14:03:14 <RRSAgent> RRSAgent has joined #wot-td
14:03:14 <RRSAgent> logging to https://www.w3.org/2021/09/08-wot-td-irc
14:04:08 <kaz> meeting: WoT-WG - TD-TF
14:04:31 <kaz> present+ Kaz_Ashimura, Cristiano_Aguzzi, Michael_McCool
14:04:40 <cris> cris has joined #wot-td
14:05:42 <dape> dape has joined #wot-td
14:05:57 <sebastian> sebastian has joined #wot-td
14:06:44 <kaz> present+ Daniel_Peintner, Sebastian_Kaebisch
14:07:09 <kaz> present+ Ben_Francis
14:07:17 <dape> scribe: dape
14:07:31 <kaz> present+ Michael_Koster
14:07:39 <kaz> rrsagent, make log public
14:07:43 <kaz> rrsagent, draft minutes
14:07:43 <RRSAgent> I have made the request to generate https://www.w3.org/2021/09/08-wot-td-minutes.html kaz
14:08:16 <kaz> chair: Sebastian
14:08:43 <kaz> Agenda: https://www.w3.org/WoT/IG/wiki/WG_WoT_Thing_Description_WebConf#Sept_8.2C_2021
14:09:37 <dape> TOPIC: Minutes review
14:09:47 <dape> -> https://www.w3.org/2021/09/01-wot-td-minutes.html
14:10:23 <dape> <Sebastian goes through the minutes>
14:15:57 <kaz> present+ Tomoaki_Mizushima
14:16:45 <dape> MMC: note that we should not forget about IANA registration for thing models
14:17:00 <dape> -> https://github.com/w3c/wot-thing-description/issues/1221
14:17:32 <kaz> rrsagent, draft minutes
14:17:32 <RRSAgent> I have made the request to generate https://www.w3.org/2021/09/08-wot-td-minutes.html kaz
14:20:34 <dape> SK: no objections to merge minutes -> approved
14:20:45 <dape> TOPIC: PRs Overview
14:21:07 <dape> SUBTOPIC: Signature
14:21:12 <dape> -> https://github.com/w3c/wot-thing-description/pull/1151
14:21:34 <dape> MMC: had long discussion ... what if IETF comes up with another solution
14:21:48 <dape> ... suggest to have an external document
14:22:03 <dape> ... spin off IETF RFC
14:22:14 <dape> ... canonicalization normative in TD
14:22:33 <dape> ... signature as external document
14:22:53 <dape> ... we should not merge PR#1151
14:23:30 <dape> ... plan to create new note and will let people know once this is done
14:25:57 <dape> ... see discussions in https://github.com/w3c/wot-security-best-practices/issues/14
14:26:34 <dape> SK: Okay, lets close PR#1151
14:27:16 <dape> MMC: Need to create signature extension
14:29:05 <dape> SK: FYI, work from Oliver will be taken over by successor with IETF background
14:29:08 <kaz> present+ Fady_Salama
14:29:26 <dape> MMC: Yes, need proper security review
14:30:11 <dape> SUBTOPIC: Add queryaction and cancelaction operations
14:30:15 <dape> -> https://github.com/w3c/wot-thing-description/pull/1208
14:30:26 <dape> Ben: naming is still WIP
14:31:02 <dape> ... distinction between query action queue and instance
14:31:21 <dape> ... subscribe to event is somewhat similar
14:31:57 <dape> ... might need "queractioninstance" ... but seems clumsy
14:32:14 <dape> s/"queractioninstance"/"queryactioninstance"
14:32:27 <dape> ... would like to get feedback from group
14:33:12 <dape> SK: no strong opinion on naming
14:33:51 <dape> Ben: unfortunately no real word in English that differs between instance
14:34:13 <dape> ... that's why I use "queryactioninstance"
14:35:48 <dape> ... see recent proposal, https://github.com/w3c/wot-thing-description/issues/302#issuecomment-915277793
14:37:50 <dape> SK: the term "instance" makes clear we deal with an instance... on the other hand the name is rather long
14:38:47 <dape> DP: proposal "queryanaction" ?
14:38:53 <dape> Ben: even more confusing ;-)
14:39:33 <dape> SK: "queryactionS" with plular?
14:39:46 <dape> Ben: was part of the original proposal
14:40:55 <dape> SK: Suggest to make a decision
14:41:05 <dape> CA: Agree, matter of taste
14:41:13 <dape> ... like the current proposal
14:41:42 <kaz> rrsagent, draft minutes
14:41:42 <RRSAgent> I have made the request to generate https://www.w3.org/2021/09/08-wot-td-minutes.html kaz
14:42:00 <dape> MMC: to me instance means one created resource
14:42:34 <dape> ... need clear definition
14:43:04 <dape> ... ok with current terms
14:43:34 <dape> MK: No objections either
14:44:10 <dape> FS: fine
14:44:28 <dape> Kaz: +1
14:44:39 <dape> SK: Looks we have a resolution
14:45:25 <dape> ... queryactioninstance, queryaction, queryallactions
14:45:48 <dape> Ben: cancelactioninstance ok too?
