Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

schema modifications for OHM #3

Open
2 of 4 tasks
andrewharvey opened this issue Jul 10, 2021 · 3 comments
Open
2 of 4 tasks

schema modifications for OHM #3

andrewharvey opened this issue Jul 10, 2021 · 3 comments
Assignees

Comments

@andrewharvey
Copy link
Collaborator

andrewharvey commented Jul 10, 2021

  • remove historicphoto and historicmap from source types, as photo and map cover this already, and historic is kind of implied being in this index.
  • remove permission_osm as for now only open licensed imagery or public domain which is deemed as acceptable for use in OHM should be included
  • consider if country_code is needed/used anywhere, given historical sources the current country might not be relevant and the historical country may not exist anymore.
  • consider if best is still relevant given multiple images of the same location but at different times could co-exist and all be the best for that place and time.
@jeffreyameyer
Copy link
Member

My apologies... 👀... looks like I've reverted a couple of your efforts there in order to get my changes to pass. I'll go back and fix correctly.

@jeffreyameyer jeffreyameyer self-assigned this Oct 7, 2021
@jeffreyameyer
Copy link
Member

@andrewharvey - what do you think about leaving 'historicphoto' and 'historicmap' as source types, in order to ease transition and transfer of layers from OSM and OHM?

For 'permission_osm' - I agree that we should probably have a type for 'permission_ohm' but would it make sense to leave 'permission_osm' for the reason above plus, it would also be true for that layer?

'country_code' - not sure what that's used for - we could still organize the layers by country codes or whatever folder/file structure we wanted to, no?

'best' - I do think keeping 'best' makes sense as it highlights the best current ground truth layer, which is super helpful.

@andrewharvey
Copy link
Collaborator Author

@jeffreyameyer

what do you think about leaving 'historicphoto' and 'historicmap' as source types, in order to ease transition and transfer of layers from OSM and OHM?

I can see your point, in the OSM ELI I'm not even sure what the threshold for historic is.

If we have October 2021 imagery and September 2021 imagery does that make the September one historic?

What if the latest imagery we have for an area is 2011, is that historic because it's 10 years old or not because it's the latest we have.

I guess on the OSM side this flag might be useful as a way of filtering out imagery which is not useful for day to day mapping, but might be useful for things like start_date which can go into OSM.

So I'm happy to restore it, I just wish we could document what it's purpose is and when it should and shouldn't be applied.

For 'permission_osm' - I agree that we should probably have a type for 'permission_ohm' but would it make sense to leave 'permission_osm' for the reason above plus, it would also be true for that layer?

I think permission_ohm makes sense for layers that we have explicit permission to use in OHM that without such permission we wouldn't be able to use. Because OHM has a much more flexible acceptable license criteria there is less need for special permissions compared to OSM, but sources like Bing might need permission_ohm.

For OSM compatibility, any layers from OHM which have permission_osm=true we probably need to filter out as that permission won't extend to OHM unless we have a separate permission, in which case we can have a copy of the layer in our index here.

'country_code' - not sure what that's used for - we could still organize the layers by country codes or whatever folder/file structure we wanted to, no?

Correct, I just hadn't yet investigated if the code here, or iD or JOSM would break if we left this field out.

'best' - I do think keeping 'best' makes sense as it highlights the best current ground truth layer, which is super helpful.

Ok so we define best as, best for the present, and don't go down the rabbit hole of trying to define best imagery at specific locations at specific points of time.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants