The Science Online 2009 Conference
On Saturday 22nd August 2009 I attended the Science Online 2009 Conference which was held at the Royal Institution of Great Britain, London. The conference followed on from last year’s event, which had the title Science Blogging 2008, but had a broader remit addressing issues such as “What is a scientific paper?”, “Author identity – Creating a new kind of reputation online”, “Real-time statistics in science” and “Google Wave: Just another ripple or science communication tsunami?” as well as blog-related talks such as “Blogging for impact” and “Legal and Ethical Aspects of Science Blogging “.
With an audience of experienced scientific bloggers present it is only to be expected that, just a few days after the conference, blog posts about the conference has already been published. So rather than repeat what has already been said, I will link to one blog post which provides links to a number of posts already published: Thoughts on the Science Online London Conference.
The Event Back Channels
This post mentions the Twitter hashtag for the event (#solo09) and provides links to the FriendFeed Science Online London group (which used the tag solondon for the Friendfeed room) and also the Flick group for the event (which used the tag solo09).
For an event aimed at scientists which focussed on innovative online technologies (as well as a talk on Google Wave several of the talks were also available in Second Life) and which discussed the implications of the online environment on traditional views on scientific papers and mechanisms for measuring the impact of scientific research in this environment it was perhaps surprising that there wasn’t more discussions of ways of preserving the online discussions associated with the conference itself which took place on Twitter, FriendFeed and Second Life, in addition to blog posts, some of which were published during the conference itself.
Preserving the Back Channel Discussion
Now although links have been provided to Twitter searches for “solo09″ I suspect the short lifespan of Twitter searches may not be well-known. Following my recent blog post containing links to the Twitter channel for UKOLN’s IWMW 2009 event I subsequently discovered that tweets disappear from Twitter’s search index in a short period of time: as reported in a recent TechCrunch article “According to Twitter’s search documentation, the current date limit on the search index is “around 1.5 weeks but is dynamic and subject to shrink as the number of tweets per day continues to grow.“”
Further information on these experiences has been published. As I have described many (but not all) of the tweets associated with the event were stored locally using the Backupmytweets, Twapperkeeper, and WTHashtag services in order to avoid the dependencies on the Twitter service.
Testing one of these services with the #solo09 hashtag I find that currently Twapperkeeper finds 1,1472 tweets for the “#solo09″ tag. This service also provides a graph of the numbers of tweets which is illustrated. To summarise in the last 7 days there have been:
- 1,435 tweets
- 193 contributors
- 205.0 tweets per day
- 29.8% come from “The Top 10″
- 16.0% are retweets
- 43.0% are mentions
- 5.7% have multiple hashtags
In addition the top contributors were:
- @kejames – 65
- @rpg7twit – 55
- @kjhaxton – 51
- @Allochthonous – 50
- @skyponderer – 41
- @brian_condon – 38
- @morphosaurus – 36
- @PaoloViscardi – 34
- @phnk – 29
- @allysonlister – 29
We might expect the science community to have a particular interest in citations and the online science community and the early adopters to have a particular interest in citations related to new collaborative and communications techn0logies such as Twitter, FriendFeed and Second Life.
In previous discussions on this topics there has been a view expressed by some that Twitter should be regarded as a transient form of communication and the loss of data should be regarded as one of Twitter’s strengths. And this might be particularly relevant when the communications relates to trivial issues or issues which time out quickly. Examples of both instances took place at Science Online 2009: the unaswered iPhone caused lots of people to complain on the various channels and the updates on the Ashes test scores are no longer relevant.
The experiences of Science Online 2009 do, however, underscore an additional challenge: the diversity of the back channels. In addition to the Twitter channel, the science community has been an adopted of FriendFeed and this was popular at the event. Discussions were also taking place on Second Life. As well as the different applications being used there were multiple variants of the event tags: ‘#solo09′ on Twitter, ‘solo09′ on Flickr and ‘solondon’ for the name of the FriendFeed room.
The conference was also faced with the question of how to display the back channel. At one point a Twitter Search screen was displayed alongside the FriendFeed display. However since the Twitter display required manual refreshing to display new tweets this was replaced by the Twitterfall software during one of the presentations. Unfortunately this last minute adjustments meant that the text on the screen display wasn’t large enough to be read comfortably by many in the audience.
Where does this leave us? I would hope that the experiences of Science Online 2009, IWMW 2009, etc. and the subsequent sharing and discussions of experiences will help to inform approaches and best practices for future amplified events. And as suggested in a recent blog post on Feral Event Data: Twitter at IWMW 2009 aren’t the benefits of preserving the (reusable) data associated with live blogging at events particularly relevant for the research community? Tony Hirst has recently given his Preliminary Thoughts on Visualising the OpenEd09 Twitter Network. And he has started “thinking about how we might start to analyse the structure of the network around the hashtag, in part so we can understand information flow through that part of the open education network better“.
Tony has written a follow-up post giving “A Quick Peek at the IWMW2009 Twitter Network” which shows who cited who on Twitter during the IWMW 2009 event. This might give rise to some interesting questions. But might the interesting observations which can be made about the IWMW 2009 event (an event aimed at Web practitioners) be of more relevance in a research context? Perhaps not -but you can only ask the questions and carry out this type of analysis if you have the data. So if there is anyone who wishes to mine the ‘#solo09′ Twitter data I hope the data I have captured is useful.