Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 JigPu : > I understand why you feel that using a way+node relation or way+direction > tag may be a hack, but "children" are also a hack. Children not only only a > single problem (that of a way being asked to take on additional metadata > that really belongs to only a portion of the way) but solves it by poorly > duplicating relations! Relations are capable of modeling the same > parent-child relationship you desire and exist precicely to relate many > pieces into a whole. While relations may already exist, I don't see how they would be the best method to handle this, since ways are children of relations, relations aren't children of ways. > I'm not saying that the "lane and lane group" proposal is perfect either. In > my opinion it abuses relations so as not to upset the status quo too much. > Personally I think > http://wiki.openstreetmap.org/wiki/Relations/Proposed/Collected_Ways does a > much better job of using relations correctly. Multiple sub-ways are related > to each other, with the relation holding the generic tags and sub-ways > containing any overriding information. If renderers supported it, it'd be > the ideal way of representing "child" ways. I disagree, I think this is another hack to try and make lanes without making lanes, we need to stop thinking about what we can do in the current framework and do come up with real solutions, and using relations for this purpose is a ugly ugly way to do it. > Somewhat tangentally, while finding the Collected Ways proposal I stumbled > across the "enforcement" relation. It's an "approved" use of relations and > acts almost identically to the stop relation proposal. Though this doesn't > mean it's not a hack, it does indicate that the community seems to be ok > with using relations in the way proposed. I never said relations couldn't be used for this purpose, however I don't think it's the best way to do it, and just because 20 or 30 people agreed with a proposal doesn't make it the best solution either, just look at the mess highway=path caused and 20 or 30 people agreed with it too. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Interesting article about Ordnance Survey on Wikileaks
http://wikileaks.org/wiki/New_Digital_Master_Map_for_Great_Britian:_Confidential_Advice_to_Ministers%2C_2009 ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 9:14 PM, John Smith wrote: > 2009/8/31 Roy Wallace : > > That sounds more or less what I had in mind, yes, except the example > osm file used relations acting as children of the way almost, why > can't ways simply have children of their own instead of using > relations? I understand why you feel that using a way+node relation or way+direction tag may be a hack, but "children" are also a hack. Children not only only a single problem (that of a way being asked to take on additional metadata that really belongs to only a portion of the way) but solves it by poorly duplicating relations! Relations are capable of modeling the same parent-child relationship you desire and exist precicely to relate many pieces into a whole. I'm not saying that the "lane and lane group" proposal is perfect either. In my opinion it abuses relations so as not to upset the status quo too much. Personally I think http://wiki.openstreetmap.org/wiki/Relations/Proposed/Collected_Ways does a much better job of using relations correctly. Multiple sub-ways are related to each other, with the relation holding the generic tags and sub-ways containing any overriding information. If renderers supported it, it'd be the ideal way of representing "child" ways. Somewhat tangentally, while finding the Collected Ways proposal I stumbled across the "enforcement" relation. It's an "approved" use of relations and acts almost identically to the stop relation proposal. Though this doesn't mean it's not a hack, it does indicate that the community seems to be ok with using relations in the way proposed. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Roy Wallace : > +1 - I agree with Tobias and Anthony. John, I know you feel that > solving the "lane" problem will solve the "stop sign" problem (indeed > it might), but the stop sign problem *can be solved independently* (by > indicating the direction or the junction to which it applies), and is > a simpler problem to solve - so I think we should go ahead with this > and vote for either the use of a way+node relation or a node+direction > tag to denote a stop sign (I vote for relation for several reasons). Which is a hack, it solves one issue out of many, it would be much better to solve most if not all of the many problems with a single solution. Coming up with a bunch of hacks that solve them independent of each other is going to be very bad for consistency. > John, perhaps we're missing something. Why do you want to add more > tables to the DB? I don't see why you think using a relation is a > "hack". You said, "just like a relation joins multiple ways, a way Because there is numerous problems that need to be solved, stop signs is just one of them, if we can tag things in lanes it would solve a good deal of them. > [should join] multiple lanes." From the wiki: "relations are basically > groups of objects in which each object may take on a specific role." http://wiki.openstreetmap.org/wiki/Relations/Relations_are_not_Categories > This sounds suitable for what you're suggesting. Have you seen > http://wiki.openstreetmap.org/wiki/Proposed_features/lane_and_lane_group, > for example? That sounds more or less what I had in mind, yes, except the example osm file used relations acting as children of the way almost, why can't ways simply have children of their own instead of using relations? This diagram explains a common way, it looks like it has 2 lanes for through traffic, 2 parking lanes and a footway http://wiki.openstreetmap.org/images/4/46/Lane_group_example1_screen_2.png ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] RR8 - Possible International Vandal (assistance required in various countries)
Hi list, Attention was drawn to the OSM user RR8 last night on IRC. It appears that they have been producing /apparently/ deconstructive edits at a high frequency since late Saturday evening (server time). Edits primarily are the reclassification of highways to a different level. This has occured frequently in the East Midlands (often including a change of reference number that cannot be confirmed from other sources), places in Ireland where under-construction motorways have been marked as opened, and a few reclassifications elsewhere, most notably Iceland. The edits appear to look constructive, but are more likely to be destructive. Ideally, someone local to the areas in question should check a few of the changes, or we could get people from the mailing list to consider the edits as a whole to decide what's to be done about them. http://www.openstreetmap.org/user/RR8/edits -- Regards, Thomas Wood (Edgemaster) ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Walking Papers was silently broken for a few days, but now it's fine.
