13:02:58 RRSAgent has joined #wot-td 13:02:58 logging to http://www.w3.org/2015/11/25-wot-td-irc 13:03:09 meeting: Thing Description task force 13:03:13 chair: Darko 13:03:17 DarkoAnicic has joined #wot-td 13:03:20 Yingying_ has joined #wot-td 13:03:23 agenda: https://lists.w3.org/Archives/Public/public-wot-ig/2015Nov/0041.html 13:03:38 kaz has joined #wot-td 13:04:13 rrsagent, make log public 13:04:18 rrsagent, draft minutes 13:04:18 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:06:13 DanhLePhuoc has joined #wot-td 13:06:39 scribenick: dsr 13:07:14 Present: Kaz_Ashimura, Dave_Raggett, Ari_Keraenen, Dan_Romascanu, Danh_Le_Phuoc, Daniel_Peintner, Darko_Anicic, Francois_Daoust, Frank_Reusch, Kevin_Gordon, Maxim_Kolchin, Michael_Coster, Sato_Yasunori_, Takuki_Kamiya, Yingying_Chen 13:07:24 present+ DanhLePhuoc 13:07:45 dape has joined #wot-td 13:08:17 Darko introduces the agenda. 13:09:47 akeranen has joined #wot-td 13:10:49 any other agenda items beyond my email? 13:11:01 [no] 13:11:12 -> https://lists.w3.org/Archives/Public/public-wot-ig/2015Nov/0041.html Darko's agenda 13:11:41 Darko says that Oliver would like to join our next call on behalf of the Security TF 13:12:20 He wants to chat about which security topics are in scope for thing descriptions. 13:12:31 If there is time today we could discuss this 13:12:44 Topic: Discussion on limitation of open world assumption for data models 13:13:02 bengo has joined #wot-td 13:13:12 dsr: W3C resource description for meta data 13:13:17 kawaguch has joined #wot-td 13:13:28 ... can be scalable 13:14:03 ... instances can override 13:14:13 ... this is a set of values 13:14:26 ... very difficult to handle with RDF 13:15:15 kaz: Dave, do you have any concrete resource at the moment? 13:15:21 dsr: not now 13:16:05 darko: that is for semantics of the language? 13:16:29 dsr: how to interpret RDF triples 13:17:20 darko: not sure about the relationship with the TD at the moment 13:17:35 dsr: if we want to build process 13:18:02 ... maybe we need to have clarity 13:18:07 dromasca has joined #wot-td 13:18:14 ... there is quite a lot of work 13:18:37 danh: what kind of semantic profiles should be applied to our work? 13:19:02 ... RDF? OWL? 13:19:19 ... we have resource constrained requirements 13:19:32 ... RDF logic profile would be enough 13:19:47 ... object oriented model might be good 13:20:16 dsr: large number of things have similar profiles 13:21:03 michael: you can't know about all possible classes 13:21:14 ... you need to restrict classes 13:21:33 ... appropriate reasoning for the actual world 13:21:57 ... OWL simply specifies domain specific details 13:22:14 dsr: if we want to formalize TD, this is something we have to address 13:22:29 danh: schema.org 13:22:52 ... mostly semantic description done using RDF 13:22:55 present+ Toru_Kawaguchi 13:23:18 ... object oriented hierarchy would be good 13:23:43 RDFS semantics http://www.w3.org/TR/rdf-schema/ 13:23:58 Darko: RDFS can be used to support closed world reasoning and may be sufficient for thing data models. 13:24:20 i/Darko:/scribenick: dsr/ 13:24:35 i/W3C resource/scribenick: kaz/ 13:24:42 rrsagent, draft minutes 13:24:42 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:25:51 Dave: the IG needs to add a task to work on use cases and examples and the requires that emerge. 13:26:37 Danh: we should consider the logic profile and the level of computational load involved so that we can see what’s practical for resource constrained devices. 13:27:36 Louay has joined #wot-td 13:28:01 Dave: I think we should aim to cover this in the IG report we’re expecting to produce early next year. 13:28:18 Dark: I will take a note to ensure that we can track this. 13:28:27 s/Dark/Darko/ 13:28:52 Dave thanks Darko for placing this on the agenda today. 13:29:06 rrsagent, draft minutes 13:29:06 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:29:25 I can help Dave with that 13:30:31 Michael: I very much agree with Dave on the need to collect the use cases and requirements, there could be some interesting examples, e.g. discovery and filtering. People should be able to add use cases as they come across them. 13:31:02 q+ 13:31:39 Darko: discovery is one of the use cases effected by the open/closed world assumption. 