W3C Logo

W3C Workshop on the Web of Things

Enablers and services for an open Web of Devices

25–26 June 2014, Berlin, Germany

Host

W3C gratefully acknowledges Siemens, for hosting this workshop.

Siemens

Thanks also to support from the European Union through the Seventh Framework Programme (FP7/2013-2015) under grant agreement n° 317862 - Compose. Compose

Workshop Report

W3C held a two day workshop on the Web of Things in Berlin on 25-26 June 2014, hosted by Siemens. The Web of Things is expected to have broad and sweeping economic and societal impact. Open standards will be critical to enabling exponential growth of the kind we experienced with the early days of the Web.

There was broad agreement at Workshop of the problems that currently face the Internet of Things. Until now, the focus has been on the devices and communications technologies. There is a growing awareness that the business opportunities will be centered on the associated services, moreover, the current situation is one of fragmentation with products being developed in isolation due to a plethora of IoT protocols and a lack of a shared approach to services.

Rather than focusing on the IoT layer, it is now critical to focus on what is needed to open up the markets for applications and services that sit on top of the IoT. The workshop attendees showed strong support for W3C to initiate new standards work to help realize the huge potential for the Web of Things as a web scale layer sitting above the IoT.

The workshop examined the opportunities for open Web standards for service platforms in the network edge (e.g. home gateways) and the cloud, along with the challenges for security, privacy and the integration with the Web of data.

The call for papers, agenda and raw minutes for 25th June and 26th June are available separately.

Table of Contents

Opening Session

The workshop was opened by our host and co-chair Jörg Heuer (see slides). He explained Siemens interest in the Web of Things and desribed how Siemens is structured in four sectors: industry, energy, health care, infrastructure and cities. We need to think across the verticals, as the Web of things activity is a lot about crossing domains. We need to "understand the elephant called web of thing" (in reference to the cartoon in the slides). We'd like to exchange our perspectives and uses of the term.

Jörg was followed by a presentation by Dave Raggett, the co-chair from W3C. Dave introduced the mission of the World Wide Web Consortium (W3C) to lead the Web to its full potential, and gave some details about the structure of W3C and how it works. The focus is on creating open standards that are associted withr a royalty free patent policy designed to encourage implementations. The Web is now 25 years old and is the most interoperable platform today, and available on many devices and operating systems. This has enabled global access to information and transformed the World. There is a huge global community of developers. Dave briefly covered the fragmentation issue facing the Internet of Things and the need to work together to define open standards and marketplaces! He said that the workshop is an opportunity to discuss:

  • Combining the Web of Devices with the Web of Data to create the Web of Things
  • Service platforms at the network edge (e.g. home hubs) or in the cloud
  • The role of open standards as a basis for open markets of services
  • What common requirements emerge from the various application domains?
  • What kinds of standards should W3C focus on?

Dave finished by introducing the structure planned for the the two days of the workshop with talks, panel sessions, break-out sessions and demos, and ending with a session on conclusions and next steps.

Talks

We planned for 20 talks, each with 15 minutes including questions and answers. The presenters were selected based upon review of their submissions by the program committee.

Sharing experiences, use cases and requirements
  • Johannes Hund, Siemens -- web technologies in smart grid and EV applications [paper, slides]

    This talk focused on virtual power plants as part of the smart grid and the various classes of stakeholders. We are looking at an extensible framework for automation based upon web technologies. The Internet of Things is in some ways analogous to social networks and necessitates good ways to push information. In our position paper we set out our interest in contributing to a common understanding of the Web of things, evaluation of related activities, identification of potential work topics for W3C and setting up a Web of things technology landscape.

  • Yusuke Doi, Toshiba -- Long Life Web of Things with EXI [paper, slides]

    This talk focused on the need for standards especially for "things" expected to have a long deployment life. XML is well suited for this now that we have EXI (Efficient XML interchange) as a binary serialization of XML that can be efficiently generated by resource constrained devices. Current discussion is looking at lightweight grammar exchange formats.

  • Masaki Umejima, Japanese Smart Community Alliance -- ECHONET Lite enables Demand Side Energy Management - IP based and IEC's open standardized interface for home appliances [paper, slides]

    An 8 year plan, starting with the Tokyo area, to upgrade Japanese elecric power meters to connect them via the ECHONET lite protocol. This is an IP based lightweight request-response protocol running on top of UDP and aimed at smart home applications. ECHONET LITE is a liaison partner in W3C.

  • Nick Allott, NquringMinds -- IOT interoperability using web technologies: lessons learned and future challenges [paper, slides]

    Nick started by observing that many IoT protocols are not IP based ("there is no 'I' in IOT"). This necessitates some kind of gateway device to connect sensors to the Internet. He quickly covered security using TLS, PKI and XACML, as well as the role of WebIDL for defining service APIs exposed by such gateways, and how this can be combined with JSON-RPC and JavaScript libraries to remote the APIs to web applications running in a web browser. To the question what are the 2-3 things we should focus on quickly, Nick answered: a simple JavaScript abstraction with good enough security, adding that he has seen things fail if security is not addressed up front.

From things to the Web of Things
  • Shigeru Owada, Sony -- Kadecot: Android Web API Server for Home Appliances and Sensors [paper, slides]

    This talk described a smart phone application (named "Kadecot") that acts as a hub for home devices. Web applications access device APIs hosted by Kadecot via the Web Application Messaging Protocol (WAMP) over Web Sockets with remote procedure calls and pub-sub. Kadecot supports ECHONET lite for discovery and communication with devices over UDP. The web application features a 3D avatar and card based user interface.

  • Ryuichi Matsukura, Fujitsu -- Service platform with Web based interface to control devices [paper, slides]

    Describes the role of a service platform as a hub connecting applications to devices, avoiding the need for applications to directly implement the IoT protocols. The service platform can be implemented as a home controller within the home, or as middleware in the cloud. A logical model of devices abstracts over vendor differences for the same categories of device. Support is included for ECHONET lite, KNX, SEP 2.0 etc. and avoids the need for devices to have web-based interfaces themselves. The approach has been applied to smart homes, residential buildings, gas stations and schools.

  • Ricardo Morin et al, Intel -- Programming Device Ensembles in the Web of Things [paper, slides]

    Ricardo started by observing that today's IoT programming model comes from embedded computing with single-purpose applications and siloed security. This doesn't scale. We need a scalable solution where everything is a resource (URL), with support for discovery, multiparty authorization and offloading computation to remote workers. Ricardo's team (Intel's Software and Service Group) is ready to participate and contribute to W3C work on those areas.

  • Dave Conway-Jones, IBM -- Node-RED: a wiring tool for the Internet of Things [website, slides]

    Node-RED is an open source browser based graphical tool for creating IoT applications via "wiring" nodes together and adding simple scripts to achieve the desired behavior. It is built on top of node.js and makes it easy to add any node.js (npm) module into the node-RED palette of nodes. A selection of nodes is available for data transforms, network protocols, IoT hardware, social networks, storage and analytics.

  • Kazuo Kajimoto, Panasonic -- Consumer Electronics and Web of Things [slides]

    Panasonic would like to contribute to work on defining APIs for cloud based platforms for Web of Things applications. These APIs can be made independent of IoT protocols such as ECHONET lite, EE-Bus, IR control and so forth. Web of Things APIs should support a variety of descriptions, e.g. RESTful HTTP, Web Socket JSON messages, and JavaScript APIs. Support is needed for enumerating registered devices, accessing device status, notifications and device control. Challenges include safety, privacy, and security. This suggests the need for access control mechanisms.

