13:48:39 <RRSAgent> RRSAgent has joined #adapt
13:48:43 <RRSAgent> logging to https://www.w3.org/2025/08/26-adapt-irc
13:48:43 <Zakim> RRSAgent, make logs Public
13:48:44 <Zakim> please title this meeting ("meeting: ..."), janina
13:49:03 <janina> Meeting: WAI-Adapt Teleconference
13:49:10 <janina> Chair: Lionel_Wolberger
13:49:17 <janina> Date: 26 Aug 2025
14:01:18 <matatk> matatk has joined #adapt
14:02:05 <Abhinav> Abhinav has joined #adapt
14:02:16 <Abhinav> present+
14:03:12 <Lionel_Wolberger> Lionel_Wolberger has joined #Adapt
14:03:14 <matatk> present+
14:03:24 <Lionel_Wolberger> present+
14:03:25 <janina> present+
14:03:29 <Lionel_Wolberger> scribe+
14:05:51 <janina> regrets+ Annalu
14:05:57 <janina> scribe+
14:06:40 <Lionel_Wolberger> rrsagent, make minutes
14:06:41 <RRSAgent> I have made the request to generate https://www.w3.org/2025/08/26-adapt-minutes.html Lionel_Wolberger
14:07:24 <Lionel_Wolberger> Topic: Symbols
14:07:53 <Lionel_Wolberger> matatk: I am ready to move the Symbols explainer forward this week
14:08:00 <Lionel_Wolberger> Lionel_Wolberger: We'll try and get you some time back for that
14:08:18 <Lionel_Wolberger> Topic: Destinations
14:10:40 <matatk> General progress update: https://lists.w3.org/Archives/Public/public-adapt/2025Aug/0004.html - I didn't mention in here that I also updated the extension to Google's new Manifest V3 API, as well as bumping the packages.
14:10:46 <matatk> rrsagent, make minutes
14:10:47 <RRSAgent> I have made the request to generate https://www.w3.org/2025/08/26-adapt-minutes.html matatk
14:16:20 <Lionel_Wolberger> Lionel_Wolberger: Is look and feel the same?
14:17:58 <Lionel_Wolberger> matatk: May go with sidebar rather than popup.
14:20:04 <matatk> Topic: Well-known destinations and machine-readable semantics
14:20:11 <matatk> rrsagnet, make minutes
14:21:10 <Lionel_Wolberger> Abhinav: I am considering how agentic AI would interpret our destinations
14:21:14 <matatk> https://en.wikipedia.org/wiki/Agentic_AI
14:21:56 <Lionel_Wolberger> ... in my experience with recent LLM models and their reasoning
14:22:32 <Lionel_Wolberger> ... take for example a flight booking: I have some conditions, an overall travel time and a date range
14:23:00 <Lionel_Wolberger> ... an LLM can parse that out into source, destination, date range and constraints
14:23:31 <Lionel_Wolberger> ... it likely has a toolkit that it then draws upon, for example an API to a flight information service
14:23:48 <Lionel_Wolberger> ... and then another tool that takes that data, compares and filters as to the conditions it received
14:24:09 <Lionel_Wolberger> ... the filtering may be iterative: one filter for price range, one for flight duration
14:24:31 <Lionel_Wolberger> ... followed by another layer that may be able to use my credit card to book it
14:25:04 <matatk> q+
14:25:45 <matatk> janina: *mentions https://innosearch.ai *
14:25:52 <Lionel_Wolberger> janina: Innosearch is an example of such a service, it uses expedia on its backend
14:27:11 <Lionel_Wolberger> ack matatk
14:27:46 <Lionel_Wolberger> matatk: While we will likely implement semantics to aid in machine readability
14:28:04 <Lionel_Wolberger> ... but I note that visual LLMs seem to be doing a reasonable job without semantics
14:28:41 <Lionel_Wolberger> Lionel_Wolberger: These visual presentations are coherent, typically thanks to underlying semantics
14:28:54 <matatk> s/seem to be doing a reasonable/are available to do the/
14:29:18 <matatk> s/job without semantics/job without semantics - i.e. more like browsing sites visually/
14:29:43 <matatk> matatk: MCP that Abhinav mentioned: https://modelcontextprotocol.io/docs/getting-started/intro
14:29:50 <matatk> scribe+
14:30:36 <Lionel_Wolberger> Abhinav: Model context protocol was put forward by Anthropic, and is commonly used by these types of agents
14:31:10 <Lionel_Wolberger> ... it is a method to expose your LLM to such external services
14:32:08 <Lionel_Wolberger> ... it enables an Agent to select the proper tool. The MCP exposes the tools to the Agents
14:36:26 <Lionel_Wolberger> ... the previous generation of web agents, that use APIs to a website, can also use this MCP
14:36:55 <Lionel_Wolberger> ... and these web agents can be called via MCP
14:38:24 <Lionel_Wolberger> ... the MCP requires authorization
14:38:46 <Lionel_Wolberger> ... Our approach to destinations is very much required in this context.
