10:01:51 RRSAgent has joined #wot-script 10:01:55 logging to https://www.w3.org/2025/01/08-wot-script-irc 10:01:55 meeting: WoT Scripting API 10:02:03 present+ Kaz_Ashimura, Cristiano_Aguzzi 10:02:07 chair: Cristiano 10:02:32 agenda: https://www.w3.org/WoT/IG/wiki/WG_WoT_Scripting_API_WebConf#January_8%2C_2025 10:02:39 dape has joined #wot-script 10:02:42 present+ Zoltan_Kis 10:04:11 present+ Daniel_Peintner 10:06:06 scribe zkis 10:06:30 Topic: past minutes 10:06:46 https://www.w3.org/2024/12/11-wot-script-minutes.html 10:08:11 CA: suggested corrections to the minutes 10:11:11 Minutes approved. 10:11:21 Topic: PRs 10:11:35 Subtopic: PR 561 10:11:53 https://github.com/w3c/wot-scripting-api/pull/561 10:11:55 i|PRs|(typos, etc., have been fixed)| 10:11:57 rrsagent, make log public 10:12:00 rrsagent, draft minutes 10:12:02 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 10:12:59 q+ 10:13:04 s|https://github.com/w3c/wot-scripting-api/pull/561|-> https://github.com/w3c/wot-scripting-api/pull/561 PR 561 - Extended return type of invokeAction()| 10:13:10 rrsagent, draft minutes 10:13:12 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 10:15:12 ZK: we need to map better the interactions (actions) to the scripting patterns 10:18:30 s/scribe zkis/scribe: zkis/ 10:18:32 rrsagent, draft minutes 10:18:33 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 10:18:56 q+ 10:19:08 ack dape 10:19:36 CA: the CG is working on the AI agent framework using WoT 10:20:18 ... we can ask feedback from there 10:21:15 CA: originally I considered using events to solve this problem, but we can step back and check other options 10:21:32 CA: we should check the new profile spec for actions 10:21:56 ... we had query and cancel actions, but not observe actions 10:23:04 q+ 10:24:39 ZK: we should look at developer best practices on exposing agentic flows and action handling and then come back. 10:25:15 CA: yes, the underlying systems could return even percentage of completion, and we should support it. 10:25:21 ack cris 10:25:59 ack DP 10:26:16 DP: showing the HTTP Basic Profile for WoT 10:26:34 ... we have invokeAction, query, cancel, queryAll 10:27:45 i|showing|-> https://www.w3.org/TR/wot-profile/#http-basic-profile WoT Profile - 9. HTTP Basic Profile| 10:27:48 rrsagent, draft minutes 10:27:49 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 10:37:13 ZK: we should have the action control object, that is ActionInteractionOutput. That should include cancel(), completed() as promises, and status as a getter. 10:37:28 ... the status format is mapped from the TD 10:38:12 CA: we should stick to the TD spec, query and cancel the action and leave it to the script to do polling or events 10:39:51 DP: in the TD we have the op values, but query is not necessarily supported for each Thing 10:41:35 CA: ok, then we can go with the promise approach for cancel() and query() 10:41:45 ZK: what would the query() return? 10:42:02 CA: it could be another ActionInteractionOutput object 10:42:13 ZK: it's too complex that way 10:43:21 ZK: it seems the query status is not defined in the TD spec 10:43:31 ... so we have a gap there? 10:43:48 ... maybe we should ask Ege 10:43:50 CA: will do 10:45:20 DP: right, let's open an issue at the TD spec 10:46:05 i|it seems the query|-> https://w3c.github.io/wot-thing-description/#form WoT Thing Description 1.1 ED - 5.3.4.2 Form| 10:46:07 rrsagent, draft minutes 10:46:08 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 10:49:59 DA: the Profile spec assumes to report the status as an object or error 10:50:09 https://www.w3.org/TR/wot-profile/#http-basic-profile-protocol-binding-queryaction 10:50:41 q+ 10:51:09 ack dape 10:51:34 q- 10:51:38 KA: the whole WoT WG needs to think how to handle the result of querying actions 10:51:55 s/https/-> https/ 10:52:12 s/queryaction/queryaction WoT Profile - 9.2.2.2 queryaction/ 10:52:15 rrsagent, draft minutes 10:52:16 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 10:52:22 q? 10:53:43 https://www.w3.org/WoT/activities/meetings/ 10:54:03 https://www.w3.org/events/meetings/5a680419-2ee9-4aab-8762-bed92519843e/20250108T100000/ 10:54:05 s|https://www.w3.org/WoT/activities/meetings/|| 10:54:12 s|https://www.w3.org/events/meetings/5a680419-2ee9-4aab-8762-bed92519843e/20250108T100000/|| 10:54:53 Topic: issues 10:55:02 Subtopic: issue 507 10:55:04 https://github.com/w3c/wot-scripting-api/issues/507 10:57:23 CA: we are considering these, but not decided 10:58:03 ZK: I agree that `getThingDescription()` is not very webby, but then we must spec how to deal with the internal slot, when to initialize etc. 10:58:37 s|https://github.com/w3c/wot-scripting-api/issues/507|-> https://github.com/w3c/wot-scripting-api/issues/507 Issue 507 - Should we refactor getThingDescription to a JS getter?| 10:58:40 rrsagent, draft minutes 10:58:42 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 10:59:07 https://www.w3.org/TR/design-principles/ 10:59:38 ZK: we should read again the web platform design principles 11:00:48 and also spec coding conventions: https://github.com/webmachinelearning/webnn/blob/main/docs/SpecCodingConventions.md 11:03:00 ... we should at least align with these. 11:03:30 s|https://www.w3.org/TR/design-principles/|-> https://www.w3.org/TR/design-principles/ Web Platform Design Principles Note| 11:03:44 rrsagent, draft minutes 11:03:45 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 11:04:17 CA: time up, meeting adjourned. 11:04:26 rrsagent, draft minutes 11:04:27 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 11:04:33 [adjourned] 11:05:24 rrsagent, draft minutes 11:05:26 I have made the request to generate https://www.w3.org/2025/01/08-wot-script-minutes.html kaz 11:06:04 Mizushima has joined #wot-script