13:32:06 Another deals with default values for properties and method arguments. 13:32:20 present+ Louay_Basbouss 13:33:33 Michael: discovery is one area, another is overrides for an instance over the definition for a class or devices (or device prototypes). 13:33:39 s/or/of/ 13:33:48 ack kaz 13:34:11 Kaz: we may want to have a joint class with the discovery task force in relation to this issue. 13:34:42 s/class/calls/ 13:34:48 Darko: any further comments on this topic? 13:34:50 s/a joint/joint/ 13:34:54 rrsagent, draft minutes 13:34:54 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:35:24 darko: how we could manage issues? 13:35:34 dsr: we can use GitHub issue tracker 13:35:59 wot IG issue tracker http://www.w3.org/WoT/IG/track/ 13:36:16 s/tracker/tracker or W3C issue tracker/ 13:36:23 rrsagent, draft minutes 13:36:23 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:37:04 Dave: suggest using the W3C issue tracker for IG and TF issues, and GitHub tracker for specific documets e.g. IG reports. 13:37:28 Michael: Github’s tracker is per repositiory 13:38:50 action: Kaz to ensure that GitHub forwards issue tracking messages to the WoT IG public list 13:38:50 Created ACTION-15 - Ensure that github forwards issue tracking messages to the wot ig public list [on Kazuyuki Ashimura - due 2015-12-02]. 13:39:31 Topic: Technical Landscape survey 13:40:37 -> https://www.w3.org/WoT/IG/wiki/Wot-TD_Tech_Landscape TD landscape wiki 13:42:16 Darko talks us through the structure of the wiki. There are tables and additional wiki pages to provide more details. He invites volunteers to help with this. 13:43:09 We also intend to switch over to GitHub for the IG report we will make from this. 13:43:31 Michael: I currently can’t update the wiki and would very much prefer to work on a GitHub doc 13:43:41 rrsagent, draft minutes 13:43:41 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:43:46 q+ 13:44:42 ack d 13:45:06 darko: what was the consensus at the f2f? 13:45:16 dsr: moving to GitHub was the consensus 13:45:57 i/what/scribenick: kaz/ 13:46:02 action: Darko to work with Sebastian on setting up a document for the landscape survey on github 13:46:02 Created ACTION-16 - Work with sebastian on setting up a document for the landscape survey on github [on Anicic Darko - due 2015-12-02]. 13:46:02 scribenick: dsr 13:46:14 rrsagent, draft minutes 13:46:14 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:46:28 q+ 13:47:12 kolchinmax has left #wot-td 13:47:50 Danh: I made a start on the wiki, but I too would prefer GitHub and to use pull requests 13:48:42 I agree with tracking contributors and splitting the work up as appropriate 13:49:44 Darko: should we just use the wiki page structure or try something else? 13:49:59 Danh: I wasn’t at the F2F and am unsure about some of the columns. 13:51:02 We need examples, so the table isn’t particularly attractive 13:52:28 Dave: I would advise a free text description with examples for each area, and to then work on using tables for summaries that show the relationships between the different approaches. 13:52:44 Danh: I agree with that 13:53:04 +1 13:53:11 Darko: okay I will prepare a document outline accordingly 13:53:56 Jaime has joined #wot-td 13:53:56 any other points to raise now? 13:54:51 Dave: we will need to be flexible about adding new approaches, e.g. on manufacturing (having just come back from Manufuture conference on smart manufacturing) 13:55:39 Topic: WoT WG items related to TD-TF 13:56:18 present+ Jaime_Jimenez 13:56:30 Darko: what about missing work items? 13:57:11 Jaime: is HATEOAS part of the work items? 13:57:29 Darko: read the work items for their definition. 13:58:11 i|what about|-> https://www.w3.org/WoT/IG/wiki/Proposals_for_WoT_WG_work_items WG items proposal 13:58:12 I don’t think that the current 2 work items mention HATEOAS. 13:58:19 rrsagent, draft minutes 13:58:19 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 13:58:20 s/2/3/ 13:58:56 Michael: there was a lot of discussion aroud HATEOAS, but should we treat it as a technology or an application feature. 13:59:53 Would we work with schema.org ... 