  • Laurent-Walter Goix, et al, Telecom Italia -- Smart social spaces: opportunities and challenges of the social web of things [paper, slides]

    The IoT & M2M are well covered with many SDOs and industry fora. However, the integration with the Web and the Web of people is still limited, e.g. siloed products for IoT, and closed ecosystems for social network sites. Social relationships between people and things, i.e. a social web of things, has lots of potential. Telecom Italia is looking to W3C for work on the web of things on use cases, abstract modeling and vocabularies, API definitions, and work on the social web on activity streams and action handlers, and support for federation across smart spaces (homes, offices, hotels, etc.)

Security, trust, privacy, provenance, access control and policies
  • John Mattsson, Göran Selander & Göran A P Eriksson, Ericsson -- Object Security in Web of Things [paper, slides]

    Fragmentation is holding back the market potential for the IoT. The Web of Things is very promising and brings new security and privacy challenges, including multi-party trust models. Flexible security solutions are needed, and W3C should address secure handling of data and policies for the Web of Things. Many pieces for this are available, some are under construction and some are missing. Note that constrained IoT devices may be limited in their security capabilities, but this can be mitigated via local gateways.

  • Oliver Pfaff, Siemens -- Authentication for the Web of Things [paper, slides]

    A discussion of security and authentication, including the role of third parties using the connected car for a case study. Authentication is needed for users, apps and devices. This will necessitate work on defining and managing identities and credentials, and their relationship to access control policies.

  • Alessio Gugliotta, INNOVA --Trust-based Discovery for Web of Things Markets [paper, slides]

    An introduction to the EU Compose project which features a cloud based market of IoT services, followed by an analysis of the role of trust for discovering and selecting services. This is based upon a combination of semantic matching using semantic annotations, and reputation management. The Compose project is applying this to three different use cases: smart cities, smart resorts and smart retail.

Semantics, linked data, vocabularies and best practices
  • Roger Menday & Neil Benn, Fujitsu -- The Linked Data Platform to Address, Describe and Interact with Things [paper, slides]

    Roger started by introducing the W3C Linked Data Platform (LDP) which features the use of HTTP to access and update RDF descriptions of web resources. Roger and his colleagues have looked at applications in healthcare. The SPARQL language can be used to query local and remote collections of RDF triples. LDP can also be used with streamed updates. Further work is needed on integrity constraints (RDF Data Shapes) and bindings to other protocols for linked data streams.

  • Phil Archer, W3C -- Building the Web of Data [paper] [slides]

    Phil surveyed the work by the W3C's Data Activity, including the Linked Data Platform, Health Care & Life Sciences, vocabulary development and coordination activities. Recently started work includes Comma Separated Values (CSV), and Data on the Web Best Practices. Work is soon to start on RDF Data Shapes and a joint W3C/OGC WG on Spatial Data. Security and privacy are key topics for consideration.

  • Michael Koster, ARM -- Information Models for an Interoperable Web of Things [paper, slides]

    Web scale interoperability is a major goal, enabling reuse and the network effect. Michael discussed the role of resource directories for collections of semantic links describing resources. Examples include CoRE RD and Hypercat. Interoperability involves consistency at multiple levels of abstraction. This needs something like HTML, but designed for machines to understand. W3C is in a good position to lead this work.

  • Amelie Gyrard et al. EURECOM -- Domain knowledge Interoperability to build the Semantic Web of Things [paper, slides]

    EURECOM have worked on reasoning over sensor data using the M3 ontology and Linked Open Rules. This builds upon the W3C Semantic Sensor Networks (SSN) ontology. The goal is a framework for the semantic web of things as a basis for helping developers to build IoT applications. Amelie would like to see standardization work on linked open rules and sensor-based domain ontologies.

  • Milan Milenkovic, Intel -- Towards a Case for Interoperable Sensor Data and Meta-data Formats [paper] [slides]

    The Web of Things will greatly expand the Internet with vast numbers of sensors and data. For this to work, we need interoperable data and metadata formats. Support for legacy systems is needed. This should start with work on use cases, requirements, naming, formats, vocabularies and best practices. Milan recommends a lean evolutionary approach, starting from cloud-end service-driven queries - such as sensor access and attribute/metadata-based search capability - and design naming, data and meta-data formats that populate the system to support such service functionality. We should start with an extensible minimal useful sensor meta-data and format definition, then grow it with use cases and deployment experiences.

  • Alessandra Mileo, Ericsson -- Semantic Modeling of Smart City Data [paper, slides]

    Alessandra started by talking about the challenges of dramatic growth of big data and its significance for smart cities. CityPulse is an EU project on real-time IoT stream processing for smart cities with a focus on semantic models as a basis for reuse. Data needs to be annotated to account for quality, provenance, privacy and other contextual information. Contextualization is tha mapping from observations to complex events. She finished by encouraging participation in the W3C RDF Stream Processing Community Group.

Breakout Sessions

The workshop included time on both days for breakout sessions. Participants were invited to submit these to a public wiki page in advance of the meeting. We asked workshop participants for a show of hands to gauge the level of interest in each proposal as a basis for room assignments. In one case, we merged two sessions: Scripting in the Web of Things, Andreas Harth and Application runtime for “things,” Ricardo Morin. We invited each session to provide a brief summary to the full workshop. The following provides a brief summary for each breakout, for more details see the full minutes.

  • Service Descriptions for the Web of Things, Dave Raggett (W3C)

    A lot of the value creation for the Web of Things will be provided by services that combine and transform data for use by other services or applications. This session will look at the potential for standards for describing services as a basis for an open market of services.

    The participants of the breakout group on Service Descriptions deem it useful if the W3C published recommendations for device vendors and service providers about how to describe Web services. These descriptions should encompass functional as well as non-functional aspects of the service, i.e. should describe its basic functionality as well as input/output data and also aspects such as QoS, access rights, etc. This recommendation should rather focus on the concept of different services within the IoT than producing a technical specification. Furthermore, the participants emphasize that these descriptions should be targeted at developers. Device vendors and service providers should prefer simple APIs and also keep the service descriptions themselves easy to understand by hiding details of the underlying protocol. Service descriptions should thus facilitate the work of developers rather than targeting challenges such as automatic service composition right away. Finally, to ensure good uptake of descriptions by the community, they should emphasize describing aspects of services that are relevant for service providers (e.g., with respect to monetization), for instance information that relates to finding a service, reviews, or payments.

  • Interaction Models for the Web of Things, Erik Wilde (Siemens)

    Many scenarios need basic interaction capabilities so that "thing events" can be published, consumed, aggregated, filtered, and re-published. This session is looking at existing standards such as Atom and emerging/evolving standards such as Activity Streams and is asking how these existing models can be used.

    What is missing from today’s web landscape?

    1. Protocol/gateway
    2. Domain models. Largely exist. Up to domain specialists to refine them. Use to create media type.
    3. Find out what is needed. (not a good answer yet). Useful to look on use cases within verticals.

    Web thrives on services: People care about what service and available transactions:

    Atom feeds used for Pod Casts and other areas currently. Very extensible. Issue was that it was XML based. Basic Atom model of how to expose services is valuable.

    Shift to Activity Streams. Already being widely deployed by Google and others.

    Possibly collaborate with W3C Social Web WG.

