[Iot-lacnog] FW: [IRTF-Announce] New Research Group chartered: Thing-to-Thing

[Alvaro Retana (aretana)]

FYI..

Estoy seguro que este nuevo grupo será de interés para la comunidad. :-)

Saludos!

Alvaro.


On 12/2/15, 3:42 AM, "IRTF-Announce on behalf of Eggert, Lars"
<irtf-announce-bounces at irtf.org on behalf of lars at netapp.com> wrote:

>Thing-to-Thing Research Group (T2TRG)
>https://irtf.org/t2trg
>
>
>The Thing-to-Thing Research Group (T2TRG) will investigate open research
>issues in turning a true "Internet of Things" into reality, an Internet
>where
>low-resource nodes ("things", "constrained nodes") can communicate among
>themselves and with the wider Internet, in order to partake in
>permissionless
>innovation. The focus of the T2TRG are on issues that touch opportunities
>for
>standardization in the IETF, i.e., it will start at the adaptation layer
>connecting devices to IP, and end at the application layer with
>architectures
>and APIs for communicating and making data and management functions
>(including
>security functions) available.
>
>Motivation
>
>A first wave of IoT standards has been completed by the IETF. Consortia
>are now
>forming to build infrastructure and industry agreements around those. This
>generates new requirements for research, based on actual usage of the
>standards
>now available.
>
>In parallel, W3C has set up an IG (Interest Group) on the Web of Things
>(WoT);
>this is operating on a similar timeline as an RG does in the IRTF. A
>parallel
>activity associated with the IETF/IRTF can help ensure the common work
>does not
>stop at the traditional boundaries of the W3C, but considers networking
>issues
>as well. IETF WGs are also active on these topics, but an IRTF RG can
>provide a
>more long-term perspective to the collaboration.
>
>Areas of Interest
>
>A number of areas of interest have been identified:
>
>  * Understanding the motivation for single-purpose silos and gateways;
>    facilitating a move towards small pieces loosely joined (hence
>    "thing-to-thing"); enabling scaling the number of applications in a
>single
>    network
>  * Deployment considerations; scaling considerations; cost of ownership
>  * Management and operation of "things"
>  * Lifecycle aspects (including, but not limited to, security
>considerations)
>  * Cooperation with W3C, e.g., on data models, formats, and semantics
>
>More exploratory areas of interest include:
>
>  * Operating "things" that have multiple masters/stakeholders (including
>    understanding role definitions of devices, owners, operators, etc.)
>  * Exploring the duality of state- and event-based approaches
>  * Aspects of distribution (cf. "fog computing"); reliability and
>scalability
>    considerations
>  * Containerization and other forms of mobile code
>
>Other Objectives
>
>Besides facilitating communication between researchers in its areas of
>interest, T2TRG pursues a number of additional objectives:
>
>  * Definition of "benchmark" or "reference" environments:
>      * to enable regular plugfests, and
>      * as a basis for repeatable, comparable research.
>  * Description of practical, real world, cross domain applications of
>    connected "things"
>  * Taxonomy, technology survey, and best practice documents
>  * Fostering collaboration with industry fora and other organizations on
>    networking of "things"
>
>These objectives will be achieved making use of a close involvement
>between the
>IETF community and the T2TRG. For the IETF, some RG documents may
>simplify the
>generation of (or even serve as) use case documents or other informational
>references. Close contact will be maintained with the IETF's IoT-related
>WGs
>and its IoT directorate, as well as related IRTF research groups.
>
>Name and Organization
>
>In the context of this RG, thing refers to a connected object that is also
>defined by its interface to the physical world. Some "things" may be
>connected
>only using a local interconnect, but the RG's focus is on "things" that
>are
>part of the Internet. Not all "things" are highly constrained, but IoT
>architectures generally need to enable the inclusion of constrained nodes.
>
>The Thing-to-Thing Research Group plays on a reminiscence to the
>End-to-End
>Research Group (END2END), which accompanied the evolution of the Internet
>for
>its first three decades. (The T2TRG is certainly not intended to literally
>exclude the communication of constrained nodes with the wider Internet or
>with
>more than one other constrained node.)
>
>The charter of the END2END RG mentioned that "the specific topics of
>interest
>(...) have changed as network research has advanced, but it is generally
>concerned with end-to-end services and protocols implemented in hosts"
>??" a
>focus similar to that of the T2TRG, except that the concern for packet
>forwarding and network building may be closer to a constrained node than
>it was
>to a traditional Internet host.
>
>Also, T2TRG will not be a closed RG, but employ an open membership policy
>towards all interested researchers, taking its clues from the way that,
>say,
>the DTNRG or the ICNRG have been and are being run.
>
>Meetings
>
>Meetings will often take place co-located with IETF meetings, and/or with
>meetings of related organizations such as W3C WoT IG. Organizing
>Workshops at
>research conferences is also envisioned.
>
>Specific meetings of the T2TRG will pick focus items that much of the
>discussion of the meeting should revolve around. The stated focus for the
>initial meetings was on three immediate areas of interest:
>
>  * Management and Operation of networks involving constrained nodes
>  * Security and Lifecycle aspects in constrained nodes
>  * REST and pub/sub: Exploring the state-event duality
>
>Future work items may include:
>
>  * Documents
>      * Guidance for designing REST-based IoT Applications ("cookbook")
>      * Security considerations for the IoT
>      * A survey of security bootstrapping approaches
>  * Activities
>      * Preparing and running "plugRESTs" (with support from the documents
>        above, in cooperation with the W3C IG WoT; preparing work on
>reference
>        and evaluation frameworks), eventually leading to:
>          * plugRESTs: lessons learned