W3C

Spatial Data on the Web Coverages Sub Group Teleconference

23 Nov 2016

Agenda

See also: IRC log

Attendees

Present
roba, phila, ChrisLittle, kerry
Regrets
scott, dmitry, bill
Chair
kerry
Scribe
kerry

Contents


Due to low turnout, this was an informal meeting.

Progress on QB4ST https://www.w3.org/2015/spatial/wiki/RDF_Datacube_for_Coverages - Rob

rob: asking about automation of TOC and how to do biblios

<ChrisLittle> * apologies for being late

<phila> Specref

phila: TOC is right, for biblio see specref (link inserted above)

roba: would be good to have some of this in boilerplate template...

phila: agrees

<phila> ReSpec

phila: points to this goude on the wiki
... would be good to have the OGC specs in specref too --- you can get all W3C refs from there

<phila> SDSVoc

phila: talked with scott about tooling last week --- there is a vocab tools meeting next week

roba: would like to avoid having a coder smuggle a vocabulary into a database --- we need to do better than windup clockwork
... graph structures
... people smuggle semantics into names in microformats
... and only one portal understands it

<ChrisLittle> +1 To Rob's Sermon!

roba: so you do search broadening and get nothing useful

<ChrisLittle> S/portla/portal/

back to qb4st..

phila: appraoches to vocabs and how to describe datasets... DCAT is not enough

roba: need ability to build graph out of standardised modules

phila: will pick up where data on th web stopped
... also includion negotiating metadata formats
... and vocabulary

roba: profiles are important for discovery...

ChrisLittle: workshop next week: is anyone from RDA coming?

phila: yes, including keith jeffery who I work with

ChrisLittle: I quite liike some of the RDA recs
... stable and persistent identifiers, archive and search mechanisms, really very important

phila: rda worls group think is for DOIs only (inc vocab terms)
... I tweeted on April 1st that all vocab terms will be dois
... NB for the minutes --- it was a JOKE!
... all uris become 404s ---- furphy!
... rants on about dois

chris: my action for reviewing QB4st is happening now

roba: would it be useful to point people at workshop to qb4st?
... active pice of work...

<scribe> scribeNick: kerry

roba: would they be interested in helping further?

phila: I can bring it up at the workshop
... barcamp session next Wed/Thursday
... exemplar of issue to describe important aspects of data in finer detail.... need something beyond a fixed schema, and making search mechanisms work so you need governance
... could advertise this process
... last thing on second day (overflow slot)

<phila> SDSVoc Bar Camp

eo-qb http://w3c.github.io/sdw/eo-qb/

roba: is it ok the mention linked data api as implemented in the UK?

phila: no normative references so rules are more relaxed.... can refer to linked data api but not a specific product or company

Progress on http://w3c.github.io/sdw/coverage-json/

phila: chris have u heard of oceanwise? a company that does GIS/Marien stuff
... was talking about the time ontology and that we are short of eyes on this
... want to see some questions asked...
... will u have time and incination for time?

s/inination/inclination/

chris: have been working on OGC to progres some of thse calendar vs CRS issues

phila: asks for video on that talk in Nottingham

chris: slides are somewhere

phila: ... has to be in a state where it is understandable .... what is there now is fine
... I will take care of respec problems

kerry: bill still to do update and review before December vote

<phila> OWL TIme

kerry: scribe talking too much

[discussing owl-time]

kerry: can put owl-time on every plenary

chris: asks what an ontology implementation looks like

phila: 2 independent people in different systems have exchnaged data using each term

roba: if we have a term in a dataset where the the namespace is declared somewhere in the metadata --- would that doo or is fully qualified uri necessary?

phila: ok if we can show that the ontology is actually encoding something that has been used for fifty years everwhere this might do

roba: this data is declared via an equivalent uri

chris; will be using webcams for a little project in Jan -- can use this as an implementation I think

scribe: tells me it snowed

chris: all the interantional climate predictions ...

phila: if we can find lots of implementations of the old owl then we can put the old stuff as normative and the new stuff as non-normative

chirs; good for plan b

Summary of Action Items

Summary of Resolutions

[End of minutes]