As regular readers know, I have been caught up in the production of OBIX 1.1 for most of last year. OBIX has world-wide use in niche locations. It has open source platforms. They do not interoperate as well as they might. To improve interoperation, and ti improve telemetry, we started work on 1.1. Then the Smart TV Alliance upped our game.
But first, a little about how 1.1 is shaping up. We broke oBIX up into smaller pieces, to make it simpler for a programmer to tell what rules they are using. With smaller pieces we can more easily say “An Application conforms only if…” This makes interoperation of different platforms much more likely.
By last May, we had isolated the core information model and interactions (OBIX 1.1) and separated out Encodings and Bindings. We added features to support large data-set telemetry. We added the capability of adding metatags to points, to support semantic sets such as Haystack and BIM. We specified Common Encodings for OBIX (XML, JSON, COAP). XML is the original encoding. JSON is much beloved by current web developers. COAP, or the Constrained Application Protocol, is on track to be the recommended communications platform for the Internet of Things. COAP is designed to translate HTTP down to the level of constrained and lossy communications. We also developed two binding specifications: REST and SOAP. The REST binding formalizes what Tridium has long provided. The SOAP binding is typified in the multi-tiered architecture used by ETSI projects and by products such as Energle.
These formal encodings and bindings make it easy to tell what the other side expects. A typical web page might integrate with a server that uses the OBIX 1.1 model encoded in XML and bound with REST.
Then I received email from the Smart TV Alliance. They liked what we were up to, but wanted us to add a Binding for WebSocket (RFC 6445). That was easy enough to do—which is why we went to the multi-part format described above. WebSocket provides for full-duplex (two-way) communications over an HTTP connection. WebSocket is a standard part of HTML5, so you are probably reading this in a device that supports WebSocket already. The Alliance proposes to use OBIX encoded in JSON and bound to WebSocket.
The initial goal of the Alliance is to create a common platform for TV apps. Today a company such as NetFlix must write an app for each TV. Under the Alliance plans, the same app would run on televisions from LG, Panasonic, Phillips, Toshiba, and Vestel. These televisions will all support HTML5 applications. I expect an explosion of TV-based apps as companies exploit the Alliance eclipse plug-in to write once, run everywhere.
The Smart TV Alliance announced their SDK at the Consumer Electronics Show, just as the latest OBIX specifications went out for public review.
Digital Signage, whose platform is essentially flat panel televisions, will be the first commercial building system to be remade. It takes only a moment to imagine how advanced signage will change with a consistent local HTML5 platform. Multi-screen digital advertising company YuMe is already a member of the Alliance. Custom development kits will need to remake themselves. Wayfinding, building directories, restaurant menus with dietary information are just a few of the applications that will change rapidly.
But that is before we consider WebSocket and OBIX. The Alliance assumes that WiFi is available. The first add-on application that comes to my mind is the smart home theater. This requires communications with lighting and with sound systems and other consumer electronics. (In possibly related news, Apple, not a member of the Alliance, pushed through an RFC draft for aggregated service discovery to support their televisions.) Mobile internet provider Obigo brings the Alliance into smart phones and PDAs. I wonder what applications will be built based on dynamic interactions between phone and television. Is this the end of hunting for the remote?
From there is it a small hop to communicating with the WiFi enabled thermostat. This is sure to intensify sparring between Honeywell’s WiFi SMart and Google’s Nest. U-SNAP can bring WebSocket to standard appliances. Smart homes might at last be here, not just for hobbyists, but for the rest of us, using existing infrastructure.
This brings App culture and App technology to smart homes, and, to me, that means Smart Energy Apps won’t be far behind. Homeowners won’t tolerate long integration requirements, so energy system discoverability is part of this picture.
It is only a matter of time before this creeps back into commercial building energy management. The predominant building system middleware is already built on OBIX. Digital signage everywhere can provide energy management platforms everywhere. Buildings will adopt new apps if the old ones do not perform as they like. Will there be Freemium Energy Apps?
Only time will tell, but it should be a wild ride for the next few years.