more nuanced aproach

[calvinmetcalf]

based on #10 this

• gives some drawbacks to sqlite based dataformats for data interchange
• suggests geojson as a preferred format for the majority of data interchange
• suggest gml for the (rare) instances where data sets can't be represented as geojson
• are more specific that geopackage is best for local editing not sending to strangers
• are more specific that file geodatabases are appropriate for local editing only in arcgis settings
• go into some of the details about why gml, while theoretically great, is terrible to use in practice most of the time.

[calvinmetcalf]

looks like my editor removed a lot of excess white space, to see the pull without the white space changes you can add ?w=1 to the pull url

[calvinmetcalf]

this also removes the superfluous '#' ins some of the anchor name properties, that was preventing some links from working, if this pull doesn't go anywhere I'm happy to pull that out into it's own pull

[jachym]

I'll look at it, thanks

[thorsten-reitz]

LGTM in general, though I might add a couple of sentences to GML afterwards. I think one of the big advantages is that using XML Schema / GML Application Schemas, we can validate the content, which in my book for a data exchange format is a big plus. That being said, I would mostly recommend GML for inter-system communication, not necessarily as a format that end users use to send files to each other.

[calvinmetcalf]

@thorsten-reitz yeah I don't want to make it sound like gml doesn't serve any purpose, I just want to steer people who are just making simpler datasets available away from it

[jachym]

thanks @calvinmetcalf

@thorsten-reitz the field is yours ;-)