Re: [Tagging] Feature Proposal - RFC - area:highway
On 5/12/2011 9:45 PM, M∡rtin Koppenhoefer wrote: 2011/5/12 Nathan Edgars IInerou...@gmail.com: On 5/12/2011 2:31 PM, M∡rtin Koppenhoefer wrote: 2011/5/12 flylowfligh...@googlemail.com: What do we do with dual-carriage ways ? Sometimes there exist paved connections between both directions. Maybe blocked by a barrier but that is no need. if they are constantly connected (no change of the paving, no physical barrier) it's actually not a dual-carriage way. If these connections are punctually you'd simply draw them explicitly and tag them as what they are (incl. turn restrictions. etc.) Well, one could have a single area of pavement with barriers placed on top to separate it into two carriageways. Hi Nathan Edgars II, that's why I wrote above: no physical barrier OK, but there's still the issue of a so-called flush median. I think in a rural area with few intersections this would be called a dual carriageway. I can't find an image, but Interstate 90 used to have one over Lookout Pass in Idaho. You can imagine http://www.youtube.com/watch?v=cPaP3K9xp3g with the concrete barrier removed. ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - area:highway
2011/5/13 Nathan Edgars II nerou...@gmail.com: OK, but there's still the issue of a so-called flush median. I think in a rural area with few intersections this would be called a dual carriageway. I can't find an image, but Interstate 90 used to have one over Lookout Pass in Idaho. You can imagine http://www.youtube.com/watch?v=cPaP3K9xp3g with the concrete barrier removed. We should make a difference between physically impossible and physically possible but legally forbidden. This is important for a series of situations, e.g. an emergency car in action could cross a road marking without problems (paying attention to surrounding traffic), while if would still be physically impossible to cross a concrete or steel barrier. Ideally our data would allow to extract this information. cheers, Martin ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - area:highway
On 5/13/2011 6:47 AM, M∡rtin Koppenhoefer wrote: We should make a difference between physically impossible and physically possible but legally forbidden. This is important for a series of situations, e.g. an emergency car in action could cross a road marking without problems (paying attention to surrounding traffic), while if would still be physically impossible to cross a concrete or steel barrier. Ideally our data would allow to extract this information. An emergency vehicle could also cross a grass median if there's no raised barrier. ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - area:highway
2011/5/13 Nathan Edgars II nerou...@gmail.com: On 5/13/2011 6:47 AM, M∡rtin Koppenhoefer wrote: An emergency vehicle could also cross a grass median if there's no raised barrier. yes, and a person can jump over a 2ft wall and climb a 8ft wall. A series of bollards is no barrier to bicycles and pedestrians, but it is to cars. You can also open a wire fence if you have pincers. You won't be able to cut metal bars with pincers though. Personally I think that it would be interesting to have these details in the map. For dual carriageways and other parallel / close by streets there is also a proposal how this data could be entered (relation area). cheers, Martin ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
[Tagging] Tagging HOV lanes that have strict times (and are completely separate from main highway)
I was just about to create some new relations for the I-279/I-579 HOV lanes to reflect the times that they are open and closed. http://www.dot.state.pa.us/penndot/districts/district11.nsf/traffic_HOV_hours?OpenForm That shows the current times when they are open and when Outbound is restricted to HOV and when it's unrestricted (to help get people out of town from PNC Park, Heinz Field, and Consol Energy Center after events). NOTE: The outbound 3:00 PM start time is incorrect on that page, it's really 4:00 PM. That's what all signage posted in the field shows and matches the end of the closed time. (I should really e-mail PennDOT about that or twitter them) As of right now, I can figure out that I would at least need 10 relations to do this correctly (5 for I-279's segment and 5 for I-579's segment). (1 Inbound; 1 closed time; 1 Outbound HOV2; 1 Outbound unrestricted; 1 weekend)x2 Also might need a few extras for the ramps too (that aren't at the end points). Also, I was planning on using outbound and inbound for the roles since that what PennDOT and everybody else around here mention for the directions (even though I-279/I-579 are both North/South Interstates). Do you think the routers (like MapQuest) will have any problem with that info or should I go with north/south? So, has anybody else dealt with HOV lanes yet? I did notice this proposal (http://wiki.openstreetmap.org/wiki/Proposed_features/hov_access), but it was rejected back in 2008. I was also thinking that hov=2+ (or should it just be hov=2) inside of the relation would work to describe the minimal persons needed to be in the car. So, any suggestions or insights would be appreciated. Thanks. ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Tags for neighborhoods / subdivisions
* Josh Doe j...@joshdoe.com [2011-05-10 23:27 -0400]: Either way I think we need to allow for admin_level or something similar to permit nesting of neighborhoods. I know, let's use relations! (Now I have two problems...) But seriously, what about a very simple contains relation? A given relation would have one member (node or way) with role parent and one or more members (nodes or ways) with role child. This would solve the hierarchy problem without assigning arbitrary numbers. A given entity could, of course, be a child in more than one relation, if a neighborhood spanned more than one containing place. I do think that this sort of thing should only be used if spatial relationships don't work: i.e. one or both related entities are nodes. -- ...computer contrarian of the first order... / http://aperiodic.net/phil/ PGP: 026A27F2 print: D200 5BDB FC4B B24A 9248 9F7A 4322 2D22 026A 27F2 --- -- Lensmen eat Jedi for breakfast. --- -- ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