Hi, I've just noticed that I broke the Walking Papers (http://walking-papers.org ) scan process with the new multi-language update, and one week's worth of new scans were not correctly handled. Sorry! I've fixed the problem (I had been relying on the existence of an Accept-Language request header, which of course every browser sends by default and every HTTP library does not) and I'm getting each of the scans back up, here: http://walking-papers.org/scans.php I hope no one was unduly inconvenienced and I apologize for the outage. -mike. michal migurski- m...@stamen.com 415.558.1610 ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sunday, August 30, 2009, Roy Wallace wrote: > On Mon, Aug 31, 2009 at 4:36 AM, Anthony wrote: >> >> You want to have two completely different ways to tag stop >> signs? What's your solution? Stop signs are directional, whether they are >> lane based or not. > > +1 - I agree with Tobias and Anthony. John, I know you feel that > solving the "lane" problem will solve the "stop sign" problem (indeed > it might), but the stop sign problem *can be solved independently* (by > indicating the direction or the junction to which it applies), and is > a simpler problem to solve - so I think we should go ahead with this > and vote for either the use of a way+node relation or a node+direction > tag to denote a stop sign (I vote for relation for several reasons). > >> That makes no sense. You want to have lanes instead of ways? >> >> Lanes are ways. > > +1 > > John, perhaps we're missing something. Why do you want to add more > tables to the DB? I don't see why you think using a relation is a > "hack". You said, "just like a relation joins multiple ways, a way > [should join] multiple lanes." From the wiki: "relations are basically > groups of objects in which each object may take on a specific role." > This sounds suitable for what you're suggesting. Have you seen > http://wiki.openstreetmap.org/wiki/Proposed_features/lane_and_lane_group, > for example? > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk > Couldn't have said it better myself. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Mon, Aug 31, 2009 at 4:36 AM, Anthony wrote: > > You want to have two completely different ways to tag stop > signs? What's your solution? Stop signs are directional, whether they are > lane based or not. +1 - I agree with Tobias and Anthony. John, I know you feel that solving the "lane" problem will solve the "stop sign" problem (indeed it might), but the stop sign problem *can be solved independently* (by indicating the direction or the junction to which it applies), and is a simpler problem to solve - so I think we should go ahead with this and vote for either the use of a way+node relation or a node+direction tag to denote a stop sign (I vote for relation for several reasons). > That makes no sense. You want to have lanes instead of ways? > > Lanes are ways. +1 John, perhaps we're missing something. Why do you want to add more tables to the DB? I don't see why you think using a relation is a "hack". You said, "just like a relation joins multiple ways, a way [should join] multiple lanes." From the wiki: "relations are basically groups of objects in which each object may take on a specific role." This sounds suitable for what you're suggesting. Have you seen http://wiki.openstreetmap.org/wiki/Proposed_features/lane_and_lane_group, for example? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] GPS gadgets can get you jailed and fined
On a plane from Paris to Boston, a guy left a GPS gadget on his seat armrest when he went to the loo. (We've all done this, haven't we?) The passengers get scared, they tell the air hostesses, the plane is diverted, the guy's arrested, he spends days in jail and is fined US $32,000! http://timescorrespondents.typepad.com/charles_bremner/2009/08/frenchman-scares-american-airlines-with-pocket-gadget-.html - L ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] [Talk-de] Statistisches Landesamt Sachse n - Straßenlisten, Einwohner
Am 30. August 2009 10:20 schrieb Florian Lohoff : > Vervielfältigen wollen wir ja gar nicht - auch nicht wirklich verbreiten ;) > Und > wenn wir das machen ist die Quellenangabe kein problem ... Wenn wir die andere > liste (OSM) ergaenzen dann machen wir das anhand von Straßenschildern und > nicht > dieser Liste. Die dient ja nur dazu einen hint zu geben wo was fehlen > koennte ... > > Aber um sicherzugehen koennte man denen ja mal eine Mail schreiben Da Dirk Stöcker in Kontakt mit dem LVA ist, wäre er dafür prädestiniert. > > Beispiel Chemnitz: > > http://www.statistik.sachsen.de/regioreg/RRServlet?function=Lesen&id=21234&type=10001&out=pdf > > > > Regionalregister Sachsen: > > http://www.statistik.sachsen.de/regioreg/ > > Und das haben die fuer jede Gemeinde? Ein traum ;) Könntest du das Ganze in die Straßenlistenauswertung mit einbauen? Ciao André ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 2:36 PM, Anthony wrote: > Lanes are ways. > "A way is a list of at least two nodes that describe a linear feature such as a street, or similar." ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 2:14 PM, John Smith wrote: > Now I know you're being a troll, however if we used lanes instead of > using ways you could easily tag that intersection you gave as an > exception rather than the average case that I presented which is a lot > more common. Tag it how? You want to have two completely different ways to tag stop signs? What's your solution? Stop signs are directional, whether they are lane based or not. I agree the case you gave is more common, but I don't think the case you gave is a problem for any of the proposed methods of tagging stop signs. > So in short the answer is to have lanes, not ways. That makes no sense. You want to have lanes instead of ways? Lanes are ways. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > On Sun, Aug 30, 2009 at 1:40 PM, John Smith > wrote: >> >> 2009/8/31 Anthony : >> >> > I guess we will have to agree to disagree, but it'd be nice if you'd >> > answer >> > my questions about how to do all those things I asked about. >> >> I don't see a point in answering any more of your questions to be >> honest, you seem to be trolling. > > You say "any more" as though you've answered any of them. You don't seem to > have thought things through. Think them through and you'll see that I'm not > trolling. Now I know you're being a troll, however if we used lanes instead of using ways you could easily tag that intersection you gave as an exception rather than the average case that I presented which is a lot more common. So in short the answer is to have lanes, not ways. This would also solve using a relation to indicate a turning restriction as you could easily show which lanes can go where, it would allow for advanced routing and you can place stop signs on all lanes effected by it. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Rendering of localized maps
