RE: W3C/OGC WG Launched!

?  Reading some of the discussion going on at GeoJSON, it seems that a big gripe is with the old style CRS definitions ...

That is part of the story. The issue is that the original GeoJSON bolted on a way to refer to a CRS, but insisted on right-hand ordering of the coordinates regardless of what the CRS definition actually said. If GeoJSON had a rule that only right-handed CRS were permitted (as noted below, OGC provides some r-h versions of well-known l-h systems), then everything would be fine. But the way the GeoJSON spec was written there was a risk of inconsistency between the CRS definition and the data. That was not perceived as a problem in the GeoJSON community, for whom the GeoJSON spec trumps everything, but was a major no-no for some elements of the geospatial standards community, particularly the 'military-industrial complex' end of the spectrum.

The stand-off has been resolved by the GeoJSON team deprecating the CRS tag in the version of GeoJSON submitted to IETF [1]. This largely satisfies the standards community as far as the technical content is concerned. And if GeoJSON is published by IETF rather than in the current ad hoc way, it also gives it enough status to be referred to normatively by other standards.

However, the GeoJSON community has now found out that the processes in IETF are at least as arcane as every other standards organization - i.e. it all works fine for insiders, but is weird and frustrating for everyone else. Their Internet Draft has failed to progress, and will expire next week [2], so they are now figuring out whether to persist with IETF, of maybe look for a forum where more assistance might be available.

[1] https://datatracker.ietf.org/doc/draft-butler-geojson/
[2] https://github.com/geojson/draft-geojson/issues/62

From: Frans Knibbe | Geodan [mailto:frans.knibbe@geodan.nl]
Sent: Friday, 16 January 2015 10:00 PM
To: Cox, Simon (L&W, Highett); andrea.perego@jrc.ec.europa.eu; John.Goodwin@ordnancesurvey.co.uk
Cc: public-locadd@w3.org
Subject: Re: W3C/OGC WG Launched!

On 2015-01-13 5:42, Simon.Cox@csiro.au<mailto:Simon.Cox@csiro.au> wrote:

Another possible consideration:



JSON-LD is being talked about as a viable serialization format for RDF data - some folk are suggesting that it will soon be the default, thus also helping with RDF's developer-acceptability problem. In which case, the issue of overlaps between LOCN, GeoJSON, and other RDF vocabularies becomes more urgent and maybe more tractable.



I am aware that the GeoJSON team has some thoughts around moving CRS support into a "-LD" extension to GeoJSON [1]. Now while the GeoJSON team are mostly concerned with the encoding, lets please make sure that the implications of this option to the RDF view is considered. I'm not yet familiar enough with JSON or JSON-LD to tease it out, so I hope someone else maybe has the knowledge and motivation.



[1] https://github.com/geojson/geojson-ld/issues/27
Thanks for the notification.
Until now I haven't had any experience with GeoJSON, but I have played with JSON-LD. As far as I know, JSON-LD is a notation for RDF graphs, so it is similar to other RDF formats like Turtle and RDF/XML. So whatever definition of CRS we come up with, it should be fine in JSON-LD.

GeoJSON on the other hand seems to be an alternative notation for geographical data. So it is similar to WKT or GML. Personally, I think geospatial data are better off making use of common models like RDF instead of using yet another domain specific model, but again I see no conflict. Perhaps it would be beneficial for GeoJSON-LD if they can use locn:crs instead of having to define geojson:crs?

Reading some of the discussion going on at GeoJSON, it seems that a big gripe is with the old style CRS definitions, as standardised by the EPSG. The main problem there is that geodetical reference systems (those using degrees instead of meters as units) usually have a different axis order, meaning that coordinates take the form of (latitude,longitude), which translates to (y,x) in the cartesian systems that are used for flat maps. This causes headaches among web developers. But I think the OGC already solved that problem (or is solving that problem?) by removing references to EPSG from CRS definitions. For example, http://www.opengis.net/def/crs/OGC/1.3/CRS84 is just like EPSG:4326, only the axis order is different. If everyone starts using those new OGC definitions, there no longer will be axis order confusion.

Regards,
Frans







-----Original Message-----

From: Andrea Perego [mailto:andrea.perego@jrc.ec.europa.eu]

Sent: Saturday, 10 January 2015 12:44 AM

To: Frans Knibbe | Geodan

Cc: LocAdd W3C CG Public Mailing list

Subject: Re: W3C/OGC WG Launched!



Very good news indeed!



I do agree with Frans that it would be important to consolidate the LOCN voc, so that it can be used as an input to the SDW WG.



I'm going to open a separate thread on this.



Cheers,



Andrea





On Fri, Jan 9, 2015 at 12:42 PM, Frans Knibbe | Geodan <frans.knibbe@geodan.nl><mailto:frans.knibbe@geodan.nl> wrote:

Hello Phil,



Excellent news! A lot has been going on in getting spatial data to

work on the web, but that work is quite fragmented. I hope the new

workgroup will serve as a focal point for many interrelated efforts.

Thank you for your contributions in making this happen! I guess it wasn't the easiest of tasks.



And I am sure the LOCN vocabulary can play an important role,

providing low-threshold semantics for location data. All the more

reason to take the job of maintaining the vocabulary seriously.



Some of the things that come to mind as things we can give some extra

attention to this year:



Finish the task of providing explanatory text in multiple languages;

Finish the proposal for adding CRS and spatial resolution to the

vocabulary; Solicit experiences from people in the field who are using

the vocabulary and see if this gives reasons for change or consolidation.



Regards,

Frans



On 2015-01-06 17:28, Phil Archer wrote:



Dear all,



As you may have seen an hour or so ago, the much talked about

collaboration between W3C and OGC is under way!



The two standards bodies issued a joint press release today and, if

you're eligible, obviously we'd be delighted if you were to join the working group.



Relevant Links:



The press release

http://www.w3.org/blog/news/archives/4287



The Working Group

https://www.w3.org/2015/spatial/



OGC have analogous pages too of course.



Something of particular relevance to the LOCADD CG is that the charter

for this new WG is sufficiently flexible that it could potentially

provide a vehicle for more formal standardisation of the LOCN vocab.



Cheers



Phil.







________________________________

Frans Knibbe

Geodan

President Kennedylaan 1

1079 MB Amsterdam (NL)



T +31 (0)20 - 5711 347

E frans.knibbe@geodan.nl<mailto:frans.knibbe@geodan.nl>

www.geodan.nl<http://www.geodan.nl> | disclaimer

________________________________







--

Andrea Perego, Ph.D.

European Commission DG JRC

Institute for Environment & Sustainability Unit H06 - Digital Earth & Reference Data Via E. Fermi, 2749 - TP 262

21027 Ispra VA, Italy



https://ec.europa.eu/jrc/



----

The views expressed are purely those of the writer and may not in any circumstances be regarded as stating an official position of the European Commission.



________________________________
Frans Knibbe
Geodan
President Kennedylaan 1
1079 MB Amsterdam (NL)

T +31 (0)20 - 5711 347
E frans.knibbe@geodan.nl<mailto:frans.knibbe@geodan.nl>
www.geodan.nl<http://www.geodan.nl> | disclaimer<http://www.geodan.nl/disclaimer>
________________________________