[Talk-us] border screwup by ToeBee needs reverting

[Nathan Edgars II]

User ToeBee has, in several changesets in February, aligned state 
borders to exact lat/long. The problem is that this is not how the 
borders are defined; instead they are based on work that the 19th 
century surveyors did with the tools they had. Two obvious examples follow:
http://www.openstreetmap.org/browse/node/158796015/history is the famous 
"Four Corners": 
http://www.deseretnews.com/article/705298412/Four-Corners-marker-212-miles-off-Too-late.html
http://www.openstreetmap.org/browse/node/158790476/history is the 
northwest corner of Colorado, which is also marked by a large monument, 
visible on aerials.


It's likely that any border node moved by ToeBee needs to be reverted. I 
tried to do this after informing him (he's currently denying there's a 
problem), but JOSM's reverter plugin is giving hundreds of conflicts. 
This damage needs to be undone.


___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Nathan Edgars II]

On 3/25/2011 3:56 AM, Nathan Edgars II wrote:

http://www.deseretnews.com/article/705298412/Four-Corners-marker-212-miles-off-Too-late.html


Note the correction to this article: 
http://www.deseretnews.com/article/705299160/Four-Corners-Monument-is-indeed-off-mark.html

I was a little hasty about linking the original.

___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Toby Murray]

On Fri, Mar 25, 2011 at 2:56 AM, Nathan Edgars II  wrote:
> User ToeBee has, in several changesets in February, aligned state borders to
> exact lat/long. The problem is that this is not how the borders are defined;
> instead they are based on work that the 19th century surveyors did with the
> tools they had. Two obvious examples follow:
> http://www.openstreetmap.org/browse/node/158796015/history is the famous
> "Four Corners":
> http://www.deseretnews.com/article/705298412/Four-Corners-marker-212-miles-off-Too-late.html
> http://www.openstreetmap.org/browse/node/158790476/history is the northwest
> corner of Colorado, which is also marked by a large monument, visible on
> aerials.
>
> It's likely that any border node moved by ToeBee needs to be reverted. I
> tried to do this after informing him (he's currently denying there's a
> problem), but JOSM's reverter plugin is giving hundreds of conflicts. This
> damage needs to be undone.

I am not saying my edit was perfect. But the position of the nodes
before my edit was even further off than it is now. The northwest node
is now 10 meters straight south of the monument where it was over 1km
east. The southwest node was off by 1km as well and is now about 400
meters west. At the end of the day, the only reason I felt I could
even make these changes is because the source for the borders in the
US isn't that great to begin with. They are all over-noded. County
borders don't align with state borders. Most of the county borders I
have worked with are visibly off by up to a couple hundred feet.
Apparently in this case I didn't look for the monuments specifically
though. My bad.

I made the map better. I did not make it perfect. And while I did
(apparently mistakenly) use lat/lon values in my work, I did take
other things into consideration. It was not a blind change.

Reverting over 850 objects because I'm off by a few meters is beyond
extreme. I might add that trying to revert such a large changeset
without even waiting for a reply is also somewhat rude. Good thing I'm
a night owl I guess.

Toby

___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Paul Norman]

The first of your examples ('015 node) appears to be more accurate than the
node it replaced in one of the ways,
http://www.openstreetmap.org/browse/node/263660932 which was farther away
from the monument (based on NAIP imagery)

In the second one ('476 node), the changeset improved the position of the
node. It wasn't aligned before. In any case, after the changeset it was only
about 8m off according to NAIP imagery.

I'd say that reverting the border fix changeset would be wrong, given the
number of problems it fixed with the borders. I'd say it was definitely
wrong to attempt to revert it over ToeBee's objections.


> -Original Message-
> From: Nathan Edgars II [mailto:nerou...@gmail.com]
> Sent: Friday, March 25, 2011 12:56 AM
> To: OpenStreetMap talk-us list
> Subject: [Talk-us] border screwup by ToeBee needs reverting
> 
> User ToeBee has, in several changesets in February, aligned state
> borders to exact lat/long. The problem is that this is not how the
> borders are defined; instead they are based on work that the 19th
> century surveyors did with the tools they had. Two obvious examples
> follow:
> http://www.openstreetmap.org/browse/node/158796015/history is the famous
> "Four Corners":
> http://www.deseretnews.com/article/705298412/Four-Corners-marker-212-
> miles-off-Too-late.html
> http://www.openstreetmap.org/browse/node/158790476/history is the
> northwest corner of Colorado, which is also marked by a large monument,
> visible on aerials.
> 
> It's likely that any border node moved by ToeBee needs to be reverted. I
> tried to do this after informing him (he's currently denying there's a
> problem), but JOSM's reverter plugin is giving hundreds of conflicts.
> This damage needs to be undone.
> 
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us


___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Ian Dees]

On Fri, Mar 25, 2011 at 3:44 AM, Paul Norman  wrote:

> ...
>
> I'd say that reverting the border fix changeset would be wrong, given the
> number of problems it fixed with the borders. I'd say it was definitely
> wrong to attempt to revert it over ToeBee's objections.
>
>
+1

I would say that a better use of our time would be in creating boundary
relations to fix the duplicated county/state boundaries.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Nathan Edgars II]

On 3/25/2011 7:49 AM, Ian Dees wrote:

I would say that a better use of our time would be in creating boundary
relations to fix the duplicated county/state boundaries.

I would say it's more important to have the border in the right place 
(at least such that all roads in one state are on the correct side).


___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Ian Dees]

On Fri, Mar 25, 2011 at 7:04 AM, Nathan Edgars II wrote:

> On 3/25/2011 7:49 AM, Ian Dees wrote:
>
>> I would say that a better use of our time would be in creating boundary
>> relations to fix the duplicated county/state boundaries.
>>
>>  I would say it's more important to have the border in the right place (at
> least such that all roads in one state are on the correct side).
>

I would disagree. No one is going to use OpenStreetMap to solve border
disputes in the US. There are higher quality datasets that come from the
people that make the rules. On the other hand, removing overlapping boundary
ways will clean up the existing OSM data, make it easier to edit and easier
to consume.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Nathan Edgars II]

On 3/25/2011 8:37 AM, Ian Dees wrote:

On Fri, Mar 25, 2011 at 7:04 AM, Nathan Edgars II mailto:nerou...@gmail.com>> wrote:

On 3/25/2011 7:49 AM, Ian Dees wrote:

I would say that a better use of our time would be in creating
boundary
relations to fix the duplicated county/state boundaries.

I would say it's more important to have the border in the right
place (at least such that all roads in one state are on the correct
side).


I would disagree. No one is going to use OpenStreetMap to solve border
disputes in the US. There are higher quality datasets that come from the
people that make the rules. On the other hand, removing overlapping
boundary ways will clean up the existing OSM data, make it easier to
edit and easier to consume.


I'm not sure what to make of this response. Why would you say we have 
borders in OSM? Would that reason be better suited by having the borders 
be in the correct place but duplicated or by having them in the wrong 
place but consolidated?


___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Ian Dees]

On Fri, Mar 25, 2011 at 7:49 AM, Nathan Edgars II wrote:

> On 3/25/2011 8:37 AM, Ian Dees wrote:
>
>> On Fri, Mar 25, 2011 at 7:04 AM, Nathan Edgars II > > wrote:
>>
>>On 3/25/2011 7:49 AM, Ian Dees wrote:
>>
>>I would say that a better use of our time would be in creating
>>boundary
>>relations to fix the duplicated county/state boundaries.
>>
>>I would say it's more important to have the border in the right
>>place (at least such that all roads in one state are on the correct
>>side).
>>
>>
>> I would disagree. No one is going to use OpenStreetMap to solve border
>> disputes in the US. There are higher quality datasets that come from the
>> people that make the rules. On the other hand, removing overlapping
>> boundary ways will clean up the existing OSM data, make it easier to
>> edit and easier to consume.
>>
>
> I'm not sure what to make of this response. Why would you say we have
> borders in OSM? Would that reason be better suited by having the borders be
> in the correct place but duplicated or by having them in the wrong place but
> consolidated?
>

I don't know why we have borders in OSM. I'm assuming they're in there
because people saw the data available and imported it (like I did with the
county borders a few years ago -- one of several imports that I wish I could
take back). It could also be because it makes for a decent thing to draw on
the map at low zoom.

I personally don't think borders that are controlled by others belong in OSM
but if others insist that the borders are there then I think they should at
least be represented with clean OSM data.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Phil! Gold]

* Nathan Edgars II  [2011-03-25 08:04 -0400]:
> I would say it's more important to have the border in the right
> place (at least such that all roads in one state are on the correct
> side).

So either fix it or *politely* ask the person who made the change to fix
the alignment.  (Trying to revert a changeset made by a committed,
experienced mapper is *not* polite.)  It looks to me like ToeBee joined
together the state and county borders, which, if Colorado was anything
like Maryland, probably differed from each other by more than the
positioning error we're currently discussing.  To me, that seems like an
improvement, and if he made some alignment errors, that's worth an
informative comment ("Hey, the county boundaries are defined by surveyed
points that only approximated an exact longitude alignment,") not an
imperious demand that it be fixed.

I made almost the same mistake with Maryland's northern boundary.  After
reading more about the boundary, I realized my mistake and realigned it to
USGS data.  I'm glad no one berated me for doing it wrong before I
corrected my own mistake.

-- 
...computer contrarian of the first order... / http://aperiodic.net/phil/
PGP: 026A27F2  print: D200 5BDB FC4B B24A 9248  9F7A 4322 2D22 026A 27F2
--- --
Every advance in civilization has been denounced while it was still
recent.
   -- Bertrand Russell
 --- --

___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] border screwup by ToeBee needs reverting

[Russ Nelson]

Ian Dees writes:
 > I personally don't think borders that are controlled by others belong in OSM
 > but if others insist that the borders are there then I think they should at
 > least be represented with clean OSM data.

Yet Another reason to have http://www.closedstreetmap.org -- renderers
need this kind of data, and it's absurd to expect them to go
collecting it from all the various and varied sources. Import it into
a CSM from which it can be deleted whenever a new dataset is available
from the original canonical source.

For the same reason I'd pull out the NYS DEC Lands, NYC Bike stands,
and refrain from importing the NYS Parks boundaries. They need to be
on the map, they need tags in OSM's format, but they don't need to be
in OSM.

-- 
--my blog is athttp://blog.russnelson.com
Crynwr supports open source software
521 Pleasant Valley Rd. | +1 315-600-8815
Potsdam, NY 13676-3213  | Sheepdog   

___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us