    Discussion:

    Q) Web services type of approach may not be effective. Devices need to be able to speak to each other without interaction of web services.

    EW) Could use activities model with web services model.

    Q) Laurent? Works well with Social Web of Things. Represents Object related activities. Rest type API. Proposed protocols to create interactions between social networks. Initially based on ATOM. New draft of Activity Streams (2.0) incorporates some extensions from ATOM. Trying to use JSON-LD as well.

    EW) JSON Activity Streams 2.0 document is a bit inconsistent currently. ITF draft will be republished as a W3C draft.

    Q) Professor Tashka (SP) How extensible is it in relation to semantics. Progmatics need to be consider as well. Does it have a time model?

    EW) Not happy with extensibility model. Nothing on commitment. Does support transaction B2B type of protocol effectively. A time model exists already.

    Thomas: Example Mashup services, Gmail with other services but this is currently closed. This and That. Uses ATOM

    EW) ATOM has about 10 activities. Extensibility is one of the reasons people like ATOM.

    Q) Laurent? Ecosystem of players (social networking players). Likely to be adopted by those who care about users.

    EW) How to expose resources so they can be used by other services.

    Q) Could push model rather than pull model work:

    EW) How you ship information around would be a pull but you could use push as well. ATOM model should work for both. Most people not interested in sensor feeds. Reports by sensors but services can then aggregate it in meaningful ways.

    Q SP) Need to be sure it is people interacting with things as well. Not just people interacting with people.

    EW) Once Social Web starts we will soon know if it includes web of things as well.

    Q)Wolfgang Dorst. Can give examples of typical web of things scenarios and what would not be.

    EW) Not a scenario … My car as one system. No external control of car. WOT – Services, tire pressure, fuel economy, smart home scenarios. Control of multiple spaces (home and office) from one service. Healthcare. Monitoring of health. Industry 4.0 (WOT could be one building block).

    Two layers: Devices provide raw stream of data which is then aggregated and then high-level activities are generated(Web Of Things). Activities could be tailored to individuals.

    Some people may need raw stream of data. Some may be interested in this (research etc). Aggregator could hold and publish raw data as well.

    Distinction should be made Web of things and internet of things. Not based on volume of data.

    Conclusions

    What are conclusions? Some interest in conversation. Not clear on deliverables. How to follow this conversation afterwards.

    Milan: What is basic problem that we are trying to solve. Configuration changes?

    EW) a) not every state change needs to be reported. B) before WOT WG, would need statement of what we are about.

    Laurent) Need to create use cases that leverage activity streams.

    EW) Look at social web working group and see what they are doing.

  • Business Models for the Web of Things, Alessio Gugliotta (INNOVA)

    Platforms for sharing and interconnecting IoT devices are already available on the Web/Market, and more (with added-value services) will most-likely appear in the short-mid term. Are they really sustainable? What is the actual value that will make them profitable? Is the Developer-Pays Business Model the only possible? Or should other stakeholders (e.g. hardware vendors, cities, large industry) pay for sustaining an open platform?

    1. Existing IoT solutions already exist. In particular in the industrial domain (e.g. manufacturing, smart building/home automation sectors), there are vertical/closed solutions. The point is to look at those business models (benefits from linking devices, collecting and storing data, create services on top of data for the companies) and use them for the WoT.
    2. WoT will provide new market opportunities, but we need standards to offer an open solution that works like existing vertical solutions (point 1) but at a larger scale. In particular, we need standards about the APIs facing sw developers and hw providers. We also need standards to cover the whole stack of WoT (e.g. security, privacy, etc.). Some standards that could fit do exist. We just need to select them, and in case just cover the existing gaps.
    3. Platform for WoT should not be just an app store, but they should be much more. For example: (i) offering richer customer experience (not only sw developers, but also non-experts with innovative ideas); (ii) enabling HW vendors to develop plug-ins for their IoT devices.
    4. Revenue models. Some ideas: partly based on SW developers (freemium approach) and leverage tracking the end-user experience to get insight and generate additional services (e.g. ads)
  • Digital Object Memories for the Web of Things, Jens Haupert (DFKI)

    Based on the idea of the internet of things each physical artifact can have a virtual counterpart. Giving such objects a digital memory they can store all relevant information about themselves along the entire life-cycle chain. With OMM we proposed a structuring object memory model and successfully passed a W3C incubator group. To leverage this approach of a cross-domain model we are interested in improving and enhancing the model to support multiple platforms, different domains and scenarios and additional object-memory related services like querying/filtering data, like privacy and access control policies and many more.

    DFKI introduces concept of digital object Memories based on W3C-XG OMM format

    • Idea: show what we’ve done at DFKI within German research projects, then take a look at workshop participants’ experiences, which topics are not yet addressed and should be
    • DFKI developed concepts and did an incubator group to evaluate if the idea can be standardized
    • Result: it could be done but the activity stopped
    • Problem: it doesn’t fit perfectly in running activities, new activity couldn’t be started because our partners not being involved in W3C

    Question by Darko Anicic: What are the competing approaches?

    Answer: GS1 has RFID standard that is set up well and used already, but it is very focused on commercial use cases. OMM approach is much more open and versatile

    Question by Darko: is GS1 focused on special businesses?

    Answer: Manufactured goods and logistics

    • EVRYTHNG approach, similar to OMM, attach memory to object and add key value pairs, but nothing else, it’s very simple. implementable in almost every device
    • OMM-XG result: too simple for some use cases, you might want more sophisticated approaches
    • OMM approach is not perfect but a first approach. DFKI has set up demonstrators using this technology; developed this in two projects, second is now finished

    Question by Pedro Reboredo: Point out differences to GS1 approach?

    Answer: Sample Scenario of production/logistics chain

    Question by Yusuke Doi: Scenarios for composed and separated objects?

    Answer: Supported by OMM (see slides)

    Question by Pedro: Why to use memory on RFID chip instead of cloud/server?

    Yusuke gives example argument for why to have it on RFID

    OMM aspects like open-loop scenarios, partitioning, meta data, defined blocks, embedded memory for integrated parts

    Question Yusuke: Synchronization between embedded objects and “mother object” after part is removed?

    Answer: It should be well defined where the synchronization is done and how it should be done to avoid for example outdated memories

    Question by Darko: How is data from DOMes normally used? How is it queried/retrieved/monitored?

    Answer: Depends on the implementation. Different memory access methods. Depending in the object solutions must differ, pizza vs. aircraft parts monitoring for example. three ways: 1) barcode, data stored on server; 2) RFID tag stores either ID in the tag, data on server or add data info on tag or hybrid system; 3) CPS

    Question by Owada: What’s the motivation of this project? Is there any specific application that helps to evaluate the idea? Why not go with proprietary systems for each company?

    Darko: As soon as it becomes complex it will be helpful to be so flexible

    Jens: It’s a research approach and we want to allow everybody to store everything they want. We have two use cases e.g. a lot of info that was present during early stage of production is lost and must be measured/evaluated again, doesn’t happen here

    Note by Yusuke: also certification of quality is a point, monitoring of logistics, get rid of parts of quality check process. Must comply with regulations of course; another interesting use case, e.g. recycling in cars, consist of many different materials and parts. Recycling company needs to know what to do with which part, which materials can be used maybe RFID can be used to carry this information.

    Question by Darko: what’s the price point?