14:00:35 I don’t thiink of HATEOAS as a separate work iten, but more as something relevant across several 14:00:40 s/iten/item/ 14:00:48 q+ 14:01:05 ack Danh 14:03:09 ack dsr 14:03:35 dsr: maybe we need to clarify the relationship between the Web architecture and HATEOAS 14:03:57 Dave: I would like to see some discussion about the relationship between the core web architecture (addresses, declarative resources, protocols) and HATEOAS. 14:04:15 s/maybe/scribenick: kaz/ 14:04:20 Darko: we need links from one thing descriptin to another. 14:04:23 i/would like/scribenick: dsr/ 14:04:30 rrsagent, draft minutes 14:04:30 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html kaz 14:04:36 s/descriptin/description/ 14:04:51 Michael: I don’t see how thing descriptions will drive all the state transitions 14:05:11 I agree with Dave that thing descriptions are analogous to HTML and includes links 14:05:35 but I think it should also include everything needed for state transitions. 14:06:16 How are state transitions driven and this is a question of web architecture 14:07:37 q+ 14:07:50 Darko: Joerg would like to see hypemedia as a thread across work items, and so how does this fit? 14:07:56 ack Danh 14:08:53 Jaime: if nobody takes ownership we should perhaps keep this for later 14:09:13 s/hypemedia/hypermedia/ 14:09:46 Danh: behaviour may be through scripting languages or state transition machines, I don’ t think we need a new work item here. 14:10:22 Darko: we don’t want to leave this to be forgotten. 14:10:47 Danh: discovery is one use case and the discovery task force should be interested. 14:10:58 q? 14:11:43 Michael: some but not all protocols would be driven by hypermedia 14:12:22 q+ to mention SCXML for explicit state transition control 14:12:43 ack kaz 14:12:43 kaz, you wanted to mention SCXML for explicit state transition control 14:13:31 q+ 14:13:32 Kaz: W3C has a standard for state transition models - SCXML, which could be of interest 14:14:07 Dave: there are a family of state transition languages we should survey 14:14:35 Jaime: perhaps this is something we can cover in joint meetings with the IETF 14:15:02 discovery driven by hypermedia, some protocols , http and coap, more than others, also scripting is driven to the extent of the underlying protocols 14:15:09 q+ 14:15:32 ack jaim 14:15:35 ack ari 14:15:52 Ari: in the T2TRG we will continue to work on HATEOAS 14:16:29 At the last joint W3C/T2TRG meeting the consensus was that we need more experience, and joint plugfests will help with that 14:16:49 also TD somehow maps to hypermedia controls, maybe as part of the same data model or a separate data model linked to 14:17:35 Darko: we could continue discussions here 14:18:12 Ari: I want us to be aware of the work in the T2TRG on HATEOAS and discuss it as needed 14:18:14 yes, agree 14:19:25 Darko asks Michael to look at which task forces could be discussing HATEOAS 14:19:57 Michael: I will have a look at this and make some suggestions 14:21:41 Jaime has joined #wot-td 14:21:59 Dave: we can cover external liaisons in the WG charter 14:22:26 Dave: We need more supporters for the work on a vocabulary for server metadata as this is criticial 14:22:44 s/criticial/critical/ 14:23:10 Darko: if you have suggestions for other work items please let us know 14:23:57 Topic: Security metadata 14:25:13 Darko: we have yet to discuss access control to metadata, and we should spend time on coming up with use cases etc. 14:26:25 Dave: I agree with that, but also think we will need to work on use cases for security metadata more generally 14:27:08 Darko: we need volunteers 14:28:34 q+ 14:28:35 Dave: I agree with making time for this in future TD-TF calls 14:29:07 Darko: okay I will check back with Oliver 14:29:10 ack kaz 14:29:31 q- 14:29:40 Kaz: the W3C Auto WG has a long list of use cases and we may want to include them in our discussions 14:29:46 Darko: okay 14:29:56 any other business for today [no] 14:30:09 … end of meeting … 14:30:15 rrsagent, make minutes 14:30:15 I have made the request to generate http://www.w3.org/2015/11/25-wot-td-minutes.html dsr 14:31:51 Sato_Yasunori has left #wot-td 14:34:25 taki has left #wot-td 14:55:52 dsr has joined #wot-td 16:12:05 dsr has joined #wot-td