Spatial Data on the Web Working Group Teleconference
31 May 2017
[General chat about whether we are quorate]
[PhilA expels Doug Shepers and Michael Cooper from WebEx]
<eparsons> https://www.w3.org/2017/05/17-sdw-minutes
<Linda> +0 was absent
<eparsons> +1
<mlefranc> +1
<ahaller2> +1
0 Not present
Resolved: Minutes of 17 May Approved
<eparsons> https://www.w3.org/2015/spatial/wiki/Patent_Call
eparsons: Invites Armin to describe where we are.
ahaller2: Apologises for the short notice.
… We have had the wide review, we had several comments that we have acted upon.
… We have based our Exit Criteria on the Time Ontology one
… Feedback received was very positive but asked for examples so we have included several now.
… These are in an appendix. Still working on them.
… Had different understandings even within the group as to how to use the ontology.
… We'll add these modelling approaches to the examples.
… Apart from that, limited changes since the previous WD
… Fixed some minor errors.
… Split out the systems capabilities module. Used to have limited implementation evidence in old SSN so we have flagged it as at risk. We can make it non-normative if no evidence.
eparsons: There seems to be a lot of change going on.
… How much is tidying up content and how much is significant.
ahaller2: Changes in the last few minutes have been minor (hash for slash)
… Making sections/sub sections etc.
ahaller2: In the last 4 weeks, very few errors. Really all about the examples.
eparsons: So the doc is stable
ahaller2: Yes
KJanowic: There hasn't been any work on the ontologies for weeks. We've been looking at the examples, the modules etc. Not the ontology
<KJanowic> Yes, but you will see that these are examples, change logs, figures, etc
<tidoust> Phil: Did you get feedback on the interleaving of SOSA and SSN?
<tidoust> Armin: Not per se. There were several comments about the lack of examples, SOSA examples, but not on the presentation itself.
<tidoust> ... We've had lengthy discussions in the subgroup about this.
<tidoust> Phil: There's a reference to SSN system way before it gets defined. Please check.
Linda: I've not been following closely, but to be clear, the last WD was on 4 May and you're saying that this new ED is not really different from the published WD.
ahaller2: True.
ahaller2: It feels different because the ToC looks so different, but if you do a diff on the doc, apart from the examples etc. you'll see little change.
ahaller2: There were 2 errors fixed in the ontologies
Linda: ReSpec is showing 2 warnings (security and HTTPS URLs)
tidoust: You don't need one for this doc, I think. There isn't one in the Time Ontology either
… And don't worry about the HTTPS thing - I'll fix that.
mlefranc: I'd like to say to Linda that all of the wide review, we asked people to look at the ED, so they've seen the latest version
[ssn-system is mentioned in 5.2.2 but not introduced by then so needs some explanation]
mlefranc: Maybe we switch the horizontal sections around
ahaller2: That would do it
phila: What is the current status of implementations?
ahaller2: We have some at GSA, who have implemented SOSA
… couple of million samples there
… Several members working on implementations. We know Siemens is looking intoimplementing it too.
… we have the SSN Usage doc from earlier that needs updating
… Only concern is around system capabilities
… hence splitting it up.
<Zakim> tidoust, you wanted to mention need to have producers and consumers in implementations
KJanowic: What is the time line on the implementations? Esp given the direction on producers and consumers
tidoust: Yes, the Director would like to see both sides. The time line is basically end of June.
… If we resolve to publish CR now, it will take a couple of weeks to arrange the Director's call. CR must be at least 4 weeks, so we would need an extension.
… If we get one, it's going to be in a dormant mode, just to let the process run.
<Zakim> mlefranc, you wanted to ask difference with dataset
tidoust: You really need to end the active work by the end of June. You should be able to show progress with implementation by then
mlefranc: Can I query the meaning of producers and consumers? So if we find evidence of producers, we need to find consumers of it?
tidoust: Not necessarily of the same dataset
mlefranc: There are lots of ways of using it
tidoust: The 3rd bullet in the exit criteria might cover this??
tidoust: If it's captured there, you're fine.
mlefranc: if I develop an ontology that extends SOSA/SSN, does that count as an implementation?
tidoust: Where would you do that? The Director is interested in what's in public and in software.
mlefranc: We've got news from IETF to create a European standard based on SSN
tidoust: If it's in the pipe, that's good, even if it's not completed.
<ahaller2> +1 for mlefranc
eparsons: A consumer doesn't have to be software or a service, it can be someone extending the ontology
tidoust: It has to match the criteria in the document.
KJanowic: Because this is an ontology, consumers will include other ontologies.
tidoust: I don't see a problem with that.
DanhLePhuoc: The WoT WG has been discussing this a lot
… So does that show usage?
… And it's also being looked at for iot.schema.org
tidoust: The transition to Proposed Rec, the implementation should already be there.
… Again, refer to the wording of the exit criteria [paraphrase]
eparsons: So in terms of a vote this evening, what's the appropriate wording?
