Skip to content

Update to CT/USA imagery - #3114

Open
Mashin6 wants to merge 2 commits into
osmlab:gh-pagesfrom
Mashin6:gh-pages
Open

Mashin6 wants to merge 2 commits into
osmlab:gh-pagesfrom
Mashin6:gh-pages

Conversation

@Mashin6

@Mashin6 Mashin6 commented Nov 13, 2026

Copy link
Copy Markdown
Contributor
  • Added new imagery from 2016
  • fixed wms url for hillshade layer 2023

@cicku

cicku commented Nov 14, 2026

Copy link
Copy Markdown
Collaborator

You should drop 2016. It was already excluded from years ago.

@cicku cicku self-assigned this Nov 14, 2026
@Mashin6

Mashin6 commented Nov 15, 2026

Copy link
Copy Markdown
Contributor Author

2016 CT ECO service is still running and tiles are being served. Why is it problematic to have it included?

@cicku cicku left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just because there is a layer running doesn't mean we need to include it.

Especially we have 2019 and 2023 already included. There is 0 value of the one from 2016.

@Mashin6

Mashin6 commented Nov 16, 2026

Copy link
Copy Markdown
Contributor Author

I still don't understand what harm does this layer bring. There are many such cases for other areas. E.g. Berlin has imagery for every year from 2026 to 2017.

The value is that some structures are better seen though tree canopies in 2016 and some in 2023 imagery. And it helps to determine if OSM object is a remnant from 2016 and now is demolished, or if it's something new that is not captured on the 2023 imagery yet.

@cicku

cicku commented Nov 16, 2026

Copy link
Copy Markdown
Collaborator

I don't usually review European submissions, they have their own rules.

Berlin and some European places may have old imageries because they captured the whole countries through multiple years due to funding issues. For USA, the included layers are mostly statewide or county imageries of the last 3 cycles. Anything older beyond should not be included unless you have a perfect reason (like WV has an imagery composite with certain areas dating back to 10-15 years ago, but it is mostly mountains), because many imagery program receives updates in 4-6 years.

value is that some structures are better seen though tree canopies in 2016 and some in 2023 imagery. And it helps to determine if OSM object is a remnant from 2016 and now is demolished, or if it's something new that is not captured on the 2023 imagery yet.

That's not a reason I will agree with. You can easily use Bing or ESRI to verify if there are old objects on the map, in fact none of them are still with 2016 in CT. Unless there is a substantial amount of areas looking better in 2016 (which is false because I committed the first ECO imagery back to 2021/2022), I won't give you a pass then.

The main consumer of this index is always the ID project - backporting old imageries will open a can of worms, like vandalism. For an affluent state like CT, publishing new imageries every 2-3 years is almost guaranteed.

If you want to map or examine building footprints, the first thing you should consider is the 2D footprints from the state, generated from the LiDAR:

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants