See also: IRC log
<kerry> regrets Jeremy, Frans, Lewis, Rachel, Ed, Danh, Clemens, Simon
<kerry> regrets Jeremy, Frans, Lewis, Rachel, Ed, Danh, Clemens, Simon
<phila> The new look BP doc
<Linda> Linda+
<laufer> hello all
<laufer> please, could anybody tell me the webex password?
<ChrisLittle> v4qtEh56
<phila> scribe: phila
<scribe> scribeNick: phila
<laufer> thank you
Topic Last week's minutes
<kerry> https://www.w3.org/2016/02/24-sdw-minutes.html
<Linda> +1
<kerry> +1
PROPOSED: Accept last week's minutes
<RaulGarciaCastro> +1
<robin> +1
<AndreaPerego> +1
<ChrisLittle> +1
<ClausStadler> +1
<billroberts> +1
+0 (absent)
RESOLUTION: Accept last week's minutes
<kerry> https://www.w3.org/2016/02/24-sdw-minutes.html
https://www.w3.org/2015/spatial/wiki/Patent_Call
kerry: I believe BP set a date for their calls
Linda: Wednesdays at 15:00 UTC
<scribe> ACTION: phila to set up WebEx for the BP task force call [recorded in http://www.w3.org/2016/03/02-sdw-minutes.html#action01]
<trackbot> Created ACTION-147 - Set up webex for the bp task force call [on Phil Archer - due 2016-03-09].
kerry: Is that every week or just
when we're not doing the joint meeetings
... There will be agendas adn minutes
<robin> Thanks Phila, I have finished the registration.
kerry: Coverages was getting close - billroberts?
billroberts: Yes, we haven't
fixed it yet. I've got some responses that I'll summarise
tomorrow.
... Main problem is that we have people in both China, Aus, USA
and Europe... so it means *someone" is bound to miss out. May
need to alternate times
... Current plan is to propose an alternating time
kerry: I recommend against that, you have alternating times and alternating weeks, that's hard
billroberts: The alternative is
to tell someone they can't take part.
... We'll go on continents
kerry: If I can't join, I can't
join
... try and get the editors there - they're the most
critical.
... Having said that, we have the same problem with SSN
... How's time going.
ChrisLittle: I failed to e-mail Simon to get the first slot. I haven't thought much as I'm rushing off to Washington shortly
kerry: SSN Doodle poll still to be filled in by some gfolks, including an editor
<kerry> SSN may be Th 8pm UTC
kerry: Please fill in the Doodle
poll as soon as possible. Please be generous with your
availability - 3 slots may not be enough.
... Plan is that all those meetings take place next week so
there won't be a call at this time next week.
phila: So this call only happens every 2 weeks now so that the other TFs can meet in between.
<billroberts> yep
<billroberts> scribenick:billroberts
<kerry> scribe: billroberts
jtandy, billroberts, denise mackenzie, maik reichert visited Beijing to discuss CEO-LD
'Coverages and earth observations in linked data'
a small amount of funding from the UK government to prmote UK-China collaboration to allow meetings to take place
University of Reading (Jon Blower and Maik Reichert) have done some very promising work on representing coverage data in JSON 'CoverageJSON'
<ChrisLittle> * jeremy now going to Korea
The meeting led to 3 Chinese organisations to commit to developing demos with reprsenting coverage in JSON
the resulting data can be presented in the browser using prototype software from U of Reading
We made good progress
kerry: has a group of final year
software engineering students interested in working in this
space
... they are just getting started this week.
... is very keen to participate in coverage calls and to keep
in close touch, in order to liaise with the students and make
best use of their time, skills and enthusiasm
... planning to use the DGGS
so backend might be different to that used by Reading
phila: will keep kerry informed on outputs from CEO-LD. There are things the students could do which would be very helpful
kerry: has asked the students to look at RDF Data Cube, which would be quite a different approach to that used by Reading
phila: note that the RDF Data
Cube approach is very verbose so may not work for large
coverage data
... want to take the RDF Data Cube's notion of pre-defined
slices to help with subsetting of coverage data. Should be
fairly easy particularly for gridded data
chrislittle: has discussed this
with jtandy and will raise it with OGC next week. Will raise in
WCS SWG
... data cube may need enhancing. A 'slice' could be too much.
Actually want a 'dice'.
... can request an n-dimensional 'tile'
... have started looking at it but not got very far yet
phila: that's it for the Beijing meeting
<phila> scribe: phila
<kerry> scribe: phila
<scribe> scribeNick: phila
kerry: Thanks to the members of that WG for joining us
<laufer> and me...
<BernadetteLoscio> thank you!
And laufer
<BernadetteLoscio> and Newton
kerry: Can you lead this bit Linda?
Linda: Not too prepared but I can step in...
Linda: I was looking through the
minutes from last time. The topics that we wanted to discuss
more were mostly around APIs
... What to write about that and maybe also something about the
DQV
... And there was something more about identifiers, but I'm not
sure why.
... I'd like to hear from DWBP on these
BernadetteLoscio: ... WE had some discussions about APIs and made some changes in the GH doc
<BernadetteLoscio> http://w3c.github.io/dwbp/bp.html#dataAccess
BernadetteLoscio: One of our worries is the intesection between the BPs.
<newton> http://w3c.github.io/dwbp/bp.html#documentYourAPI
BernadetteLoscio: No 25 says
document your APIs which is related to SDW 29 that says APIs
should be self decribing
... They seem to be more generic, so we were thinking - if the
BP that we have is enough for you, or what change should we
make?
... Or should we add a new one to incorprate your needs
... Our 26 is similar to your 28
Linda: Looking
... Our 28 says you should have a convenience API that allows
users to ask basic questions - make simple queries
... because APIs tailored to a simple goal
BernadetteLoscio: I think we don't have this kind of detail
<annette_g> no , we don't have that in dwbp
BernadetteLoscio: What we were discussing, for e.g., this notion of data granularity isn't clear for us. Is that the same as subsetting data?
Linda: Yes, making a subset of the data
BernadetteLoscio: How do you see this?
<annette_g> *the line is noisy
Linda: I think that some of the
content that we have in SDW on APIs is more detailed. It's up
to you how much of our content you'd like to use.
... I know that JT offered to work with you on that
<annette_g> *it's better now
<ChrisLittle> * the panting dog is back on audio
annette_g: There are some things
that we're thinking about changing in DWBP. Maybe we can make
those changes and then let you know
... Then SDW can decide whether you want to keep it.
... We looked at your self documentation info. We might say
that self-documenting is a way to do that.
... And we may offer something on subsetting
... We can make those changes and then you can take a look
Linda: That sounds like a good
approach to me
... If you make some changes, then please let us know,
especially Jeremy, Payam and me
... If you're adding more content on APIs then we can lose some
of ours
<annette_g> *I don't hear any sound
BernadetteLoscio: I talked about
the identifiers section at the end of the meeting
... I was reading the SDW doc. You have some BPs for
identifiers that are similar or that cover things that we don't
cover
... And I was wondering ... what I'm worried about is that we
have BPs that are similar in 2 different documents. Maybe we're
going to have a common audience
... I think for e.g. you have BP1 on gloablly unique
identifiers and we have BP11 on persistent IDs for datasets. We
don't have anything like your BP2
... So is there a way for us to refer to your doc? Can we
cooperate more closely?
... For e.g. BP2 is not specifically spatial
<newton> (we are without numbering in our BPs for while, so, for now, its easier to see the index on this page: http://w3c.github.io/dwbp/bp-status.html )
Linda: What we tried to do in SDW
is that it's an extension of DWBP
... We want to avoid having similar BPs.
... What we have is a first draft. There are thing there that I
don't think are specifically spatial, so if we can move some of
our BPs to your doc, that makes sense to me
BernadetteLoscio: Thank you
Linda: Then we'd have to discuss which ones.
BernadetteLoscio: I think we have
to have this discussion in our WG too of course. It woujld be
good to align them.
... We discussed this in our WG - how our BPs can be extended
in specific domains anda SDW will be one of those.
Linda: That's interesting as
we're alreadyt doing and we can describe how to do it.#
... So at the moment, when we think there's an overlap, we make
a reference to you
annette_g: I was just going to say that SDW has used references, that's fine - I don't think we need to do anything special with that
BernadetteLoscio: I was saying that the idea of using references is all right but they have some BPs that we don't haev but that maybe we shuold have
annette_g: +1
BernadetteLoscio: The ones they have should be more specific
annette_g: Sounds like everyone agrees with that
<Zakim> phila, you wanted to makea boring process comment
<BernadetteLoscio> ok... i understand
phila:
Talked about need for implementation evidnce requirements in DWBP and the fact that SDW is a joint Wg with OGC which *may* have an effect - certaunly we need to be aware of it if moving content
BernadetteLoscio: Did you find a gap in our doc about linking data?
Linda: hard for me to answer as Jeremy wrote that. It says that it extends your BP. It talks about linking at data level, you ytalk about dataset level
phila: That sounds like the BP I wrote needs more work
BernadetteLoscio: Not sure we havae one about link creation
<scribe> ACTION: phila To look at data item linking cf SDW BP on this topic and see if more needs to be done [recorded in http://www.w3.org/2016/03/02-sdw-minutes.html#action02]
<trackbot> Created ACTION-148 - Look at data item linking cf sdw bp on this topic and see if more needs to be done [on Phil Archer - due 2016-03-09].
BernadetteLoscio: We dodn't talk about links because we dodn't want to be biased to RDF
annette_g: Were SDW thinking about hypermedia APIs for that or something else?
Linda: I don't have that in my mind
ChrisLittle: A lot of spatial data is complex, lots of objects in a box anbd you want to pick those out. That's not really hypermedia
annette_g: We've been adding a little more about hypermedia APIs and we are trying to say mnore about that
ChrisLittle: I don't think we've talked about streaming at all. If we do then we would get into hypermedia
annette_g: Our discussion isn't about streaming, it's about an API that returns links as well as data
kerry: There was one more topic that, about data quality. Is there sometehign we can do about that. Do we need anotehr meeting?
BernadetteLoscio: I think it's better to meet about that separately with those editors (who aren't here)
Linda: I could get involved with that. We have to make some recommendations about precision and accuracy and I think the DQV could help us with that.
<scribe> ACTION: phila to put Linda in touch with Antoine and Riccardo [recorded in http://www.w3.org/2016/03/02-sdw-minutes.html#action03]
<trackbot> Created ACTION-149 - Put linda in touch with antoine and riccardo [on Phil Archer - due 2016-03-09].
kerry: No joint meeting next
week, but the sub groups should meet
... Ed and I will try and join those calls
... We can use this IRC channel and the agendas and minutes can
go in the same system as we use for the group.
... The meetings are open to all, they're just specialised to
SDW.
<ChrisLittle> +1 to kerry re minutes irc etc
kerry: Any more comments?
<ChrisLittle> bye
<billroberts> ok thanks, bye
<annette_g> bye, and thanks!
<RaulGarciaCastro> bye
kerry: Then that will do for now.
<LarsG> bye and thanks
<Linda> bye
kerry: Thanks to DWBP folks for coming
<robin> bye
<laufer> bye
<newton> bye, thanks
<BernadetteLoscio> bye, thanks
BernadetteLoscio: Pleasure to be here thanks for having us