W3C

- DRAFT -

WoT IG/WG

12 Apr 2017

Agenda

See also: IRC log

Attendees

Present
Kaz_Ashimura, Michael_Koster, Ari_Keranen, Daniel_Peintner, Dave_Raggett, Feng_Zhang, Johannes_Hund, Kunihiko_Toumura, Matthias_Kovatsch, Michael_McCool, Niklas_Widell, Takeshi_Yamada, Keiichi_Tokuyama, Uday_Davuluru, Yongzing_Zhang, Yingying_Chen, Michael, McCool, Achille_Zappa, Masato_Ohura, Zoltan_Kis, Katsuyoshi_Naka, Kazuaki_Nimura
Regrets
Chair
Matthias, Michael, Yongjing
Scribe
Yongjing, Kaz

Contents


Agenda

<inserted> scribenick: kaz

Matthias: Osaka meeting and TF reports

Ari: management interface discussion today?

Matthias: not sure given we don't have Zoltan

Ari: unfortunately, can't make the Osaka meeting

Matthias: we can continue the discussion on the list

Osaka F2F

<kaz> scribenick: yongjing

Matthias: anyone from Panasonic?
... deadline for F2F registration? regarding dinner and room

Ohura: Apr 21st is the deadline

registration site

Matthias: how many people can be invited? what's the room size for the open day?

Ohura: the capacity is 100 people
... dinner shares the same deadline

Matthias: currently 21 attendees, 2 remote participants

kaz: can accept late registration for the meeting?

Ohura: ok

McCool: late registration is not applicable to dinner reservation

Osaka f2f wiki

Matthias: reminder for TF leads: put breakout agenda for F2F
... Matthias: if multiple slots needed, please reserve multiple bullets
... last week of Apr should freeze the agenda input
... binding template heavily rely on liaison. do we have input by Osaka? OCF for instance?
... what's the ocf plugfest status?

Koster: it's ongoing. don't exact status now. Would work with Zoltan et al.
.... will drive that in the next weeks

Matthias: we dont have many representitives to report TF status

<inserted> scribenick: kaz

Yongjing: open day and plugfest in parallel?

Matthias: better to concentrate on plugfest on Monday

Yongjing: ok
... may bring some information on oneM2M for Open Day

McCool: can bring some example of OCF devices

Koster: we can see what we can within a few weeks

Scripting TF report

<inserted> scribenick: yongjing

Matthias: ask Zoltan to discuss the management i/f

Zoltan: on Monday scripting call, discussed mgmt i/f
... reluctant to call it 'mgmt i/f'

<kaz> Scripting minutes

Zoltan: would be very useful to discuss based on common use cases..

scripting API sit on top of security

McCool: API to install the scripts need to be protected.

Zoltan: we don't have it anymore

we should revisit some assumptions, in security document

we could discuss security architecture, provisioning ... in F2F or in prior

Ari: a lot of attraction of LWM2M in addition to OCF/oneM2M

Zoltan: we have looked in to LWM2M recently
... security TF will first clarify what are the use cases, actors, ... then think of the underlying platforms

<kaz> 2-runtime(servient) example

local thing and remote thing

kaz: Matthias pointed out this discussion was related to synchronization discussion, and I'd agree to that view
... it might be useful to think of some models including version management capability like git
... runtime B on the right side above may manage actual TD properties using git to handle multiple requests from more than UA runtime
... and think about when/how to merge which request based on the context
... will respond to the email discussion as well

Dave: emphasize the need of collecting use cases

<dsr> some use cases: https://lists.w3.org/Archives/Public/public-wot-wg/2017Apr/0001.html

jhund: we have collected micro use cases for scripting API
... question is can we add scripts at run time?
... this topic may impact plenary
... we should pick up this topic in Osaka

Zoltan: a script is always a deployment scripts, we don't deploy script anymore (?)

Matthias: the two things you're mentioning are equivalent

<dsr> (a thing is often more than just a script)

Matthias: the original assumption is that scripts is always locally creating things

<dsr> +1 to Matthias on embedding the management interface via a thing

Zoltan: it's misunderstanding of my intention
... using existing mechanism avoiding creating additional mgmt API

Matthias: define the problem clearly first, then discuss in the joint call among different TFs

kaz: we can start with the above two run time model proposed by Johannes

Dave: we should first have concrete use case as the discussion ground

McCool: do we have use case template to follow?
... create repository in github/wiki
... suggest to have one session in F2F on use case

Matthias: concerned putting independent use cases together is meaningless
... we need more input on the topic we just discussed for the f2f.

McCool: scripting api, use case can be two agenda items for next call.

[ adjourned ]

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2017/04/12 14:25:58 $