W3C

– DRAFT –
WoT Plugfest

13 November 2024

Attendees

Present
Ege_Korkan, Fady_Salama, Jan_Romann, Kaz_Ashiura, Kunihiko_Toumura, Larry_Robinson, Luca_Barbato, Mahda_Noura, Michael_Koster, Michael_McCool, Roman_Binkert, Tomoaki_Mizushima
Regrets
Sebastian
Chair
Ege
Scribe
JKRhb, kaz

Meeting minutes

Agenda Review

<kaz> agenda for today

Ege: Sebastian is at a fair that is why I am taking over
… everyone is already aware of the patent policy
… we are using IRC for the chat and the queue

Minutes

<kaz> Nov-6

Ege: we will quickly take a look at the minutes
… but everyone should have already received them
… I will quickly go over them, but everyone should have already read them
… last time we talked about VPN and the demo presentations
… the PRs
… any remarks regarding the minutes from last week?
… Hearing none, then the minutes are approved, thank you, Fady, for taking minutes last time

Logistics

<kaz> WoT Week wiki

Ege: The list of partitipants has grown
… there are more people from Siemens, also someone we met in the context of the CG
… already a great amount of people
… even excluding remote participants, should be about 40 people

Ege: please be aware that the event on Wednesday needs additional registration
… also heard people ask whether there is a "perfect integration" between the hotel and the event
… there isn't, you just need to mention that you are going to attend the event, if there are any issues, contact us

Status of the VPN

<kaz> PR 604 - Basic Infrastructure Information

Ege: I needed to make sure that we have enough network switches and so on
… the number of ports is not a concern, but we need to make sure that people will be able to quickly reconnect on Wednesday
… but I would recommend bringing your own switch on Wednesday
… also have very long Ethernet cables, so there should not be an issue
… then regarding the network, if we want to conduct a video call, you should connect to the Siemens guest network
… you can contact Sebastian, Mahda, or me to get access, please don't use it for heavy networking
… still need to test whether the VPN will work, it is supposed to "trap" people in the VPN when they join the network

McCool: Just a comment: If you need to have a server on the public network, you can use an SSH tunnel and host it via AWS, for example

Ege: Good point, will add that

McCool: Need to make sure that security etc. is verified, just need an external computer, then you can connect to something like Digital Ocean

<kaz> PR 604 - Basic Infrastructure Information

Ege: If there is nothing else, then I will merge PR 604 that includes this information now

Merged

<kaz> PR 589 - Update vpn_softetherbridge.md : to briged interaface will assiged with physical interface's MAC address.

Ege: Next up is PR 589 with the updated VPN instructions
… it was opened by someone external, and Toumura-San asked them about it

Toumura: I asked them whether the procedure they describe is very necessary, not needed for a working VPN

Ege: Okay, then we are not going to merge it

Validation of TDs

<kaz> PR 605 - CI test

Ege: As you might have noticed, the TD validation is currently failing, with TD 605 I was able to fix it, it is also much faster now
… please submit now the TDs of your implementations
… I would suggest that we merge the PRs asynchronously, as long as they do not overwrite others
… we can also link the TD of the implementation in the table, I added one already. Note that the IP addresses are going to be different for the live devices

<kaz> PR 606 - add TUM devices part 1

Ege: In PR 606 I also added additional TDs for TUM devices
… they have been provided by Fady and Roman
… this the direction I would like to go so that we can have an organized overview of the submitted TDs

Ege: Are there any objections to merging TD submissions asychronously?

No objections

Ege: Then we can proceed, probably no resolution needed for that

Presentations

<EgeKorkan> https://github.com/w3c/wot/blob/main/PRESENTATIONS/2024-11-wot-week/2024-11-27-WoT-Conference-Presentation.pptx

Ege: Everyone should provide one slide for the presentation on stage
… there is also going to be a poster template, which you will be able to put behind your table during the demo

New PRs

Ege: There are two new PRs which don't add TDs

<kaz> PR 602 - More Siemens Devices

Ege: there is one from Siemens (PR 602), these add devices with temperature and humidity sensors
… we will also have a gateway that can expose other protocols northbound via HTTP and WebSockets
… enabling to test us whether we can map concepts to HTTP without losing semantics
… (notices that there is a merge conflict)
… okay, if there are no objections, then I would do the conflict resolution later and merge the PR

<kaz> PR 603 - Small robot arm and conveyor belts from TUM

Ege: The next PR (603) adds more TUM devices to the table, a small robot arm and two conveyor belts
… also seeing no objections, so I will apply the same procedure to this PR

