Please stop mapping like this, what you are contributing is not as per the HOT task or the actual content on the ground. Please ask your instructor or mapping party organiser to stop the activity and discuss with the HOT-OSM mailing list.

Don't map like this; it's wrong and will need to be reverted. Can you tell your instructor or mapping party host that this is inappropriate and to stop the activity.

Don't map like this; the task doesn't require forests

Why are you mapping like this, it has no relationship to the imagery or HOT task

Why are you mapping forest? This little or nothing to do with the imagery or task

Ah the building for the Reepham already seems tagged as a pub etc

way/317410025/history causes a lot of errors in keepright - is this better modelled as highway=proposed or not even as a highway at all (just part of a relation)?

http://keepright.ipax.at/report_map.php?zoom=13&lat=-25.58087&lon=-48.5838&layers=B0T&ch=0%2C191&show_ign=0&show_tmpign=0

This causes a lot of errors in keepright, and the main highway should probably be highway=proposed, not highway=planned. Specifically way/296110059

Ah neat. If you are keen, there's a few others via osm.wiki/Import/South_Australian_Roads that we know of - there's a generated .osm file with the details

Ah nice!

What's "Household Code" ?

Ah hah! Thanks

What's the error osmose is coming up with? The wiki suggests that combination of tags (direction=clockwise, highway=mini_roundabout), and in Australia that's consistent with the traffic flow...

Feel free to revert/model it better if you have on the ground observations that are more accurate than the GPS traces and satellite imagery; however *even the original way had it documented that it had been filled in*.

Routers really shouldn't direct riders through derelict canals without some kind of other structure (bridge, etc), as it usually means they'll fall into a great big empty dry former waterway that looks something like:

https://www.google.com/search?q=derelict+canal&source=lnms&tbm=isch&sa=X&ved=0ahUKEwj115urx67KAhUBHaYKHTVrBDUQ_AUIBygB

As it stands, I really doubt it should be modelled at all - the strava error reports indicate that cyclists aren't aware of a former canal at all; suggesting its not even visible to them.

http://labs.strava.com/routing-errors/#10000000/15/-1.29036/52.95923

The path was intersecting the waterway; but there was clearly not a bridge etc.

I split the waterway either side of the bike path to better reflect was is on the ground/apparent from GPS traces, and satellite imagery.

Given that the canal itself is filled in; I don't think it should be mapped - at least not if its indistinguishable from the surrounding terrain.

However, someone clearly when to the effort of mapping it, so splitting it seemed to be the least harmful approach.

Do the various offramps intersect with the Cross city tunnel? It's also at layer -2 and showing up in missing-junctions checks on keepright

Ditto http://keepright.ipax.at/report_map.php?schema=50&error=71070414

There's a couple of routing errors with this one; like http://keepright.ipax.at/report_map.php?schema=50&error=71074868 - could you take a bit of a look?

I'm removing these.

This intersects a lot of buildings; also sourced from digitalglobe - which is more accurate? The existing building data, or road?