Changeset: 50598666
added runway sections
Closed by 4b696d
Tags
created_by | JOSM/1.5 (12450 de) |
---|
Discussion
-
Comment from rwrev
Some anomalous data appears to be part of this changeset.
**way id="510548248" indicates self as aeroway=runway, and lists length of "11095 ft", but the referenced nodes <nd ref="1523113554"/> <nd ref="3402826496"/> appear to cover less than a couple hundred feet, was this to be a blast_pad or displaced_threshold that is not currently marked as such? If marked as blast_pad/displaced_threshold the length should be corrected or removed.
**way id="510548245" This way indicates <tag k="ref" v="07L/25R"/> is marked as a displaced_threshold, and the v="?" should only reference the runway (end) it belongs to.
**way id="510548249" is indicated as blast_pad but is labeled with both runway designators, <tag k="ref" v="06R/24L"/>, it should have only the runway designator for the runway it is part of.
**way id="510548250" lists <tag k="length" v="10285 ft"/> but its nodes appear to cover only a few hundred feet. The length should be corrected or removed. It is also labeled k="ref" v="06R/24L" but marked as a displaced_threshold which is only on one of those runway(end)s, and should have v="the end it is part of".
**way id="510548251" is marked as blastpad by lists <tag k="length" v="10285 ft"/> while its nodes appear to cover only a few hundred ft. The length should be corrected or removed. It is also labeled with <tag k="ref" v="06R/24L"/> but should have only the designator for the runway (end) it is part of.
**way id="510548252" is marked as a blast_pad but labeled with <tag k="ref" v="06L/24R"/>. It should only be labeled with the runway (end) it is part of.
**way id="138935869" is marked as a blast_pad but labeled with <tag k="ref" v="07R/25L"/>. It should only be labeled with the runway (end) it is part of. -
Comment from 4b696d
I fixed some things but I don't see why thresholds and blast pads should not carry the full name of the runway.
-
Comment from rwrev
A 'single runway strip' is actually two separate runways for air traffic.
A (single) blast pad or threshold is only on one runway (one end of a strip), not both ends. OSM data seems to only record one strip to represent two otherwise logically separate (for air traffic) runways, and additionally in renderings uses the combined label, when only the applicable part of those combined labels should be used on the runway (strip end) to which it belongs - at least to follow US FAA guidelines. But the blast_pad/displaced_thresholds for each runway (end) are (have to be, since they are disjoint?) recorded separately, and hence really only the runway (end) to which they belong should be used to label them. I am by no means an authority, but you can see that I do have an opinion. :)
It seems OSM could have actually represented the runways separately, with two ways (one for each logical runway) that listed the representing nodes in opposite order. But so far, having only been working with three US states, this does not seem to have been done anywhere. Maybe there's something I'm not aware of that makes it a bad idea. -
Comment from 4b696d
1. OSM will never be accurate enough for air traffic.
2. There is the rule that for one physical object there should only be one element in OSM.
Have a look at the wiki: https://wiki.openstreetmap.org/wiki/Tag:aeroway%3DrunwayYour account is quite new so I suppose that you are new to OSM. Some new members tend to make drastic changes to the data because they think that things are done wrong, but usually those changes to more harm than good. To begin please observe how things are done and check the wiki if you are not sure. You also have the option to ask on your local mailing list or the forum.
-
Comment from rwrev
a)accuracy - perhaps not, but nice to have as representative as reasonably possible
b)new user - yes
c)one phys obj - may have read that before, suspected something of the sort
d)FWIW - how things are done - you can double check me (tired of counting), but appears that pattern established by displaced_threshold (prior to changes made by uid=1420318) in California/Colorado was to label the (single) runway end or leave unlabeled, and often, though not exclusively, to indicate via node with single point, rather than a way and multiple points. (In current Mass. data, all blast_pad/displaced_threshold entries/changes are relatively recent by uid=1420318, no 'earlier history' to examine in current data [don't know what DB may have elsewhere.])
some Stats
Colo.,
displaced_threshold 6 single label, 5 unlabeled, 1 dual labeled,
blast_pad, none prior to recent changes by uid=1420318.
California
displaced_threshold 12 single label, 12 unlabeled, 4 dual labeled (all by uid=1420318)
blast_pad 10 - all 'recent',1 by uid=3176959, others by uid=1420318 -
Comment from 4b696d
Just so that you don't get me wrong, 'new user' was not supposed to sound negative.
To be honest this blast_pad and displaced_threshold thing was not widely spread. I stumbled about it at some point of time and I then added way more than there have been before to airports around the world. There is no rule for these appendixes of runways but they are officially part of the runway, that is why I added them as ways and not as nodes.
The wiki says that runways should be tagged with both numbers. I know that some runway ends have been tagged only with one number but there were just a minority on a global scale.
If you don't know taginfo, check it out. You can display the tags on a map if you click on 'Overpass Turbo' in the top right. It is a cool and very useful tool. https://taginfo.openstreetmap.org/tags/?key=runway&value=displaced_threshold#overview -
Comment from rwrev
Perhaps the wiki should be changed to clarify that where appropriate parts (blast_pad/displaced_threshold/?) of a runway (end) should be labeled with the (singular) runway(end) they are part of. There were, IIRC, at least 5 different users with some changes dating back some 5-or-so years assoc'd with the earlier displaced_threshold entries in california that exhibit (hopefully accurate) singular labeling - perhaps some are still available and could be consulted for what guidance they were using. Per your guidance to 'observe how things are done', their earlier input would indicate singular labeling of those items is appropriate.
-
Comment from 4b696d
The wiki can be changed but there should be discussions about it and first of all these runway sections needed to be added in the first place. There have been suggestions a long time ago: https://wiki.openstreetmap.org/wiki/Talk:Aeroways#Sections_of_a_runway
I suppose that those guys just tagged the runways in a way they thought was appropriate. Did you have a look at this at a global scale?
'observe how things are done' - of course. That is still my suggestion to new members and most of the time it is the right way to go but what I did not mean was observe how things were done 5 years ago. There are a lot of deprecated features that might not be wise to use anymore. Keep in mind as well that OSM if full of errors.
Back to the runway sections. In my opinion these ends are part of 'both' runways as you would call it, because in emergency use of a runway they may serve as overrun areas. That is to say that all sections might be used in both directions, but some only in emergency use of a runway which is still a valid use.
Ways (14)
- 07R/25L (510548244), v1
- 07L/25R (510548245), v1
- 07L/25R (510548246), v1
- 07L/25R (510548247), v1
- 07R/25L (510548248), v1
- 06R/24L (510548249), v1
- 06R/24L (510548250), v1
- 06R/24L (510548251), v1
- 06L/24R (510548252), v1
- 07R/25L (138935869), v5
- 07L/25R (138935876), v8
- 07R/25L (333172248), v3
- 06R/24L (409186724), v2
- 409201141, v3
Relations (1)
Nodes (7)
Welcome to OpenStreetMap!
OpenStreetMap is a map of the world, created by people like you and free to use under an open license.
Hosting is supported by Fastly, OSMF corporate members, and other partners.
https://openstreetmap.org/copyright | https://openstreetmap.org |
Copyright OpenStreetMap and contributors, under an open license |