SDW WG Weekly

<scribe> scribe: chaals

<jtandy> +jtandy

<phila> Agenda: https://www.w3.org/2015/spatial/wiki/Meetings:Telecon20150128

<eparsons> http://www.w3.org/2015/spatial/wiki/Meetings:Telecon20150128

EP: Minutes have been published.

<phila> Last week's minutes

More introductions

<Simon> +q

EP: If you didn't introduce yourself last eek, please queue up now

GeorgePercivall: OGC staff, joining for the first time

Christine Perey: Joining from Switzerland

MattP: On SPARQL group

Kevin: Sent email introduction to the list.

AH: Question on minutes. Attendee list may not be complete -is that a problem?

<jtandy> (agreed - it seemed incomplete)

EP: Don't think so...

PA: If people are not listed please tell me and I will correct that.

… Please check and let me know.

<aharth> phila: please add me :)

[In IRC you can type "Present+ MyName" (no spaces!) and you will be added]

<ramon-alcarria> phila: please add me too

<jtandy> (jtandy not listed in last week's minutes)

<ioannis> chaals - I attended last week's call

SimonCox: Introduced mysefl by email yesterday. Last week I was slacking on the beach.

ChrisLittle: Chair of MetOcean domain group, Met office in UK.

<DanhLePhuoc> +35385740aahh is me, DanhLePhuoc

BVL: Been a W3C member for linked data work.

… will send intro to the mailing list.

Review the charter

-> http://www.w3.org/2015/spatial/charter Our charter.

EP: The charter identifies what we are trying to do. I don't propose changing it now, but want to focus on the deliverables - especially use cases and requirements.

… It is important to focus on solving real problems. The best approach is to work on use cases, and find the requirements that come from them.

… Is there a view on the scope and size of successful use case work?

<phila> chaals: You start by laying out all the UCs that people can think of

<phila> ... then you review the UCs, requirements then drop out almost by magic

<phila> ed: Any idea of the right number? A couple of dozen?

<phila> chaals: That's a good number. It's about talking about problems that people actually have rather than invent UCs that never actually arise. There is not ideal number

i/chaals: you/scribe: phila

<KJano> Use cases wiki

<scribe> scribe: chaals

<KJano> yes, will contribute

<Simon> Yes, will contribute

EP: We have a couple of use cases already, I plan to add a few from the perspective of a mas-market search provider. Are there others expecting to add use cases?

<Linda> yes, probably

<cory_henson> yes, will contribute

<LarsG> yes, I will contribute a use case

<ocorcho> UPM (OEG) will contribute

<jtandy> +1 to work toward f2f

<Simon> +1

EP: What about timelines. Should we ask for them to be done before the face to face?

+1

<KJano> +1

<AndreaPerego> +1

<Payam> +1

<Clemens> good for me

<BartvanLeeuwen> +1

<Linda> +1

<JoshLieberman> There will be use cases to contribute from the OGC Testbed 11 activity and they will have to be done by March

<MattPerry> +1

<LarsG> not sure I can make f2f

<armin> +1

EP: Will talk about the face to face, but think it is a good timeline for getting use cases together.

<armin> can't make the f2f :-(

<ChrisLittle> +1

<KJano> I would propose that each use case includes an example

… I think it makes sense to create use cases that are not related to you but that you can consider through discussions with other organisations.

<Simon> I also won't make f2f but plan to dial-in

… There may well be a task required to boil down the use cases somewhat.

<AndreaPerego> +1 to KJano

<KJano> same, here will dial-in for the f2f

<jtandy> (lost audio ... will dial back in)

… If use cases don't appear we might try to give action items.

… Is it useful to do other work in parallel?

… any views?

<ocorcho> I prefer getting first the use cases

<KJano> in the w3c ssn-xg we collected related and relevant work in the wiki and that turned out to be very useful

JL: My experience is that statistical approach to use cases isn't very effective. It works better to let deliverables and use cases inform the other.

… The most valuable use cases connect a real activity with a reasonable scope.

EP: Suggesting more qualitative than statistical approach. But is there work to do before we get use cases together?

JL: Yes, think there is some in the charter.

SC: Don't think we should postpone the discussion. Should get use cases from real backgrounds for the discussion

… but we can run the discussions in parallel. But should have a set of use cases before we make decisions.

EP: Support that.

<Zakim> phila, you wanted to talk about BPs

<phila> Template for Best Practises group use cases.

