Re: [OSM-talk] Better Bing imagery in Potlatch?

[Erik Lundin]

Thanks, this was helpful. My JOSM had max zoom level set to 17. Changing 
it to 18 or 19 gives a significant increase of the resolution. Just a 
small difference between 18/19 in this case.


/Erik

2011-05-28 16:07, Tom Hughes skrev:

On 28/05/11 13:08, Maarten Deen wrote:


Does Potlatch have access to better BING imagery? I'm looking at Walbeck
in Germany
 and
in Potlatch 2 the imagery is a lot better than in JOSM. In the imagery
settings I have bing:bing for Bing Sat.


My guess is you have the max zoom set too low in JOSM. IIRC it defaults
to 18 but bing can go higher than that, depending on the area.

Tom



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

[OSM-talk] Wanted: Web application that connects notes to an OSM slippy map

[Erik Lundin]

Hi,

My family is planning a Europe tour this summer, and the planning is 
performed on a wiki. When brain storming about different destinations it 
would be of great value to be able to add a new suggestion through 
selecting a place on a map and attaching a note. I can think of many 
situations where such an application could be useful. It doesn't 
necessary has to be notes _on_ the map, but just something that links an 
object to a position on the map. Perhaps a list beside the map where the 
position on the map is highlighted when selecting an object in the list, 
and vice versa.

If something like this exists, please let me know. Otherwise this would 
be a killer application for somebody to write.

Regards,
Erik

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

Re: [OSM-talk] Announcemen: Multilingual Country-List

[Erik Lundin]

I think there's a good point of having the redundancy of name:xx tags 
even when it is the same as the name tag, because that makes the 
translated names more safe. For the Swedish translations I saw several 
names that were marked as OK even though they weren't translated. It's 
too easy to read through the names and mark them as OK without thinking 
too much.

/Erik

Peter Körner skrev:
> No, just mark it as ok. If there's an existing one with an identical 
> name you may also delete the needless translation.
> 
> I'll add some highlight to those needless translations, soon.

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

[OSM-talk] DraganFly RC helicopters

[Erik Lundin]

I know RC helicopters as potential aerial photo source has been 
discussed before, but today I read about some helicopter from DraganFly 
and didn't find anything about them on the lists. They seem to be very 
interesting for our purposes.

http://www.draganfly.com/

/Erik

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

Re: [OSM-talk] DraganFly RC helicopters

[Erik Lundin]

I don't really understand what you mean by "open source" here. If you 
buy an RC helicopter, aren't you free to do whatever you want with it?

/Erik

Stefan de Konink skrev:
 > If you want to go Quadcopter I strongly suggest to go for a real open
 > source project. Because you will not realise how much it will annoy you
 > if you have to change something but can't do it because the platform is
 > closed.


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

Re: [OSM-talk] DraganFly RC helicopters

[Erik Lundin]

Ach so... Thank you for the explanation! :)

Wish I had more time to spend on such a project.

/Erik


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

Re: [OSM-talk] Trace type

[Erik Lundin]

Ed Avis skrev:
> I think it would be better to specify the model of GPS device you used, which
> can then be matched against a database of model quirks including such things
> as whether it undulates the geoid correctly.

This approach maybe would be useful to differentiate between tracks made 
with GPS devices and those (in the future) made with devices supporting 
the Galileo system. AFAIK Galileo will have a bit higher precision than GPS.

/Erik

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

Re: [OSM-talk] more accurate open-source gps hardware

[Erik Lundin]

Nice idea. Alas, it seems to be too good to be true, as the first 
comment points out.

Alex S. skrev:
> http://blog.makezine.com/archive/2009/11/diy_real_time_kinematic_gps.html
> 
> 
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk


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

Re: [OSM-talk] openmaps.eu

[Erik Lundin]

Yeah! Maybe we have to increase our efforts to announce that we exist? I 
joined OSM in May 2008, but would have done it earlier if I had found 
out about it before.

/Erik

Andreas Labres skrev:
> They seem to be reinventing the wheel, somehow...

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

Re: [OSM-talk] openmaps.eu

[Erik Lundin]

I just found their links page (http://openmaps.eu/links) where they link 
to OpenStreetMap, so they are at least aware of our project.

/Erik

Erik Lundin skrev:
> Yeah! Maybe we have to increase our efforts to announce that we exist? I 
> joined OSM in May 2008, but would have done it earlier if I had found 
> out about it before.
> 
> /Erik
> 
> Andreas Labres skrev:
>> They seem to be reinventing the wheel, somehow...

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

[OSM-talk] DIY-streetview.org

[Erik Lundin]

Hi,

I just found this very interesting site, with guys working on an open 
source streetview photographing system:

http://www.diy-streetview.org/

Yes, I know of Openstreetview etc., but it's hard to find people 
building the camera rigs.

Cheers,
Erik

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

Re: [OSM-talk] Motion sensors for navigation

[Erik Lundin]

Talking about tunnel tracking...

I have thought of a solution were you have a GPS device together with an 
electronic compass sensor. In the tunnel, the position is calculated by 
knowing the last position, the speed (either from the speedometer in a 
car or keeping the same speed as before the tunnel) and the compass 
direction. Since the error in this case increases exponentially with 
time (right?), it would help to drive through the tunnel in both directions.

/Erik

David Earl skrev:
> On 23/03/2010 12:21, John Smith wrote:
>> On 23 March 2010 21:56, David Earl  wrote:
>>> I wonder whether they might be interested in offering us some units for
>>> testing?
>> Using inertial navigation the accuracy deteriorates considerably over time.
>>
>> http://en.wikipedia.org/wiki/Inertial_navigation
> 
> Yes, obviously. The problems we have are exactly the ones they are 
> trying to solve though - urban canyons and tunnels - where we lose the 
> GPS accuracy for a short while. I don't think they're suggesting they 
> are a substitute for GPS, just to augment it where GPS can't reach. If 
> they are looking at military applications, the accuracy has to be an 
> order of magnitude better than we would need as is their GPS.
> 
> David
> 
> 
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk

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

Re: [OSM-talk] Motion sensors for navigation

[Erik Lundin]

That's correct, but the earth's magnetic field is often the dominating. 
If a classical compass can point out the earth's magnetic field, why 
shouldn't it be possible with an electronic sensor? Quick fluctuations 
could be filtered out.

/Erik

John Smith skrev:
> On 25 March 2010 10:20, Erik Lundin  wrote:
>> I have thought of a solution were you have a GPS device together with an
>> electronic compass sensor. In the tunnel, the position is calculated by
>> knowing the last position, the speed (either from the speedometer in a
>> car or keeping the same speed as before the tunnel) and the compass
>> direction. Since the error in this case increases exponentially with
>> time (right?), it would help to drive through the tunnel in both directions.
> 
> Magnetometer sensors are sensitive to all sorts of magnetic fields,
> not just the magnet field of the earth and so aren't very accurate, to
> prove this point someone made a metal detector for android phones a
> while back and it really does detect metal, or at least the magnetic
> field around metal.

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

Re: [OSM-talk] Sister Projects / possible data source

[Erik Lundin]

I don't know if I have got this discussion wrong, but I have seen at 
least two community projects that are trying to build databases of 
locations of different wireless networks: http://openbmap.org/ and 
http://opencellid.org/


/ Erik

[1] http://opencellid.org/

2011-08-07 23:50, John Smith skrev:

On 8 August 2011 07:46, Richard Weait  wrote:

As I remember it from previous discussions, wifi locations are
somewhat transient for OSM.  Cell tower locations are likely from
government databases are they not?


Google etc estimate location of towers by using data handsets expose.


Given that this would be a 'complete' dataset from another source, why
duplicate (and eventually synchronize) it with OSM, when it should be
possible to display them together in a mash up?


If there is GPS data involved that might be more useful than the tower
estimations etc.

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


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

Re: [OSM-talk] Image of the week points to wrong week

[Erik Lundin]

There was an edit [1] to get the correct year belonging to this week, 
since week number 1 of 2013 started before year 2012 ended. There is a 
syntax for this in MediaWiki, see [2]. However, the trouble seems to be 
that using that syntax, the week number is only available zero-padded, 
while the images have names that aren't zero-padded. I guess the easiest 
solution is to rename "Template:Iotw image/2013-1" to "Template:Iotw 
image/2013-01".