14:38:59 <Lionel_Wolberger> ... Consider (a) AI Agents (b) tools (c) MCP interface
14:39:26 <Lionel_Wolberger> ... these agents are struggling to find the relevant pages, via DOM parsing or visual inspection to find the navigation
14:40:04 <Lionel_Wolberger> ... In my example, if a WKD clarifies that a website has an endpoint for flight search, the agent will go to that endpoint to get its results
14:40:12 <Lionel_Wolberger> rrsagent, make minutes
14:40:13 <RRSAgent> I have made the request to generate https://www.w3.org/2025/08/26-adapt-minutes.html Lionel_Wolberger
14:42:07 <janina> q+ to suggest we tell W3C payment processing about our linkset approach
14:43:40 <matatk> q+
14:45:00 <Lionel_Wolberger> janina: We should chase this at TPAC.
14:45:35 <Lionel_Wolberger> Abhinav: In summary, this WKD concept is supporting the current transition to Generative AI optimization
14:47:09 <Lionel_Wolberger> janina: Because we are making the URLs declarative
14:49:25 <Lionel_Wolberger> janina: Accessibility is more and more important, as accessibility relies on semantics and regularities that also benefit the LLMs
14:54:36 <Lionel_Wolberger> matatk: The 'destination' is the linkset
14:54:45 <Lionel_Wolberger> ... checkout is more of an action than a destination
14:54:53 <Lionel_Wolberger> ... it's a thing that you do, rather than a place that you go
14:55:08 <Lionel_Wolberger> ... Abhinav explained this as dependent on whether you have something
14:55:10 <Lionel_Wolberger> q+
14:55:22 <Lionel_Wolberger> ... so we may need to revisit action v destination
14:55:35 <Lionel_Wolberger> ... also, janina brought up payments
14:55:46 <Lionel_Wolberger> ... there's payment for the final item, what you are buying; there's also paying for API usage
14:55:52 <Lionel_Wolberger> ... web payments is for the former not the latter
14:56:56 <Lionel_Wolberger> Lionel_Wolberger: I like it that 'actions' would be useful for LLMs. that could motivate us to tackle that next
14:57:27 <Lionel_Wolberger> Abhinav: Should we add this LLM agentic consideration to the explainer?
14:57:30 <Lionel_Wolberger> matatk: +1
14:57:33 <Lionel_Wolberger> Lionel_Wolberger: +1
14:57:52 <Lionel_Wolberger> matatk: A short addition to the use cases section is a good place to start
14:59:34 <Lionel_Wolberger> zakim, end meeting
14:59:34 <Zakim> As of this point the attendees have been Abhinav, matatk, Lionel_Wolberger, janina
14:59:36 <Zakim> RRSAgent, please draft minutes
14:59:38 <RRSAgent> I have made the request to generate https://www.w3.org/2025/08/26-adapt-minutes.html Zakim
14:59:44 <Zakim> I am happy to have been of service, Lionel_Wolberger; please remember to excuse RRSAgent.  Goodbye
14:59:44 <Zakim> Zakim has left #adapt