Re: [m2m-iwg] W3C Device API

["andypiperuk@xxxxxxxxx" <andypiperuk@xxxxxxxxx>]

Rick, I completely agree with your analysis here. There are already a wide range of standards and protocols. Accept the diversity, demand openness and access to the data, and be prepared to connect and transform.

When I talk about MQTT I *always* point out that it is just one protocol and that others e.g. HTTP and many more, also exist - but have different strengths and weaknesses. I also point out that it is particularly well-suited for certain M2M type scenarios and mobile, and being easy to use and data-content agnostic it is good for acting as glue between others. Therefore, having Open Source implementations makes it a good way to enable things to interoperate, but ultimately there will frequently be the need to connect and transform between formats and protocols. This is actually how it came about - as an effort to break data out of the "SCADA prison" ~12 years ago.

As well as e.g. the W3C looking at this, there's recently been a TSB (Technology Standards Board) exercise around IoT in the UK, looking to sponsor a few technology pilots in this same area with the goal of driving "standardisation".

I'm quite sure that we'll hear about yet more of these activities at Le Web next week, too...

On Thu, Nov 29, 2012 at 1:27 PM, Rick Bullotta <rick.bullotta@xxxxxxxxxxxxx> wrote:

It *should* be of interest, if for no other reason than reinforcing the reality that there will likely never be any single standard or even a few dominant ones. With the wide number of industry groups creating their own formats and standards, as well as a growing number of commercial products on the market that are unlikely to change their formats/protocols in any significant way, I think it would be most valuable to accept and embrace this diversity and reality, at least at the data representation level.  The nature and type of M2M applications is going to be ever expanding, and any attempt to standardize too much at the application layer would, I think, always be out of date and potentially restrictive of innovation. On the contrary, there are a number of areas of the M2M stack that are becoming rapidly commoditized, and in those cases, standardization would be more important than innovation.  To me, that would include functionality such as provisioning, billing, monitoring
 , etc... - clearly a great target for a standardized API and protocols.  And perhaps some high level standardization or "best practices" for device security and identity management (a big gap today).  I do feel, however, that despite the virtuous intent of standardizing data formats and protocols, it is unlikely to ever be able to stay ahead of the innovation curve and to gain any dominant market momentum.

-----Original Message-----
From: m2m-iwg-bounces@xxxxxxxxxxx [mailto:m2m-iwg-bounces@xxxxxxxxxxx] On Behalf Of Werner Keil
Sent: Thursday, November 29, 2012 8:08 AM
To: m2m-iwg@xxxxxxxxxxx
Subject: [m2m-iwg] W3C Device API

Dear all,

There seems to be other standardization bodies, especially W3C dealing with several Device APIs primarily to be accessed via

http://www.w3.org/2011/07/DeviceAPICharter
http://www.w3.org/2009/dap/

Since it is mostly for HTML5 and Web Applications, I am not sure, if it's of interest or HTML5/_javascript_ are considered beside the other languages like C,Java or Lua, just wanted to mention it for now.

The focus both of the existing drafts (driven by e.g. Nokia) at W3C and an earlier study by the Open Ajax Alliance show, that these were so far intended to support traditional phone type handsets with device categories limited to simple examples like battery power, etc. Much like what Java Mobile Sensor API was created for.

More complex and versatile types of sensors and devices (you may have blood pressure in a Medical use case or Temperature inside a container for Logistic/Transport, just to name 2) are vaguely mentioned as "Generic Sensor API" but so far no work has been done there in the W3C WG.

Werner
_______________________________________________
m2m-iwg mailing list
m2m-iwg@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/m2m-iwg
_______________________________________________
m2m-iwg mailing list
m2m-iwg@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/m2m-iwg

--
Andy Piper | Farnborough, Hampshire (UK)
blog: http://andypiper.co.uk   |   skype: andypiperuk
twitter: @andypiper  |  images: http://www.flickr.com/photos/andypiper