W3C home > Mailing lists > Public > public-wot-ig@w3.org > September 2016

Re: Is the conformance to the Web architecture principles what distinguishes the IoT and the WoT ?

From: Dave Raggett <dsr@w3.org>
Date: Sun, 25 Sep 2016 18:54:36 +0100
Cc: public-wot-ig@w3.org
Message-Id: <C31AAC74-0B0E-4961-9DC0-8D1AD1FA4EB3@w3.org>
To: Maxime Lefrançois <maxime.lefrancois.86@gmail.com>

> On 25 Sep 2016, at 14:06, Maxime Lefrançois <maxime.lefrancois.86@gmail.com> wrote:
> After this exiting week at TPAC, I would like to kindly ask you for some piece of advice about how to define the exact difference between IoT and WoT for a paper I'm currently writing.

The IoT doesn’t have a single definition, but loosely speaking, it concerns the means for applications to interact with the physical world through connected devices that expose sensors and actuators. In essence, the Web of things provides an abstraction layer above the IoT, and based upon W3C’s framework for linked data. This allows you to describe things and their relationships independently of how the devices are connected.

The Web of Things simplifies application development and enables interoperability across platforms which may use a variety of protocols and standards. You can make the analogy with the Internet itself, which defines an abstraction layer (IP addresses and packets) together with the network technology independent IP sockets API. The Internet and the Web stimulated exponential growth in services and innovation in business models. We want to repeat this process for the IoT.

The starting point is the idea of things that stand for physical or virtual entities, whether connected or not. These are exposed to applications as software objects with APIs corresponding to the thing's properties, actions and events. This builds upon more than three decades of experience with event driven object oriented software. These software objects can represent local resources, e.g., sensors and actuators attached to the device running an application script, or remote Things, where the object acts as a proxy for resources on another device.

The approach is based upon the fundamentals of Web architecture [1]
URIs for identifying things and their descriptions
A variety of protocols for accessing things, since no one protocol will be appropriate in all contexts
Metadata for describing things as a basis for interoperability and discovery, and playing an analogous role to HTML for Web pages
URIs can be used to access machine-interpretable descriptions of things. These descriptions enable the automatic generation of scriptable objects whose interaction capabilities correspond to those of the thing the object stands for. The metadata can cover semantic models and domain constraints, including relationships between things. 

The application logic using the software objects can be hosted on the same device as the thing, or on other devices such as a local hub or a cloud platform. Application developers are shielded from the implementation details of how objects are coupled to things, allowing platform developers to choose the transport protocols and communication patterns best suited to the context.

The table in the white paper shows how the abstraction layer for things fits into the communications stack. The web of things deals with the higher layers and the IoT with the lower layers (transfer, transport, network). For more details, see the attached copy of the white paper as distributed at last week’s Industry of Things World conference in Berlin. This version of white paper includes some refinements to the HTML version produced by the IG along with reformatting as a Word document for printing purposes.

[1]    http://www.w3.org/standards/webarch/

   Dave Raggett <dsr@w3.org <mailto:dsr@w3.org>>
Received on Sunday, 25 September 2016 17:54:52 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 25 September 2016 17:54:53 UTC