That’s weird. Usually people just change what they don’t agree with, but not revert the whole changeset. Did you ever contact the user for discussions?

Wow!

One question. Does that OpenDroneMap fully utilize multi core cpus?

my suggestion would be to put the corrected imagery into the offset imagery database in JOSM. That’s pretty much the only way people can know that the imagery has been aligned to the best accuracy available at this moment.

And good luck with China’s claim all over the South China Sea. #$@!

Camera make & model? Reliability issues? Tips? Good work!.

U guys are nuts… :)

Warin, I believe that the latest version of the firmware gets that 30cm in much less time than that, although high accuracy is not reliably available every single time. I prefer processed stationary gps points (from nmea logger) on a visible landmark instead of gps tracks since that seems to be a better measure of accuracy. Looking forward to what this new development in gps positioning can do to make OSM/mapping better.

Piggybacking on this… so how do you include roundabouts in relations? Do you break them up into sections or do you just bypass them and just use the next road segment?

welcome to editing!

You’ll find out that those offset will become part of the game soon enough. What probably happened was that the buildings were traced on older imagery compared to what you are currently seeing. And there is nothing to prevent a newer imagery in the future to have different offsets of the building that you traced now.

Even if you are able to obtain precise gps coordinates for a point, and then re-aligned the imagery, you’ll still find that things don’t really line up somewhere else. We have to live with that imperfection, close is good enough!

In your situation, I would do one of these: If there are already a lot of buildings traced, I’ll locally realign the imagery to fit those buildings and start tracing other things that needed tracing. This way, everything will be offset by a constant value. Perhaps in the future when I really feel confident that I have a good reference coordinate, I’ll try get an agreement from others to shift the whole area (and imagery) to the more accurate coordinate. (don’t know how practically relevant when applying offsets to city sized area, I traced villages mostly) If only a handful of buildings had been traced, might just ignore them. In JOSM you can filter out buildings to declutter the map.

In most cases, I just said fuck it. :) hehe

I finally limit my edits to POI and street naming esp. in Vespucci. Even when using a tab, there’s no satisfactory way to draw roads, geometry etc.

Can we just bypass this “tagging freedom” thing and just create an approved list?

Aye, same thing happened here in Malaysia. We (our less than a dozen mapper heh) kinda suspended using the trunk tag for the moment.

Some of us wanted to us the technical side of the definition, which eliminates a lot of current trunk tagging and reclassified them to primary.

I argued for using trunk for major intercity connecting roads (most are of the primary type in classification) that were built during the British times before we started building modern tolled motorways. These are usually given a ref number of value less than 10 since they are the earliest built intercity roads.

No consensus, therefore we just freeze the tag usage.

A very welcomed news.

Although I dread the alignment and realignment of imagery/data layer that is bound to happen. There is no conveniently accessible way to tell all mappers that the imagery of an area has been aligned correctly by such and such on this date through surveyed known point, (or gps tracks or some other ways).

JOSM map offset database plugin isn’t widely used in my area (I’ve not come across any), and no idea if ID accesses the same resource.

How do you guys in well developed maps cope with this issue?

…Trusting a wonky smartphone satnav and compass when using Mapillary. Openstreetcam handles the wonky satnav part really well, but it’s even worse with the compass since you can’t currently correct it manually after uploading…

Unless your are talking about Openstreetcam only, compass heading can be corrected in Mapillary, but you have to use the web interface which I found to be slow. About time for an app/plugin that can edit those details before uploading to the Mapillary server. Regards.

thanks for the tip… will try that if it happens again.

would you kindly write a JOSM plugin for that “jumping nodes along a way”? That’s a feature I really like about Potlatch, it makes easy work of aligning a 20km stretch of rural highway to newly available sat image.

Wow… salute!

Terrorist symphatizer, you! ;)

Impressive! as darth vader would say..

How fast does the map load in osmand on lower spec android? Still usable?

hahaha… tapi kalau dia ada rss feed, jenuh la hang menunggu..