[Talk-GB] New Bing Imagery
Andy Townsend
ajt1047 at gmail.com
Wed Aug 19 10:17:47 UTC 2020
On 19/08/2020 10:11, Stephen Colebourne wrote:
> And now I can see Amazon mappers using an iD variant
> that doesn't have the offset and moving all the roads as a result:
> https://osmcha.org/changesets/89549551?aoi=758c7f2b-faca-44e5-acd2-0cb8c33034bd
> https://www.openstreetmap.org/changeset/89549551
If that's happening at all, please comment on the changeset explaining
the problem. In English urban areas OS OpenData StreetView is a pretty
good guide for alignment and if people (especially people doing a lot of
editing) are not taking into account different imagery offsets then
that's just wrong.
If, after being told about the problem, people are ignoring the advice
and just banging in roads with incorrect alignments then please email
the DWG at data at osmfoundation.org about it. However if people haven't
ever been told what the best practice for a country (especially if they
edit in lots of countries) is it's not really fair to criticise.
One other thing that I have noticed is that the imagery behind the "AI"
detection that they have been using in some places in the UK is among
the worst available, and I've suggested where I've seen stuff added on
that alignment that it's wrong and needs fixing.
> This is going to keep happening so long as OSM has multiple image
> sources and multiple editors. Frankly I'm amazed that this isn't a
> solved problem.
I don't think that it ever will be a solved problem, because the level
of accuracy that we're worrying about in OSM has changed over the
years. About 12 years people were adding features from out of copyright
maps such as NPE that sometimes might have been 200m out, or adding
stuff with a bit of guesswork from Yahoo imagery that might be similarly
inaccurate. Now we're worried about 2m here or there, in 10 years time
it'll be blades of grass(!).
>
> Having done some mapping across the country recently, it seems like
> Bing is offset to the previous best imagery across the country, but by
> varying amounts. Is there really no solution that can be applied to
> the source Bing layer?
If there is, it'll be Bing that does it rather than us. When Bing was
"new" to OSM in 2012 or so there were lots of odd offset issues in Bing
imagery and these were sorted out over time. I suspect that that will
happen again.
To be honest, I'd never trust any imagery anywhere until I've seen how
it compares with other available sources in that area (GPS traces, OS
OpenData StreetView, etc.).
Best Regards,
Andy
More information about the Talk-GB
mailing list