annotatAR lo-fi augmented reality for the mobile web

Week 9 Delinquency

Overthinking and under-doing in October. Next steps.

How I spent October

I’ve had a very fruitful October. Unfortunately, this activity has not been focused on my own projects and thus annotatAR has been left on the wayside for far too long.

While I’ve been a bit delinquent in updating both this project blog and working on the project itself, I have accomplished much rumination on the final outcome for this project.

I somewhat switched focus at the beginning of October, and began to think of the project as a sort of meta-documentation tool whose primary end-user would be me. This focus hasn’t sat well with me though, as the spirit of the project has been to build experientially what has already been ongoing virtually.

Many questions remain to be answered:

  • Who is the audience for the mobile site?
  • In what context will they discover it?
  • Will the site display current or historic tweets?
  • What controls will the end user have over the tweets being display?
  • Will the website generate a database of tweet data?
  • Will multiple hashtags be available at a particular location?
  • Will the end user be able to change tweets?
  • Will there be an administrative site where an admin can set up a new database entry for a geolocation / hashtag pairing?

I will attempt to clarify my thinking on these topics in this post.

Audience

My current thinking on the purpose of annotatAR is to fully embrace the art project aspect. Rather than thinking that this project is a tool to make other works, I would like to present it as a work in itself, a generative social media artwork.

In this case, the audience / end user becomes those interested in radical use of digital spaces, Twitter and social media, and contemporary activist movements.

The limitations of this direction are that the primary purpose is no longer documentation of an event in process. The Twitter stream API will likely not be useful - it will be necessary to utilize historic tweet data, such as has been gathered by Occupy Research.

Context

The working context for annotatAR will be a digital exhibition of some kind, where the audience finds out about the work through some third party - much like the Invisible Monument project. The artwork will be available for a specified timeframe during which someone can go to the appropriate location and experience the augmented reality app on their Android mobile device.

This work might be highlighted in a launch event or tour of some kind. Once annotatAR has a first version, I can begin to look at fellowships and opportunities to take the project further and share it with communities.

Stream vs REST vs Historic

The Twitter API has two versions: the stream which affords access to real-time data as it is posted - and the REST version that allows queries based on various parameters.

For the conceptual purposes of this project, the stream is ideal, as it expresses the contemporality of dual sites: online and in the “juice” of physical space. However, the stream depends highly the volume of tweets being generated in a given timeframe; if the event is generating a low level of interest on Twitter, than annotatAR will have sparse data to work with and be less effective as a visual presentation.

My deployment at the Invisible Monument launch utilized a blend of REST and stream in order to overcome the limitations of the stream API.

As the purpose of annotatAR has solidified around an experiential artwork, it becomes imperative to use Twitter API in a way that best expresses concept of the piece. The idea of synesthesia becomes an important concept - as we walk through the juice, the ideas and social sites of Twitter are invisible but present. annotatAR interacts with the visual sense to encode the virtual component of a multi-sited co-occurring event.

The use of historic data, where it exists, affords a slightly different expression of this co-occurrence. Instead of expressing contemporality, the visual metaphor would be more archeological in tone - that of unearthing and re-experiencing an event in the past, like an historical marker expressed in virtuality.

Hashtag / Geolocation

Should multiple hashtags be associated with a specific geolocation?

So doing might dilute the strength of the connection between a juice space and a twitter hashtag settlement - but it would also afford a more nuanced understanding of historicity of a space.

End user controls

Given the above discussion, the end user will have some minimal controls - to select a hashtag (if there are multiple available at the geolocation), to pan around, to take a screenshot, and to scroll through time.

Desktop site

The desktop site will be quite minimal - a brief description of the project and map of the geolocations available. A sign-up for newsletter of project updates.

Questions

What historic tweet data is available? - this needs to be answered ASAP to properly scope this and future versions.

Future directions:

  • Funding for a tweet aggregation service such as GNIP
  • Development of a robust administrative layer
  • Development of a way to share screen captures
  • Integration with the Invisible Monument project (iOS)
  • Development of a robust data collection system to gather data on contemporary events

More immediate goals for annotatAR (still) include:

  • Bug fixing - hashtags on the stream seemed not to work on the website (though the stream worked on the local version)
  • Fluid accelerometer use
  • URL routing for hashtag display
  • Automation or parameterization of annotatAR UX elements (ie age til die-off)
  • Pulling in historic tweets from Occupy Research (or other source)
  • Single-page desktop site