Talk:Highway Tag Africa
What is the point?
No objections to your page, but what is your point? It is describing what we already have set. People mapping "wrong" by classifying primary roads as tracks because they are unpaved? This is what we call mapping for the renderer (map a feature as something else than it is in order to get the rendering you want) and is simply bad style. It is basically indice of an intermediate state because there aren't lots of mappers in these areas and the default "main" style doesn't differentiate between paved and unpaved roads. --Dieterdreist (talk) 12:25, 8 January 2014 (UTC)
- For more discussions about this, you are invited to come and discuss on the Hot Discussion list.
- From the Mali Activation experience in 2013, the Humanitarian OpenStreetMap Team decided to adopt a classification more appropriate to african countries.
- The original page Mali Highway tag was later renamed to Highway Tag Africa.
- Below are links to discussions on the HOT distribution list about this highway typology.
- Rename to Highway Tag Africa Central-African-Republic-Bangui-taken-over-today-by-the-rebels
- Kenyan or African OSM community? Kenyan-or-African-OSM-community
- These wiki pages recommend to use this highway classification: Senegal Highway tag and WikiProject Central African Republic
- -- pierzen (talk) 20:09, 8 January 2014 (UTC)
There's an empty photos column. Could we use some Bing (or another) imagery snapshot to get the idea ?
I assume that most of the mappers that read this page map from imagery, so that could be useful to have some examples...
Maybe there's some licencing issue, like Bing is only allowed for tracing in OSM or something ?
- While it might be easy to represent road infrastructures in northern countries, there is less homogeneity in africa. We could simply remove the photo column.
- pierzen (talk) 16:26, 18 April 2014 (UTC)
The suggestions in this article could probably be applied generically to other countries, even developed ones. But a definition for trunk ways in Africa is currently missing.
I don't have deep knowledge of differences within the continent, but apparently in Nigeria and Ghana trunk ways are associated with national routes ("A" routes in Nigeria and "N" routes in Ghana). This would be the same criteria adopted in various other countries (Argentina, Paraguay, Bolivia, Peru, Canada, Australia, etc.).
The article also describes these roads as connecting place=*, whose definition may depend on administrative/political status or, if that is not yet established, based on total population. Because of that, I think it makes sense considering trunk ways those (and perhaps only those) that connect places that are considered . If an assignment between OSM place definitions and local administrative/political status is missing, I think it makes sense to define a "metropolis" as a city with more than 1 million people.
This part seems to have been changed. In the translation I made, there is more focus on ford=yes versus flood_prone=yes. Where the first is for normal river crossings where you have to just drive through the water, and flood prone for exceptional situations. The focus in the English version on flood_prone is a bit weird to me, as ford is used 10 times more than ford... Joost schouppe (talk) 14:19, 22 January 2016 (UTC)
- I think that Joost is saying and what I too think + that it is simple and clear: Crossings of any waterway=* when they don't have a bridge=* or a tunnel=culvert or other tunnel=* for that matter -- ie. when the road crosses the waterway in the same level (in principle level=0 but zero is never used as a layer value as it's the deault -- should simply be tagged with ford=yes (on either then node of the road that crosses with the waterway or on the bit of the road that crosses a waterway that also has waterway=riverbank drawn around it.
- As for the low-level or flood_prone bridges that may be at times under water or those that are even regularly submerged, e.g. during rainy season or other intermittent rains, one should use the bridge=low_water_crossing tag (as usual) for the bit of road that is on top of that bridge structure. As it is instructed (and makes sense) one can add flood_prone=yes to the bridge. (It's also instructed that one could use ford=yes but I think that is simply wrong as the "ford" in question is only intermittent/seasonal and would hence require intermittent/seasonal=yes to tell the reality -- and that would get messy as it could be seen as to imply that the bridge is intermittent/seasonal as is the case with ice roads in a similar-ish manner.)
- For more details and discussion about these bridge page #Values section and the talk page of the article are good reading.
- I'll edit the page as explained above. Just wanted to explain the edit here so there's no further confusion. --JaakkoH (talk) 21:12, 13 April 2016 (UTC)
- Note: I tweaked both the Other crossings a bit more than noted above. The low_water_crossing, while technically not a bridge (see bridge=* discussion page for details is in practice a bridge and I moved it there -- explaining it additionally in the flood prone part. I also made some general (but IMO not that major) edits to structure of the text for (hopefully) improved clarity. Please check if you think it's better or worse now in your opinion. --JaakkoH (talk) 22:33, 13 April 2016 (UTC)