> Are the tiles of the localized maps re-rendered at all? I've finished > adding persian country names to the OSM db last week already and the > persian map still shows the state from when it was initially set up [2]. > I think this question rather goes to Ævar Arnfjörð as he set up the > localized map servers (to my knowledge). I'm tired to repeat this: no, they aren't. State is of 2009-07-15 and it will stay there until s/o with root-perms will start working on cassini / ortelius. Peter ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > I guess we will have to agree to disagree, but it'd be nice if you'd answer > my questions about how to do all those things I asked about. I don't see a point in answering any more of your questions to be honest, you seem to be trolling. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 1:40 PM, John Smith wrote: > 2009/8/31 Anthony : > > > I guess we will have to agree to disagree, but it'd be nice if you'd > answer > > my questions about how to do all those things I asked about. > > I don't see a point in answering any more of your questions to be > honest, you seem to be trolling. > You say "any more" as though you've answered any of them. You don't seem to have thought things through. Think them through and you'll see that I'm not trolling. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 1:22 PM, John Smith wrote: > 2009/8/31 Anthony : > > > Right, and at an intersection which has a turning lane, there is a > > restriction on a per-lane basis. You can only turn left from the turning > > lane - you can only go straight (or possibly right) from the other lanes. > > So do you propose splitting the way at every turning lane? > > How would using relations solve this either? I never suggested using relations to "solve" it. I don't think it needs solving. You cross the roads at a 90 degree angle and ignore the turning lane. People know you have to get into the left hand turning lane to make a left hand turn (a jughandle would look altogether different as the jughandle itself would constitute a new way). Right now turning restrictions already require a relation, though this relation could actually be avoided if you allowed "lane" to be a "way" (you'd simply have nodes for the lanes that connect and no nodes for the lanes that don't connect). Each lane would already be one-way unless you're talking about a road where both directions share the same lane. > I guess for the most part we don't. We could just split the way up into > > multiple lanes as necessary and then merge them back together. Still no > > need for an extra table or children, though. > > I guess we have to agree to disagree then, you seem to be trying to > fit things into the existing frame work for the sake of it and it's an > ugly hack. > I guess we will have to agree to disagree, but it'd be nice if you'd answer my questions about how to do all those things I asked about. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > Right, and at an intersection which has a turning lane, there is a > restriction on a per-lane basis. You can only turn left from the turning > lane - you can only go straight (or possibly right) from the other lanes. > So do you propose splitting the way at every turning lane? How would using relations solve this either? > I guess for the most part we don't. We could just split the way up into > multiple lanes as necessary and then merge them back together. Still no > need for an extra table or children, though. I guess we have to agree to disagree then, you seem to be trying to fit things into the existing frame work for the sake of it and it's an ugly hack. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 1:09 PM, John Smith wrote: > 2009/8/31 Anthony : > > What happens when the road goes from four lanes to six? Should this be > > recorded at every intersection which has a turning lane? > > I really need to stop feedint the trolls. Lanes only get tagged if > they differ from the way to need it, the values of the way cascade to > the lanes. Right, and at an intersection which has a turning lane, there is a restriction on a per-lane basis. You can only turn left from the turning lane - you can only go straight (or possibly right) from the other lanes. So do you propose splitting the way at every turning lane? In any case when switching from 4 to 6 they should be split > already anyway. Which, if you read above, is what I said. Of course, on a 2 lane, 2 direction road with a painted median which turns into a 3 lane road with a turning lane, this isn't likely to be done. > Once there was enough demand for it, I'd probably introduce "lane" as a > type > > of "way" and use relations to tie together lanes in places where lane > > changes are allowed. But I think we're quite a ways off from needing > that. > > Why do we need to use relations at all? I guess for the most part we don't. We could just split the way up into multiple lanes as necessary and then merge them back together. Still no need for an extra table or children, though. > You are trying to push a square peg in a round hole rather than doing > it properly. > I'd say you're the one who's doing that. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > What happens when the road goes from four lanes to six? Should this be > recorded at every intersection which has a turning lane? I really need to stop feedint the trolls. Lanes only get tagged if they differ from the way to need it, the values of the way cascade to the lanes. In any case when switching from 4 to 6 they should be split already anyway. > Once there was enough demand for it, I'd probably introduce "lane" as a type > of "way" and use relations to tie together lanes in places where lane > changes are allowed. But I think we're quite a ways off from needing that. Why do we need to use relations at all? You are trying to push a square peg in a round hole rather than doing it properly. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 12:51 PM, John Smith wrote: > 2009/8/31 Anthony : > > > And that's part of what's wrong with it. You still haven't explained how > to > > handle stop signs on bi-directional, one lane road. You haven't > explained > > how to handle lane-changes. You require ways to be split every time the > > number of lanes changes (though I guess any system with "lanes=*" does > > that). You don't explain how to record where lanes can be changed and > where > > they can't. You require adding children to every single bi-directional > road > > in the world that merely has a stop sign. > > I'm still unsure where you get this idea that I think a way needs to > be split, just like a relation joins multiple ways, a way joins > multiple lanes. What happens when the road goes from four lanes to six? Should this be recorded at every intersection which has a turning lane? > > It is a reason to do it. But it's a reason to do it right. > > How would you do it? > Once there was enough demand for it, I'd probably introduce "lane" as a type of "way" and use relations to tie together lanes in places where lane changes are allowed. But I think we're quite a ways off from needing that. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > And that's part of what's wrong with it. You still haven't explained how to > handle stop signs on bi-directional, one lane road. You haven't explained > how to handle lane-changes. You require ways to be split every time the > number of lanes changes (though I guess any system with "lanes=*" does > that). You don't explain how to record where lanes can be changed and where > they can't. You require adding children to every single bi-directional road > in the world that merely has a stop sign. I'm still unsure where you get this idea that I think a way needs to be split, just like a relation joins multiple ways, a way joins multiple lanes. > It is a reason to do it. But it's a reason to do it right. How would you do it? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Path in JOSM preset and motorcar=yes/designated