PhilA: I think use cases are going to be slightly ahead, but when writing a use case it is useful to have a clear idea of where we are heading.

… The work we do has influenced the use case, helping to write them in a way that helps do the work.

… These documents can be iterated over time.

<Zakim> jtandy, you wanted to suggest that use cases would be good to include real data within their examples (we can use these for test cases later)

<ChrisLittle> q

… But having some idea of where you are heading enables you to write use cases well.

JT: When writing use cases, focus on tangible things and using real data is helpful.

EP: It is good having a link to a real customer.

<ThiagoA> I´m back

CL: My use cases reflect different aspects - more of a scenario you can break down into use cases. Sometimes scenarios give some framework to contextualise several use cases, which help define something of a roadmap as you shake it down to concrete requirements.

EP: Sometimes we come to use cases from our own perspective, and need to get the different views included. With a breadth of use cases we might be able to mitigate the biases.
... Suggest we work toward the deadline of the first face to face.

First F2F meeting

<phila> OGC TC details

EP: Will be held jointly with the OGC meeting in Barcelona in March. has everyone been able to see the details needed to register? If you are not an OGC member everything should be publicly available.

… Our meeting will be all day on Wednesday. We'll work on the agenda over the next few weeks.

<Simon> sorry - was typing

… It is good to get people to see each other.

<AndreaPerego> s/HAs/Has/

<JoshLieberman> Simon, thought it was castanets...

CL: OGC is very busy, and there are several things going on in parallel. So it will be difficult to make it. We should pin down the agenda to help.

EP: Think it will be a learning experience.

<JoshLieberman> Possibility of F2F on the Friday, after TC activities?

… There is a different tempo of meetings between OGC and W3C. Latter has whole-day [NdelT or multi-day] meetings.

<Simon> In support of ChrisLittle statement - OGC TC not good for more than 1/2 day meeting

CL: GML and JP? are scheduled so far, 3D working group, …

<Simon> For real work, other timings should be considered.

EP: There isn't much we can do about it. There is an ideal of running joint meetings, but this will be a learning experience. Maybe we will end up holding meetings at the end of the week...

… are there issues with getting registered?

<AndreaPerego> Tentative OGC TC agenda, including SDW WG f2f: http://www.opengeospatial.org/event/1503tcagenda

EP: That's the agenda. I cannot make call next week, but my co-chair will.

<Zakim> jtandy, you wanted to ask what level of detail you need for the f2f

JT: Is the level of detail enough to triage what I have added to use cases, or should I add more?

EP: Not sure. Think they may be a little light, but don't want to ask for massive amounts of effort. It might help if people add detail to them as we go - perhaps for now we should aim for depth, and flesh them out later.

… What is your sense?

JT: For requirements to magically be clear we are going to have to add some more detail to what we have. But recommend we start aiming for breadth, and look at what appears to have common issues, then elaborate.

… rather than elaborating everything in parallel.

EP: Agree

+1 to Jeremy

<BartvanLeeuwen> +q use the why how what principle for UC

<Zakim> phila, you wanted to be boring about zakim names etc.

<BartvanLeeuwen> +q to use the why how what principle for UC

<Zakim> BartvanLeeuwen, you wanted to use the why how what principle for UC

BVL: Wanted to add the idea of using the How Why What. Use cases can get technical. We need to know what needs to be solved and why, rather than How...

… and we need to make sure the use cases are actually in scope.

administrivia

PhilA: Useful approach is to have specific questions we want to ask.

<phila> are specifically relevant to data published on the Web;

<phila> encourage publication or re-use of data on the Web;

<phila> can be tested by machines, humans or a combination of the two.

… would need to be amended for this group, but the approach is really helpful.

EP: Sounds great. Can you append that to the Wiki page?

PhilA: If we agree on what the questions need to be for this group…

First question - for this group we want…

<phila> PROPOSAL: That the first scope question be ... Is the use case specifically about Spatial data on the Web?

+1

<BartvanLeeuwen> +1

<eparsons> +1

<Linda> +1

<AndreaPerego> +1

<KJano> +1

<LarsG> what about temporal data?

<MattPerry> +1

<jtandy> +1

<ChrisLittle> +1

<Payam> +1

<Clemens> +1

<KJano> spatial always includes spatiotemporal

<Ingo__> +1

<IanH> +1

EP: There is a danger that we get tied up in the nuances of wording...