14:47:51 <dape> ... separate issue with meta interactions, suggest to create separate PR
14:48:32 <dape> ... update PR#1208 with the aforementioned terms
14:49:19 <dape> ... separate PR for meta interactions (top level forms, Issue#1200)
14:49:22 <mjk> mjk has joined #wot-td
14:49:51 <dape> ... -> https://github.com/w3c/wot-thing-description/issues/302#issuecomment-915307097
14:50:40 <dape> CA: Plan looks good
14:50:43 <dape> ... later I will make another proposal
14:50:59 <dape> SK: Okay, let's look at it later
14:51:30 <dape> SUBTOPIC: fix: maxItems,minItems are xsd:nonNegativeInteger
14:51:36 <dape> -> https://github.com/w3c/wot-thing-description/pull/1212
14:52:20 <dape> DP: added missing fixes and checked with Victor
14:52:39 <dape> ... IPR issue because of original author
14:54:04 <dape> Kaz: marked as non substantial
14:54:11 <dape> SK: let's merge
14:54:40 <dape> TOPIC: Yet another action model
14:54:52 <dape> SK: Cristiano please go ahead
14:55:09 <dape> CA: It is a patchwork of the other proposals
14:55:26 <dape> <Cristiano goes through slides>
14:56:06 <dape> ... actions may be more complex than other interactions
14:56:17 <dape> ... query, cancel et cetera
14:56:38 <dape> ... there can be an action queue
14:56:44 <dape> ... or a single action
14:56:56 <dape> ... we have different use cases
14:57:04 <dape> ... * webthings.io
14:57:38 <dape> ... * BRAIN-IOT
14:58:20 <dape> ... * OPC UA
14:59:14 <dape> ... at the moment we have 3 proposals (Victor, Ege, Ben)
14:59:35 <dape> ... dynamic TDs by Victor
15:00:11 <dape> ... issues: caching, security, no static description
15:01:01 <dape> ... w.r.t. new ops and URI templates we had concerns about how consumers understand it
15:01:10 <dape> ... *New* Proposal
15:01:25 <dape> ... leverage on thing models and action TDs
15:02:01 <dape> ... idea is to use ThingModel to describe further operations ... after invokeaction operation
15:02:46 <dape> ... after invokeaction a virtual resource is created and a TD could describe the new thing
15:04:03 <dape> ... action does not return TD by itself
15:04:38 <dape> ... a variable "fills" the mapping
15:05:05 <dape> q+
15:05:58 <dape> DP: how to find variable ?
15:06:06 <dape> CA: there is mapping field
15:06:12 <dape> ack dape
15:06:47 <dape> MMC: Note about absolute pointers in mapping
15:07:21 <dape> CA: JSON pointer refers to output one gets
15:07:46 <sebastian> q+
15:08:00 <dape> CA: about green field
15:08:27 <dape> ... core profile could work without model
15:08:40 <dape> ... model could still be a validation hint
15:10:36 <dape> CA: Pros
15:10:55 <dape> ... no id tracking, embedded in ThingModel
15:11:15 <dape> ... lightweight
15:12:23 <dape> ... backward compatible
15:12:46 <dape> ... flexible (green and brown field devices)
15:13:29 <dape> ... compact
15:14:49 <dape> CA: Cons
15:15:15 <dape> ... new op types needed? Yes, need some but not all
15:16:52 <dape> ... queues might be a static resource
15:17:03 <benfrancis> q+
15:17:28 <dape> SK: I like the proposal a lot
15:17:44 <dape> ack sebastian
15:19:02 <dape> ... action to me means being kind of a produce.. with new functions
15:19:24 <dape> ... anhow, would keep output and input for core profile
15:19:52 <dape> ... but have a way to indicate or point to the TD
15:20:19 <dape> ... in general I like the idea a lot
15:21:09 <dape> ... additional features like cancel/monitor action is in a separate TD
15:22:03 <dape> CA: having "model" in action "output" tries to solve 2 issues, but we can put "model" a level up
15:22:04 <dape> q?
15:22:05 <sebastian> q?
15:22:08 <sebastian> ack b
15:22:15 <dape> Ben: Clever idea
15:22:27 <dape> ... I would say action is not a thing
15:23:03 <dape> ... maybe better to have ActionDescription than TD
15:23:32 <dape> ... moreover, a TD would not be self-contained any longer
15:23:56 <dape> ... biggest question: how to deal with multiple consumers?
15:24:23 <dape> CA: core profile could live without URL
15:24:36 <dape> ... brown-field devices would not be supported
15:24:42 <dape> Ben: case: 2 consumers
15:24:48 <dape> ... 1. consumer invokes action
15:25:01 <dape> ... 2. consumer wants to monitor
15:25:26 <dape> ... how does separate consumer get the TD for the action
15:25:33 <kaz> q+
15:26:23 <dape> CA: Need action queue
15:26:41 <dape> ... and infer thing description
15:27:54 <dape> Ben: wonder about relation between BASE_ADDRESS in example
15:28:28 <dape> CA: action return value needs to have information about the variable
15:31:07 <dape> Ben: Ok, got general idea
15:32:34 <kaz> q?