    Answer: barcode “nothing”, RFID depends on the tag from 0,1 cent to 10 euro per tag with hardware cost like readers, depends on the technology you use and application you want to realize; large factories might need thousands of point to read this info

    Question Darko: did the W3C group produce an implementation?

    Answer: formal report, DFKI developed a OMM-library + Object Memory Server for Java and C#, not public but can be distributed, not that well documented at the moment

  • The Web of Things Friendly Label, Dominique Guinard (EVRYTHNG)

    After a brief report on the last 5 years in all things WoT (see Web of Things community blog), in this session we would like to discuss the "Web of Things Friendly" label. The idea of the WoT label is not to become a standard but a set of simple guidelines. The main goal of these guidelines is to ensure that new or existing IoT devices and services are designed with interoperability at the Web level in mind. The ultimate goal being the boostrap of a community label that helps selecting the right standards and tools to achieve interoperability.

    General concepts for the WoT Label

    Core principles

    Much needed to bring back the WoT to what it is supposed to be, to how it was intially defined: an application layer for the IoT built on Web protocols.

    WoT != IoT, WoT builds on top of IoT to create an application layer ensuring:

    • Interoperability
    • Creating applications for IoT devices and services is made accessible
    • IoT devices and services are accessible through well-known Web tools

    WoT Architecture framework used for the WoT Label:

    • Accessibility
    • Findability
    • Security and sharing
    • Composability

    We need several layers of WoT friendliness, like the 5 star model of the semantic Web.

    Accessibility

    • REST either directly or through a smart gateway
    • Pub/Sub with bridges to Web protocols: Websockets
    • JSON
    • HTML representation I should be able to use my Web-browser

    Smart Gateway pattern

    • A Box that connects to the devices, this is what should be WoT-friendly not necessarily the nodes themselves.
    • Any application can use any proxy without configuration.

    Findability

    • Discovery: low level, UDP in the browser? -- Finding devices
    • Web-wide discovery: findability -- Finding resources or services, semantics corelink format, json-ld, IPSO, free form configurable things, JSON-ld
    • Devices and gateways could be crawlable, but not by default.
    • Idea: Creating a Schema.org for the WoT on Webofthings.org
    • Idea: Enscapsulated Web Object: metadata + data in a JSON object

    Sharing & Security

    • Security: at the minimum Web security (TLS) at the Gateway layer, ideally end-to-end
    • Sharing: Reuse social networks (Social Web of Things)

    To look at, related:

    • Hypercat: 1248.io
    • Semantic hyperlinks
    • Open Remote

    Misc Q&A

    Is it a problem to not have end-to-end HTTP connectivty?

    not necessarily

    Does the WoTness extend to the protocols beyond the gateway?

    probably not

    Is the hue lamp a wot-friendly product:

    Yes because of the gateway, not because of the bulb. But gateway + bulbs are a product as a whole.

  • Privacy in the Web of Things, John Mattsson (Ericsson)

    How do we protect end user privacy in the Web of Things? This will be an extremely important issue. Even now data from a smart power meter reveals a large amount of privacy sensitive information: when did the family awake, when did they go to bed, when are they on vacation, how many people are home, etc. What happens when we have sensors everywhere? Some privacy issues can be solved by cryptography, but most cannot. Privacy policies will come from individual persons, from enterprises, and from regulators. How do we handle them in a secure and standardized way?

    13 people attended. Agreement that this is important, a make or break issue for the web of things. Also agreement that it is a very hard issue to solve. Combined data can identify a person even if individual data do not. Little technical enforcement can be done. Trust and reputation will play an important role. Possible to do legal enforcement. Need to find that a breach has occurred. Different opinion on Regulation:

    1. Regulation is the only way. Regulate heavily and make it cost. No technical enforcement possible.
    2. Regulators often regulate to specific and misses the bad things. Users take best decisions. Need freedom of contract. Very different regulations in different countries, and complicated with legal action when users and services are in different countries.

    Send privacy policy with data. Privacy flags. Problematic in device-device communication if the user has no contract with the second device. Right to revoke access is a minimum requirement. Short term licensing of data might be one option. Data transfers needs to be tracked. Automatic detection of breaches. Users need to know who is tracking them. There is a place for simple rules (like creative commons) to be communicated with the user.

    Conclusion: W3C should create a technical mechanism to pass policy data around in the Web of Things.

  • Scripting in the Web of Things, Andreas Harth (KIT) and Ricardo Morin (Intel)

    Andreas: The Web of Things will consist of many components that provide access to individual sensors and actuators. The topic of the breakout session is scripting languages for combining such atomic components to provide elaborate functionality.

    Ricardo: The WoT will benefit greatly from defining a common application execution engine standard, utilizing Web programming languages such as ECMAScript, and a set of standardized APIs. A good example of this kind of runtime is the Node.js framework , which has already being proposed for supporting WoT applications. The work in this area may include defining standard profiles to cover APIs for various device categories (e.g., bare sensors, constrained microcontrollers, edge gateways, anchor devices). The application runtime will also need to define a standardized security model as well as user interaction approaches for constrained devices (e.g., control panels, dashboards). In this area, there seems to be a high synergy potential with the Web Applications (WebApps) Working Group that should be pursued. Due to the close synergy with browser technology, it seems very appropriate for W3C to address the need for standardization in this area.

    1. There was general agreement that standardization of a WoT Runtime is highly desirable
    2. This session recommends starting a W3C activity or working group to tackle this task
    3. This session recommends inviting several companies with partial or full implementations and/or high level of interest to participate in the activity or working group
  • The Web of Data We Want, Phil Archer (W3C)

    The Web of Things combines the Web of Services with the Web of Data. What should that Web of data look like? What's already available, what's coming? Where are the gaps? Data dumps? Vocabularies? APIs? What's with this Hypermedia thing, isn't that just HTTP (yes, but, you know, marketing and all that). In this breakout session, I'd like to explore what the W3C Data Activity should be doing to support the Web of Things.

    This session began by asking participants what standards they most wanted W3C to work on to support the WoT. Much of that work is already under way, or planned, notably Activity Streams, RDF Shapes, Spatial Data (including Semantic Sensor Networks), security, and annotations (for POIs etc.). It was reassuring that so much of what is needed is being worked on, however, some areas are not currently under consideration and probably need to be looked at to make the Web of Things a reality. These are:

    • reasoning and rules - we have RIF for rule interchange, but there is a gap for simple rule languages aligned to Web of Things use cases;
    • RDF streams;
    • semantics of constrained devices, including binary and text-based data flows;
    • naming and query conventions, metadata, Web scale discovery;
    • URIs in the WoT.

    In short: there is plenty of work that needs to be done.

  • Application Layer Protocols and Data Encoding for Constrained Devices, Hauke Petersen (Freie Universität Berlin)

    The vast majority of machines intended to run standard protocols for Internet of Things & Web of Things are expected to be very constrained in terms of memory and power capacity (see RFC 7228 for instance). In particular, generic data encoding and request-response protocols used in this context must fit these constraints. Thus, alternatives to XML, JSON and HTTP are desired. The idea of this session is to have an open discussion on available alternatives that would fit constrained devices, including, but not limited to, combinations of protocols such as CoAP, CBOR, EXI etc.