Martin Koppenhoefer wrote: > 2009/8/30 Christoph Eckert : > >> If a car can use it, it's a track, not a path. > > +1 > > > Anyway, I have removed motorcar from the list of options, as tagwatch > > does not show significant use (10 times motorcar=no in Europe). If the > > removal triggers some complaints, I'll reenable it. > > thank you. I think it was a good decision to remove motorcar as option > from path as Pieren said: if cars can use the way, it's at least a > track. Sometimes paths are signed with prohibition signs for motorcars, in which case it does make sense to tag the paths with motorcar=no. Cars don't have to be able to use the road to put that sign there. Ben ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 11:45 AM, John Smith wrote: > 2009/8/31 Anthony : > > No, it's a redesign of the whole system. A system which wasn't made for > > per-lane routing information. > > It wasn't designed with relations either, but they now exist too. > > In any case I've only suggested a minor change to add children to ways > and how this could be represented as one table for the DB and > additional tags for ways. Not exactly redoing things from scratch. > And that's part of what's wrong with it. You still haven't explained how to handle stop signs on bi-directional, one lane road. You haven't explained how to handle lane-changes. You require ways to be split every time the number of lanes changes (though I guess any system with "lanes=*" does that). You don't explain how to record where lanes can be changed and where they can't. You require adding children to every single bi-directional road in the world that merely has a stop sign. And you seem to be opposed to splitting ways in situations where there is a painted median. Yet you offer no explanation of how to handle the situation in the example I provided, where all the commercial solutions split the way. On Sun, Aug 30, 2009 at 11:50 AM, John Smith wrote: > As for per-lane routing, funnily enough people actually listed this as > a reason to do it, so you can have smarter routing. > It is a reason to do it. But it's a reason to do it right. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Path in JOSM preset and motorcar=yes/designated
On Sun, Aug 30, 2009 at 4:56 PM, Craig Wallace wrote: > And on a related note, it would be useful to have motor_vehicle in the > presets for highway=track. > It would save having to set both motorcar and motorcycle, as both are > usually prohibited on most tracks around here. Not in many countries. I couldn't give a track default access restriction in France for instance. It's really depending on the region. Pieren ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Rendering of localized maps (was: Problems with the Multilingual Country-List)
Am 27.08.2009 18:08, Peter Körner: >>> [3] http://cassini.toolserver.org/tile-browse/ >> >> The Danish map[1] is simply a blank map at all zoom levels. What gives? >> >> [1] http://cassini.toolserver.org/tile-browse/browse-da.html > > It is just not yet rendered. The tiles will get in the render-queue as > you try to see them. Take another look - it has filles already! > > Peter Are the tiles of the localized maps re-rendered at all? I've finished adding persian country names to the OSM db last week already and the persian map still shows the state from when it was initially set up [2]. I think this question rather goes to Ævar Arnfjörð as he set up the localized map servers (to my knowledge). Claudius [2] http://cassini.toolserver.org/tile-browse/browse-fa.html ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 John Smith : > 2009/8/31 Anthony : >> No, it's a redesign of the whole system. A system which wasn't made for >> per-lane routing information. > > It wasn't designed with relations either, but they now exist too. > > In any case I've only suggested a minor change to add children to ways > and how this could be represented as one table for the DB and > additional tags for ways. Not exactly redoing things from scratch. > Sorry it'd take 3 tables, one for lanes, one for nodes on those lanes and one for lane tags. As for per-lane routing, funnily enough people actually listed this as a reason to do it, so you can have smarter routing. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > No, it's a redesign of the whole system. A system which wasn't made for > per-lane routing information. It wasn't designed with relations either, but they now exist too. In any case I've only suggested a minor change to add children to ways and how this could be represented as one table for the DB and additional tags for ways. Not exactly redoing things from scratch. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 11:19 AM, John Smith wrote: > 2009/8/31 Anthony : > > What solution is better? Your lane-based solution doesn't work if there > is > > only one lane with bi-directional traffic. The solution of adding a node > > and a direction would be second best, but I think it's clunky adding > > multiple extra nodes instead of one relation. > > You really haven't read or understood the details I've put forth. > > As far as I can tell my solution does work, you treat the lanes as > children of the way, so a bi-direction way with traffic while 1 way, > is 2 lanes, and being able to tag those independent of each other is > the key to all this. Please re-read my previous posts on this matter. Please re-read my previous posts on this matter, regarding the road I live on. It is bi-directional, but it does not have two lanes. Are you suggesting we pretend it does? > Are you sure that's always true? I'm certainly not going to adhere to > that > > restriction. If you have a single way, there is an assumption that you > can > > > The you will be fighting every other mapper out there, it's clear that > physical barriers are used, not painted ones so you will end up with > edit wars when people confused with how you've gone off on your own > and it doesn't match what everyone else is doing. We'll see. I just took a look at the section of the OSM map related to the example I gave earlier, and it's too much of a mess for me to fix right now (it has a major highway listed as one way in the wrong direction). But when I fix it, I'm going to use separate ways, because that's the only way to get the routing data to be correct. If someone is going to revert me, I hope they will invent a method to get the routing data correct first. > relations or some other method to tie the lanes together in areas where > lane > > changing is allowed (lane changing allowed between way X, Y, and Z from > node > > A to node B)? > > But that's a hack to work with the existing framework, ie tagging for > software which isn't supposed to be done. No, it's a redesign of the whole system. A system which wasn't made for per-lane routing information. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] waterway=lock
