140Char

Last night I spent a fair bit of time chatting about Twitter with a friend in the publishing industry, as we talked about how useful we find it, and how it has replaced some of our usage of email and Facebook. We’re both around 30, and we’re both mixing professional and personal use to connect with work contacts and friends.

And yet, sat on the train home surrounded by 10+ teenagers chatting away, there was not a single Twitter mention – overhearing them without trying to eavesdrop, my ears naturally picked up the 5 or 6 mentions of Facebook.

Anecdotal experiences are always interesting, but I’ve also been following the spread of Twitter surveys like the Pear Analytics ‘pointless babble’ whitepaper. By categorising 2000 tweets in English and in the US and putting them into buckets for News, Spam, Self-Promotion, Pointless Babble, Conversational and Pass-Along Value, they concluded that Pointless Babble makes up 40.55% of tweets, followed by Conversational and only 3.6% are news.

Many places simply repeated the study, but two people I respect a lot have responded:

There’s a great post by Stephen Fry, pointing out that Twitter was never advertised as anything other than a means to connect to people.

‘The clue’s in the name of the service: Twitter. It’s not called Roar, Assert, Debate or Reason, it’s called Twitter. As in the chirruping of birds.’

And the always well-reasoned research mind of Danah Boyd looks at whether the fact that conversation, both online and offline, tends to be social, is actually a good thing, anyway – and our obsession with trying to claim some measure of perceived value

‘I vote that we stop dismissing Twitter just because the majority of people who are joining its ranks are there to be social. We like the fact that humans are social. It’s good for society.’

Well worth reading…

Yesterday, Chris Hambly of Social Media Mafia circulatd the link to an online research paper about Twitter. In brief, this research downloaded the last 3201 tweets of about 300K users.  Twitter keeps only our last 3201 tweets.

200K were effectively  non-users and the remainder provided a sample of network connections for an ‘average’ period of 7 months.

This article caught the imagination of the Twittersphere.  One research finding in particular caught people’s eye: that we don’t have conversations with everyone in our Twitter stream.

I would have thought that’s obvious myself.  I am also uneasy about the use of the expression ‘average’ without co-reporting the distribution of the data.

This is fun research though. The authors, Huberman, Romero and Wu (2009) make the point that Twitter offers a brilliant opportunty to uplift digitalised network data and learn something about social networks.

That’s my ambition.  I want to become a lot more fluent in social network theory.  I’ve sent up a wiki & forum, Netinkers, where like-minded people can congregate, initially to parse the Huberman article and then to add others and to suggest lines of research.  We may not have the time and resources to do the research ourselves, but students might.  They are always on the lookout for ‘doable’ projects.

See you over there when you feel in a scholarly mood. The password is ‘playful’.

Cli.gs, a short url service with analytics, has released stats and analysis of 10.2 million tweets and 2 million links to see which domains are most used.

Unsurprisingly Tinyurl leads the way as the default shortening service automatically provided by Twitter.

Interestingly the data scrape being analysed contains data from 8 million users – a higher number than most people have assigned to Twitter. And from that figures comes the figure that tinyurl provides 75.09% of shortened links.

The next is is.gd with 7.67% and my own favoured choice of bit.ly in third with 4.84%.

Cli.gs itself is in a creditable 10th, with 0.35%

From all links, Twitpic is the third most popular, with blip.fm and brightkite also in the top ten.

Go and see some more interesting information at the original post on the Cli.gs blog.

And there’s a huge list of shortening services on the Microblogging Tools page.

While I was offline and on holiday, I started thinking about permanence and the ‘half-life’ of content and ideas. Is print a better medium for archiving content, in the event all electronic systems are turned off or destroyed in a catastrophe? Or is the electronic medium far better for simply getting an idea into the world, where it may take on a life of it’s own?

Originally it was a comparison between print and blogging, but reading some recent posts from bloggers listing referrals, I noticed how many were crediting large numbers of referrals from Twitter as a direct result of friends and followers retweeting their messages.

That struck a chord, as I tend to be online between 9am and midnight (sometimes 1 or 2am) so I get to see various timezones conversing – and I wonder how much I miss when I’m sleeping or avoiding microblogging to concentrate on other tasks. Unless I set up RSS feeds for every possible keyword, and constantly trawl through archives, there’s always going to be a huge amount of content that is published when I’m not there to see it – and the same it true for almost everyone using microblogging.

Which is why microblogging is probably the best test of a concept or an idea. Not only do you have to fit the pitch into 140 characters, but for it to reach a significant number of relevant people around the globe it needs to be retweeted by a significant number – easy if you have thousands of followers, but unless you’re in the top few Tweeters, for example, it’s purely down to the strength of your idea.

That’s much different to blogging, as tools like RSS are effective at archiving days or weeks of content which can be scanned fairly quickly and efficiently. And it’s obviously different to the print mechanisms of old.

It’s another benefit of the microformat over the longform for getting quicker feedback and response on an idea – although I’d hesitate to drop an entire simply down to a lack of responses – simple requests for messages I’ve posted have had a response rate of around 8% of my followers within an hour or so, which is pretty damn impressive, but if the idea is relevant to the other 90% it might take a while to reach them!

Note: Thinking about it after finishing this post, there may be the subconcious influence of this post by Dave Cushman, regarding responses on Twitter and Focus Groups vs Communities of Purpose.