Chris Cameron of RWW asks Open or Closed: What’s the Best Path for Mobile Augmented Reality?
The best path is open, no question about it. This “browser” model is the only way an AR app like Layar (or Wikitude) can survive beyond the gimmicky “app” life cycle. Just ask yourself: do I want to install the gazillionth “metro maps” app when visiting Paris, or just type “metro” in my standard AR browser and get the results (remember, the thing is location aware)?
So.
Going further, I envision a future for mobile AR where geo-tagged data points will be browsed in a generic way. By that time, the geo browser or whatever you will call it becomes a basic functionality of any mobile device, likely pre-installed and hopefully also downloadable from several vendors.
How Open is Open?
The currently available AR browsers (Wikitude, Layar) both have an open model – what they call the API, but in essence it is a formatting definition of a POI list, just like html is used for standard web browsers. At the same time both a closed with regards to their catalog and discovery model, in other words, AR data can only be consumed through either browser once a source has been explicitly registered and thus becomes part of their catalog. This make sense for now, as the market is very young and there are no other mechanisms for discovery or established yet. Remember when the www had only a few thousand sites, catalog services like Yahoo! made sense, they became obsolete after the number of web pages became too huge to be cataloged and search engines started to become the new way for discovering relevant results (and so Yahoo! transformed itself into a destination site including a search engine).
Another key aspect of the www is that there’s no need to explicitly add your content to some kind of catalog, just put it up and it will be found by aggregators (note: this is an evolving goal, we’re not there yet and have to do with SEO for the good or bad for a while – Linked Open Data is one of the answers here).
Following this analogy, this is what we need for AR to really become a mature way to “browse the world”:
- better UX for the mobile hardware (eventually glasses and lenses)
- open standard for AR data publishing
- open standard for AR data queries
- standardized AR browsers / clients
The path to real Open AR
The good news is that there are a lot of lessons learned from the regular www to be applied and a lot of existing open specifications and protocols which can be used. To mention a few:
- publish using linked open data standards (semantic web, e.g. RDF – at least for metadata to support discoverability)
- support a real time experience from the start (to avoid kludges like Twitter afterwards)
- support social graphs from the start (AR is about you and your social network too)
- use open standards for the transfer protocol, XMPP makes a lot of sense here
Two months ago, Tish Shute presented at the Mobile Monday conference in Amsterdam and introduced the AR Wave initiative to me. This looks like a very strong contender to evolve into the open AR web. If you’re interested in this matter you really should read up on this initiative!
Oh, and for now and the next few years we should really be very happy with our catalog based AR browsers, Layar and Wikitude are paving the road as innovator and evangelizer for a whole new industry. You can be sure they will keep stretching the experience to the limits as currently imposed by the handsets and immature technology.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=cc93dc27-c7b9-4ce3-8dea-4b4ede398d2b)
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=44d7fd32-da0d-4f97-afbc-072829909d26)