Introduction to Argon

Summary of “The Argon AR Web Browser and Standards-based AR Application Environment” by Blair MacIntyre, et al.

This paper from ISMAR 2011 is an introduction and explanation of Argon and the related underlying and supporting technologies. Argon is an ongoing project to implement AR exploiting the ubiquity of mobile phones (specifically, at least for now, iOS devices from Apple) and widely accepted web standards.

Philosophically, Argon is the spearpoint for the goal of mobile AR for everyone. Politely follow standards, prove that the web can work for AR, adapt from existing standards, invite multiple streams to coexist (how democratic), and get past the greedy idea that AR will only work in a high performance single POV paradigm to be realistic or useful. The way in which Argon lives within and on top of web technologies means that work can be done faster and with familiar tools while exploring a space of applications which is currently underexplored (there’s still no killer app yet).

Schematically, Argon is something like a framework, embodied in code as a browser. It hosts multiple isolated layers of AR content in the form of “channels” which are essentially each a session of web based communication.

Argon is motivated by a desire to have a unified environment that enables multiple applications to coexist in one reference point. The goal is to provide a layer for abstraction where developers can leverage their talent and experience with well known programming languages, libraries, and protocols to enable rapid and effective authoring.

Stated, the three major goals are (1) “develop a ‘window system’ of AR” displaying independent programmed containers. (2) Build on mobile web tech to exploit the power of web author expertise, deployment platforms, services, content, and the ubiquity of mobile web devices. And (3) create an AR ecosystem that lives in concert with existing web standards as they are discovered, implemented, and adapted to changing tech.

The paper goes on to explore and explain some of the specifics of the implementation. In particular, the authors demonstrate how some limitations of KML motivate additions in KARML. Offered is anecdotal evidence of time and effort savings gained in extending off-the-shelf web server and development infrastructure by designers and experienced programmers alike.

As a criticism, I had expected to see a more detailed discussion of future work and sometimes wonder why this isn’t a bigger part of a paper on an ongoing software framework project. Overall, the paper increased my level of interest in Argon and especially in Argon2.

Comments are closed.