Re: [Talk-us] National Park boundaries

2012-07-28 Thread Daniel Sabo
If you're OK with using non-rendering tags you should use the established boundary=protected_area instead of something US specific ( https://wiki.openstreetmap.org/wiki/Tag:boundary%3Dprotected_area#Protect_classes_for_various_countries ). If you want it to render the convention I've been using for

Re: [Talk-us] Proposal: delete census-designated place polygons

2010-11-11 Thread Daniel Sabo
I would oppose deleting them. They do have real world significance because they represent community boundaries in unincorporated areas, and the name that you would use to search for an address these communities. McKinleyville, CA (http://en.wikipedia.org/wiki/McKinleyville,_California) is as muc

Re: [Talk-us] Proposal: delete census-designated place polygons

2010-11-11 Thread Daniel Sabo
e are also cases where the tags are out of sync between point and polygon. On Nov 11, 2010, at 9:21 PM, Nathan Edgars II wrote: > On Thu, Nov 11, 2010 at 11:23 PM, Daniel Sabo wrote: >> I would oppose deleting them. They do have real world significance because >> they represen

Re: [Talk-us] Proposal: delete census-designated place polygons

2010-11-12 Thread Daniel Sabo
>> but the polygon is still a much better approximation that you would get with >> just a node. While a node be able to tell you "unincorporated the stuff >> north of Arcata is McKinleyville" it wouldn't convey that "west of 101 and >> south of the river is not McKinleyville". > Problem is if Ar

Re: [Talk-us] Proposal: delete census-designated place polygons

2010-11-12 Thread Daniel Sabo
Resending to the list. On Nov 12, 2010, at 12:37 AM, Nathan Edgars II wrote: > On Fri, Nov 12, 2010 at 3:14 AM, Daniel Sabo wrote: >>> If it's not in GNIS, for example, that means it's not on USGS or other >>> government maps, which narrows down the field consider

[Talk-us] boundary = national_park in the US

2011-01-07 Thread Daniel Sabo
boundary=national_park gets stretched around quite a bit mapping the US, it's being used for: National Parks National Forests National Monuments National Preserves National Recreation Areas State Parks State Forests (not sure if any of these have actually been imported). I've seen admin_level use

Re: [Talk-us] boundary = national_park in the US

2011-01-09 Thread Daniel Sabo
The problem I see with using protected_area right now (since it's not an accepted/rendered tag), is that an object can't have two values for they key "boundary". National forest objects can be huge mulitpolygons (e.g. http://www.openstreetmap.org/browse/relation/1273907 ), having a boundary=pro

Re: [Talk-us] boundary = national_park in the US

2011-01-09 Thread Daniel Sabo
Thanks for your input to help sort this out. So after more reading, I guess protected_area will work for these. When I first looked at it protectedplanet.net was confusing me because it doesn't show national forests, but at least according to wikipedia National Forests qualify as Category VI.

Re: [Talk-us] Relations, cycle routes, shapefiles

2011-02-04 Thread Daniel Sabo
What are you going to do when the route is part of more than one state highway or bike route? You can't do a db query for ref:highway:ca:0, ref:highway:ca:1, ref:highway:ca:n without doing expensive string comparisons, and you can't explode a delimited list of refs without breaking the one key =

Re: [Talk-us] Relations, cycle routes, shapefiles

2011-02-05 Thread Daniel Sabo
On Feb 5, 2011, at 7:14 AM, Craig Hinners wrote: > Sure, relations get you an additional degree of normalization. And using > relations to carry route/network tags gets the job done, granted. But at what > cost? > > I've yet to hear a convincing argument that justifies the additional > comple

Re: [Talk-us] Relations, cycle routes, shapefiles

2011-02-05 Thread Daniel Sabo
Using semicolons brings us back to "impossible to query without string manipulation". I agree with you that multiple values per key would have been a better design for many things, it still wouldn't solve the fact that there may be a set of keys (e.g. names) associated with each ref rather than