Info

A trade journal of a still-emerging field, written by Adam Tinworth.

Posts from the Media140 Category

Andrew Lyons

Andrew Lyons – Ultraknowledge

The dataeconomy is about turning information into a usable asset – and an engaging experience. So, there’s a reason to develop new business models. Lyons invested £100 in a quiz at the beginning (the money was the rewards) – he might get contacts, a drink, anything out of it at the end – but he’s trying something new.

He’s showing off his Twitterwall product, which draws out user icons, tweets and stats. Which came up with a 404. Oops. OK – working again. Lots of data sifting about the people who have tweeted with the #media140 hashtag – can this be used to identify the most influential people at the events? The product can build a relationship wheel to show you who is connected to whom amongst a Twitter community – and its being demonstrated with my relationships… gulp

They also work with publishers. They have archives – but now they need to think about how to wrap it up in new visual stimulus. They can create news walls for every sub-section of a publisher’s website. They content’s already there – they’re just giving a new, enticing interface. Here’s an example. You can search visually for things you’re interested in. It’s more of a discovery engine than a search one, perhaps – it’s about finding things you don’t know you might be interested in.

David McCandless

David McCandless – Information is Beautiful

Oh, look. David McCandless. I’be blogged his talks once or twice recently. He’s kicking off with his billion dollar-a-gram. He scraped all the data for it from journalism sites, but he did it manually. Did it via searches on billion dollar amounts on sites. Drew the proportional shapes in Adobe Illustrator. And then played Tetris with them until the fitted together.

He sketches ideas for his visualisations, trying to match the style of the graphic to the feel of the data being conveyed – triangles are more conspiratorial, for instance…

Spreadsheets underlie a lot of his work – word processors are linear, while spreadsheets are multi-directional – and you can bring in live elements. They’re very tidy, and they can be the start of a great design. Visualisations go through many drafts, as he works on the details of layout with the person commissioning it. The designer on the time travel visualisation left after the 13th of 14th draft. McCandless then realised that time travel was a chaotic thing, and that a visualisation that embraced that would work – and allowed you to see crossover points between various time-travel movies… Sometimes the form of the visualisation only emerges during the work – you discover the important data as you go.

What doesn’t work? Too much spaghetti – not shaping the data with a story – just creates a mess. Circular diagrams often make little sense. Cartograms (based on geography) are over-used and the information is difficult to get out.

McCandless’s bif FAIL:

McCandless FAIL

Can you figure out why?

 

(Live-blogging – prone to error, omission and typos…)

opendata.jpg

Rufus Pollock from Open Knowledge Foundation

Wouldn’t it be nice to know where our money goes when we pay taxes?

You need lots of stuff – government spending (local and regional), region codes, company data… Much of it locked up or in difficult formats like PDF…

We need raw data and we want it now…

Open = freedom to use, re-use, redistribute – but it must be non-personal. More and more businesses are being built on data (cough Reed Business Information cough) Data is non-rivalrous – if I give you a copy of my data, I haven’t lost access to it (like my shoes or car). Information systems are the most complex systems we’ve ever built – and how do we deal with complexity? We break it down into bits. But if those bits are closed, it’s very hard to put it all back together again.

Why open? Other people may come up with the best ideas of how to use the data – and we move towards a read-write society, where we no longer just passively consume information.  Too much information is now locked up in closed systems like Google and Facebook? How do we change from that? How do we build an ecosystem around that outside these walled gardens?

Lots of back-end work needed…

Simon Rogers

Simon Rogers – The Guardian Datastore

Inspired by James Cameron – not the Avatar one – he was all about telling human stories and making things real, and Florence Nightingale, who made visualisations of troop casualties as well as nursing…

Guardian busy collecting and using data for journalism – why not share that? Lead to the setting up of the data blog. No longer journalists labouring in isolation, but others can pitch in and do things with the data.

Showing a bunch of examples – MPs Travel Expenses, the expenses crowd-sourcing – one person did 29k pages! Interest rapidly declined after first day. Sometimes too much data can swamp the story. Coins Data Explorer – built both for journalists and others to use.

