Re: [OSM-talk] Divided/Non-Divided Intersection
At 2010-07-22 08:25, Dylan Semler wrote: On Thu, Jul 22, 2010 at 11:08 AM, Dylan Semler dylan.sem...@gmail.com wrote: On Thu, Jul 22, 2010 at 5:57 AM, andrzej zaborowski balr...@gmail.com wrote: On 11 July 2010 10:23, Maarten Deen md...@xs4all.nl wrote: If you have a layout like this (use a fixed-width font): | | A+-+ | +B C+-+ | | And you want to go from B to A, why would routing software say go straight on and not go right, then go left? My opinion is that it is a routing software issue after all... And option is to map it like this: | | A+ | |\| | +B |/| C+ | | | How about grouping all of the nodes of the intersection into a relation? Routing software can treat it as a single intersection and the map can reflect how the roads are actually laid. Actually, I see there already is a proposed feature to use relations to handle routing instructions for complicated turns[1] [1]http://wiki.openstreetmap.org/wiki/Relations/Proposed/Turn_hints But these are not complicated intersections - they are quite common perpendicular intersections of two roads, and mappers have drawn them in all possible ways. I don't think it's reasonable to ask mappers to have to jump through yet another hoop to map something so common - they simply aren't going to do it. I spend a totally unreasonable amount of time mapping turn restrictions (mostly no-u-turn) as it is, and even that is hard to justify. -- Alan Mintz alan_mintz+...@earthlink.net ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
Hi, sorry to resurrect an old thread, but I haven't seen it answered in the end. On 11 July 2010 10:23, Maarten Deen md...@xs4all.nl wrote: John Smith wrote: On 11 July 2010 06:43, Chris Dombroski cdombroski+...@icanttype.org wrote: I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right Isn't that a problem with the routing software, not the data? Not IMHO. If you have a layout like this (use a fixed-width font): | | A+-+ | +B C+-+ | | And you want to go from B to A, why would routing software say go straight on and not go right, then go left? My opinion is that it is a routing software issue after all. When a road segment is just about 5m long, the software should just skip it in the driving directions and look at the angle between the A and the B roads. It can go as far as doing what Alan said, i.e. join all of the nodes in a junction (a concentration of nodes where roads meet) into a single node, possibly in the preprocessing phase. I don't agree that this should be done in OSM data, it will prevent more clever routing direction being given when more clever routing software is written (e.g. software telling your car where to stop to wait for green light, which lane to take etc). The lengths need to be chosen carefully because in e.g. pedestrian routing, 5m may be significant (say you're looking for an exit from a maze). A car can't even make turns that tight, and you're interested in the bigger picture when you reach a junction, the routing is not clever enough to get you through a junction anyway. And option is to map it like this: | | A+ | |\| | +B |/| C+ | | | Cheers ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
On Thu, Jul 22, 2010 at 5:57 AM, andrzej zaborowski balr...@gmail.comwrote: On 11 July 2010 10:23, Maarten Deen md...@xs4all.nl wrote: If you have a layout like this (use a fixed-width font): | | A+-+ | +B C+-+ | | And you want to go from B to A, why would routing software say go straight on and not go right, then go left? My opinion is that it is a routing software issue after all. When a road segment is just about 5m long, the software should just skip it in the driving directions and look at the angle between the A and the B roads. It can go as far as doing what Alan said, i.e. join all of the nodes in a junction (a concentration of nodes where roads meet) into a single node, possibly in the preprocessing phase. I don't agree that this should be done in OSM data, it will prevent more clever routing direction being given when more clever routing software is written (e.g. software telling your car where to stop to wait for green light, which lane to take etc). The lengths need to be chosen carefully because in e.g. pedestrian routing, 5m may be significant (say you're looking for an exit from a maze). A car can't even make turns that tight, and you're interested in the bigger picture when you reach a junction, the routing is not clever enough to get you through a junction anyway. And option is to map it like this: | | A+ | |\| | +B |/| C+ | | | How about grouping all of the nodes of the intersection into a relation? Routing software can treat it as a single intersection and the map can reflect how the roads are actually laid. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
On Thu, Jul 22, 2010 at 11:08 AM, Dylan Semler dylan.sem...@gmail.comwrote: On Thu, Jul 22, 2010 at 5:57 AM, andrzej zaborowski balr...@gmail.comwrote: On 11 July 2010 10:23, Maarten Deen md...@xs4all.nl wrote: If you have a layout like this (use a fixed-width font): | | A+-+ | +B C+-+ | | And you want to go from B to A, why would routing software say go straight on and not go right, then go left? My opinion is that it is a routing software issue after all. When a road segment is just about 5m long, the software should just skip it in the driving directions and look at the angle between the A and the B roads. It can go as far as doing what Alan said, i.e. join all of the nodes in a junction (a concentration of nodes where roads meet) into a single node, possibly in the preprocessing phase. I don't agree that this should be done in OSM data, it will prevent more clever routing direction being given when more clever routing software is written (e.g. software telling your car where to stop to wait for green light, which lane to take etc). The lengths need to be chosen carefully because in e.g. pedestrian routing, 5m may be significant (say you're looking for an exit from a maze). A car can't even make turns that tight, and you're interested in the bigger picture when you reach a junction, the routing is not clever enough to get you through a junction anyway. And option is to map it like this: | | A+ | |\| | +B |/| C+ | | | How about grouping all of the nodes of the intersection into a relation? Routing software can treat it as a single intersection and the map can reflect how the roads are actually laid. Actually, I see there already is a proposed feature to use relations to handle routing instructions for complicated turns[1] [1]http://wiki.openstreetmap.org/wiki/Relations/Proposed/Turn_hints ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
