Tubestrike Crowdsourcing Experiment

I have been working with BBC London for the past few weeks, helping to support them with some new social media initiatives. (It should be noted they already have a strong foundation – see @BBCTravelAlert as a great example of an engaged twitterfeed). A couple of weeks ago when the London Tube strike was announced I thought it would be interesting to try out the new Ushahidi crowdmap platform. It was launched about a month ago and is an attempt to do for crowdsourced mapping what wordpress has done to blogging….basically make it foolproof.

I had seen TBD’s map in DC, and thought it would be a perfect way of easily visualising a lot of information about the strike. We’re now one hour into the strike and I’m writing this while checking the Transport for London website and updating the crowdmap with new information. So far the only updates we’ve posted have been based on official data. Hopefully tomorrow, people will start sending us their own reports via text, twitter, audioboo and the webform on the site.

The hope is that the map will become the host for photos, audio clips and commuter experiences rather than simply parroting the official information, already available on the Transport for London site. That will depend on whether the map takes off and people want to help us.

I hope so. Crowdmap is a great site, easy to use, and aware of the all the issues of verification and information management which could make this type of journalism a minefield.

Hopefully you’ll take a look at our map even if you don’t live in London and aren’t affected by the Tube strike, and will see what a great resource this can be for anyone interested in collaborative journalism projects.

7 thoughts on “Tubestrike Crowdsourcing Experiment

  1. Interesting idea. Does the service/platform let you define data attributes for reports so you can enrich it with e.g. Mode (bus, tube, walking) and controlled locations (Liverpool station??) and maybe status “avoid, recommend, chaotic”? That would make it much more powerful as a resource to help commuters (in this instance)

  2. HI Rob
    Sorry for delay in responding – yesterday was absolutely full on! I’m going to blog about it today, but your points are really good. The crowdmap platform was only launched a month ago and so far only us and have been deployed. We will feeding back our experiences about what worked and what could be improved.

    You could search by category and we had trains, tubes, roads, buses. During the day we added river and then station closures as we knew that was a big issue. Each category was a different colour so you could just search for station closures which were orange dots.

    The status element is a really interesting idea as some people were giving us positive information, like ‘there’s a rack full of Boris Bikes at Euston’ but it felt bad giving it a red dot when it was a good thing or suggestion.

    The other issue is having better information right on the screen rather than having to click through to a new screen.

    But considering it was the first time we or the audience were using the platform, we were really pleased.

  3. Claire,
    if it helps we might be able to determine whether our ability to scan posts and provide instant categories and summaries of content may be helpful to you. We employ semantic algorithms and simple maps with our internet marketing software.

    The supplier of our main server component has developed the technology over the last 10 years in Canada and his Chief Scientist is English. We would be able to understand each other, possibly without too much difficulty!

    We employ two concepts that contribute to the building of Resonance maps, the concept of a employing behavioural perspectives, we call them Neuropersonae to a ‘Story Lens’ which determinse the facet of the perspectives we employ for our maps. Google (with quotes) “Story Lens, SpeedSynch, Twitter Resonance”.

    Resonance as a concept is fairly easy to understand and unless content and people ‘resonate’ closely the engagement may not be sufficient to create value for publisher or reader alike.

    Exemplars may be found at @ResonantView or @SpeedSynch on Twitter and for those who must know the ‘secret of the semantic algo’ Google (with quotes included) “Any words used in similar spaces share meaning.”

    I suspect this approach would save you considerable stress, alert you to patterns quickly and allow you to adjust categories on the fly according to the needs of people and content alike.


Comments are closed.