    1. The use of XML or JSON based data formats and the corresponding binary formats EXI and CBOR for interaction with the Internet
    2. internal, hard-coded data formats for intra-node communication
    3. common opinion seems is that EXI and CBOR should be sufficient for the constrained internet for end-to-end communication scenarios
    4. gateway: application and architecture dependent, there will be multiple solutions as appropriate to the needs
    5. security: transport vs. application layer have to be chosen dependent on the application
  • Think Robot, the next smart object, Redouane Boumghar (Dacteev)

    The robot can be seen as a set of objects that are connected by its own operating system. When a robot has to interact with its environment and with humans we hit issues equivalent to those of the Web of Things. These robots will take different forms; autonomous car, home majordomo, the museum guide, your cart at the supermarket.

    A short review of robot platforms and operating systems will give an idea of what has been done so far in terms of how things are connected. I'd like to show and discuss theh different kind of data robots exchange. The data can look very different to what is mostly imagined for connected objects but models can be mutualized.

    Working on multi-robot cooperation, I think the robotics world and the Web world should be interacting more closely. This is a discussion about what may be a far future but it is always good to start building the present with the future in mind.

    The concept of the decision loop in robotics where data transits and is transformed or generated through the three main points of the loop: Perception - Decision - Action. Robotics is a large domain where different disciplines interact: sensor data acquisition and modelisation, autonomous planning, electronics, semantic of the human machine interactions, probabilistic reasoning, simulation and hybrid simulation (interaction between real robots and simulated ones), inter-process communication, and a lot more.

    A lot of ideas have already be shared around the web of things about the modelisation of data and services which help the processing between the perception and the decision which are in fact only a third of the decision loop.

    During this breakout session, we have discussed about enlarging this focus to actuators in order to be able to close the decision loop with standards by the modelisation of actions. Actions are the features of actuators. Actuators are a key concern when drawing the interactions people can have with objects. We should not only consider visual interactions but also auditive, olfactory and tactile interactions.

    Models always to be thought in pairs with their environment. The action model should interact with the environment model. The data models should interact with the services models, etc.

    When considering sets of attached objects (such as robots) we should consider not only the composition of services but also their decomposition accordingly to constraints within the set of objects. Models must facilitate the decomposition and recomposition of services involving pieces in sets of objects.

    The Web of things is like Robotics, it covers a great variety of disciplines. A future Web of things working group should take this into account in its composition, as it is essential to have strong coordination with other groups working on standards that would define the models of the world which is the objects' environment and also the humans' one.

Demonstrations

We had demonstrations of web of things technologies on both days of the workshop. The following descriptions were provided by the demonstrators.

  • David Conway-Jones (IBM) Node-RED - a wiring tool for the web of things.

    The Node-RED demonstration consisted of several aspects. It was all running locally on a laptop. Firstly I demonstrated the use of the simple web based UI (built using HTML canvas and the d3 library), to compose simple flow style event streams. These were then deployed to the running node.js instance. Next more complex flows were demonstrated showing the use of higher level nodes, such as Twitter, sentiment analysis, xml parsing and web sockets. By deploying these it was possible to show easy it was to create a geographic map of twitter sentiment, or a live dashboard of energy usage. Finally by using Node-RED itself as a web server it was possible to demonstrate serving a complete presentation from Node-RED with live interactive data and visualizations within the presentation pages. You can view a 5 Minute demo (showing some of the basic capabilities).

  • Kosuke Nagano (ACCESS Company) Connect devices to the Web

    HaaP : Gateway demonstration; control ECHONET Lite devices from browsers on your smart phone.

  • Ricardo Morin (Intel) - Using remote Web Workers to dynamically off-load computationally intensive tasks

    Description: The demo shows a subset of the use case described in our position paper Programming Device Ensembles in the Web of Things, Section 4, Illustrative Example. The demo consists of a node.js-based Digital Video Recorder (DVR) application written in JavaScript running on an Intel Galileo microcontroller board which uses remote Web Workers to dynamically off-load computationally intensive video analytics algorithms to both a LAN-attached MacBook Air acting as a server, as well as to the Amazon Web Services (AWS) Cloud. The demo shows the value of our proposal to extend the W3C Web Worker standard to support remoting.

  • Mitko Vasilev (Cisco)

  • Benoit Herard (Orange) sensoNet

    A low cost, arduino-based open source sensor network enabler.

  • Nicholas Allott (NquiringMinds)

    The UbiApps demonstration was intended to illustrate how a distrusted, interoperable but secure IOT ecosystem can work using web friendly protocols. Importantly most of the protocols used are pre-existing standards. The demonstration covered the following areas

    • URI addressing: using a single Uri for the identification of an IOT sensor, but working over both cloud and locally routed peer to peer networks
    • Device and service discovery, using a simple JavaScript wrapper, devices and service can be discovered from the webcontext
    • Remote invocation, which works over multiple bearers and multiple application protocols. All however abstracted through interoperable protocols
    • Security: a robust peer to peer security architecture based on PKI and XACML technologies
    • Driver model: a simple drive model that adapts both proprietary and disparate IOT standards under a unified web access layer
    • IOT management: a comprehensive dashboard for the management display and data agregtation of IOT streams

    Live demos are available from www.nquiringminds.com.

    screenshot of Ubi Toolbox

  • Dominique Guinard (EVRYTHNG) EVRYTHNG Web of Things API

    Our demonstrator was illustrating how the EVRYTHNG Web of Things API can be used to give physical objects a virtual identity and an accessible API. To show this we presented two use-cases, one with a street lamp fitted with an embedded device (Flyport) and sensors directly connected to the EVRYTHNG platform making it accessible and manageable directly from the Web. We then showed how this was giving the street lamp an API on which all kinds of applications could be developed such as a mobile monitoring app or a defect reporting app for good citizen! We finally showed how the EVRYTHNG cloud can also be used to give non-electronic objects an identity and API through QR codes, barcodes, NFC or image recognition. We showed this by connecting a can of soda to our engine where it started generating data for each scan and delivering personalized content to the mobile phones interacting with the can. Developers can apply for free API accounts at: dev.evrythng.com

  • Andreas Harth (Karlsruhe Institute of Technology) On-the-fly Integration of Static and Dynamic Linked Data

    The demo shows how to use a rule-based language (Linked Data-Fu) for accessing and integrating disparate web data sources in the Smart Cities domain. The sources follow the Linked Data principles and thus are accessible via polling in a request/response model. The demo shows that it is feasible to access and integrate dynamic data from sensors that expose access to their data via web APIs.

  • Robert Kleinfeld (Fraunhofer FOKUS) glue.things Platform as a Service (PaaS) for integrating the Internet of Things with the Internet of Services

    glue.things is Platform as a Service (PaaS) which provides all the necessary technological components, organized into a coherent and robust framework covering both delivery and management aspects of smart objects, apps and their integration. Connect TVs, cars, wearable computing devices, and all of the consumer and business tools to the Internet. Let they talk to each other. Easily build apps for them and share the data of connected devices with the developer community. Finally, distribute apps on an open and scalable marketplace.

    glue.things demonstrates how to connect devices with the platform by using Node-RED as visual composition tool. Connected devices appear as nodes in Node-RED. Via click and drop devices can be easily connected with Web services. Users can define condition statements, triggers and actions for compositions of device and Web service data streams. With the support of CloudFoundry, final compositions can be deployed as gluethings service in the cloud.

