That’s just super-cool thing. Here is the comparison of tags from old and new leaf property schemes, clearly demonstrating that argument about re-tagging being a bad idea is wrong:

Question is, what exactly you call “the Greenwich meridian” - several things sharing this name. EPSG:8901 is a zero longitude line and it is called “Greenwich meridian”. But historical Greenwich meridian mark is not a zero anymore, as explained above.

@Warin61,

While I do agree, that in the general case it could be hard to map something on the other side of the Earth, in certain cases is could be quite simple (however, it is still impossible to reduce uncertainty to zero). In Texas, they have quite detailed aerial imagery. Using it and looking for a fence, dividing private land from the street, it is possible to split Oak Circle into two pieces: Oak Circle itself and private driveway, going deeper into a property, mentioned above. There are many situations where we can’t really rely on imagery, but this doesn’t seem to be a case.

I’m aware of certain “cult of the survey” and negative attitude towards “chair mapping”, but every type of mapping is equally useful and effective if used within its limitations.

No derogatory terms, no rude wording, just slight irony. You really call it “biting” (keeping in mind it’s anonymous note)?

And please, keep in mind, Monowi already exists here, so this note is definitely misplaced.

Since OSM is a database, while maps are only the derivatives of it, “labeling” is not completely correct term. Labeling presumes, that certain properties are assigned to an object to get a certain visual representation of it. However, in OSM, tags are the parts of the semantic description of any object. And then, map renderer can use that description in conjunction with representation rules (style) to label objects on a map.

Splitting streets into pieces is completely normal practice. Some streets are split into tens of pieces to properly reflect lanes, turn restrictions, mass transit routes, different surface, speed limits, lighting, overhead trolleybus wires and other features. So, don’t hesitate to split it when necessary.

@d1g,

Идите-ка вы с wikidata на форум в соответствующую тему. Это тут вообще никаким боком.

@d1g,

Вас откуда-то выпустили, что вы как с цепи сорвались? Этот топик - про конкретный тег, который можно использовать правильно, а можно - неправильно. Не более того. А не про систему обозначения всего, что растёт (и не только).

То, что вы тут понаписали в свойственной вам манере - вопрос иного уровня, который можно было изложить без словесного фонтана, одной фразой: “Я за введение и использование системы обозначений landcover.”

И, замечу, в данном конкретном случае значение landcover практически эквивалентно значению, которое вкладывается в https://wiki.openstreetmap.org/wiki/Tag:natural=heath, потому что heath - это подмножество класса Shrubland (51) по NLCD 92 или A24 по LCCS. А классу 322 Corine Land Cover он вообще тождественен.

@pnorman,

Thank you. What do you think, does it make sense to create a Wiki page with explanation like this to be able to point on it every time someone starting an argument related to this topic?

@PlaneMad,

First of all, how did you turn “quoting some documentation” into “poking”?

Mapper (including a newbie) always bears the responsibility. It is possible to enforce certain rules and to implement some pre-upload validation (just like it’s working in JOSM). But it is impossible to completely replace mapper’s responsibility (at least until we can’t replace mapper himself with an artificial intellect completely).

Good example of it is usage of wrong tags to map certain objects - syntax is perfectly correct in this case, but wrong tag is used instead of proper one (I mean, https://wiki.openstreetmap.org/wiki/Tag:amenity=waste_disposal instead of https://wiki.openstreetmap.org/wiki/Tag:amenity=recycling or otherwise, for example). You can not fix this problem with validation.

And speaking of (mapping) parties - since mapping is among those “highly structured occupations”, drawing attention of people with autistic spectrum disorders, it’s correct to assume, that there is certain amount of people among mappers, who, due to their psychological condition, do not really like when totally unknown person tries to sound like he’s a friend or to invite them to participate in any group activity. In this case, neutral logical explanation with documentation references work way better than “friendly atmosphere”. As I said before, socially mature responsible person usually does understand neutral comments, so it should work for almost everybody.

