W3C

– DRAFT –
WoT-WG - TD-TF - Slot 1

18 December 2024

Attendees

Present
Cristiano_Aguzzi, Daniel_Peintner, Ege_Korkan, Jan_Romann, Kaz_Ashimura, Michael_Koster, Tomoaki_Mizushima
Regrets
Mahda, McCool, Sebastian
Chair
Ege
Scribe
dape

Meeting minutes

Minutes Review

Ege: first slot, https://www.w3.org/2024/12/11-wot-td-minutes.html

Jan: Small typo in repository

Kaz: fixed

Ege: Anything else?
… none -> minutes are approved

Ege: second slot, https://www.w3.org/2024/12/12-wot-td-minutes.html

Ege: Any remark?
… none -> minutes are approved

Proposals Folder

Ege: see PR w3c/wot-thing-description#2062
… not clear what the proposals folder is about
… created root README
… about feature that doesn't fit into a single issue/PR
… indicated different status
… in progress, rejected, paused, accepted, retired

Daniel: Do we plan to remove proposals again?

Ege: Good point, we could...

Kaz: documentation is very useful
… wondering when and how to put description
… content should move to WoT process/transition

Ege: Correct
… main question is when does it start to be an actual issue

<kaz> example of the "strategy" repository

Ege: I would say it can be part of the overall process (if the group decides to work on it)

Ege: I will add this information
… a possible outcome could point to the project management tooling

Kaz: For today we can record this point
… separate issue is also fine

Ege: created issue 2063 to keep track

Ege: No more remarks on PR 2062 --> merging

BACnet mediaType

<EgeKorka_> w3c/wot-binding-templates#357

<EgeKorka_> w3c/wot-binding-templates#390

Ege: Similar in Modbus
… in binding document there is no contentType
… this would mean JSON
… but BACnet is not JSON
… it is going to be octet stream
… it is somewhat redundant information
… Dohan added const octet-stream for contentType
… there is explanation in PR as well
… simple PR but there are possbile "issues"

Koster: I think it is correct
… it feels appropriate

Cristiano: I think PR is harmless
… I am okay with moving on with this one
… with connection proposals we can more easily handle this
… about Modbus
… could declare contentType application/JSON
… driver will read all the bytes and interpret it as JSON
… not sure if this is the same in BACnet, I don't think so..

Ege: I don't know
… but then we should not use const ... default should be enough
… but I know BACnet is rather strict
… I will ping Dogan

Ege: merging PR 390

Kaz: I was wondering about smart building "companies" ...
… they are interested in BACnet also
… we can ask them about their approach

Ege: Yes, we should get in touch with them

Kaz: I think they are using BACnet devices

Koster: In Sweden, BACnet is going to come as requirement

Kaz: I can try to ask Takenaka people. Also the Nordic Smart City CG guys.

Ege: That would be very useful

Binding Registry

<EgeKorka_> w3c/wot-binding-templates#378

Ege: We talked about last Thursday
… like copyright rules et cetera
… we decided that there is no need to follow W3C rules
… Jan also asked about document being freely available
… this has several dimensions
… free to read does not necessarily mean free to implement, etc.
… Dimensions: Reading the binding document, reading the protocol specification, implementing a device/Thing, implementing a Consumer application/driver, and building a commercial product with it, making a statement about your product's supporting that binding.
… e.g. BACnet binding is free to read, but that does not mean the BACNet spec
… there are all this legal frameworks
… Koster raised the point, we should not limit it to be "free"
… which would end up being IETF only
… in IANA registration you need a summary document
… we could require such a summary document to be freely available

Koster: I think the summary of the discussion is good
… we should talk about different kinds of restrictions
… technical vs commercial use
… commercial use is out of scope
… Matter is about using the trademarks

Kaz: I agree with M. Koster
… what we can/should do is listing which protocols and ecosystems to be used for WoT like the "WebCodecs Codec Registry" shows.

<EgeKorka_> WebCodecs Codec Registry

Ege: Yes, I wanted to show that

Ege: Basic what is in the registry
… that is something we can do
… another point is the review process
… reviewer should be able to read the binding document ... maybe through liaison
… or through member companies
… if there is no such possibility I think we need to reject it

Koster: Minimum explainer document could be enough, right?

Ege: I would be uncomfortable
… we cannot check the requirements

Koster: I see
… we have to discuss that more
… what we require

Ege: I will try to split the points mentioned
… copy right rules
… look and feel
… reviewer access
… binding availability

Ege: Does anyone have strong opinions on these points?

Jan: Reviewer access to protocol as well?

Ege: Yes, or must be expert already

Cristiano: Implications
… implementing consumer might encounter licensing issues

Ege: Yes

Cristiano: I am still skeptical

Ege: I wish we don't have that situation
… but it is already the case for BACnet

Cristiano: important for generic consumers
… plugin concept

Ege: will discuss this next year further

Ege: What we need to discuss is also the status change
… we need some implementation experience etc.
… to become "current"
… I don't think we gonna do a lab test

AOB

Ege: Not hearing anything.. merry christmas

<kaz> [adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 240 (Tue Dec 10 03:59:59 2024 UTC).