<mlefranc> * /s/IETF/ETSI
<tidoust> [I note there's a recent comment from Dirk Jan Venema that needs an answer too: https://lists.w3.org/Archives/Public/public-sdw-comments/2017May/0024.html]
<mlefranc> * @tidoust --> we will make an example specifically to address his question
phila: It might be... That the Editor's draft at @@@ be published as a Candidate Recommendation subject to, switching sections 5 and 6 around
phila: It might be... That the Editor's draft at @@@ be published as a Candidate Recommendation subject to, switching sections 5 and 6 around, and the Exit Criteria be updated in line with advice from the Director
KJanowic: If we're listing what will be done, then the examples in the appendix might be improved.
<mlefranc> * but this is non normative section, tidoust said yesterday we can do it
tidoust: That comment came today, it's going to be addressed.
phila: Is that going to lead to a change?
mlefranc: There's one use case here that means one of us can write a use case to cover this.
phila: So you're talking about adding another example to the appendix?
mlefranc: Yep.
PROPOSED: That the Editor's draft at http://w3c.github.io/sdw/ssn/ be published as a Candidate Recommendation subject to, switching sections 5 and 6 around, and the Exit Criteria be updated in line with advice from the Director
PROPOSED: That the Editor's draft at http://w3c.github.io/sdw/ssn/ be published as a Candidate Recommendation subject to, switching sections 5 and 6 around, and the Exit Criteria be updated in line with advice from the Director
<Zakim> tidoust, you wanted to insist once more on exit criteria just in case
tidoust: Just to insist on the exit criteria once more. This is what the Director will look for - you'll have to prove it.
<KJanowic> Thanks, we will look at the wording one more time
tidoust: I had a call with the Director earlier today (about the Time Ontology) and he was asking about how each bullet would be proved.
<ahaller2> http://w3c.github.io/sdw/ssn-usage/
ahaller2: I am worried by the 'and' there... that makes it difficult
… If we look at the usage of the old one, evidence might be missing for some classes and properties
… I'm worried about making it too hard.
mlefranc: For the 2nd point, I'd say something like it's an OWL ontology - every class must not be equivalent to OWL nothing.
… No 2 classes conflate and...
… and in the 3rd bullet, 2nd sub bullet... I think it's implied
… Since you can say that if a sub class is used, so is its parent class
KJanowic: May I propose that we do the wordsmithing outside the meeting
<tidoust> Phil: Changing the exit criteria is quite an important decision. The group has to be comfortable that the exit criteria can be met.
<eparsons> I'm back I think...
mlefranc: I'm not talking about modifying the exit criteria, just the way that people will understand what they say.
… I'm fine with it, as long as we change in line with Director's adavice
ahaller2: We don't want to change the exit criteria, Maxime was just saying that one point subsumes another.
<KJanowic> [I have to teach a class in 5 min, if there would be a vote on moving SSN/SOSA to CR, I would vote +1]
KJanowic: Maybe this due to the technical language that we're using. We have to show implementation evidence. We're discussing how to word what is there
Linda: Are you confident that we can prove implementations?
ahaller2: If it's 4 for each class or property, we might need to make some non-normative
Linda: Then you might want to make those bits non-normative
eparsons: If we progress this, you have a month to work on the implementations, during which time there is the potential that some bits may fall out.
<tidoust> phil: If you say here are 4 consuming bits of software, and there's one class that is used only once, I doubt that would be a problem. The Director is able to make a judgement call.
<tidoust> ... Goal is to prove usefulness.
Linda: I don't fee that confident about voting on this. It feels a little last minute and rushed.
… I don't want to stop it, but I'm not comfortable.
eparsons: Well, you can vote zero.
ahaller2: On Linda's comment - can we make it explicit that the WG has 5 days to comment on the vote?
Linda: Can that go in the wording of the vote
PROPOSED: That the Editor's draft at http://w3c.github.io/sdw/ssn/ be published as a Candidate Recommendation subject to, switching sections 5 and 6 around, and the Exit Criteria be updated in line with advice from the Director. These changes to be made ASAP and the WG informed that it has been done. Any objection should be raised within 5 days of that notification.
<ahaller2> +1
<mlefranc> +1
<eparsons> +0
<Linda> +0
<DanhLePhuoc> +1
phila: Notes that Raúl sent a +1 in his mail
<joshlieberman> +1
<mlefranc> *(and two other +1 recorded in the mailing list)
<tidoust> [KJanowic dropped from IRC but I would assume a +1 from him as well]
Resolved: That the Editor's draft at http://w3c.github.io/sdw/ssn/ be published as a Candidate Recommendation subject to, switching sections 5 and 6 around, and the Exit Criteria be updated in line with advice from the Director. These changes to be made ASAP and the WG informed that it has been done. Any objection should be raised within 5 days of that notification.
eparsons: So Armin, the quicker you can get that done, the better. But it's still tight.
ahaller2: I'll do it right after the call.
eparsons: Any more questions?
Linda: I think there were more on the mailing list?
Linda: Bill supported it
<joshlieberman> woo - hoo
<ahaller2> thanks to everyone!
eparsons: Good job, editors - lots of work has gone into this.
<joshlieberman> bye
<Linda> bye
<ahaller2> bye
<eparsons> night all !!