August 2014

Traffic Island Disks

A radio show about music, people and spaces.


We walked the streets looking for people listening to headphones then talked to them about the music, their day, and the location while recording whatever they happened to be listening to. The result was a show that offered a glimpse into people’s soundtracks to the city, and how they used them to move through their day.

Host Mikey and postman Kwame in Brixton

Traffic Island Disks started out with a regular slot on London’s Resonance FM station in 2003, then grew into an internationally touring live Internet radio show. In 2004 we designed a mobile system with custom software for live mixing, editing and streaming, and teamed up with local presenters to stream the show via wifi connection from the streets of cities starting in Newcastle then touring to Bratislava, Vienna, Sofia, Aarhus and San Jose.

Credits and Links

Traffic Island Disks Read More »


Enables large groups of passers-by to engage in ongoing spontaneous conversations.

Heckle at the National Theatre
Heckle used in the foyer of the National Theatre

Heckle augments conversations by letting people catch up with what is being talked about visually – providing a user-contributed flow of relevant texts, tweets images, videos, websites overlaid on a live video feed of the event.

Developed with The People Speak to augment their public conversational performances, Heckle lets participants use a web app to contribute to a live visual summary of the conversation so far, so that new people can get involved at any time in the course of a discussion.

Heckle also provides a post-event visual timeline of the conversation, so over the last 7 years, The People Speak have been building a searchable, participant-annotated video archive of spontaneous conversations.

A post-event visual summary
A post-event visual summary
Schematic of a Heckle set-up
Schematic of a Heckle set-up

Credits & Links

Heckle Read More »

Heckling at Ontologies

Heckling at Ontologies explored the relationship between TV-viewers’ text chat and TV broadcaster media metadata.

Heckling at Ontologies Research Poster
Heckling at Ontologies Research Poster

Heckling at Ontologies is a collaboration with fellow MAT student Toby Harris exploring the dissonance between top-down and bottom-up approaches to media metadata generation. The project started with two datasets:

1. A semantic description of Season 4, Episode 1 of Doctor Who

This was one of the outcomes of Toby’s industrial placement at the BBC, which developed an ontology for rich description of BBC programme content and then used it to create a highly detailed summary of a specific episode of Doctor Who.

2. User-generated tweets ‘heckled’ at the screen by viewers

As part of an industrial placement at BT, I worked with The People Speak to develop a new version of Heckle, a web service that gathers tweets, images, videos and texts ‘heckled’ to a shared screen by viewers of a TV show, building up a visual summary of the mediated conversation. These two data sets were analysed to discover the dissonances and regularities between Toby’s top-down and my bottom-up descriptions of the same TV show.

The two data sets were then superimposed on a live stream of the video. You can download the source of the demo and grab copies of semantic data on the ontoheckle github page.

Links & Credits


Heckling at Ontologies Read More »

Conversational Annotation

Conversational Annotation explored the interactional dynamics and potential of ‘Social TV’ audiences. 

Conversational Annotation research poster
Conversational Annotation research poster

As part of an industrial placement with BT I worked with The People Speak to develop Heckle, a twitter-like web-service that captures the comments, asides, and discussion generated by an audience to annotate video content.

Analysis of of these mediated conversations revealed the degree to which people’s comments related to the TV content (characters, plot developments and production entities), and to what extent they related to the organisation of activity in the local context (crisps, drinks, sofa arrangements etc.), and looked for differences in frequency, peaks and other pragmatic detail evident in the data indicating to what extent this could function as media metadata.

A closer qualitative approach assessed the its amenability of this data to conversation analysis, looking for evidence of turn-taking sequences, and various forms of conversational repair.  ‘Conversationality’ in these empirical terms interrogated the technology and situations of ‘Social TV’, providing a critical view of the over-applied term ‘social media’ often used to refer to these forms of mediated communication.

There’s a working copy of the Msc thesis based on this research available here.


Links & Credits

  • M.Sc. project co-supervised by Pat Healey at Queen Mary University of London and Andy Gower at BT.
  • Thanks to all the participants in the Dr. Who screenings, Richard Kelly, Toby Harris, and The People Speak.
  • This research was funded by the Digital Economy programme through the Media and Arts Technology Doctoral Training Centre at Queen Mary, University of London

Conversational Annotation Read More »

The Dicshunary

The Dicshunary collected and shared vocabularies used by one person, family, or other micro-language group.

Screenshot from the Dicshunary

Users added their neologisms, definitions and redefinitions, and could create sub-lexicons to host on their own websites. They could even download the code and run their own version.

The dicshunary installed on trash technologyInstalled as a kiosk on recycled computers reclaimed from local skips, dumps and store cupboards, the Dicshunary would capture local place names and micro-dialects from areas near the galleries, schools and public spaces in which it was exhibited.


Credits & Links

The Dicshunary Read More »