  • Scott Walsh (Plantronics) Web Connected Contextually Aware Wearable Device Streams via JavaScript APIs on a Chromebook

    This demonstration was from the PLT Labs group within Plantronics. The PLT Labs built Wearable Concept 1 device (WC1) is a fully functioning smart Bluetooth headset that has been modified to include a nine-axis sensing capability (gyroscope, accelerometer and magnetometer). The WC1 also has device APIs allowing command and control of the device and the ability to stream all of the internal sensor data over Bluetooth. In the demonstration the WC1 device was shown connected to a Chromebook over Bluetooth. The demonstration leveraged the Chromium experimental JavaScript Bluetooth APIs) in conjunction with the PLTLabs WC1 device's Bluetooth APIs to stream head orientation (quaternions), wearing state (on/off head), pedometer, and free fall data to a Chrome Packaged App using only JavaScript. For further details on PLTLabs and the WC1 device see http://pltlabs.com/concepts/.

  • Jens Schmutzler (Technische Universität Dortmund) WoT-based Vehicle to Grid Communication and Conformance Testing

    The Communication Networks Institute of the University of Technology in Dortmund presented a demonstrator for the Vehicle-to-Grid Communication Interface that will manage the charging process of Electric Vehicles in the future. From a power grid's point of view, Electric Vehicles are scalable non-persistent storage devices connected to the grid and may be seen as sensors (metering points) as well as actuators (demand response / reverse charge units). However, only in large scale distributed deployments a real benefit can be derived from Electric Vehicles for the grid. In that sense, the application domain of Electric Vehicles and Charging Infrastructures aligns well with IoT and WoT paradigms. The demonstrator showed how W3C technology like Efficient XML Interchange (EXI) is used in the context of the V2G Communication Interface, that was just recently published as international standard in ISO/IEC 15118. W3C's EXI web technology was chosen for its extensibility as well as for its suitability for resource constraint target platforms and communication channels. The demonstrator was prepared as part of the research project eNterop funded by the German Federal Ministry for Economic Affairs and Energy.

  • Thomas Amberg (Yaler) Web-enabled Bluetooth Low Energy Watcher

    Yaler enables Web access to embedded devices behind a firewall, NAT or mobile network router. This allows devices to provide Web services. In this case the device (ARM Cortex M4, NETMF) provides a RESTful Web service for the configuration of Web-hooks (outgoing HTTP calls). A Bluetooth Low Energy module attached to the device detects the presence of BLE tokens in the same room and calls a Web-hook for each arrival or departure. The demonstration shows a possible solution for the integration of an Internet-connected device with existing Web applications.

  • Sebastian Käbisch (Siemens) µRDF Store with Efficient RDF

    The demo shows the use of an efficient semantic repository (the µRDF Store), to manage RDF triples on a resource-constrained microcontroller (based on an ARM Cortex-M3). The µRDF Store uses the benefits of the W3C Efficient XML Interchange (EXI) format such as type-aware encoding, low memory usage, and fast processing. RDF data are exchanged in a standardized XML-based format, with the highly efficient EXI representation. Thus, the µRDF Store makes semantics feasible on even very constrained devices and enables the exchange of end-to-end semantic information based on established and standardized W3C technologies.

  • Charalampos Doukas (CREATE-NET) Bridging HTTP REST with M2M protocols like MQTT

    The purpose of the demo is to present how COMPOSE components can be used for enabling bidirectional communication between web applications with embedded devices that use M2M binary protocols like MQTT. The demo consists of a mobile web-based application that is using HTML5 for reading the tilt sensor of a smartphone directly from the browser, and an Arduino board connected to COMPOSE platform using a WiFi shield. The Arduino communicates using MQTT and is equipped with a colourful (RGB) LED and a distance sensor.

    Each time the phone tilts, a rectangle on the phone's webpage changes colour and the same colour is also displayed on the LED on the Arduino. When the distance sensor is activated on the Arduino, the background colour of the webpage on the phone changes instantly to reflect the event. The webpage communicates through REST calls for sending the colour codes and through WebSockets for receiving the distance sensor events from the Arduino device. The demo is shown in this short video

    COMPOSE provides all the components for protocol bridging and communication and also the libraries for developing such web-based and embedded applications.

Panel Sessions

The workshop included three panel sessions.

Panel session on core technologies
  • Moderator: Jörg Heuer, Siemens
  • Markus Isomäki, Nokia [slides]
  • Charalampos Doukas, CREATE­NET
  • Charles Eckel, Cisco [slides]
  • Milan Patel, Huawei
  • Matthias Kovatsch, ETH Zurich

Jörg noted that this panel is about asking what are the core technologies, and what will enable these technologies? He then introduced the panelists and invited them to state their positions.

Charampalos talked about how to bridge the Web and the IoT, citing work in the Compose project. W3C could work on standards for service platforms. Matthias would like the Web of Things to begin at the device level.

Markus is interested in the interaction between Web clients and constrained smart objects. The CoAP protocol looks very promising. Could W3C work on browser API for web apps to access devices via CoAP?

Milan focused on reuse based upon core technologies like LTE (mobile networks) and consistency of data sets across use cases, as well as the role of gateways. Security and privacy are important, but should not be a barrier to innovation.

Charles discussed the need for communications between applications and the underlying network to attain the desired performance goals, and to allow applications to adapt to changing networking conditions.

Charles noted that the ability to monetize web applications will be a driver. The ability to mash data and services from different sources is key, according to Milan, who sees opportunities in health, sports, disability, smart energy and so forth.

Markus noted that there is a lack of open standards in industry right now - too many silos. The ability to combine data from different vendors will be important for valued added services. Matthias added that a lot of people have already noticed that silos don't work, and cited work by the OMA on M2M standards that build on top of CoAP as a reaction to this observation.

What is your definition of the Web of Things?

Matthias responded that the Web is the application layer for the Internet, and the Web of Things plays the same role for the IoT. The other panelists agreed, Milan said that the Web of Things will add value to the IoT. Charalampos added that the Web of Things is using technologies that browsers can communicate with.

Where do you see balance between API and protocol technology?

Jörg then observed that many of us agree that APIs are needed to attract developers, whilst others emphasize protocols -- where do you see balance between API and protocol technology? Markus replied that the Web has worked with just a few protocols, but for constrained devices, we may need to add further protocols such as CoAP to the Web platform. Wolfgang Dorst noted the importance of response time in relation to sensors and actuators. Charles added that this is why he had mentioned network constraints. Some things are time critical, whilst others can be handled more leisurely. Charalampos noted that some decisions need to be made close to the device (at the network edge).

What technologies do you see as the part of the Web of Things?

Charalampos noted current work on semantic descriptions of sensors as a basis for reuse of services. Matthias commented that we can learn from the Web, e.g. easy ways to discover using linking, many ways to describe services (microformats, schema.org, etc.). We'ved talked a lot about data, but things can have embed logic as well. This relates to work on constrained versions of scripting languages for low powered devices and the means to upload scripts to devices. Charles talked about passing data to the cloud for scalable services.

What is the granularity of things?

It was agreed that things can be composed as needed, at the same time allowing you to drill down to the lowest sensor.

What are the expectations in regards to W3C?

Matthias said the Web grew thanks to interoperability of core components like HTML. We need the equivalent for the Web of Things - good representations & Internet media types that are reusable across many domains, not just for each domain. A CoAPI API would be worth considering.

Markus added that there is a lot of relevant work already under way, e.g. WebSockets for asynchronous communication, WebRTC for peer to peer and real-time, and CoAP for constrained devices. A new area of work would be for describing services and resources.

