Web of Things Interest Group Teleconference


09 Dec 2015

See also: IRC log


Johannes_Hund, Sebastian_Kaebisch, Joerg_Heuer, Darko_Anicic, Toru_Kawaguchi, Katsuyoshi_Naka, Yingying_Chen, Daniel_Peintner, Dan_Romascanu, Frank_Reusch, Michael_Koster, Oliver_Pfaff, Takuki_Kamiya


Sebstian: This is the last TD TF meeting in this year.
... the agenda is:
... 1. Plugfest logistics
... 2. Proposed plugfest topics
... 3. Tech Landscape

Plugfest logistics

Sebastian show the slides for plugfest logistics.

Sebastian: it was agreed to have the preparation day one day before the plugfest. It will be Jan 25th.
... this is the plan for next plugfest. any questions?


Sebastian: it was on wiki page as well.

<jhund> wiki page: https://www.w3.org/WoT/IG/wiki/F2F_meeting_2016,_January,_26th_%E2%80%93_28th,_France,_Nice#PlugFest

Sebastian shows the wiki page for plugfest plan.

Sebastian: you need to input your name, your organization, your application that you would show. I just copied the table from last plugfest.
... please everyone put your information on it.

Proposed plugfest topics

Sebastian shows the slides for the plugfest proposal.

Sebastian: 1. proposal is like in Sapporo, using TD to some thing discovery;
... 2. to cover the security aspect.
... 3. we need some API on Things.
... 4. from the joint meeting with IRTF, use PlugREST.
... I need some volunteers to write some howto.
... what we need to do in the plugfest.
... I am asking Darko, Oliver, and Matthias to help.
... I will do the first one.
... The 1st idea is how we use the TD to do central-based discovery.
... an example: We have down here different kinds of things.
... each of them has a TD.
... upper layer we have TD repository.
... then we have one kind of client.
... it is able to request some kind of TD.
... How does it look like in interaction?
... The thing is going to do the thing registration.
... to the TD repository.
... in the client slide, open source project, which you could download it.
... 2 ways to do registration: one is thing does it by itself; another is others do it.
... the major goal is to have different TDs in the repository.
... there is a client to look for different kind of things.
... the idea is the TD repository will provide different TDs.
... we need to do howto on the TD.
... it will be on the wiki page of W3C in following days.
... Darko, do you want to share yours>

Darko: Yes. idea is to provide how-to on the Things DI.
... this is very preliminary docs.
... I will show you how the docs will be like and get your comments.Currently it's incomplete yet.
... Here I summarized what this repository can be used for. It can be used by human and things.
... It is also possible to discover certain things based on TD.
... invoke actions on certain things.
... can create, update and delete TD from repository.
... currently we have the reading and querying functionalities.
... if you want to query repository, you need to provide some data like this: URI, IP address, etc.
... there is a payload example.
... This is a kind of response example to the query.
... This is a JSON-LD file.
... for others like updating or deleting TD, we will fill in.

Michael: Resource directory is exactly the same as your repository in our state machine partner.
... Hypercat, SPARQL also used.
... resource directory also has some basic query.

Darko: will you join us on this work?

Michael: sure.

Darko: we will communicate this how to with you.
... we will work together to make examples.

Sebatian asked Oliver to show the security part.

<michael> Resource Directory; https://tools.ietf.org/html/draft-ietf-core-resource-directory-05

Oliver: less than 8 weeks remaining
...add secure communications but do not limit security enabling to that
...Security proposal to adopt IETF ACE Working Group Architecture
... focus on authorizing and authentication
... add AM and AS components to the plugfest
... runs in a less constrained environment
... there will be also a HowTo

<michael> ACE: https://datatracker.ietf.org/wg/ace/documents/

Oliver: also a cheat sheet
... running AM and AS instances (provided by Siemens)

<michael> another REST client:https://chrome.google.com/webstore/detail/advanced-rest-client/hgmloofddffdnphfgcellkdfbfbjeloo?hl=en-US

Oliver shows a demo.

Oliver using a REST client to demonstrate security APIs.

<michael> https://www.getpostman.com/

Oliver interacting with AS and AM to create authorization for a request to the RS

<michael> CoAP client is copper: https://github.com/mkovatsc/Copper

<michael> https://addons.mozilla.org/en-US/firefox/addon/copper-270430/

Oliver: tomorrow we will look at this same thing but more details.
... we can bring it to life by end of next week.

