Query about resurvey intervals for road surface, no. of lanes

[hamishmb]

Hi there,

I've been mapping with StreetComplete for a few years now, and have set the resurvey interval to the least frequent option. However, I do find it asking me about road surfaces and numbers of lanes again every now and then - is this something that should be being resurveyed, as these are really unlikely to change with any kind of frequency?

[matkoniecz]

I guess it depends? Maybe resurvey frequency can be lower for surface=asphalt roads? Or altogether skipped for such cases?

Even surface=asphalt can change (for example some roads in city center where I live got changed to surface=sett for decorational/restoration reasons). But maybe surface overlay can be considered as covering this?

I do find it asking me about road surfaces and numbers of lanes again every now and then

is it certainly resurvey? not asking about missing data?

[mnalis]

is it certainly resurvey? not asking about missing data?

Yeah, I don't think the lanes=* are being resurveyed at all in StreetComplete if they are already tagged. Can you provide an example where that happened to you @hamishmb?

Road surfaces are being reasked, but with "Less frequent" preference it happens only every 24 years (for paved roads), or every 12 years (for unpaved roads) -- so it is unlikely that this is what is happening to @hamishmb (StreetComplete itself is only about 9 years old, so even if they've been using it from day one, StreetComplete still shouldn't have asked twice about same road if it still has the surface=* tag).

It is more likely that something else is the problem, e.g.:

• there is conflict between surface and track_type so that is why quest is being re-asked (e.g. Infinite loop of surface and track type quests #6272)
• someone has removed the roads and recreated them without surface (so they are in fact completely new roads without surface details)
• there are multiple roads nearby and it only looks like quest is being asked about the same one
• there are spurious road duplicates in OSM database sharing the exactly same nodes.
• user is misremembering
• there is a bug in a code which only happens very rarely, so they are the first to notice that problem
• something else

But without having an example where this behaviour allegedly happened, we cannot determine what seems to be the problem. Of course, sharing an example might be privacy concern, in which case we can help only as much as suggesting what @hamishmb should look for (e.g. finding their changeset which does the change, looking at a history of the way and its tags before and after StreetComplete change, etc).

[hamishmb]

Okay, makes sense. Unfortunately I can't remember exactly where it happened, and the general area is near where I used to live so I probably don't want to share it publicly on here. Thanks for clearing up how often it's meant to be happening though.

[mnalis]

No worries! If you do stumble upon the issue again, do remember the location and ask us again, so we can provide instructions how you can gather enough non-identifiable information to let us reproduce the problem somewhere else (or, alternatively, determine why it isn't the problem but expected behaviour, so we can document it in SC FAQ)

[hamishmb]

Okay, will do.

[BalooUriza]

I don't think StreetComplete should be asking number of lanes until it also asks access and turns for those lanes, so no recheck necessary. Incomplete lane tagging is the same as wrong lane tagging.

[matkoniecz]

Incomplete lane tagging is the same as wrong lane tagging.

no, it is not

for example tagging lane count without tagging turn lane info is 100% valid and fine