Peter Childs schrieb: > 2009/8/28 : > >>> On 27/08/09 12:13, Jack Stringer wrote: >>> >>> lock=yes lock_name=Withrington Bottom Lock >>> When you are tagging a way, you can't use name= because that will >>> already contain the name of the canal. Hence lock_name=. >>> >>> >> Why would you want to repeat the name of a canal on its individual nodes? >> Isn’t that repeating the mistake of the TIGER node tags? >> >> >> > > Read it again. > > on a node > > lock=yes > name="lock name" > > or on a node > > waterway=lock_gate > name="lock gate name" > > and on the way between the lock gates > > waterway=canal > name="canal name" > lock=yes > lock_name="lock name" > > The Canal way will need to be split at the lock gates, (or in some > cases where a diversion starts (due to some rivers going over weirs) > while there is a lock for boats. > > Peter. > I really like this methode! let's do it that way :) -- Mario ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] revert request
Hi, Richard Weait wrote: > Would somebody run two reverts for me please? At it right now... Bye Frederik ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > What solution is better? Your lane-based solution doesn't work if there is > only one lane with bi-directional traffic. The solution of adding a node > and a direction would be second best, but I think it's clunky adding > multiple extra nodes instead of one relation. You really haven't read or understood the details I've put forth. As far as I can tell my solution does work, you treat the lanes as children of the way, so a bi-direction way with traffic while 1 way, is 2 lanes, and being able to tag those independent of each other is the key to all this. Please re-read my previous posts on this matter. > I don't see the practical application. We're a long way from having our > cars drive for us using micro-level OSM data and some advanced positioning > system. If you want to set up a system that can handle that kind of detail, > be my guest. But it's not needed for something trivial like stop signs and > maximum speeds. THere is many reasons I've already outlined why it's needed, if you don't grasp the concepts please ask for clarification but stop trying to send things off in a different directions that do not address the issues previously outlined in this thread. > Are you sure that's always true? I'm certainly not going to adhere to that > restriction. If you have a single way, there is an assumption that you can The you will be fighting every other mapper out there, it's clear that physical barriers are used, not painted ones so you will end up with edit wars when people confused with how you've gone off on your own and it doesn't match what everyone else is doing. > I'm not against it. I just think it's a lot of work for very little > benefit. And there are probably better solutions. If you really want to > map every single lane of a road, why not make the lane the base unit and use I'm not trying to map individual lanes when there is no need to, I'm suggesting a system that can tag individual lanes if needed. You are the only one suggesting anything different. > relations or some other method to tie the lanes together in areas where lane > changing is allowed (lane changing allowed between way X, Y, and Z from node > A to node B)? But that's a hack to work with the existing framework, ie tagging for software which isn't supposed to be done. By having the ability to tag individual lanes if needed would stop us from needing to make such bad hacks just to stay withing the current framework. You really haven't grasped the concept I've put forward, please re-read and ask for clarification rather than hijacking the thread. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] revert request
On Sun, Aug 30, 2009 at 11:05 AM, Ian Dees wrote: > On Sun, Aug 30, 2009 at 9:38 AM, Richard Weait wrote: >> >> Would somebody run two reverts for me please? >> My import failed in changesets 2311851 and 2308610 >> >> Nodes were imported for woods polygons but not the ways. fail. > > Did you get an error from the server? I also saw this happening for an > import I did a few weeks ago. > just "Error uploading changeset:404" from bulk_upload.py ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 10:27 AM, John Smith wrote: > Why is a relation the best solution? What solution is better? Your lane-based solution doesn't work if there is only one lane with bi-directional traffic. The solution of adding a node and a direction would be second best, but I think it's clunky adding multiple extra nodes instead of one relation. Regardless if it's common sense or not to slow/stop it's a legal > restriction, however it only applies to one direction similar to a > traffic calming device that only applies to one direction, however in > both case it doesn't apply to the entire way, it just applies to a > single lane of the way. > > Just like there is motorway lanes that have different maxspeeds > depending what lane you are in, it would be pointless to do this in a > relation, but being able to tag lanes with individual restrictions it > would solve a number of issues, not just stop signs. I don't see the practical application. We're a long way from having our cars drive for us using micro-level OSM data and some advanced positioning system. If you want to set up a system that can handle that kind of detail, be my guest. But it's not needed for something trivial like stop signs and maximum speeds. > > Whoever provided Google with their data split the way in the example I > > provided, and does so in pretty much any situation where you need to be > in a > > particular lane in order to get where you want to go. In my opinion > paint > > which disallows lane changes should be equivalent to a physical gore area > as > > far as mapping applications are concerned (with allowances for > > approximations in micro areas and lenience when everyone is going > straight > > and it doesn't really matter). > > At this stage only physical barriers are drawn as separate ways Are you sure that's always true? I'm certainly not going to adhere to that restriction. If you have a single way, there is an assumption that you can travel freely between the lanes. If I have to get into the right lane 100 meters before the intersection, I want the way split at that point, not at the intersection, regardless if the barrier is concrete or paint. I suspect that's the way things are already done at least some of the time. It's common sense, really. > , that > isn't really the point of this thread, the point is the best way to > tag nodes or other restrictions on a per lane basis. I'm not against it. I just think it's a lot of work for very little benefit. And there are probably better solutions. If you really want to map every single lane of a road, why not make the lane the base unit and use relations or some other method to tie the lanes together in areas where lane changing is allowed (lane changing allowed between way X, Y, and Z from node A to node B)? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] revert request
On Sun, Aug 30, 2009 at 9:38 AM, Richard Weait wrote: > Would somebody run two reverts for me please? > My import failed in changesets 2311851 and 2308610 > > Nodes were imported for woods polygons but not the ways. fail. Did you get an error from the server? I also saw this happening for an import I did a few weeks ago. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Path in JOSM preset and motorcar=yes/designated
On 30/08/2009 08:24, Christoph Eckert wrote: >> If a car can use it, it's a track, not a path. >> > I better should resist to abuse my dictatorship for tagging politics ;-) . > Anyway, I have removed motorcar from the list of options, as tagwatch does not > show significant use (10 times motorcar=no in Europe). If the removal triggers > some complaints, I'll reenable it. > There seems to be something wrong with those Tagwatch stats, as Tagwatch lists over 4000 uses of motorcar=no for highway=path for Germany. Though I suspect many of them are set by people just because its listed in the JOSM presets. I would agree with its removal anyway from the prests anyway, as it just causes confusion. Maybe it would be more useful to have the option for motorcycle or motor_vehicle instead? And on a related note, it would be useful to have motor_vehicle in the presets for highway=track. It would save having to set both motorcar and motorcycle, as both are usually prohibited on most tracks around here. Craig ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] revert request
Would somebody run two reverts for me please? My import failed in changesets 2311851 and 2308610 Nodes were imported for woods polygons but not the ways. fail. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] waterway=lock
On 28/08/09 13:07, wynnd...@lavabit.com wrote: >> On 27/08/09 12:13, Jack Stringer wrote: >> >>> lock=yes >>> lock_name=Withrington Bottom Lock >> >> When you are tagging a way, you can't use name= because that will >> already contain the name of the canal. Hence lock_name=. >> > Why would you want to repeat the name of a canal on its individual nodes? > Isn’t that repeating the mistake of the TIGER node tags? You wouldn't. The tagging scheme above is for the *way* between the two gate nodes. That way will already have e.g. "name=Grand Union" so you need a lock_name tag to avoid a clash. Gerv ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/31 Anthony : > Unless I'm misreading this, anyone going from 4th Avenue onto Rodeo Drive > has to stop, and everyone going from Roxton Ave onto Rodeo Drive has to > stop. That, in my opinion, has nothing to do with lanes. I'm not sure why > it's so important to model in the first place (you'd have to slow way down > and just about stop whether there was a stop sign there or not), but a > relation would probably be the best solution here. I'm not sure what the > local law is, but where I live the requirement to stop is determined by the > relation of the two roads, not the presence of the stop sign. The stop sign > is just a reminder. Why is a relation the best solution? Regardless if it's common sense or not to slow/stop it's a legal restriction, however it only applies to one direction similar to a traffic calming device that only applies to one direction, however in both case it doesn't apply to the entire way, it just applies to a single lane of the way. Just like there is motorway lanes that have different maxspeeds depending what lane you are in, it would be pointless to do this in a relation, but being able to tag lanes with individual restrictions it would solve a number of issues, not just stop signs. > Whoever provided Google with their data split the way in the example I > provided, and does so in pretty much any situation where you need to be in a > particular lane in order to get where you want to go. In my opinion paint > which disallows lane changes should be equivalent to a physical gore area as > far as mapping applications are concerned (with allowances for > approximations in micro areas and lenience when everyone is going straight > and it doesn't really matter). At this stage only physical barriers are drawn as separate ways, that isn't really the point of this thread, the point is the best way to tag nodes or other restrictions on a per lane basis. > The example you gave has none of that, of course, but I see no reason to > split the way. You are the only one talking about splitting the way, I just want to tag a subset if there is a reason to. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 9:37 AM, John Smith wrote: > However I think this is the most common case: > > > http://maps.google.com/maps?f=q&source=s_q&hl=en&sll=40.111982,-83.089069&sspn=0.001553,0.004506&ie=UTF8&cd=1&geocode=FY4gBwIdU7bx-A&split=0&ll=34.018261,-118.321327&spn=0,359.981976&z=17&layer=c&cbll=34.018064,-118.321328&panoid=SINsqoKL6-IG3u4muEvxfA&cbp=12,15.67,,0,9.88 > > Main way with both side roads having only 2 lanes have a stop sign > before the main way. Unless I'm misreading this, anyone going from 4th Avenue onto Rodeo Drive has to stop, and everyone going from Roxton Ave onto Rodeo Drive has to stop. That, in my opinion, has nothing to do with lanes. I'm not sure why it's so important to model in the first place (you'd have to slow way down and just about stop whether there was a stop sign there or not), but a relation would probably be the best solution here. I'm not sure what the local law is, but where I live the requirement to stop is determined by the relation of the two roads, not the presence of the stop sign. The stop sign is just a reminder. > > Would you split the way? Is there a physical separation? Does paint > count? > > No idea, what do commercial companies do in that situation? > Whoever provided Google with their data split the way in the example I provided, and does so in pretty much any situation where you need to be in a particular lane in order to get where you want to go. In my opinion paint which disallows lane changes should be equivalent to a physical gore area as far as mapping applications are concerned (with allowances for approximations in micro areas and lenience when everyone is going straight and it doesn't really matter). The example you gave has none of that, of course, but I see no reason to split the way. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/30 Anthony : > Of course, if you want to add up the instances of one-lane, bi-directional, > roads with stop signs, and the instances of undivided roads with stop-signs > which only apply to some lanes going in the same direction, I'm sure there > are many many more of the former. I doubt that, but without proper tagging or a different data set to compare with, we won't ever know. However I think this is the most common case: http://maps.google.com/maps?f=q&source=s_q&hl=en&sll=40.111982,-83.089069&sspn=0.001553,0.004506&ie=UTF8&cd=1&geocode=FY4gBwIdU7bx-A&split=0&ll=34.018261,-118.321327&spn=0,359.981976&z=17&layer=c&cbll=34.018064,-118.321328&panoid=SINsqoKL6-IG3u4muEvxfA&cbp=12,15.67,,0,9.88 Main way with both side roads having only 2 lanes have a stop sign before the main way. > Would you split the way? Is there a physical separation? Does paint count? No idea, what do commercial companies do in that situation? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Path in JOSM preset and motorcar=yes/designated
2009/8/30 Christoph Eckert : >> If a car can use it, it's a track, not a path. +1 > Anyway, I have removed motorcar from the list of options, as tagwatch does not > show significant use (10 times motorcar=no in Europe). If the removal triggers > some complaints, I'll reenable it. > thank you. I think it was a good decision to remove motorcar as option from path as Pieren said: if cars can use the way, it's at least a track. cheers, Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 9:01 AM, John Smith wrote: > 2009/8/30 Anthony : > > It was you who suggested that a stop sign is applicable to a lane not a > > way. I'd say, like Tobias, that it is applicable to a way and a > direction. > > No, you stop at a stop sign which is a point on a way, direction only > matters if you can't tag this point on individual lanes. I could just as well say that lane only matters if you can't tag this point in individual directions. Of course, if you want to add up the instances of one-lane, bi-directional, roads with stop signs, and the instances of undivided roads with stop-signs which only apply to some lanes going in the same direction, I'm sure there are many many more of the former. > > exceptions I can think of this is especially appropriate because at the > spot > > where the stop sign applies to one lane and not the others the road is > > divided by a painted median and changing lanes is not allowed. There are > > probably a small number of exceptions where this is not true, but > splitting > > the way in those cases is harmless. > > I think you are confusing what I said, assuming you have a way which > is a lane of through traffic that runs in each direction, the stop > sign only applies to one of the 2 lanes, not the entire way. It also applies to one of the 2 directions. There aren't supposed to be people going in the wrong direction in the wrong lane, so this is a degenerate case. > I really don't see how it's less complicated to use the physical rather > than > > the logical. It's actually much more complicated when you get into the > > micro areas and you start adding straight lines through a large > intersection > > instead of curved left turns. > > Because most ways are a simple case of 2 symmetrical lanes, one in > each direction, so simplier because it's half the work to make 2 lanes > by making a single way. And in all the common cases, that's exactly what you do. > Mapping a way becomes more easier than lanes > if there is 3 or 4 or 10 lanes. > And in 99% of cases, if there is no restriction between changing lanes, that's what you do. What would you do in this case: http://maps.google.com/maps?f=q&source=s_q&hl=en&sll=27.965755,-82.546307&sspn=0.000776,0.001194&ie=UTF8&ll=27.966151,-82.546337&spn=0.000776,0.001194&t=h&z=20 We have a six-lane road with all the lanes going in the same direction, separated into groups of three lanes by nothing but paint. There is a traffic light for the left three lanes, and no traffic light for the right three lanes. As you see, whoever provided the map for Google split the way. Would you split the way? Is there a physical separation? Does paint count? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/30 Anthony : > It was you who suggested that a stop sign is applicable to a lane not a > way. I'd say, like Tobias, that it is applicable to a way and a direction. No, you stop at a stop sign which is a point on a way, direction only matters if you can't tag this point on individual lanes. > Where I live we do not have multiple lanes. You park on the side and you > drive around parked cars in the middle. With very few exceptions, stop > signs don't apply to lanes, they apply to ways and directions. In those few > exceptions, it probably makes sense to split the way. In the case of the That would confuse a lot of people that are used to ways being physical. > exceptions I can think of this is especially appropriate because at the spot > where the stop sign applies to one lane and not the others the road is > divided by a painted median and changing lanes is not allowed. There are > probably a small number of exceptions where this is not true, but splitting > the way in those cases is harmless. I think you are confusing what I said, assuming you have a way which is a lane of through traffic that runs in each direction, the stop sign only applies to one of the 2 lanes, not the entire way. The rest of your email seemed to keep making the same flawed assumptions about what I said. > I really don't see how it's less complicated to use the physical rather than > the logical. It's actually much more complicated when you get into the > micro areas and you start adding straight lines through a large intersection > instead of curved left turns. Because most ways are a simple case of 2 symmetrical lanes, one in each direction, so simplier because it's half the work to make 2 lanes by making a single way. Mapping a way becomes more easier than lanes if there is 3 or 4 or 10 lanes. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 8:40 AM, Anthony wrote: > I really don't see how it's less complicated to use the physical rather > than the logical. It's actually much more complicated when you get into the > micro areas and you start adding straight lines through a large intersection > instead of curved left turns. > That was backward. It's more complicated making the curved left turns, of course, compared to just connecting nodes in ways that logically connect. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 7:57 AM, John Smith wrote: > 2009/8/30 Anthony : > > True but the example did not show why that was needed. Looking at > > commercial mapping software, the typical solution is to split the way. > > This whole discussion started because of people wanting to do weird > things to show how a stop sign was applicable to a lane not a way, > there is also cases of other restrictions like maxheight and maxspeed > effecting a lane but not both lanes. It was you who suggested that a stop sign is applicable to a lane not a way. I'd say, like Tobias, that it is applicable to a way and a direction. Where I live we do not have multiple lanes. You park on the side and you drive around parked cars in the middle. With very few exceptions, stop signs don't apply to lanes, they apply to ways and directions. In those few exceptions, it probably makes sense to split the way. In the case of the exceptions I can think of this is especially appropriate because at the spot where the stop sign applies to one lane and not the others the road is divided by a painted median and changing lanes is not allowed. There are probably a small number of exceptions where this is not true, but splitting the way in those cases is harmless. I'm not sure how common maxheight and maxspeed are different when applied to a single way. Maxheight would probably be best accomplished by splitting the way. For maxspeed I'm less sure of what the best solution is, but I'm sure there are many. In most cases, assuming one is free to change lanes, the best solution is probably to simply use the maxspeed of the entire way, regardless of lane. If there are restrictions on changing lanes, the way should be split. If there is a desire to keep the full information, "lanespeed" could be introduced, but