Loading deck

Web of Things
-

reaching out to the physical world

The essence of WoT

"If you cannot explain it simply,
you have not understood it well enough"

- Albert Einstein

Implementations
&
Standards

A chicken and egg problem?

WoT IG F2F Nice:

Quo vadis, Web of Things?

In medias res:

How we started - the WoT Kickoff Workshop, Berlin

Key findings
of the WoT Kickoff

  • The time for IoT is now 
  • Avoid the n+1 - problem
  • Bridging of the silos is #1 priority
  • High diversity of IoT
    (Is both a blessing and a challenge)

TPAC Santa Clara:
finding common patterns

  • collect use cases
  • group by domains 
  • identify patterns
  • find technology candidates

WoT F2F, Munich:
divide and conquer

  • "the web was not invented for cat videos"
    • we cannot find all possible use cases
    • focus on building blocks
  • set up task forces for building blocks
    • APIs & Protocols
    • Thing Description
    • Discovery
    • Security & Privacy

WoT F2F, Sunnyvale:
modelling the Web of Things

  • find a simple, yet powerful first consensus
  •  Thing model
    • Properties
    • Actions
    • Events
  • Architectural deployment models

TPAC, Sapporo:
kicking the tires

  • first exploratory implementations
  • first plugfest
    • interop of different implementations 
    • proof ease of implementation
    • proof of interoperability

WoT IG F2F Nice:

Quo vadis, Web of Things?

Why plugfest?

Early Implementations:

  • provide reality-check and feedback for standardization

  • spark off open-source pojects

  • Provide decision support in technical questions

“We reject kings, presidents and voting. We believe in rough consensus and running code.” - Dave Clark

An Interest group &
specs for their plugfest:

"How far do you want to go?"

  • Thing model
  • Context for description in  JSON-LD
  • Resource model for RESTful Architectures
  • Protocol bindings (CoAP, HTTP, WS)
  • Scripting API
  • Security model, protocol & token formats
  • Thing Registry spec

Proposed: A Technical Spec*
that is not a standard

  • call to evaluate in implementations
  • achieve Interop and early features
  • BUT:
    • consider it as experimental
    • expect it to change
  • provide feedback to standardization
  • Provides concrete materialization for discussions

* = concept exists e.g. in C++ Standardisation