[openstreetmap/openstreetmap-website] Report rate limit settings in capabilities API call (Issue #4380)

Martin Raifer notifications at github.com
Wed Dec 6 22:48:33 UTC 2023


> I don't think this will buy you much, since you don't have all the information available on client side to do the same calculation that will be done on the server.

Well, yeah, but one can relatively easily re-implement an approximation which gives accurate results for the vast majority of cases, i.e. a user with no active reports and blocks. That's more than good enough to solve the problem outlined above.

> At one point we included this information in the API error message, 

The main point I'm trying to point out here is that an API error message is too late to handle this situation gracefully. Not only might an affected enthusiastic user potentially loose hours of work, but in worst case they might end up discouraged from further mapping and leaving the project.

A client software can also much better decide how to gracefully represent this information: As beginner mappers will most likely be unfamiliar with how "changes" are counted by the OSM API, the information in the API error message would not have helped them much. The editor, however, can count these numbers in the background and, in case when a mapper is about to reach the limit, show an information box with information about the rate limit. This could include basic information of why the limit is in place, and perhaps even give some mapping tips for things to do in the meantime.


-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/4380#issuecomment-1843812878
You are receiving this because you are subscribed to this thread.

Message ID: <openstreetmap/openstreetmap-website/issues/4380/1843812878 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/rails-dev/attachments/20231206/219307f3/attachment.htm>


More information about the rails-dev mailing list