WoT Discovery

<kaz> scribenick: cris

Previous minutes

<kaz> Sep-21

McCool: any objection for publishing the minutes?
... ok published
... any updates?
... ok we have 3 PRs

Christian: Did somebody reach out the JSON-LD community?

McCool: yeah I contacted them, but I have to ping them again

Directory service for PlugFest

McCool: Does anybody knows where Farshid is?

Andrea: he should be here today

McCool: I was hoping to do some directory testing. it is plugfest week
... We need a directory service online.
... Probably, we could follow the last time instructions... but it would be better if Farshid can do it
... I can send an e-mail directly
... in today's plugfest call we draft some scenarios that may impact also discovery topics
... first I'll try to annotate the currently available TDs with oneDM
... then we can import them inside our directory to test different queries
... I'd like to test also geospatial queries

<inserted> wot-thing-description issue 941 related to geolocation information withing TD

McCool: I'd use schema.org to annotate our TDs
... Siemens will probably provide a map to visualize location
... about Directories, it would be nice to have an endpoint to support geospatial queries.
... A basic method might be to have a center point and a range. TDs should be filtered out if they are outside this range
... geospatial discovery is much harder. We can start with directories
... we should meet soon with geospatial group

<McCool> https://github.com/w3c/wot-testing/issues

McCool: Another thing is to filter tds using time data (i.e. give me only the TDs published on hour ago)

Kaz: if we can't get to Farshid this week, we may use Toumura-san's work or NHK's directory.

McCool: yeah, but we need more advanced search mechanisms (queries using JSON Path).

Kaz: yes, just as a fallback

McCool: if don't have a directory working, is there something else we can do in this PlugFest?
... what about improving the node-red plugin?
... I'll fill an issue about this topic
... Also as a fallback, I'll try to bring up the directory myself
... if people wants to test other features, please create an issue so that we can track it.

PRs

<inserted> PR 74

McCool: pr 74 is being merged, any objections?
... merged

<inserted> PR 73

tomura: PR 73 minor editorial changes

McCool: it is better to use ReSpec database than local references
... seems fine
... any objection to merging?
... merged

<inserted> PR 47

McCool: PR 47 is far bigger
... I didn't have time to review it, however I'm proposing to merge it. We are going to review the entire document later.
... there are some missing references

Andrea: I am not sure which references to use for JSON Path

McCool: I see no problem
... let's merge this
... there might be some tab expansions, we have to look at it later
... oh there's also the combination scheme. We have to wait until it is merged in the TD spec
... anyway we'll leave it for now
... I'm merging the PR, any objections?
... ok merged.
... let's do a quick look to the current working draft

<inserted> preview for PR 47

McCool: we have a decent abstract
... a nice table of contents
... Use Cases is empty
... probably we can remove use cases section. At least for FWPD

Kaz: suggests we look at the latest github.io version instead of the pr preview

<kaz> latest draft

McCool: Architecture section needs more work

<kaz> Section 4. Architecture

McCool: Exploration Mechanisms as well

<inserted> Section 6. Exploration Mechanisms

McCool: security section should have a TODO item
... ok I think we are ready for the FWPD, we only need to change the empty sections to Editor's notes
... does anybody have any objection ?
... kaz do you think this is ok for FWPD?

Kaz: yes, it is!

McCool: I'll do a clean up and we'll proceed
... any objections?
... ok no objections

<McCool> proposal: publish current version, with formatting changes to mark incomplete sections using editor's notes formatting, as FPWD.

RESOLUTION: publish current version, with formatting changes to mark incomplete sections using editor's notes formatting, as FPWD.

McCool: Toumura san, you have this really nice diagram of the entire process. could you post it?

Toumura: yes

<kaz> Issue 53 on Toumura-san's diagram

McCool: it is a great diagram, we can start from there and maybe fix some minor issues
... once we have FWPD, is pretty easy to publish

Kaz: that's true
... Will the diagram be included in the next version? not the FWPD?

McCool: I think so
... in the next week we are going to cancel the Discovery call. We have a conflict with the F2F schedule
... there're also some other conflicts so the next three calls are canceled
... we may do Oct 19 call, only if we really need
... kaz also have a conflict for the 19th

Publication schedule

McCool: the only problem is that we'll not have a meeting to merge the next PR that I am planning

Kaz: regarding the publication procedure, we need to bring the draft to the main call on Wednesday, Sep. 30 to get the whole WG approval

Cristiano: can't we use Github review process to decide if the PR can be merged?

McCool: yeah, I'll create a PR, send an email, and wait a week before merging. I aspect that people will give their reviews on github

AOB

McCool: anything else?

<kaz> meetings on Oct 5, Oct and Oct 26 will be cancelled

McCool: ok, please don't forget the F2F activities, there are some intresting meetings. like joint meetings, one with DID and another with Web&Networks.

<McCool> https://github.com/w3c/wot/tree/master/PRESENTATIONS/2020-10-online-f2f

McCool: if you are resposable for presentations, please upload them on github wot repository

<kaz> [adjourned]