Sebatian: we have also another TF which concentrates on API.
... Daniel will also give some first idea.
... how the API would look like in next plugfest.

Daniel: it's as mature as security.
... idea is to use WebIDL.
... we also will provide javascript APsI.
... we encourage to try out other languages.
... features are based on previous plugfest in sapporo.
... WebIDL and Javascript APsI will be provided by end of next week.
... I can shortly show the javascript APIs made in september.
... very simple APIs that allows you to set or retrieve properties.
... in addition, there will be a howto provided once the APIs are shared.

Michael: is the WebIDL machine-readable?

Daniel: WebIDL is just to provide the methods like prevously javascript APIs provide.

Michael: so it's the standardized format of scripting APIs.
... there will be a WebIDL coming along with TD for everyone to use. Cool!

Johannes: WebIDL is what W3C uses to describe the APIs.

Michael: just wondering general description for general interaction pattern. Interesting.

Sebastian: last topic is plugREST. Please Matthias share with us.

Matthias: this is the so called plugREST by IRTF.
... background info is IRTF is the counterpart of IETF.
... the slides are on github.
... we focus on hypertext driven apps.
... we describes REST apps by URI, media types, link relation types, form relation types, and optionally locations.
... we choose the lighting use case for the PlugREST.
... there is also a draft of this use case on IRTF web page.
... the plan of PlugREST is to define basic components like lights and controllers.
... on the long run we hope to interconnect with other app domains.
... imaging you have the TD and we can get the TD and interact with the thing.
... the client communicates with resouce directory. Then we can follow links to get the TD.
... they are actually decentralized.
... the purpose is to find IoT implementation on this.
... what we implemented is a CoRE HAL to support forms and media types.
... we have the hypermedia client.
... we have the Java base and Javascript Runtime.
... this is an example in JSON.

Matthias shows an example for hypermedia client.

Matthias shows an example of JS in Actinium.

Matthias: the plan for Nice F2F meeting is to make the tech demo to find more implementers.
... to discuss how to harmonize W3C and T2TRG approaches.
... sources not ready yet. We will make public this year.

Michael: We use different media types. but we don't make application specific media types.
... the easiest way is we go to the media directory and query some media types we want.

Matthias: we can have generic media types that have no semantic info in it. Then you coud group them to get specific media type.
... how to balance them is the key point.

Michael: 3rd part is the schema.org.
... there are many different ontologies.
... this is the work on progress and we should continue. We need to do more test to see what we should use.

[some discussion on the generic media types and app specific media types]

Ari: is there any minimum subset that I could implement before PlugREST because I have very limited time.

Matthias: yes. I can try to give you the pointer of it.

Tech Landscape

Sebastian: there is already some info on wiki about plugfest.
... All presentations will be put on it.
... please input your name and organization and the topic you would like to implement into the table.
... please input by end of next week.
... I would like to take a look at the tech landscape now.
... Darko has already transferred the tech landscape from wiki to github.

Darko: We decided to move the landscape to the github in Sapporo meeting.
... everyone can see it and edit it.
... you can see the table containing different consortium and other information.
... We provided some initial structure. Maybe you have better suggestion for the struture. We could discuss it.
... we need more volunteers to work on it.
... if you are now willing to do any part of it, please put your name and we then can move on.

Sebastian: what is the plan? should we put the name now or give several days?

Darko: we can fill it right now if some volunteers are here.

Ari: please put note on wiki that the latest version is on github.

Sebastian: Good point. let's set the timeline to be the end of week.
... Happy Christmas! See you next year!


Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2015/12/10 06:00:19 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.144  of Date: 2015/11/17 08:39:34  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/SPARCL/SPARQL/
Succeeded: s/information/organization/
Found Scribe: Yingying
Found ScribeNick: Yingying
Found Scribe: yingying
Inferring ScribeNick: yingying

WARNING: No "Topic:" lines found.

Present: Johannes_Hund Sebastian_Kaebisch Joerg_Heuer Darko_Anicic Toru_Kawaguchi Katsuyoshi_Naka
Found Date: 09 Dec 2015
Guessing minutes URL: http://www.w3.org/2015/12/09-wot-td-minutes.html
People with action items: 

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report

[End of scribe.perl diagnostic output]