15:33:50 <dape> DP: ActionTD being a real TD makes it possible to use any existing tool
15:34:16 <dape> Kaz: TD is a mechanism to describe thing capability
15:35:09 <dape> ... model view and controller is not clear to me in this proposal
15:35:44 <dape> CA: we do not embed control
15:35:53 <dape> ... it is a meta model
15:36:44 <dape> Kaz: Should clarify the entities and features
15:37:43 <dape> ... and clarify our expectations
15:39:07 <dape> q?
15:39:11 <dape> ack kaz
15:40:48 <benfrancis> q+
15:41:34 <dape> Ben: w.r.t. Charter... the charter should cover this proposal under the current charter
15:42:00 <kaz> s/model view and cotroller/the separation and assignment of Model-View-Control/
15:42:11 <dape> ... a new resource is created once an action is invoked
15:42:24 <dape> ... in my proposal it is in TD
15:42:38 <dape> ... in Cristianos proposal it points to new TD
15:42:43 <dape> ... both seem valid
15:42:54 <dape> SK: How can we proceed?
15:42:55 <kaz> s/Cristianos/Cristiano's/
15:43:14 <dape> ... suggest to get some experience with the new proposal
15:43:32 <dape> ... maybe in PlugFest... will happen in some weeks
15:44:08 <dape> CA: Where to put the slides and the information?
15:44:26 <dape> ... experiments are valuable
15:45:09 <dape> SK: Use Ben's proposal as working assumption
15:45:23 <dape> ... or come back to the new proposal
15:46:05 <dape> MMC: in PlugFest we can add prototype
15:46:34 <dape> ... I like the hypermedia approach
15:46:36 <dape> q+
15:46:41 <dape> ack benfrancis
15:47:19 <dape> ... node-wot prototype would be fine
15:48:03 <dape> ... maybe both approaches are complementary
15:50:03 <dape> DP: Suggest to wait for Ege/Victor also since they provided a proposal also
15:50:19 <dape> ... using Ben's proposal as working assumption is fine
15:50:26 <dape> ... add slides to GitHub
15:50:58 <dape> SK: ask Cristiano to upload slides or create new issue
15:52:04 <dape> ... have some more discussion also next week
15:52:20 <dape> ... but maybe stick for the time being with Ben's proposal
15:52:26 <benfrancis> cris: Note there are some existing proposals in https://github.com/w3c/wot-thing-description/tree/main/proposals - your slides could go there?
15:52:47 <dape> CA: I like also the comment that we may need both in the end
15:53:28 <cris> benfrancis: I was thinking about that, I'll post them there :)
15:53:53 <dape> SK: Let's give anybody a chance to take a look the next days
15:53:59 <dape> CA: Makes sense
15:54:08 <dape> TOPIC: Issues
15:54:26 <dape> SUBTOPIC: Describing initial connection
15:54:30 <dape> -> https://github.com/w3c/wot-thing-description/issues/878
15:54:53 <dape> SK: 1. aspect, define forms optional
15:55:26 <dape> ... 2. forms on root providing endpoint information
15:56:04 <dape> ... avoids cluttering the local forms with the same information over and over again
15:56:33 <dape> ... if global forms contain enough information no local forms needed any longer
15:56:55 <dape> SK: objections/concerns w.r.t. this change ?
15:57:44 <dape> ... might run into interop issues ... since local forms are mandatory at the moment
15:59:08 <dape> ... however, we can define a process that creates an 1.0 interoperable TD
15:59:17 <benfrancis> q+
15:59:22 <dape> ack dape
16:00:12 <dape> Ben: proposing local forms optional always or depending on protocol like websocket
16:00:33 <dape> SK: would not say that for a specific protocol..
16:00:37 <dape> MMC: ECHONET
16:00:48 <dape> CA: modbus is another
16:01:03 <benfrancis> ack benfrancis
16:01:26 <kaz> q+
16:01:28 <dape> MMC: breaking compatibility might be an issue
16:01:46 <dape> ... but there are not that many 1.0 devices
16:01:55 <dape> ... so it seems okay to me
16:02:10 <dape> Kaz: Agree with McCool
16:02:26 <dape> ... suggest to see what happens during PlugFest
16:02:42 <dape> SK: old implementation will not accept the new TD
16:03:05 <dape> ... I suggest to put a note in the spec mentioning this fact
16:03:25 <dape> ... are we still OK to make local forms optional?
16:03:28 <dape> q?
16:03:32 <dape> ack kaz
16:03:51 <dape> Kaz: Suggest to see the impact in PlugFest
16:04:29 <dape> MMC: Can add it as "at-risk"
16:04:43 <dape> ... suggest to go ahead
16:05:18 <dape> SK: new op value "open"
16:05:38 <dape> ... meant to indicate that it is meant to open connection
16:06:22 <dape> ... see https://github.com/w3c/wot-thing-description/issues/878#issuecomment-915370832
16:07:26 <kaz> [adjourned]
16:07:31 <kaz> rrsagent, draft minutes
16:07:31 <RRSAgent> I have made the request to generate https://www.w3.org/2021/09/08-wot-td-minutes.html kaz