On Sun, 11 Jul 2010 22:03:02 -0700, Alan Mintz alan_mintz+...@earthlink.net wrote: At 2010-07-10 18:39, John Smith wrote: On 11 July 2010 06:43, Chris Dombroski cdombroski+...@icanttype.org wrote: I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right Isn't that a problem with the routing software, not the data? That's what I would think. The problem may be even worse when ways need to be split to accommodate turn restrictions, if the intersection is modeled as four separate points. I like bringing everything together to a single intersection point because that's what it (topo)logically is - a single intersection controlled by a group of signals operating together (or stop signs with drivers co-operating). I disagree with that view. A map is a representation of how it looks on the ground, not how the road is topologically made up. Or are you also removing bends and curves in roads because they have no topological meaning? The first example Nathan gave is IMHO a perfect example how not to map. Regards, Maarten ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
At 2010-07-11 23:44, Maarten Deen wrote: On Sun, 11 Jul 2010 22:03:02 -0700, Alan Mintz alan_mintz+...@earthlink.net wrote: I like bringing everything together to a single intersection point because that's what it (topo)logically is - a single intersection controlled by a group of signals operating together (or stop signs with drivers co-operating). I disagree with that view. A map is a representation of how it looks on the ground, not how the road is topologically made up. I believe that is exactly backwards. A picture is used if you want to see what an area physically looks like. The style of map that is being rendered by Mapnik, OSMarender, etc. has always been a type of map (in a cartographic sense) that is a diagram of how roads and other mapped features connect with each other, using a defined set of symbols, linetypes, colors, etc. I suppose it could be argued that a topo map might blur this distinction, though it still uses symbology and lines to define the elevations instead of rendering the actual features. It could be argued that splitting the roads into separate one-way ways is a hack for the purpose of convenience (usually to reduce the number of turn restrictions required). I do think that this is an issue of personal style, and that most of the variations I've seen are reasonable. Or are you also removing bends and curves in roads because they have no topological meaning? I remove very little. I draw bends and curves to make the map follow on-the-road GPS navigation, and to look nice. Yes, it's not a seamless intellectual framework for the universe :) -- Alan Mintz alan_mintz+...@earthlink.net ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
John Smith wrote: On 11 July 2010 06:43, Chris Dombroski cdombroski+...@icanttype.org wrote: I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right Isn't that a problem with the routing software, not the data? Not IMHO. If you have a layout like this (use a fixed-width font): | | A+-+ | +B C+-+ | | And you want to go from B to A, why would routing software say go straight on and not go right, then go left? And option is to map it like this: | | A+ | |\| | +B |/| C+ | | | Maarten ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Divided/Non-Divided Intersection
Maarten Deen wrote: John Smith wrote: On 11 July 2010 06:43, Chris Dombroski cdombroski+osm at icanttype.org wrote: I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right Isn't that a problem with the routing software, not the data? Not IMHO. If you have a layout like this (use a fixed-width font): | | A+-+ | +B C+-+ | | And you want to go from B to A, why would routing software say go straight on and not go right, then go left? And option is to map it like this: | | A+ | |\| | +B |/| C+ | | | I use either this or a slight modification where ways A and C meet on the left side and B continues across the right side. Usually I decide based on whether A/C or B is a more major road, and thus what classification the pieces in the middle should be. In some US counties, the imported TIGER data treats an intersection of divided highways as a single node: http://www.openstreetmap.org/?lat=39.785241lon=-75.546328zoom=18layers=B000FTF This is IMO a horrible solution, especially when channelied left turns are added between the main lanes. I changed a nearby stretch of US 13 to keep the main lanes straight, and it looks much better: http://www.openstreetmap.org/?lat=39.70937lon=-75.56229zoom=17layers=B000FTF ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
At 2010-07-10 18:39, John Smith wrote: On 11 July 2010 06:43, Chris Dombroski cdombroski+...@icanttype.org wrote: I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right Isn't that a problem with the routing software, not the data? That's what I would think. The problem may be even worse when ways need to be split to accommodate turn restrictions, if the intersection is modeled as four separate points. I like bringing everything together to a single intersection point because that's what it (topo)logically is - a single intersection controlled by a group of signals operating together (or stop signs with drivers co-operating). -- Alan Mintz alan_mintz+...@earthlink.net ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Divided/Non-Divided Intersection
What is the the best/accepted way to map an intersection where a road changes from divided to non-divided at the intersection if the intersecting road is also divided? i.e: =||- I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right (Originally posted on help.openstreetmap.org, where it was suggested that this would probably spawn a discussion so a mailing list would be better) ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
On 11 July 2010 06:43, Chris Dombroski cdombroski+...@icanttype.org wrote: I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right Isn't that a problem with the routing software, not the data? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Divided/Non-Divided Intersection
To a certain extent, that's probably true. I know my Garmin gives strange directions at some intersections, but of course I can't see the raw data to see if the fault lies there. I just suspect that's the case. John Smith deltafoxtrot...@gmail.com wrote: On 11 July 2010 06:43, Chris Dombroski cdombroski+...@icanttype.org wrote: I ask because I think this is the cause of stupid GPS directions at times make a left, followed by a slight right Isn't that a problem with the routing software, not the data? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