Regards,
Erik

[1] 
http://wiki.openstreetmap.org/w/index.php?title=Template:Image_of_the_week&diff=prev&oldid=848636

[2] http://www.mediawiki.org/wiki/Help:Extension:ParserFunctions#.23time

2013-01-01 19:35, Pierre Béland skrev:

In page http://wiki.openstreetmap.org/wiki/Main_Page, Image of the week
points to Template:Iotw image/2013-2.


The first week image do not appear today. Is this the zero day bug?
Somebody nows how to correct?
Pierre


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



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

Re: [OSM-talk] Image of the week points to wrong week

[Erik Lundin]

Another way to get the week number that's not zero-padded is to bring 
back the {{CURRENTWEEK}}, but then different syntaxes are mixed, which 
is a bit ugly. I did this as a workaround just to get it working, but 
we'll have to figure out the best long term solution.


Regards,
Erik

2013-01-01 19:52, Erik Lundin skrev:

There was an edit [1] to get the correct year belonging to this week,
since week number 1 of 2013 started before year 2012 ended. There is a
syntax for this in MediaWiki, see [2]. However, the trouble seems to be
that using that syntax, the week number is only available zero-padded,
while the images have names that aren't zero-padded. I guess the easiest
solution is to rename "Template:Iotw image/2013-1" to "Template:Iotw
image/2013-01".

Regards,
Erik

[1]
http://wiki.openstreetmap.org/w/index.php?title=Template:Image_of_the_week&diff=prev&oldid=848636

[2] http://www.mediawiki.org/wiki/Help:Extension:ParserFunctions#.23time

2013-01-01 19:35, Pierre Béland skrev:

In page http://wiki.openstreetmap.org/wiki/Main_Page, Image of the week
points to Template:Iotw image/2013-2.
<http://wiki.openstreetmap.org/w/index.php?title=Template:Iotw_image/2013-2&action=edit&redlink=1>


The first week image do not appear today. Is this the zero day bug?
Somebody nows how to correct?
Pierre


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



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


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

Re: [OSM-talk] Question: How to describe groups of islands -> proposal draft

[Erik Lundin]

All established relations 
(http://wiki.openstreetmap.org/index.php/Relations#Established_uses_of_Relations)
 
seem to use k='type' v='something' to indicate what kind of relation it 
is. Later somebody may want to add a relation that describes eg. a group 
of lakes. I'd suggest we follow that template and include .

What's the difference between way and area in this context? Wouldn't 
area be a more natural way to describe an island?

/Erik

Hendrik T Voelker skrev:
> Am 13.09.2008 um 19:17 Uhr schrieb Hendrik T Voelker:
> 
>> just a short question:
>>
>>   How do I describe a group of islands, like e.g. the Ertenholmen (consists 
>> of
>>   about 13 islands, islets, and archipelago in the Baltic Sea.
>>
>> I would expect to use the relation for that
> 
> It seems there is not yet any solution for this. So how about the following:
> 
>   
> 
> 
> 
> 
> 
> 
> 
> 
> ...
>   
> 
> The immediately arising question is, is "group" enough, or would you also
> think that an
> 
>   
> 
> is required? I personally thing that is unnecessary redundancy, but then, who
> knows.  Maybe someone can point out why it is not.
> 
> For choosing 'archipelago' as value, maybe someone has a better alternative. I
> am no native English speaker.
> 
> TIA
> 
> Hendrik
>   

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

[OSM-talk] Upload of relation from JOSM fails

[Erik Lundin]

Hi,

When I try to upload any change to relation 36947 (route E 18) from 
JOSM, the answer gets

upload to: http://www.openstreetmap.org/api/0.5/relation/36947...connected
got return: 412 with id 36947
...

It seems to be the combination of JOSM and this certain relation that 
fails, because I managed to edit the relation from Potlatch, and JOSM 
has no problems with for instance relation 20773 (route E 45).

Has the relation got corrupted somehow, or is it a bug in JOSM?

Regards,
Erik

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

Re: [OSM-talk] Upload of relation from JOSM fails

[Erik Lundin]

Frederik Ramm skrev:
> The reason for this error is that the API will reject (with 
> "precondition failed") any relation which contains a way where one of 
> the nodes in that way is deleted.

> There may be other options, perhaps it is possible to download all 
> relation members in JOSM and check which is flagged "incomplete" 
> afterwards.

