Presentation for OGC

[cholmes]

Just gave an overview for OGC Members meeting, that others might be interested in:

https://docs.google.com/presentation/d/15QiUpxT9eoATkyqzk302I_sPVCN5T4src2_XFic7uGQ/edit#slide=id.g22b969406e6_0_348

[cholmes]

We had a good small group - it was more ad hoc since the SWG isn't quite formed and I didn't get us on the agenda. In general people thought it looked good, we went pretty deep into the details. Some of the feedback was:

• projjson is our most controversial choice (which we knew). Esri was at the meeting and fully understood why we made the choices - their engine. There will be a json crs thing from OGC at some point, and it'd be ideal if that was there and we could use it. But it seems like there's a reasonable solution for Esri to be able to adopt it, by adding an extra (non-standard) metadata field of WKT.
• There were questions about the choice of long/lat for the bounding box, vs a bounding box in the CRS. Other specs have two fields, one for the long/lat one, one for the crs native bounding box.
• The 'edges' field was not understood initially, with the thought that the CRS should specify it. But I think it made sense to people when thinking about geojson, how this is about the interpolation of data between two points. I probably should have jumped to our detailed explanation in the spec more quickly, but we might also consider trying to make this a bit more clear in the first table.

Overall the feedback was quite positive. Hopefully we'll see some implementations from ESRI and Safe FME, and some data in GeoParquet from Ordnance Survey. Google was also there and supportive.