Milan suggested W3C survey technologies to see what the Web of Things can reuse, or which need adaption to use them. Charles added that he hoped that W3C didn't work on creating a whole bunch of new protocols. W3C should instead focus on APIs, data models and push mechanisms to replace inefficient polling. Charalampos would like to see W3C organizing more events like this one, that are open to developers as it is critical to listen to their needs.

Panel session on who is doing what (IoT projects and standardization bodies)

What standards are needed and how to drive them? Which standards organizations are doing what and recommendations for how W3C should collaborate with them.

Each of the panelists introduced themselves with a single slide. Richard presented a long list of relevant organizations. He cited reducing food waste in supply chains, and energy use in transport as use cases for the Web of Things in connection to the Industrial Internet Forum. He added:

  • There are plenty of standards already that directly support WoT applications and construction, we should avoid creating competing standards (there was a bit of discussion about existing identity standards, e.g., GS1 for example; and about low-level communications standards like OASIS MQTT and OMG DDS).
  • It's important to base standards work on use cases and prototypes ("testbeds") which provide real support for requirements for standards, and point at likely new (disruptive) products and services.
  • The Industrial Internet Consortium (IIC) exists specifically to help its members define, build, roll-out and learn from testbeds.

Eric introduced GS1 which focuses on standards and solutions for supply and demand chains, globally and across sectors. GS1 has a very large data model and interested in semantic markup and URIs for product codes. GS1 is involved in W3C work on data on the web best practices.

Mike introduced the CEA (Consumer Electronics Association) which focuses on devices in the home and other consumer specific locations, adding that the CEA work at the device level not at the cloud level. He says: as an example, we are working on sleep monitoring wearables. We are also looking at XML and EXI as a starting point as it is so well supported. We have a great interest in cyber security for IoT devices. We're active in IPv6 deployment including profiles of hosts and routers. We are primarily US based but expanding globally. We are interested in the meeting zone between devices and the web of things.

Ingo introduces the Kantara initiative which focuses on identity management for things. Different kinds of devices have different kinds of identifiers. We've set up a identities of things discussion group as part of Kantara, which itself grew out of the Liberty Alliance and work on single sign-on. His slide has three levels from discovery at the bottom, then control and finally at the top, coordination. This is mirrored with object identifiers, discovery, authentication, authorization, policies and identity management. Ingo discusses the relation to OAuth. He says: we're looking at possibilities for registries, etc. We intend to feed our ideas into existing standards organizations. The Kantara initiative is open to companies and individual membership.

Istvan introduces the relationship of the GSMA to the web of things, including what we're calling connected living, which focuses on helping network operators to get the best out of M2M. We held a recent meeting in Shanghai on a number of application areas. The GSMA is focusing on 4 main topics: connection efficiency, future IoT what does this mean to operators, remote M2M provisioning, and business enablers. Concrete discussions around SIMs for M2M and how operators can create new ecosystems and business. G&D and Gemalto have a big interest. Standardization around storage, security, and embedded SIM.

Nick stated: I've been involved in a number of industry associations and research projects. A couple of observations - we should use existing standards where possible (assuming they are royalty free). Second, there are 2 camps with data and functional perspectives. We need to define a way to name objects that will be applicable to devices with and without IP addresses. This impacts on URI naming, can we enable URIs that are independent of the IoT technologies, e.g. whether the device is accessed over WiFi, BLuetooth, 6LoPAN, or is locally attached to the hub, etc.?

In regards to security, we need focus on explicit consumer centric models, and a hub/gateway is a natural way to approach this. I very much support open source implementations that precede standardization. Specific suggestions: URI schema that can resolve to non IP devices; discovery API; sensor APIs that embed the domain semantics for use cases; and a data base API for storing data for access to historical values. We need good enough security appropriate to the distributed model. Nick shows a slide (Wot is the Scope?) that has a script that receives IoT data. What is the context in which this executes? We need to support legacy devices, so drivers are key and can run on the hub. What is in scope for W3C, e.g. the details of how devices can talk to one another.

What are your views on the roles of different standardization organizations, and the idea that W3C focuses on the application layer and web technologies?

Richard noted that there several pub-sub protocols, and this is likely to continue given that they address different needs. W3C's job should be to bridge the gaps where good solutions are missing. He noted that GS1 has been effective with standards for identification (Parties and Product), but ISO's work on legal identifiers for organizations by contrast looks likely to fail. Standards work needs to be solidly based on real world implementation experience. Eric added that interoperability is an area where different orgs have a shared interest to collaborate.

Mike said everything is getting connected, and the IoT is encroaching on everyone's domains. We're looking at how to partner. Thomas added the question is also about acceptance by the target communities. If standards are perceived as too heavy they won't be accepted. Istvan responded it is all about collaboration with other SDO's. We've recently signed a liaison agreement with W3C and are proud about that. We want to bridge SIMs to web services and applications. Nick said anything W3C does needs to work with existing and well established standards, e.g. MQTT and CoAP. I am not sure that W3C can help much with device management.

Claes invites questions from the audience.

What can we say about the architecture, and what we can standardize?

For example the role of a gateway/hub to proxy for the actual device. This avoids the need for W3C to deal with constrained devices, right or wrong? Mike responded Zigbee, Z-wave or Bluetooth all have well defined profiles, enabling higher level abstractions involving simple data models. One of the challenges is moving away from the few really large companies who currently deal with this and opening it up. Richard added the reality is that there will be multiple architectures that address different needs.

There are billions of devices. will standardization succeed? (Kemal Delic, HP)

Richard replied that global standards have proven to be very effective, e.g. the number power connectors across the world is relatively few.

Claes asks Eric to elaborate about how interoperability relates to semantics? Eric replied that some guidelines will help to encourage interoperability. Richard added that this will help but won't make the problem go away.

What about Nick's proposal that URIs be extended to support non IP devices -- any suggestions?

Claes then asked about the challenge of identifiers and authentication. Nick said that practically, if I want to address something in my house, the closest is my router's IP adress which changes regularly. We need a scheme that is persistent.

Mike added that privacy always comes up in regards to this subject. Nick where do you see this fitting in? Nick replied that it is a matter of how can obtain these URIs and gain access to them. Having the address isn't enough to get the data. Richard then said that there may be devices that you don't want the rest of the world to know about. This is a very hard problem particularly in regions like the EU with strong privacy laws. Concern about identifiers being handed off to 3rd parties with a resulting loss of privacy. Nick responded that it is essential that there be role based access control in place. Thomas added XRIs are cool as you can embed domains within them (talks about books and libraries).

Eric said we're looking at the URI approach with GS1 GTIN+ on the Web Work Group. I assume Nick is talking about a specific instance of a TV, right [yes]. Jeff Jaffe (W3C) said I'm familiar with the problem of forum shopping. There are real challenges for interoperability. I see a lot of technical clutter getting in the way.

How do we stop the competing so called standards with an answer other than you don't?

Richard replied that is only possible within a single company. We should address this by focusing on bridges and cross SDO coordination. We've been successfully doing this for cloud related standards, but it is a significant effort. Nick added that SDO staff may be rewarded for covering new areas, which is a potential problem. The way to stop this is to move forward quickly. It is common to see companies coming together to promote something rather than actually working on standards.

Richard added that companies want to promote their own approach to reduce their costs. Charalampos said public URIs are subject to DDOS. Nick added I want to be able to ask which of the devices in my house support a given service. Phil Archer (W3C) stated that whether a URI is for a product or a service is a matter of design. There are good and bad identifiers. Is the cultural expectation that a URI is a web page a barrier? Richard responded no, most people aren't even aware of the term.