Then he showed us an Excel spreadsheet of Afghanistan data, and showed how it could be used via a pivot table to feed a visualisation. It’s probably quite straightforward, if you have the first clue about Excel, which I don’t. 🙂

Oh, and if you really want to get attention – do a Doctor Who visualisation

media140 Meetup
And so to the first media140 meetup, over in the City. Unusually for this sort of event, I’ve only recognised two faces (oh, three) here so far.
It’s a grand total of eight months since I went to my first media140 event, and the mood and feel has changed completely. There’s been a progressive move from the social media enthusiasts and evangelists to those who are trying to figure out how to integrate it into their businesses – and that’s great. That’s social media growing up.
But I am feel rather that the questions and answers I’m hearing aren’t really adding to my sum of knowledge. In fact, there are things being said that I’d say were objectively wrong. “homogonised output on Twitter”? Uh, no. Luckily some voices are being raised in protest. But this is social media as a tool applied to existing processes rather than as a thought process, as a means of conversation. 
Charlie Osmond from Fresh Networks is being provocative by suggesting that the social media mantra of “go where the conversation is” is wrong. But his argument rather relies on the idea that the only conversation worth having is the marketing one. He gave the example of a vicar, who does his evangelism from the pulpit, but also goes to village fetes, etc. And that trying to evangelise at the fete would be a poor idea. And he’s right in that. However, just trying evangelising from the pulpit if you haven’t built a relationship with your parishioners by going and chatting at the fete…
He’s a social media pragmatist rather than a social media purist, he suggests. I might ge tempted to insert the word “cynic” in place of “pragmatist”. Certainly one could argue the methods he are advocating are cynical rather than genuine engagement…  🙂
Ooh, best idea of his bit: the last thing you want to do when launching a community is just open the doors. If you just open the doors of a restaurant, and it’s empty on a Friday night, you have a problem. And that’s bang on. And that brings us back to the vicar at the fete, talking to folks so they will come back to his sermons. 
And he’s predicting a social media backlash. Which is pretty safe, as it’s already started. 🙂 And, yes, I think he was being deliberately provocative up front, as he is talking what I would consider the talk, with sensible words about the fact it’s going to take at least a year to build a community, and get genuine engagement.
And with that, I’m sneaking off home.
Nice cocktails from Ping Pong, though…

Hemlock is an open-source environment for building social applications, with a gaming focus. There’s a showcase of uses on the site.

AudioBoo – a low coast, efficient way of generating audio reports. The platform collects pictures, geodata and so on at the time of recording, and pushes it into a social, connected, embeddable environment. Currently, it’s on the iPhone, with Android and web versions coming soon. And they’re adding an API which will allow others to build clients. 
Podcasting was too complicated, suggests SEO MArk Rock. The idea of AudioBoo was to make audio publishing as easy as possible. In the short term, they’ll monetise through pro services, from editing, to managing team contributions. 10 licences sold for the pro services – and it’s not built yet!

David McCandlessDavid McCandless (@mccandelish) has a book coming out. He didn’t beat around the bush about that. He showed us the UK and US editions. But he’s also interested in the growing visual literacy of the population through the growth of the web. 

If you come into the social space trying to sell something “you’re a bit of a dick, really”. What you can do is bring a gift – interestingness. If you’re interesting, you’ll get followed and you’ll get retweeted. And the most interesting gift is the revealing gift. 
(This is a really hard session to liveblog, BTW. It’s all about great, fascinating graphics.)
And I think what he’s showing, and this should be of particular note to journalists, is how limited and uninspired our graphical representation of data is right now, and how illuminating it can be when done right. Asking the real-time web the right questions, and you can find some fascinating information – but give back. Give back into the social space what you create from that data.
Wonderful talk. Go read his blog.