So, your argument doesn’t sound strong at all.

@Warin61,

OSM principles are not “view points” or “opinions” by definition. There are, indeed, certain cases with some space for interpretation. Mistakes are often understandable, but OSM does have rules. Like, value of contact:phone key must be in international format (including country code); opening_hours key has it’s own format and must not contain stuff like “from five to three”; streets on different levels (going by and under the bridge) must not have common node. If someone needs to be addressed with “Dear Sir, …” instead of receiving brief quotes from documentation not to feel offended or addressed in rude manner, that’s just immature and oversensitive. (To learn, what’s “rude”, go read some comments to pull requests from Linux kernel developers.)

@Tomas Straupis,

I’d say, that after ten years things are changing - quantity transforms into quality (lack of quality, actually). If there is, say, one person who did some nonsense edit per square kilometer per week, it’s 0.001% of data in the same area, so it’s okay to allow this person to learn by fixing his edit by himself. And it was the case for a long period. But if there are twenty people doing the same per square kilometer per week, it’s already 0.02% and so on. Nothing new, except the scale of situation. While number of active, experienced and responsible mappers doesn’t grow equally fast.

Since there are more and more users, who have no understanding of (often - unwritten) principles of OSM, certain paradigm shift becomes more and more justified - correction without any waiting period, but with more or less detailed explanation of what’s wrong with this edit.

Waiting certain period serves for teaching newbies by allowing them to fix their mistakes by themselves. This approach is based on assumption, that every contributor wants to be involved in further editing and wants to develop himself into a better contributor, which isn’t necessarily a case - some people just want to try editing, or to “add a couple of things”. But data quality is still a primary objective of OSM, so, there is no much sense in assuming it automatically anymore, because leaving the whole bunch of wrong objects makes map data worse, while actual users of data (third-party companies, rendering style maintainers, etc) don’t really care about wrong data, kept for the purpose of education.

Based on my experience of answering tagging-related questions, I’d say, that if you thing that something in OSM is obvious/intuitive without reading tag documentation, you most likely have wrong idea about it. OSM uses natural language for keys and values, and especially if you are native English speaker, you can feel like you got everything from meaning of this word. But concepts of OSM often have a bit different or stricter meaning than words you know. So, it’s better to check documentation first.

Good examples are masts/towers and bridges/tunnels.

There is certain important thing about OSM being a database first, then - a map. Without explaining this dualism, it’s hard to give anyone systematic understanding of many things, including the most common newbie question: “why this particular thing is not displayed on the map?” It also helps to understand why map tiles and data are licensed differently, as it was mentioned above.

So, it’s okay to tell people that OSM is a map, but then you should either explain that “map” is something larger than just Standard style on osm.org or give people some information about database, which is the source of all OSM-based maps.

@Vincent de Phily,

Even better contribution to existing OSM tracks storage can actually be helpful, without any new service. As I said, if data collection is under the full control of an app (or its part) intended exactly for OSM, quality can be improved dramatically.

@Zverik,

As I said several times before: you can not read or you reading headlines only. It was in this diary entry from the beginning:

This screenshot was taken by OSM member luiswoo using Bing satellite imagery and Yandex GPS point cloud, which can’t be used for OSM mapping.

You know, how I’m always concerned about proper license, but I’m not going to target completely unaccountable people, since it’s impossible to prevent their unwanted activity with warnings.

@escada,

Yandex tracks layer can only be used in their own collaborative mapping project, they don’t allow any third parties to use it. However, it always shows only tracks not older than a week. So I wanted to bring it as an example of amount of data that can be collected by popular navigation/mapping mobile apps.

If you don’t want to master your Zen meditation skill, it’s better to use supervised semi-automated tracing/classification methods, as it was previously suggested.

Oh, light beams and angels… I see. OpenStreetMap is very, very diverse…