IRC log of voiceinteraction on 2024-10-23

Timestamps are in UTC.

14:21:12 [RRSAgent]
RRSAgent has joined #voiceinteraction
14:21:17 [RRSAgent]
logging to https://www.w3.org/2024/10/23-voiceinteraction-irc
14:21:28 [ddahl]
meeting: Voice Interaction
14:21:33 [ddahl]
chair: debbie
14:21:40 [ddahl]
scribe:ddahl
14:22:30 [ddahl]
agenda: https://lists.w3.org/Archives/Public/public-voiceinteraction/2024Oct/0006.html
14:31:18 [ddahl]
present: debbie, gerard, debbie
14:32:16 [ddahl]
review changes to Architecture document
14:32:49 [ddahl]
https://lists.w3.org/Archives/Public/public-voiceinteraction/2024Oct/0003.html
14:33:05 [ddahl]
for issue 52
14:35:48 [ddahl]
debbie: could just say a little more about the difference between traditional and GenAI systems in 3.1
14:37:44 [ddahl]
dirk: should we use another word besides "traditional"?
14:38:00 [ddahl]
we can't think of anything better
14:44:33 [ddahl]
dirk: redraws Figure 3
14:45:22 [ddahl]
dirk: Section 5.1 is just about traditional systems
14:46:34 [ddahl]
debbie: why is LLM in Figure 3?
14:47:11 [ddahl]
dirk: could gray out LLM to show that it's not involved. the same for Figure 4 (output)
14:47:46 [ddahl]
debbie: should say something about why the LLM boxes are grayed out
14:47:53 [ddahl]
dirk: agrees
14:49:49 [ddahl]
dirk: would like to add the word "agentic" because that's an established term
14:50:17 [ddahl]
debbie: I agree
14:50:24 [ddahl]
dirk: will add that
14:51:58 [ddahl]
debbie: should we say something about GenAI at the beginning (Section 1) about how these specifications apply to GenAI
14:52:43 [ddahl]
Action: Debbie to add that in Section 1
14:54:25 [ddahl]
dirk: should we add a use case for GenAI, like an avatar
14:56:36 [ddahl]
dirk: chat application with a GenAI system while you're driving
14:57:40 [ddahl]
action: debbie to write up a sample use case
14:57:58 [ddahl]
action: dirk to write up a sample use case
15:04:45 [ddahl]
topic: follow-up from Dirk's presentation last time
15:04:54 [ddahl]
https://docs.google.com/presentation/d/13ONyFA96wsWqnEV3jCTvHl6ze7rCspjZuhZ9bU66U50/edit?usp=sharing
15:05:39 [ddahl]
dirk: it seems like the taxonomy introduced last time was ok but data types were missing
15:06:01 [ddahl]
...should we mention those categories in the document?
15:06:35 [ddahl]
debbie: I don't think so, interesting, but going off on a tangent
15:07:55 [ddahl]
dirk: could try to contact author and see if he has anything to add
15:08:29 [ddahl]
topic: assistant terminology
15:08:54 [ddahl]
debbie: list discussion had a lot of opinions
15:09:25 [ddahl]
dirk: this term, "personal" has two meanings, knows about you or nothing is shared
15:10:01 [ddahl]
debbie: I never thought it meant nothing is shared
15:11:00 [ddahl]
dirk: "personal assistant" would require a lot of changes
15:11:22 [ddahl]
s/personal assistant/changing personal assistant
15:11:48 [ddahl]
dirk: can add something to say that personal doesn't mean nothing is shared
15:13:56 [ddahl]
dirk: will make a suggestion for text about why we call these personal assistants
15:14:28 [ddahl]
debbie: it doesn't affect anything in the architecture or specs
15:14:42 [ddahl]
rrsagent, format minutes
15:14:43 [RRSAgent]
I have made the request to generate https://www.w3.org/2024/10/23-voiceinteraction-minutes.html ddahl
15:14:51 [ddahl]
rrsagent, make logs public
15:17:45 [ddahl]
dirk: we're not making any assumptions about where the component is running, and we may suggest not to share data unnecessarily
15:17:54 [ddahl]
debbie: that's kind of a best practice
15:19:37 [ddahl]
debbie: we can postpone reviewing GitHub issues until next time
15:19:44 [ddahl]
rrsagent, format minutes
15:19:45 [RRSAgent]
I have made the request to generate https://www.w3.org/2024/10/23-voiceinteraction-minutes.html ddahl
15:48:30 [ddahl]
present+ dirk
15:48:37 [ddahl]
rrsagent, format minutes
15:48:38 [RRSAgent]
I have made the request to generate https://www.w3.org/2024/10/23-voiceinteraction-minutes.html ddahl
15:52:48 [ddahl]
ddahl has left #voiceinteraction