Patrickov's Comments
| Changeset | When | Comment |
|---|---|---|
| 122875457 | over 3 years ago | Let me rephrase. I removed that platform because it caused certain problems (e.g. people adding relations to them arbitrarily) *and* had become an eyesore after new buildings are built over it. That doesn't mean other platforms of similar kind necessarily need to go, and other multipolygons whose creators had bothered to create are, of course, very good. I simply don't have that leisure, and besides areas still need pathing, I believe? I did take extra care to actually survey. Admittedly it's because I found out that you were the creator, although you would probably still complain even if not. I never satisfy you. |
| 122875457 | over 3 years ago | (Re: 'highway=platform' is not the same as for 'highway=bus_stop') That's exactly why drawing a single platform in the middle caused problems. As I said before, when the platform was around, other mappers (or is it just you?) put relations to *the* platform, making it look like anyone can board any bus anywhere, when it's simply not true. I also noticed that some mapper(s) used "platform edge" to denote specific places for designated routes. IMHO that's more applicable to railway platforms, but for buses (where people board at a single door on the bus) it's not accurate either. Other bus terminals (e.g. Nina Tower next door) simply put boarding spots as stops, and bus companies do put up poles at designated spots, so it's an accurate depiction. Even "open space" terminals like Tsing Yi AR Station specifically spot the boarding places instead of drawing a horrific platform area and using platform edges. I fail to see why I cannot do the same here. |
| 122875457 | over 3 years ago | I did NOT map for the renderer. As said, I *actually* did a survey myself and all pathing and stop positions (stop posts) were put down as they are *at the ground*. Conversely treating the entire area as a single platform is not accurate at all. You mentioned multipolygon but let me tell you, mapping all those columns would need many, many more "inner" shapes. Paths both represent the ground feature more accurately and cause much less intrusion to the surroundings. |
| 122875457 | over 3 years ago | Sorry, I retract the word "hard to read". It only hurt my eyes, but I do think it's inaccurate and caused quite some problems. |
| 122875457 | over 3 years ago | The platform is bad or wrong for at least three reasons. 1. It does not address all the columns and structures in between. In effect that's not a single platform, but two corridors with a third in between for much (but not most) of their lengths.
I have been wanting to get rid of the platform for quite some time, but after I found out it's you who made it, I KNEW you would complain if I simply do it without some backing, so I actually went to the place and did a survey myself to ensure my initiative is justified. About the source, I only looked up HKBUS for the KMB reference numbers; every other bit was from my own outing and records, so I admit that I should just have stuck to "self-survey". |
| 122567153 | over 3 years ago | Re 1.
Re 2 - 3.
That said, I have read osm.wiki/Hong_Kong/Transport/Road and it seems that "unclassified" was reserved for bad roads, which I don't know why and don't agree (Probably you don't too). But in any case, maybe let's just keep all of Wing Shun Street and Wing Tak Street as tertiary as a compromise. Re 4.
|
| 122567153 | over 3 years ago | Both "covered=yes" and "tunnel=yes" are rendered the same. It's easy to distinguish the two though. As example, Wing Shun Street under Tsing Tsuen Road is a tunnel; while Kwai Chung Road under Tsing Kwai Highway is covered. IMHO the Tsuen Wan Road over Wing Tak Street seems too open to make the road under it qualify as covered. Also, previously the covered path extended well beyond the actually covered part. |
| 122567153 | over 3 years ago | "Wing Tak St is the main access from Tsuen Tsing Interchange." >> No it is not. It has a restriction of length which means buses and HGVs are not legally allowed to use it. |
| 122567153 | over 3 years ago | "Actually you should not use standard as the criteria for classification. Function in the road hierarchy is used."
|
| 122567153 | over 3 years ago | JOSM is not always the best tool to use. Using browser is often more convenient as they displays points and paths better IMHO. |
| 122254313 | over 3 years ago | In any case using housename as the estate part sounds very wrong to me. You can proceed to change the tags if you like, but I refuse to comply myself and can only accept not to add anything under addr tags. Sorry. |
| 122254313 | over 3 years ago | addr tags certainly needs to be improved then. Actually most of the time we simply cannot fill in anything in the default UI. |
| 122254313 | over 3 years ago | In Hong Kong most (public) housing estates have the word "邨", which indeed is an alternative form of "村", meaning "village". I cannot find any documentation which support your rationale. Can you give a link to support your case? |
| 122174422 | over 3 years ago | Replaced with something else. The signs there asked cyclists to dismount so re-pathed to reflect the regulations there. The resulting short stretches is the ridiculousness on the part of the related authorities but if the structures are like that so be it. |
| 122254313 | over 3 years ago | Awww you again. I'll take a look at the description documents and see how they are done. For the "addr:place", again, I changed tag because it seemed more appropriate than the previous one ("addr:housename"), which itself means individual houses. There are buildings in, say, Mong Kok which would fit "addr:housename", but I don't think buildings of a housing estate, some of them separated by named streets, qualify. |
| 121609483 | over 3 years ago | As a matter of fact, the foliage does not span across the road. |
| 121609483 | over 3 years ago | There are too many things to fix and it is take me a day or two. I do aim to add tree areas back where they should afterwards, but I can't stand seeing those polygons going across roads and paths for no good reason. Also, most of them are not natural wood areas -- they are artificially placed alongside paths or roads in a very orderly manner. I do think tree rows are more accurate description of them than , at least for the roadside areas. My upcoming change will conditionally preserve wood areas, but my interpretation of the area is that they are not actual woodlands and should be changed to something else wherever applicable. |
| 121550877 | over 3 years ago | Sorry I did not notice I did that kind of change. This change was mostly about setting up the bus route. Can you please point specifically to where your concern is about? |
| 121609483 | over 3 years ago | Frankly the reversal action by you is very rude. |
| 121609483 | over 3 years ago | It's the Openstreetmap team who are wrong by refusing to render shrubberies. Please join me and many other contributors at the discussion at https://github.com/gravitystorm/openstreetmap-carto/issues/4473 |