W3C

– DRAFT –
WoT Discovery

28 November 2022

Attendees

Present
Addison_Phillips, Andrea_Cimmino, Ege_Korkan, Fady_Salama, Kaz_Ashimura, Kunihiko_Toumura, Michael_McCool, Tomoaki_Mizushima
Regrets
-
Chair
McCool
Scribe
Ege, McCool_

Meeting minutes

Minutes Review

<kaz> Nov-21

McCool: minutes are approved

Wide Review

<kaz> i18n issues

Issue 321

<kaz> Issue 321 - Registration information timezone handling

McCool: how does this issue look like Addison?

Addison: This is fine from my point of view

issue 320

<kaz> Issue 320 - Question is id in 7.2.2.1.1 the same as TD id?

McCool: the id's are also in URLs to identify TDs

Addison: if non ascii chars can be percent escaped, then it is fine. So this is the boundary between URLs and their text representations

Addison: do you know what format to expect? If you assume uris, then you can encode. If iri then it would not work

McCool: we need better wording around this, but the consensus is no change necessary

Issue 316

<addison> Issue 316 - Specify Sorting Algorithm

McCool: created a PR

<kaz> PR 441 - Use Unicode code points for sorting orders

McCool: (goes through the changes)
... does this PR address the question?

Addison: yes
... but not all the programmers use UTF-8

McCool: Unicode should be a more generic term

McCool: any objections to doing those two things?

<McCool_> proposal: merge PR #441 and update CR candidate correspondingly.

McCool: (types resolution)

RESOLUTION: merge PR #441 and update CR candidate correspondingly. McCool will create and merge a PR to update the CR candidate.

McCool: I hear no objections

McCool: that should be the end of internationalization issues

TD i18n issues

wot-thing-description Issue 1750 - Implementation of text direction from language tags

Ege: very specific language we added

Ege: we have tried to implement this in the playground since we have a visualization capability
… this assertion however is still needed and we need a means to guess the direction
… it is optional but people still need to implement this for our spec

Addison: we tell people NOT to do this, since, ouch
… still wrong some of the time
… if you provide metadata it's better
… in practice you have first-strong, can use metadata to fix problems, guessing in wrong half the time, so...

Ege: so in the end, if we really need to do language guessing
… would it be ok to remove this assertion?

Addison: could make spec permissive...

McCool: but still need two implementations even for optional (e.g. permissive) assertions

Ege: would it be ok for us to remove this assertion?

Ege: which one should have priority?

Addison: guess by language first, then by first strong
… there are a number of things about bidi isolation in our documents
… but this is not really about your spec; is if there is a string in your spec, how to feed it to a display?

Ege: in this case, instead of a full lookup table, can we do a subset?
… e.g. just the most common RTL languages

Addison: users would appreciate a more complete list, but can use the link provided in the issue

https://github.com/unicode-org/icu/blob/3654e945b68d5042cbf6254dd559a7ba794a76b3/icu4j/main/classes/core/src/com/ibm/icu/lang/UScript.java#L1337

McCool: what about languages that may be either, i.e. Azeri?

<kaz> i18n wg's page on Languages using right-to-left scripts

Addison: leave it off the list, use first-strong in that case

McCool: is there a CSV file or something?

Addison: no, because recommendation is not to do this...

McCool: (sigh)

Ege: clean solution is to use just metadata
… but we want to keep backward compatibility

Kaz: So we need to change the TD 1.1 spec?

McCool: I think Ege only asked for help with implementation

Kaz: maybe something editorial, to guide implementors?

McCool: that makes sense, Ege can you create a PR about it?

McCool: Thank you Addison

<kaz> (Addison leaves)

Horizontal View Checklist

<kaz> Review comment tracker for WoT Discovery

McCool: should we fix these red boxes?

<McCool_> Normative references checker for WoT Discovery

McCool: we are inspired from Hydra. This is an informative reference though

McCool: these are marked but I do not understand it

Kaz: these are just recommendations from the reference checker right?
… so we can review them and decide which suggestion to be applied

McCool: (writes github comment)

McCool: would anyone object to changing it to "inspired by"

<kaz> transition issue 465 - CR Request for Web of Things (WoT) Discovery

<McCool_> https://github.com/w3c/transitions/issues/465#issuecomment-1329329272

<McCool_> proposal: McCool to affect changes to CR snapshot as described in https://github.com/w3c/transitions/issues/465#issuecomment-1329329272 (removing section references via the datatracker, changing "derived from" to "inspired by" for hydra refs, leaving others alone)

RESOLUTION: McCool to affect changes to CR snapshot as described in https://github.com/w3c/transitions/issues/465#issuecomment-1329329272 (removing section references via the datatracker, changing "derived from" to "inspired by" for hydra refs, leaving others alone)

IANA Registration

<kaz> Issue 403 - Refine DNS-SD Name Requirements and IANA Consideration

McCool: we have gotten confirmation but it says that no txt keys are registered

Toumura: there are many pages

<kaz> IANA Registry

McCool: ok we find wot in the list

McCool: but it is wrong even though I have asked for it in the original request

McCool: we can maybe do it right now

<McCool_> https://www.iana.org/assignments/service-names-port-numbers/service-names-port-numbers.xhtml?search=wot

McCool: the problem is not having TXT keys

McCool: we need to close the meeting

meeting adjourned

Summary of resolutions

  1. merge PR #441 and update CR candidate correspondingly. McCool will create and merge a PR to update the CR candidate.
  2. McCool to affect changes to CR snapshot as described in https://github.com/w3c/transitions/issues/465#issuecomment-1329329272 (removing section references via the datatracker, changing "derived from" to "inspired by" for hydra refs, leaving others alone)
Minutes manually created (not a transcript), formatted by scribe.perl version 196 (Thu Oct 27 17:06:44 2022 UTC).