User Scenarios

<EgeKorkan> https://github.com/w3c/wot-testing/blob/main/events/2024.11.Munich/Readme.md#wip---user-scenarios

Ege: So now we can come to the more interesting part and do some brainstorming on user scenarios
… is part of the document I linked
… I think NSGI-LG(?) is not relevant for Smart Cities at the moment, so I would remove it for now

Kaz: I am okay with your top-down approach for now, but I think we should clarify which devices are going to be used for which purpose
… starting with these listed items is fine, but we also need to ask about people's expectations
… some scenarios might not be listed down below that would fit the existing devices, so we should ask the people what kind of application or service they want to use

Ege: Commenting for Siemens, the Openflexure Microscope is more about long-running actions
… the others are more related to smart home or smart city, sold to building managers

Kaz: Maybe we should ask the people present what they want to use their devices for

KNX IoT devices from Siemens

McCool: Just noticed KNX IoT

Ege: They are more like a profile for CoAP + other technologies

McCool: So this could be assigned to "Technical Purposes", indicating that we test devices like KNX

Ege: (adds KNX as a bullet point)

Hitachi

McCool: The same is probably true for Hitachi

Ege: Should I add it?

Toumura: Yes

Siemens Sentron PAC

Ege: (Adds other Siemens to the "Technical Purposes")

Siemens TargetV

Mahda: There is also TargetV as well as sayWoT!, supporting OPC UA and other protocols

TUM

Ege: Roman, what are you from TUM going to show?

Roman: Big Robot Arm that is going to interact with hololenses, also Hue devices

Ege: Also interesting for standardization: There two alternative security definitions that can be used in one device, but only one is officially documented, quite interesting

Fady: Regarding the Hololens, there is the possibility to do visual interactions, but only HTTP is supported at the moment, could also be possible to do mashups

Ege: I think the small robot arm and conveyor belt is also interesting for mashup scenarios

Luminem

Ege: Luca, what is the purpose for your devices?

Luca: (is currently away)

Jan: I am in contact with him, I am going to operate devices locally, run wot-rust on ESP32, BPi

Armor Safe

Ege: I am in contact with Larry Robinson
… going to check interoperability with IgniteRetail ControlCenter

Hitachi - revisited

Toumura: Going to test discovery

Thingweb dart_wot

Jan: Going to test interoperability, identify shortcomings and improve implementation

Intel

McCool: AI Service integration

Ege: Generally, we should think about how to link high-level to low-level stuff
… all possible because of interoperability on the layers below
… so I would ask if anyone is going to put multiple things together, also considering that we are already testing interoperability

McCool: Can we glue things together ourselves? I was thinking about the retail scenario
… dealing with an AI-based scenario, working together with Ignite Retail
… ideal would also be a smart home where you can operate things based on an LLM that reads in TDs
… complicated to do it more generally, since it is based on Python, therefore I cannot use the Thingweb implementation

Ege: There is also wotpy that you could use, it gets updated

McCool: Want to keep it as simple as possible, using multiple protocols would be a stretch-goal

Ege: We can use sayWoT!, for example, that is acting as sort of a bridge to HTTP

McCool: If I could use that it would be ideal

Next steps

Kaz: I think before the actual plugfest, we need to clarify which device (Thing) and which application (Consumer) fulfills which technical purpose

Ege: I think we should add a more detailed description

Kaz: We could have a separate markdown file for each one

McCool: Could link to that from a shorter description

Ege: Let's merge this PR one then and create a folder called "Scenarios" where we can put the more detailed descriptions of the scenarios that are to be tested during the plugfest
… so we can say, let's create one Markdown file per scenario you are working on
… we can pick up the scenario mentioned by Fady for that one

McCool: I think we need a directory under that directory, and then we can link from the main README file
… and then people can add their own scenarios

<kaz> 2024.11.Munich/Scenarios area

Ege: (creates a directory and READMME file for a VoiceControl scenario)
… (adds involved devices and applications)
… (adds a preliminary Mashup Explanation for the scenario, that centers around a "McCool 9000" consumer application)
… okay, I guess there could then be a results section

Ege: would ask people to create a MD for their scenario like this

VoiceControl example

Kaz: just to make sure, have you commited your changes to 2024.11.Munich/Readme.md ?

2024.11.Munich/Readme.md

Ege: yes (as above)

Kaz: I'll ask the ECHONET guys about their ideas too

Ege: ok

[adjourned; TD call starts at 10mins past]

Minutes manually created (not a transcript), formatted by scribe.perl version 238 (Fri Oct 18 20:51:13 2024 UTC).