A year ago I wrote a post entitled Hashtags for the ALT-C 2009 Conference in which I suggested use of a session hashtag in addition to the event’s hashtag in order to be able to differentiate tweets related to the numerous parallel session which were taking place at the conference. My suggestion for minting the session hashtag was simply to use the session’s code which was listed on the conference Web site – #s321 for the session I ran, for example.

It would be an understatement to say that, at the time, this suggestion didn’t receive a favourable response, with the following comments being made:

  • Sorry Brian, but I do think this scheme is too complicated for the lightweight Twitter approach“;
  • I really think this is trying to make Twitter something it isn’t. The very thing that people appreciate about Twitter is its lightweight nature and this is simply over complicating things“;
  • When you first started suggesting multiple hashtags, I think I assumed it was a bit of a comedy experiment. Now, it’s becoming clear that The Librarian Is Too Strong In You.“;
  • Way too complicated, messy, and just so damn cluttered“;
  • I’m in agreement with those that suggest this is over-complicating things – mainly because I struggle to see the problem it’s solving“;
  • Sorry Brian, I’m with the others here. Twitter is for catching the ‘buzz’”.

There were six negative comments with only one supporting, although in a somewhat lukewarm fashion, my suggestion:

In the past I’ve generally argued against multiple hashtags – agreeing with the comment that they introduce complexity. However, given the size of ALT-C, and the number of concurrent sessions, I have some sympathy with the issue that Brian raise

A follow-up post on “I Want To Use Twitter For My Conference” provided suggestions on use of Twitter to support events but avoided mentioning use of session hashtags. Chris Gutteridge, however, made a suggestion in a comment to the post: “At Dev8D2010, at the end of February, I plan an experiment of assigning each location a hashtag, then publishing an electronic form of the schedule so the twitter can be merged into each session via location+program data.” Chris also pointed out that use of session hashtag at IWMW 2009 “fell apart in small sessions in IWMW because nobody advertised them and people didn’t care enough to go to a webpage to check.

The suggestion that session hashtags could be processed by software was interesting. I also agreed with Chris’s implied suggestion that there was a need to promote session hashtags more effectively.

Summarizr statistics for hashtags used at IWMW 2010 eventSo at this year’s IWMW 2010 event we used a session tag more consistently throughout the event (#P1-#P9 for the plenary talks and #A1-A9 and #B1-B9 for the parallel sessions) and ensured that the chairs of the plenary talks encouraged participants to use the session tags when tweeting.

Did this work?  The Summarizr statistics for the #IWMW10 event hashtag provides details of the top 10 tweeted hashtags, as illustrated.  This indicates that the most widely discussed session was session #P8 – the group session on Doing the Day Job.  Whilst it is true that the session lasted longer than the other plenary talks (it consisted of three plenary talks) it is also true that this session included a rather controversial talk which generated much discussion on the Twitter back channel.  Looking at the usage of the other session hashtags we can see that Paul Boag’s talk on “No money? No matter – Improve your website with next to no cash”  – this does not surprise me as Paul’s talk was widely acknowledged to be the most inspirational and did generate much discussion after the event as well as on the Twitter backchannel.

It was also interesting to observe that #remote hashtag which was also widely used. We had previously stated that we would “treat the remote audience as first class citizens” and use of that hashtag seemed to be effective in communicating with those who were watching the live video stream remotely.

The session hashtag can also enable tweets about a particular talk or session to be further analysed. Although the comment had been made that the  “obsession with tracking, capturing and archiving everything to the nth degree just doesn’t fit with Twitter” in reality we are now seeing that a strength of Twitter lies not just in “catching the ‘buzz’” but also in the  interoperability the service provides. A good example of this is the way in which Martin Hawksey’s iTitle Twitter captioning service combines a twitter stream with a video of the talks.  And whilst this particular example is meant to illustrate how tweets can be reused, and is not  specifically related (currently!) to session hashtags (the tweets are integrated using a timestamp rather than a session hashtag) I am still convinced of the benefits of this lightweight approach to disambiguating tweets at large events.  A session tag was useful, for example, in my final conclusions about the session. When I gave my thoughts on the Doing the Day Job session, for example, the Twitter community could exploit the simplicity of the #P8 hashtag rather than attempting to coin a hashtag based on the title of the session or the speakers’ names.

Is the case for use of session hashtags at large conferences proven? After all if beer drinkers can make use of the beerspotr syntax which ranges from:

@beerspotr pint:x  (if you’ve spotted a pint)

through to:

@beerspotr bottle:x y%  (where y is the ABV of the beer)

@beerspotr bottle:x pub:y  (where y is the name of the pub)

I´m sure sober Web techies are capable of using two tags with no additional syntax required!