Hi,

The Bing and Esri imagery shows the rivers essentially completely filling the areas.

The Intermittent tag could well be applicable here. In that case I think one would have to be careful to mark the minimal size separately, since there’s always a decently large shape of water around the center of the river.

-Tómas

Hi gscholz,

The area of the river changes frequently - this shape is very roughly a "normal" size of the river, somewhere between its peak in late spring and its minimum in the autumn / winter. It could perhaps be tagged as intermittent, but I think retagging the area as entirely shingle could be a bit drastic.

Best regards,
Tómas Ingi

Hi,

This area is indeed relatively barren; however, it is (sparsely) covered with young birch trees, as can be seen here: https://ja.is/kort/?x=590112&y=386075&nz=12.40&ja360=1&jh=15.7

I originally tagged this as natural=scrub as it felt the most appropriate for a sparse, young forest; maybe there's a better one?

Best regards,
Tómas Ingi

Hi,

Most Icelandic sources (and people, for that matter) call Akureyri a town ("bær"); this seems like a change that should be discussed with the Icelandic OSM community https://lists.openstreetmap.org/listinfo/talk-is if there's a compelling reason for the change.

Best regards,
Tómas Ingi

Hi,

I noticed that all the roads are now tagged as having shoulders for bicycles in this and other changesets. I don't believe that's the case for any of them; do you know why this happened?

Hi,

Thank you for sharing the criteria you use. I agree with you for the most part, and think they're very reasonable and well-presented.

I do still have some thoughts on this exact case:

I agree that I wouldn't map a pedestrian street with kerb-less sidewalks as more than one separate way. On the other hand, I don't necessarily see anything wrong with doing so, and if someone else had mapped out the sidewalks separately, I would hesitate to remove them. Same goes for segregated pedestrian and bicycle paths, even if there's no physical barrier between them.

Looking around on OSM there are several instances where this is the case. Some examples around Stockholm include e.g. Barkarbyvägen and nearby streets, Birger Jarlsgatan, and Sankt Eriksgatan north of Kungsholmen. In Iceland, the standard has been to always map separate ways, even when there is only a painted line (assuming that the bicycle part of path has two distinct lanes).

Taking Sankt Eriksgatan as an example, there are some sections where the paths are separated by a kerb, but large sections only have different surfaces. That being said, it could be argued that the nature of the path doesn't change - kerb or no kerb, there is still one distinct bicycle path and one distinct pedestrian path. Whether or not it "is" one path or two is more of a subjective question in this case, and I don't think either way of mapping is necessarily more or less correct than the other. In such a case, I would perhaps not map them out separately myself, but I would always leave separately mapped paths in place if they show more detail than can be captured by a single way. These details could include marked crossings (as you mentioned with oneways, these might not be stricly "real" in a legal sense, but are followed by many bicyclists and pedestrians), slightly different geometries, sparse obstacles between the paths, etc. I can see the appeal of reducing the complexity of the mapped paths and agree that there is a balance to be struck between mapping in full detail on one hand and having a reasonable and maintainable database and end product on the other hand; in this case, I would see the separate mapping as providing useful detail, rather than being just clutter.

Seeing as the surrounding streets around Tulegatan all have the paths mapped separately, I would suggest keeping them so for now. If there is a consensus in the future to map all of these sorts of paths consistently as a single way, I would welcome that change.

Does this sound reasonable?

Hi,

I see your point that the oneway tag does not cause confusion in these cases, and I agree with you and the community.

I would be interested in seeing what community discussions exist around mapping paths as one or more ways. Is there a general consensus on where the exact boundaries between mapping one and two ways lie, and is it generally accepted that merging separately mapped ways improves the quality of the database?

The point "easily traversible" especially stuck with me - a kerb is easily traversible on foot (even on wheels, if it's low enough), and so is a meter of grass; but I imagine such a large separation would almost always be mapped as two paths. I feel there must be some other criterion that is used in practice. What are your criteria when judging this, and what criteria has the community found works well?

Best regards,
Tómas Ingi

Hi,

I would argue in the case of the paths along Tulegatan / Råsundavägen that they would be better off marked as separate paths - besides the different surfaces, they are clearly separated by a line of cobblestone. As they are tagged now, there is a significant amount of information that has been lost: The surface of the footway is not tagged, the entire path is tagged as one-way (even for pedestrians!), and there is a stretch where the path starts from Råsundavägen tagged "foot=designated", even though pedestrians are not expected to walk from the road onto the path there.

What are your thoughts on this?

Best regards,
Tómas Ingi

Hi,

Currently, the standard in Iceland is to follow the Icelandic Road and Coastal Administration's classification of national highways in the countryside for the highway type. A change in this schema should probably first go through the Icelandic talk page, and be done on a nationwide level.

I do agree that this way should not be shown as a driving alternative to highway 60/61. If there is a way to add the specific information that the road is not maintained during winter (as is common with high-altitude secondary highways in Iceland), that would be very useful.

Best regards,
Tómas Ingi

Hi, and welcome to OpenStreetMap!

When you add the Chinese translations, make sure you add them under the "name:zh-Hans" tag for simplified characters or "name:zh-Hant" for traditional characters ("name:zh" can also be used with a slash between simplified / traditional). Changing the "name" tag loses the original name of the objects, so right now there is no Swedish name on these roads!

Hi AVWB, and welcome to OpenStreetMap!

It seems you've copied some features from the town of Vogar here and added them to a new location - was this intentional? If so, please note that the OpenStreetMap database is only meant for current real-world information.

If you want to test out features in a sandbox-like environment, it's possible over at https://api06.dev.openstreetmap.org/.

I've deleted these features here for now, so you don't have to worry about anything :)

Best regards,
Tómas Ingi

Hæ,

Hvar má finna upplýsingar um nafnið á húsinu? Er þetta eiginlegt heiti, eða er það kallað Costa Del Sol af íbúum þess?

Hi!

I have some thoughts on this retagging, mostly in terms of deciding what is a part of the station and what not. As an example, why is this (node/10379253912) node a part of Keleti, but not this (way/228385724) area? This seems to be a rather arbitrary distinction; an artefact of how the different landuse=railway areas happened to be split.

I can see how mapping railway stations as areas could be useful in some circumstances. That being said, how would you justify the choice in where to map the boundaries of the station, both in this context and in general? Would you agree that the borders extend too far in this case, or is there a sound reason for them?

That's a good point; I suppose local is better for the time being then.

Hi,

I think this route could even be considered a regional cycling network, since it connects nearby towns (similar to the routes in the capital area). What do you think?

I see!

I haven't seen any other name for this waterfall, so it may well be correct.

I would be interested if you know where the name comes from; I ask mainly because it does not appear to be of an Icelandic origin.

Hi,

What exactly do you mean by local name?

Hi,

That's the hope! We'll see if it works out :)

Thankfully there have been others who have contributed significantly to the landcover (see Ben97 for example), and it's always motivating to see the progress when the map updates!

Alright, no worries!

Hi,

What source did you use for this change? I rode this path today and it's definitely not paved.

Best regards,
Tómas