Two and a half years ago or so, I cobbled together a Yahoo Pipe that created a caption/subtitle feed file from a hashtag based search on Twitter. The idea was that if folk were using a tag to tweet around a video event, such as a conference video feed or TV broadcast, and a recording of the same video feed was then uploaded to Youtube, you could watch original “live” Twitter stream as captions on the original recording (Wikipedia: Twitter subtitling). Since then, I’ve pushed the thinking along a bit and Martin Hawksey has pushed the code along, demonstrating among other things how to create twitter based iPlayer captions (Martin: maybe we need to tell ‘the BBC’s “King of subtitles”, @amcp [h/t @nevali]?! ;-) and “as-if live” tweet capture to fold back twitter updates from a viewing of the replay into the original social stream.
Now it seems as if there’s a new app on the block, that whilst not directed directly at the TV replay market, cerainly lists it as a use case: Rewinder [via Follow tweets about recorded TV shows as if you were watching live, with Rewinder, h/t Laura James]. The idea appears to be that you can (I wonder if this service is built on top of the (metered) Datasift?): watch your timeline or a hashtag as it happened in realtime!
When Martin and I were bouncing round ideas for Twitter subtitles, we were both Twapperkeeper as the backend archiver, but following Twitter’s API licensing changes earlier this year, Martin has been taking the Google hacker approach and looking at all sorts of ways of using Google spreadsheets for archiving tweets. As well as running a hosted service, Twapperkeeper’s developer was (I believe) also funded by JISC to develop an open source version, reflecting JISC’s continuing interest in exploiting – and archiving – social media backchannels and community support, particularly around events and (increasingly) JISC projects. Andy Powell’s Summarizr is one service that demonstrates what we can start to glean from tag archives. (ThinkUp App is another open source app that allows you archive your social media activity.)
Although I only spotted it today, it appears that Hootsuite acquired Twapperkeeper a couple of days ago (congrats @jobrieniii). Although JISC managed to help get the open source YourTwapperKeeper code released, I wonder whether plans are in place surrounding any continued access to data in the Twapperkeeper archive following the sale, especially given the legal wranglings between Twitter and Twapperkeeper regarding API access to Twapperkeeper data earlier this year? (For the record, I think it’s great that Twapperkeeper got acquired, and wonder whether this might lead JISC to consider whether or not there may be opportunities in occasionally acting as an angel or venture funder? I don’t think this is entirely without precedent: wasn’t 3i originally UK government backed?)
Quite by chance, another TV’n’Twitter post passed through my feeds today: Trendrr launches social media tools for TV networks. I guess the operators of social media dashboards that are based around Twitter are all getting twitchy (hmmm… Twitter, tweet, twitch… So what’s a twitch then? Reminds me of the tweverything days, when we twere all going to become tweachers…), and so I guess we may see a flurry of sector specific dashboards appearing (I wonder if @briankelly is currently seeking out VC funding for an education industry social media monitoring spinoff, notwithstanding my recent mutterings about What’s the Point of Social Media Metrics??!;-)
Just in passing, if you’re interested in the sorts of Twitter based analysis that are possible, see: Engaging News Hungry Audiences Tweet by Tweet: An audience analysis of prominent mainstream media news accounts on Twitter, or this tutorial – Twitter Research Methods – from the folk at Mapping Online Publics (who also put together this First (Twitter) Map of Australia).
And finally, as a much as a note to self as anything else, here’s an interesting social network analysis walk through based on a Twitter data set: Combing Through the Infovis Twitter Network Hairball. I’ve just started playing with the dexy.it automated documentation framework, and this recipe looks like something it might be quite fun to try to automate…
PS related – see the comments on Joss Winn’s response to this post: Universities as Venture Capitalists. And quite by chance, via Paul Stainthorp, I notice a Research And Enterprise [Office] Merger at Lincoln. Which makes me wonder again about whether it’s Time for TechCrunch, Academic?, and whether there are any other UK HEIs out their with blogging enterprise and innovation offices? Hmm… If I wanted to find out the bit of the university that tried to make it money from spinouts, where would I look in the Linking You model (URI conventions for HE)?