault lines in the forgoing explanation.
There are at least two other axes where conversations sit along:
Synchronous-Asynchronous: I start an IM conversation because I can see you are online. If you were not online I would have used email; BUT, if I thought that you weren’t going to respond to my email inside two weeks then I would not have sent it at all – an emails utility times out. Read newsgroups and you see that the responses frequently come back over months or even years - so newsgroups are extremely asynchronous. Note that email still sits in the middle. Slightly orthogonal to this article, but if I want to force a synchronous conversation I’ll call you. That’s great, but typically we want asynchronicity because it allows us hold multiple concurrent conversations. This is the fundamental form of modern collaboration, and any approach that is not predicated on this will fail (are you listening Second Life?).
Narrow/broad casting: An IM is a conversation with YOU. I know I can invite others into a conversation through most IM tools/protocols but typically I do not. As such, IM is strongly narrowcast. Email can start off as a conversation with you but can quickly be expanded to involve others. Newsgroups are open in nature and thus broadcast. To labour the obvious: email is yet again in the middle.
What about directionality along these axes? Certainly successful conversations become more asynchronous because they involve more people. But hang on Jasper, you’re mixing your axes! Isn’t the ‘more people’ term a simile for broadcast? And what exactly is a successful conversation?
I don’t know but it is probably measureable: Length, number of spawned conversations, number of participants, number of outcomes, actions etc. And yes I mixed my axes. Successful conversations are probably formal, asynchronous and broadcast.
So the question beckons: What place do Facebook and Twitter occupy on these axes?
I think they line up with email here; conversations remain a jumble, and yet have an order and threading that is more sophisticated than IM, but less so than a newsgroup.
I think that Facebook and Twitter are rather more synchronous than email because conversations are within a broader stream that leaves conversations behind if they do not remain fresh.
This is an interesting one because Facebook is DEFAULT PRIVATE, whereas Twitter is DEFAULT PUBLIC. So although they can move along this axis, they tpically reside here. So like a political party, grabbing the middle ground is probably the right strategy. Twitter and Facebook sit in that middle ground like email and that is why I think they are genuinely important (despite personal reservations I have with respect to the range of use-cases that they support).
There are weaknesses in this line of argument. Are newsgroups really broadcast when there is a single place you read them? Just because the door is open and you don’t know all the people in the room, it does not follow that you are broadcasting. I’ve played with the metaphor of coupling as an alternative. In newsgroups we are coupled strangers because we haunt the same forum. But on Twitter I am typically broadcasting but loosely coupled.
There is also an important relationship between broadcast and formality; how I talk and present argument is different if lots of people can see my chatter, than if I’m talking to my inner circle i.e. I automatically move to a formal mode when broadcasting.
What we are with witnessing with Facebook, Twitter, FriendFeed, Tumblr et al are baby steps to better modality of conversation. What no one has yet demonstrated is a viable approach to shifting those modalities as the conversation demands. There is a big opportunity for anyone that really cracks this.
2 years ago