Last keynote of the day: Bernard Desarnauts of Glam Media and Tinker.com talking (in theory) about where the money is.
Media is every-fragmenting. Blogs are growing as are social networks. Traffic is fragmenting around topics, and traditional sites are drawing a small and smaller amount of people’s time. The traditional places where advertisers spent money are becoming less and less relevant. 
So, the answer is distributed media – go where the audience is. And that’s brought them 12m uniques worldwide – by focusing on mid to long tail.
Tinker – instead of following people, you follow topics. They use semantic analysis to find relevant content wherever it is. Beyond that, they create widgets that contain the conversation within them. You don’t force people to come to Tinker – you can push the conversation out to where they are. 
Has been used by many media brands. They use algorithmic and manual filtering to curate the flow of content. They have direct API access to the Twitter “firehose”. Most customers only want the Twitter data. 
[This feels a bit like a sales pitch rather than a case study]

Gordon suggests that there’s a time when you have to stop listening and start doing. People expect a quick response to problems, just because people can communicate them. Will suggested that the Twitter behaviour was chasing down towards the playground in the previous session
Drew pointed out that companies have people dealing with e-mail and letters – why should Twitter get special treatment? [Answer: not given – it’s a public channel, so that problem, if unsolved damages your reputation]
I take my last comment back – Candace just made that very point.
Will asks if we will have to deal with a world where all our employees have accounts and voices? [Yes]
Richard points out that you need a corporate account, because, if people have a problem, they don’t search for your name, they search for the brand. And that’s true, but Drew also compellingly argued that people will quite happily discuss a brand elsewhere, and companies need to accept that, and that they no longer have total control over discussion of their reputation.
Is there a danger of your employees misrepresenting your brand? Will: “Yes, but they do that every day anyway.”
Drew suggested that the way HR should approach this is just suggest that you behave as you would at a work party. Will followed up to point out that there is an HR / internal comms angle to this, and it is important that social media is the responsibility of more than one department…
Is people taking their social graph with them when the leave a problem? Candace recommends the “@” approach – [email protected], for example. [This raises the issue of separate personal and business twitter accounts, which I guess won’t get discussed, as it’s all over]

Jess GreenwoodI’m uncomfortable with some of this panel’s message. There seems to be a distinct undertow of “how do we control the message through Twitter. The examples we’re being given seem to be using Twitter as just another channel, maybe to give a veneer of engagement, without a reality to it.

ASOS seems to be the notable exception right now, but let’s see how the panel develops.
Amelia Torode from VCCP suggests that it’s about the right people doing the right things…
And this panel has almost been completely hijacked by the Twitterfeed… Running jokes about the guy from Innocent, Ted Hunt.
Some suggestion that Twitter is a passing fad, and it will be replaced by something else next year, but that feels like someone who isn’t really engaged.
Look, the  real lesson of this panel was that, if a panel is insufficiently interesting, using a Twitterwall will sabotage it. Fascinated people don’t undermine sessions with jokey tweets, they tweet reaction to people’s statements – or just tweet the statements. Despite the best efforts of chair Jess Greenwood of Contagious magazine, this is actually a classic counter to the sentiment that I saw emerging at the start of the panel – about controlling the message. With so many people able to publish now, you can’t control the message – you can just join in.

Ciarán NorrisFormer colleague Ciarán Norris is going to tell us how to listen.

Use your ears?
🙂
First up: why should we listen on Twitter? Exponential growth in the last year, for one. We’ve had a growth of people at the creator end of the scale. First you had know HTML, then you had to have a blog, and now you just need Twitter, a mobile phone and maybe a cameras. 
And because of that, Google is now your company front page, and other people’s work maybe on the front page and more interesting than yours. Tools like ViralTracker allow them to follow how pieces of work spread across the web. Delicious is still a good way of seeing how people view your brand. Tweetdeck gives you the ability to understand and monitor Twitter across multiple accounts. Tweetfunnel allows multiple people to manage one account. Tweetmeme tracks how links are spread around Twitter. They’re adding analytics (which I’m testing and it’s cool). Twitterfall, great way to check what people are saying around a brand.
But: some perspective. 2% of Iranians have access to Twitter, so it’s the “liberal intelligentsia talking to the liberal intelligentsia in the west”.
Some good examples: