Maxar imagery not working (was "Maxar is blurred in ID" and other similar topics)

More seriously, the specific Potlatch issue is mentioned here.

Getting this imagery error in the JOSM for the past few days. I tried everything people suggested to do but the imagery still shows error. Can anyone tell me how to fix this?


See #23025 (maxar tiles cannot be downloaded - invalid connect ID) – JOSM

(currently nothing can be done as an user, JOSM developers are waiting for updated key for Maxar imagery)


JOSM is returning errors instead of tiles for the “MAXAR Premium Imagery” tile layer today.

Are other experiencing this? If not, what is likely to have gone wrong with my internet connection?

Error message:

Error downloading files: Login failed - an invalid connect ID was specified.Credentials[username=null, connectId=((something that looks like a unique ID))]

The api keys have been invalidated and new ones haven’t been distributed yet.


yes but according to the HOT Discord server…


See Maxar imagery not working (was "Maxar is blurred in ID" and other similar topics) - #7 by SimonPoole

As an editor who likes to work on remote areas, this is quite a massive hit.
Maxar was my go-to most of the time.

I’m not quite active here and mostly read the weeklyOSM news, so not sure if I missed something, but is there some effort from the Foundation or community to obtain reliable imagery for those remote areas?
I’m sure that this is not cheap, but maybe we can get some kind of fundraiser if there is a deal?


This is going to cause plenty of problems in Zambia, it is the go to imagery there as it is the most up to date by a long way. I fear new contributors to HOT projects will start to delete items mapped previously using Maxar that do not appear in other imagery. This is particularly so around large towns and cities where development is fairly rapid.

1 Like

Maxar imagery is a great resource that many mappers appreciate. It is useful in many places; crucial in some.

We’ve been in regular touch with them about API key rotation and other security mitigation efforts. We understand their concerns that a small number of people are using the OpenStreetMap API key to steal imagery for use for non-OpenStreetMap usage.

We have not heard much from Maxar on the current situation, but are trying to connect and resolve the issue.


I have seen pricing of about 10 $ / square kilometer. And license of imagery data would be anyway incompatible with OSM and would be impossible to use it for mapping.

No idea how much they would charge for commercial access on special rules allowing use for OSM mapping. And data storage rules required by commercial entities paying for it are quite strong, to the point that each mapper would likely need to pay for own access.

So about 10 $ / square kilometer for each mapper?


Alright…so no fundraiser…
Is there someone we could ask nicely?
Maybe Bing for some updates? Do they even have updated data on the unprofitable parts of the world?

1 Like

There are attempts, see:


I wouldn’t mind having to do an Oauth login and copy paste a unique API key into JOSM, but that seems like it would just lead to an increase in the number of bogus OpenStreetMap accounts.

I’m assuming any solution based on per-user keys would be a bit more user friendly. While not a panacea it would make misuse of keys at least trackable.

Just as a hint to avoid misuse:
In Osmand you get free map downloads when having contributed to OSM considerably in the last weeks.

I wouldn’t mind if I would have to undergo some procedures to have access to Maxar imagery.
In some regions these were definitely the most actual images.

That might work if we robustly checked every edit by new users, but we don’t. For a small niche thing like Osmand you can get away with it but I think for something like satellite imagery we might find the Canadian wilderness getting filled with bogus but plausible trees to ge the edit number up. That’s if the scrapers care enough about the map to at least try to avoid harmful edits, which I doubt they would unless they’re overlaying OSM.

Edit: the requirement to be a pre-existing editor is also unfriendly to new mappers, which might be harmful long term.

With OsmAnd’s “bitcoin” scheme, OSM saw a few people trying to game the system by trying to move nodes back and forth. The mapping wasn’t really “incorrect”, just “unnecessary”.

The idea of tying imagery access to contributions reminds me of a Wikimedia Foundation program called The Wikipedia Library. It offers Wikimedians institutional access to premium subscription databases such as JSTOR,, and Gale Cengage. Access to most of the resources is based on the number of contributions, age of the account, and recent activity; some resources like have a separate manually reviewed application process. Participants agree to cite their sources when using TWL. On a technical level, TWL endorses individual logins or provides OAuth-style integration using the databases’ existing authentication schemes.

To my knowledge, this program hasn’t resulted in any significant nonsense editing on Wikipedia, but if that ever happens, blocking the user from Wikipedia would also block the user from TWL. The main downside of this approach is that it depends on volunteers to (minimally) vet applicants to one of the sources that don’t provide automatic access.

Even so, I agree that Maxar is currently such a basic resource that putting it behind a bureaucratic process could impact the project’s growth in some regions. This is a wake-up call for some regions to work to further diversify the available imagery resources in ELI, to the extent that this is possible.


One of the quirks in this mini-drama is that it is quite unclear if Maxar is still being paid to provide the imagery or not. As some may remember this all goes back to a backroom deal in the aftermath of Facebooks chaotic attempts at adding ML generated data to OSM. See New aerial imagery from Digital Globe? - OSM Help