Re: [Tagging] unfinished railway of historic importance

2012-03-06 Thread Richard Welty
On 3/6/12 9:22 PM, Steve Bennett wrote: On Mon, Mar 5, 2012 at 10:54 PM, Frederik Ramm wrote: I just found the idea of saying "this is a railway - a never-built one, but a railway nonetheless" a little extravagant. umm, not never built. never completed. and in this case, a never completed rail

Re: [Tagging] unfinished railway of historic importance

2012-03-06 Thread Bill Ricker
On Tue, Mar 6, 2012 at 9:22 PM, Steve Bennett wrote: > I wouldn't agree with > planned-but-abandoned features being stored except in unusual > circumstances. > Key distinction is planned-but-never-built (county plat book fantasy "roads"), vs built, used, then abandoned (subca

Re: [Tagging] unfinished railway of historic importance

2012-03-06 Thread Steve Bennett
On Mon, Mar 5, 2012 at 10:54 PM, Frederik Ramm wrote: > Nah, it was rather about what the priority is. A constrution site could > always be annotated with "this is planned to become a hotel", even though it > isn't a hotel; and a cut could always be annotated with "this was once > planned to becom

Re: [Tagging] Feature Proposal - Voting - lanes General Extension

2012-03-06 Thread Tobias Knerr
Martin Vonwald wrote: ~ 11.500 cycleway:left/right ~ 10.000 footway=left/right, 22.000 if you count "both" (same proposal) ~ 4.500 footway:left/right/both:* As far as I understand, those are ways next to the carriageway. If they are mapped as tags on the highway=* way, rather than as separ

Re: [Tagging] This needs to be nipped in the bud

2012-03-06 Thread Morten Kjeldgaard
On 06/03/2012, at 10.43, Pieren wrote: On Tue, Mar 6, 2012 at 10:03 AM, Ross Scanlon wrote: Definitely not how to map an intersection. AU list have had several discussions on this and it's junk mapping. I still believe that mapping each lane is easier than using verbose and encrypted tags

Re: [Tagging] This needs to be nipped in the bud

2012-03-06 Thread Paul Johnson
On Tue, Mar 6, 2012 at 1:43 AM, Pieren wrote: > On Tue, Mar 6, 2012 at 10:03 AM, Ross Scanlon wrote: > > Definitely not how to map an intersection. AU list have had several > > discussions on this and it's junk mapping. > > I still believe that mapping each lane is easier than using verbose > a

Re: [Tagging] Feature Proposal - Voting - lanes General Extension

2012-03-06 Thread Martin Vonwald
> The much more relevant precedent are existing attempts to tag lanes. One > example is indeed lanes:forward/backward. But there are other examples for > existing lane tagging which are also documented on the wiki, and used more > frequently than your example according to taginfo: > > ~ 11.500 cycl

Re: [Tagging] Feature Proposal - Voting - lanes General Extension

2012-03-06 Thread Tobias Knerr
Martin Vonwald wrote: 1) The objective part: How is it done currently? Take a look at my first example in the proposal - it's using maxspeed. How is maxspeed currently tagged? According to the wiki maxspeed:forward and maxspeed:backward should be used. What tells use taginfo? The forward/backward

Re: [Tagging] This needs to be nipped in the bud

2012-03-06 Thread Pieren
On Tue, Mar 6, 2012 at 10:03 AM, Ross Scanlon wrote: > Definitely not how to map an intersection.  AU list have had several > discussions on this and it's junk mapping. I still believe that mapping each lane is easier than using verbose and encrypted tags (probably that need some special tools on

Re: [Tagging] This needs to be nipped in the bud

2012-03-06 Thread Ross Scanlon
On 06/03/12 03:32, Frederik Ramm wrote: Hi, On 03/05/2012 08:12 PM, Nathan Edgars II wrote: http://wiki.openstreetmap.org/wiki/Lanes_and_complex_intersections_visual_approach User Cmuelle8 insists on adding it to http://wiki.openstreetmap.org/wiki/Lane_tagging_comparison#A_visual_approach as

Re: [Tagging] Feature Proposal - Voting - lanes General Extension

2012-03-06 Thread Martin Vonwald
> However, routers already need to handle country-specific traffic rules > anyway (for things like default maxspeeds, implied access rights, and so > on). This is not usually the case for renderers. > > Therefore, wouldn't it make more sense to pick renderers as the application > category that can