<Simon> Located temporal

<Ashok_Malhotra> +1

<ThiagoA> temporal data on the web ?

<armin> we do need use cases for the time ontology

<LarsG> +1, just wanted to know...

<DanhLePhuoc> +!

<DanhLePhuoc> +1

<LarsG> resolved for me

<phila> RESOLVED: That the first scope question be ... Is the use case specifically about Spatial data on the Web?

<phila> 2nd q is encourage publication or re-use of data on the Web;

<KJano> so does the resolution include spatiotemporal or not?

[Phila uses this as a teaching moment]

[Spatial includes spatiotemporal since all spatial data includes spatiotemporal data]

<KJano> yes, thanks

PA: What's the equivalent question for this group?

JT: This question was specific to the data on the web group. For spatial data you get a very amorphous requirement...

… don't think this question gets tight enough for our purposes.

EP: THis is for data you are trying to share. No?

JT: On a higher level, yes.

<jtandy> ok

… OK with the question in its broader sense, but not sure it is specific enough to help.

<jtandy> (sorry - poor network connection I think)

[+1 to parking this question for now]

BVL: Question with scope - the *public* web?

<Percivall> Is the use case specifcally about how spatial information is linked with other data on the Web?

<JoshLieberman> f+1 to Percivall

BVL: I think DWBP group is specific to public web, I want to use web technology that may not be public.

<cperey> +1 to remark ed made

KJ: We shouldn't be overly specific on this call.

<cperey> this is not about public web

… Should be driven by use cases.

… if we have great use cases for something we should do it…

<JoshLieberman> breathers please mute

CL: Don't see distinction between public web and others. But the issue is tightening the scope. We want things published, but we want them to be interoperable.

<Zakim> chaals, you wanted to suggest that the private web is still the web.

CMN: +1 to Chris

<cperey> +1 to consistent use of Web technologes

<Zakim> phila, you wanted to talk about spatial/Geospatial

<BartvanLeeuwen> +1 as well ;)

PhilA: No difference between public and closed web.

<KJano> just to make my point more clear: lets not use this call to sort out whether spatial also includes geospatial, or even geo-data on the web as such or whether by web we only mean the public web,etc. Lets be driven by the uses cases and derive requirements from them.

… Someone mentioned spatial and geospatial - the point is that we were talking about non-geo use cases.

[+1 to non-geospatial use cases being in scope]

<KJano> +1

<eparsons> +1

<LarsG> +1

<ocorcho> +1

… we cannot tackle things we don't have someone working on. Decisions are made by people who turn up…

GP: Use case about spatial information linked to other information on the web

<Simon> +1 Percivall

<BartvanLeeuwen> +10 Percivall

EP: How about not trying to be too prescriptiive up front - work on them first, then check them.

<Simon> Linking is distinguishing feature of 'the web'

<KJano> +1 for use cases first

<AndreaPerego> +1 to Ed

<jtandy> (phila's point that "can be tested by machines, humans or a combination of the two." is good)

EP: OK. Let's go do work on use cases.

… Any other business?

[Adjourned until next week]

<jtandy> thanks chaals for scribing; thanks ed for chairing

<eparsons> thanks chaals

<DanhLePhuoc> +35385740aahh is me

<cperey> bye bye everyone

<ioannis> [IPCaller] is me

<KJano> -1

<cperey> -1

<aharth> +1

<armin> +1

<eparsons> +1 to later

<LarsG> -1

<AndreaPerego> +1

<Simon> +1 1/2 hour later

<ChrisLittle> -1

<cperey> -1 late enough

<Linda> -1

<JoshLieberman> +1

<cory_henson> +1

0

<BartvanLeeuwen> 0

<ocorcho> 0

<jtandy> also late enough for me

<Payam> 0

<BartvanLeeuwen> why ?

<jtandy> +0

<BartvanLeeuwen> ;)

i/[ipcaller is me]/Phil, can we move the call half an hour later?

<Simon> Good *morning* everyone

<BartvanLeeuwen> thx bye

<ChrisLittle> bye

<AndreaPerego> Bye bye

<Payam> bye

<Linda> bye!

<ocorcho> bye

<KJano> thx bye bye

<ioannis> bye

<armin> bye

<MattPerry> bye

<LarsG> good *night*

<Clemens> bye

<JoshLieberman> bye

<Ramon_Alcarria> goodbye everybody!

<eparsons> By all