Panel on conclusions and next steps
  • Moderator; Dave Raggett, W3C
  • Philipp Hoschka, W3C [slides]
  • Milan Milenkovic, Intel [slides]
  • Ryuichi Matsukura, Fujitsu [slides]
  • Milan Patel, Huawei
  • Dominique Guinard, Evrythng
  • Laurent Walter Goix, Telecom Italia, [slides]

The session started with a presentation by Philipp Hoschka, Deputy Director of W3C. W3C's standards (W3C Recommendations) are produced in Working Groups. He introduced the distinction between Working Groups, Interest Groups and Community Groups and proposed the formation of a Web of Things Interest Group, citing the existing Web and TV Interest Group as a model. The Interest Group establishes requirements, and hands these off to existing Working Groups or to new Working Groups as appropriate. The Web and TV Interest Group consists of several Task Forces for different topics, and with different Task Force leads. Interest Groups have minimal IPR requirements making it easier for companies to get approval to join and participate in their work. If collaborative work on specifications is to be done prior to launching a Working Group, then this could be done in narrowly scoped Community Groups. The requirements for an Interest Group include:

  • Charter
  • Chair(s)
  • Task force leaders
  • Document editors
  • Contributors

What are your thoughts on what is the key to breaking free from product silos?

Dominique answered that you need to make it as simple as possible but not too simple. You need to make things simple to use, select the right tools etc.

What are you suggestions for how to bridge the gap between different cultures of stakeholders, e.g. Web hackers, linked data community etc.?

Laurent suggested that JSON-LD was a step in the right direction. Dave wondered if one way to approach this is via task forces focusing on different areas. Milan Patel said that you need to identify the use cases where the communities need to interact with them. Milan Milenkovic added that you need test beds to inform standards work based upon practical experience, but you can also benefit by learning from the academic community. Ryuichi noted that we should collect opinions from device manufacturers.

What are your priorities for W3C to tackle the WoT?

Milan Milenkovic responded that this workshop has met my expectations, what seems to be needed etc. I am trying to practice what I preach, deploy some actual prototypes etc. informing the thinking as you go. The field is so broad, defining the standards is really hard. Milan Patel emphasized the need to identify some use cases where you can understand the commonalities between industries. You need to identify what we already have that can be used, as this is more important than creating new standards. Yesterday we emphasized that whatever we do we need to do it fast. The W3C should issue some Best Practices. These are the technologies that can be used for the WoT, with guidance for how to use them.

Laurent agreed, adding what is the real focus of the community - that needs to be found. We should scout for what is already there. We could have a Task Force on data modeling, another on service modeling etc. I'd also like the ensure that the user is in the loop, as we shouldn't forget that we have machine to person communication. We should aim to link worlds. including security and privacy.

Ryuichi said the definition of the "Things" is important. We need to established a shared view of the whole architecture.

Dominique added that he would like to see a mix of what everyone has said. A set of guidelines and BPs would be useful: We published a White paper on that in 2008 (further publications). It was downloaded many times. There is a need for something simple to explain the core. We have a lot of people in the room who could do that. We need to react very quickly as it will happen with or without W3C. The standardization method is more for going beyond the current, into a sustainable future.

Dave asked for opinions from the audience.

Redouane Boumghar responds I led a session on robots. We didn't talk too much about modeling actuators, only services. Shadi Abou-Zahra (W3C) said: excellent workshop, congratulations. I do think that the work should be at W3C for its high commitment to accessibility, internationalization etc. We're here to help you create the data, the vocabs, the services etc including accessibility use cases, helping establish requirements to make sure Web is available to everyone.

Jörg Heuer said: the diversity of people here is amazing. There are a lot of perspectives. We need to pay attention to structure to talk about the same topic. It's important to discuss the different backgrounds and applications to make sure we're talking about the same thing. Dave finished by thanking everyone for coming and participating. Thanks to especially to Siemens and Jörg for hosting us!

Next Steps

This section describes proposals for next steps and follows on from the discussions in the closing Panel on conclusions and next steps.

The Web of Things is expected to develop into a huge market of services, and W3C is well positioned to develop open standards around service platforms, security, privacy, and integration with the Web of data for a Web of services. This positions W3C to define standards at the service and application layer, complementing the role of other standards development organizations and industry fora that are focusing on the device layer and the array of communication technologies for accessing sensors and actuators, that form the Internet of Things.

The workshop gave a strong message of support for W3C to initiate work on standardization for the Web of Things. The foundations include RESTful HTTP and pub-sub protocols, but the detailed requirements vary across the use cases, e.g. for latency, throughput, transactional robustness and so forth. Building upon these foundations, the workshop identified the need for standards for Web APIs that abstract away from the protocols, including the wide range of IoT technologies used at the network edge to connect to sensors and actuators. Through standardization, we can encourage re-use of APIs and data models.

The closing session of the workshop raised a proposal to launch a W3C Interest Group. The existing W3C Web and TV Interest Group provides a model for what we can expect for the Web of Things Interest Group. It has established requirements and handed work off to existing Working Groups or launched new Working Groups where appropriate. The Interest Group consists of several tasks forces for different topics with different leads for each. W3C Interest Groups work on gathering use cases and requirements, surveying existing work, and identifying gaps, general introductions and best practices. There are no intellectual property commitments as a precondition of participation. A consequence of this is that Interest Groups can't work on specifications. However, Interest Groups can produce charters for specification work in W3C Working Groups.

The proposed Web of Things Interest Group will require:

  • A formal charter
  • Chair(s)
  • Task force leaders
  • Document editors
  • Contributors

The W3C staff will drive the process of chartering the Interest Group through reaching out to a wide range of interested stakeholders. This consultation will seek to clarify the mission for the Interest Group, the target topics and industry priorities, and opportunities for liaisons with other related standards development organizations.

Some of the potential topics for task forces include:

  • Use cases and requirements for different application domains
  • A scalable architecture for the Web of Things
  • Survey of existing practices and standards relevant to the Web of Things
  • Requirements for open markets of services for the Web of Things
    • Data modeling
    • Service descriptions and dependencies
    • Discovery and trust management
    • Bridging the gap between the Web of Things and the IoT
    • Relationship between app/service layer and network layer
    • Monetization
    • Provisioning and lifecycle management
  • Security and Privacy for the Web of Things
  • Requirements and best practices for integrating the Web of data
  • The role of social relationships between people, places and things
  • Collaborations with other standards development organizations

The W3C welcomes feedback on this report, which should be sent to Dave Raggett <dsr@w3.org>

Participation

There were 119 registered participants.

The following shows the different categories of organizations who provided submissions to the workshop.

Companies
Startups
Universities
Research Institutes
Standards Organizations, Industry Associations, and Government Institutions etc.
Nationalities

Based upon the registration data we have the following distribution of nationalities:

  • 28 Germans
  • 15 French
  • 12 Japanese
  • 11 American
  • 10 British
  • 6 Italian
  • 6 South Korean
  • 4 Swiss
  • 3 Austrian
  • 3 Swedish
  • 1 Bolivian
  • 1 Bosnia and Hercegovina
  • 1 Brazilian
  • 1 Chinese
  • 1 Finnish
  • 1 Greek
  • 1 Hungarian
  • 1 Indian
  • 1 Norwegian
  • 1 Portugese
  • 1 Spanish

Note: that in hindsight we should have asked for country of residence rather than nationality!