I'm not sure what the practical value would be. This ignores the probably more common situation where speeds are different in different directions, though. Technically, if U-turns are not allowed, you probably should split the way, but that might be impractical if the situation is very common. And if U-turns are allowed, on a way with different maxspeeds in different directions, well, I showed a possible solution above. This seems very uncommon. On Sun, Aug 30, 2009 at 7:57 AM, John Smith wrote: > 2009/8/30 Anthony : > > Or traffic cones, or a painted median. Should the ways model the > physical > > or the logical? Again looking at commercial mapping software, the answer > is > > logical. The question they ask is whether or not you're supposed to be > able > > to make turns or change lanes, not what the physical road description is. > > My only experience in this area is with OSM and someone somewhere > obviously decided to do it based on physical, it's certainly less > complicated that way, however we just need to tag some lanes in a > minority of cases for a particular reason, but most of the time all > lanes on the way will be symmetrical and tagging the ways should > cascade down to lanes unless something overrides it. > I really don't see how it's less complicated to use the physical rather than the logical. It's actually much more complicated when you get into the micro areas and you start adding straight lines through a large intersection instead of curved left turns. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/30 Anthony : > True but the example did not show why that was needed. Looking at > commercial mapping software, the typical solution is to split the way. This whole discussion started because of people wanting to do weird things to show how a stop sign was applicable to a lane not a way, there is also cases of other restrictions like maxheight and maxspeed effecting a lane but not both lanes. > Or traffic cones, or a painted median. Should the ways model the physical > or the logical? Again looking at commercial mapping software, the answer is > logical. The question they ask is whether or not you're supposed to be able > to make turns or change lanes, not what the physical road description is. My only experience in this area is with OSM and someone somewhere obviously decided to do it based on physical, it's certainly less complicated that way, however we just need to tag some lanes in a minority of cases for a particular reason, but most of the time all lanes on the way will be symmetrical and tagging the ways should cascade down to lanes unless something overrides it. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
On Sun, Aug 30, 2009 at 12:20 AM, John Smith wrote: > 2009/8/30 Anthony : > > > > > > > > > > > > > > > > > > That solution would fail to allow us to attach nodes to individual lanes. True but the example did not show why that was needed. Looking at commercial mapping software, the typical solution is to split the way. > Dual carriage ways are physically separated, regardless if that's a > strip of grass or by barriers so I don't see that as an issue in the > same context as being able to tag lanes individually. Or traffic cones, or a painted median. Should the ways model the physical or the logical? Again looking at commercial mapping software, the answer is logical. The question they ask is whether or not you're supposed to be able to make turns or change lanes, not what the physical road description is. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/30 Peter Childs : > Another reason this is a bad idea, is that its difficult to access, > search index and process. I think lanes need there own records, like > ways, nodes and relations. There would be some absolute limitations we will have to work around, but while it would make these things easier to do I don't think that should be the real reason we put lanes information into it's own table separate to ways. > We needs some way to map thinks like the pavement having 3 steps in it > where the road is flat, and this is likly to be difficult to push > chairs and wheel chairs and hence they need to cross the road to avoid > the steps. (Currently they are all on the same way and splitting > it just gets more and more difficult to get reasonable accuracy and I'm not sure my suggestions about lanes would fix this, since you are tagging a physical way you would still need to split, maybe most of the meta info could move to a relation rather than being on each individual way segment. This is how major highways are being dealt with, it not only reduces redundency but makes errors easier to spot/fix. > scale (Too much on top of each other). I'm not even sure I like the > way we currently map dual carriageways. Dual carriage ways are physically separated, regardless if that's a strip of grass or by barriers so I don't see that as an issue in the same context as being able to tag lanes individually. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag lanes, not ways, was: Deprecating the use of Tag:highway=stop in favour of Key:stop
2009/8/30 John Smith : > 2009/8/30 Anthony : >> >> >> >> >> >> >> >> > > That solution would fail to allow us to attach nodes to individual lanes. > > This is just another hack to try and accomodate the existing framework > and I think it's a bad idea to try and cram multiple values into a > single keypair. > > We need a proper solution to do this properly, at some point OSM is > going to need to move to more micro mapping, lanes isn't really a > micro mapping issue, it's a subset of a way issue and we should treat > it as such. > Another reason this is a bad idea, is that its difficult to access, search index and process. I think lanes need there own records, like ways, nodes and relations. We needs some way to map thinks like the pavement having 3 steps in it where the road is flat, and this is likly to be difficult to push chairs and wheel chairs and hence they need to cross the road to avoid the steps. (Currently they are all on the same way and splitting it just gets more and more difficult to get reasonable accuracy and scale (Too much on top of each other). I'm not even sure I like the way we currently map dual carriageways. Peter. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Path in JOSM preset and motorcar=yes/designated
Hi, > Hi, I would really appreciate if the JOSM presets dictators could > remove the option about motorcar in the JOSM presets dialog. > > We get on the french ML more and more complains about the confusion > between unclassified, tracks, footway, path, etc. If in addition, the > JOSM presets about highway=path suggests that a path is also for > motorcars, we will never stop ! I agree there is some fuzzyness between unclassified, track, footway, cycleway and path. But this fuzzyness cannot be removed just by changing the JOSM presets. > If a car can use it, it's a track, not a path. I better should resist to abuse my dictatorship for tagging politics ;-) . Anyway, I have removed motorcar from the list of options, as tagwatch does not show significant use (10 times motorcar=no in Europe). If the removal triggers some complaints, I'll reenable it. HTH, ce ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