I downloaded all members in JOSM and saved to a file, and then wrote a 
script that checked that all nodes referenced from ways existed. 
Surprisingly I didn't find any references to non-existent nodes.

Except from that, I've cleaned the relation's ways from duplicate nodes 
etc., so there's no validation errors from the validator in JOSM. 
However, the problem is still there.

Erik

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

Re: [OSM-talk] Upload of relation from JOSM fails

[Erik Lundin]

Hi,

Frederik Ramm skrev:
> I downloaded the relation and saved it to a file. The file had 666 
> lines. Need I say more ;-)?

To get all members you have to open the relation editor and choose to 
retrieve all incomplete members. Doing this gives me a file with 26288 
lines.

> Using my "binary search" technique mentioned above, I found that the 
> relation upload fails as soon as you include way 4258639. It succeeds if 
> that way is removed from the relation (which I now have done).

Thanks! I have now modified my script to find references to non-existent 
ways as well, and this works for the mentioned way.

> After that I tried to analyse what was wrong with way 4258639. Easy: It 
> has been deleted more than a week ago. I'm surprised you were able to 
> download the relation in JOSM; it should have been flagged "incomplete".

Can't help with this, but I'm using the latest JOSM.

Erik

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

Re: [OSM-talk] Upload of relation from JOSM fails

[Erik Lundin]

Hi,

Frederik Ramm skrev:
> I downloaded the relation and saved it to a file. The file had 666 
> lines. Need I say more ;-)?

To get all members you have to open the relation editor and choose to 
retrieve all incomplete members. Doing this gives me a file with 26288 
lines.

> Using my "binary search" technique mentioned above, I found that the 
> relation upload fails as soon as you include way 4258639. It succeeds if 
> that way is removed from the relation (which I now have done).

Thanks! I have now modified my script to find references to non-existent 
ways as well, and this works for the mentioned way.

> After that I tried to analyse what was wrong with way 4258639. Easy: It 
> has been deleted more than a week ago. I'm surprised you were able to 
> download the relation in JOSM; it should have been flagged "incomplete".

Can't help with this, but I'm using the latest JOSM.

Erik

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

[OSM-talk] Article at Good Gear Guide

[Erik Lundin]

Hi everybody,

I ran into this article at Good Gear Guide: 
http://www.goodgearguide.com.au/article/273270/project_turns_geodata_collection_into_party

Erik

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

Re: [OSM-talk] Can not remove a way from a relation

[Erik Lundin]

Hi,

I had a similar problem earlier this month. The correspondence can be 
found in the mail archives: 
http://lists.openstreetmap.org/pipermail/talk/2009-January/thread.html#33157

The problem is that JOSM gets an error while uploading a relation that 
references a non-existent way. I wrote a little perl script to find 
relations referencing non-existent ways or ways referencing non-existent 
nodes. Using the script I found that the relation (21359) contains a 
reference to way 8135282, which was deleted on January 5th 
(http://www.openstreetmap.org/browse/way/8135282/history). I tried to 
remove the way from the relation, but get the same error while uploading.

Last time it was Frederik Ramm who removed the reference to the 
non-existent way 
(http://lists.openstreetmap.org/pipermail/talk/2009-January/033167.html), 
so maybe he knows how to do.

Erik


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

Re: [OSM-talk] Can not remove a way from a relation

[Erik Lundin]

Shaun McDonald skrev:
There could be another way that has been deleted, or there is a node 
that is reference by a way that is missing.


Does your script look at those?


Yes, it does. I tested to remove a couple of ways and nodes, and the 
script detected all faulty references. The script is attached to this 
mail. It's not fancy (ie. it has no xml parsing), but I think it's 
working. You just have to run it on a file containing the osm data of a 
relation and related ways and nodes. To achieve this, select to download 
incomplete members in the relation editor of JOSM.


Erik



findBadReferences.pl
Description: Perl program
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Can not remove a way from a relation

[Erik Lundin]

Frederik Ramm skrev:
> I made the relation editable again by removing way #29370058 from it. I 
> have not investigated why this way causes trouble; it probably 
> references a deleted object or something like that.

I compared the way downloaded to JOSM and 
http://www.openstreetmap.org/browse/way/29370058. It turned out that in 
JOSM only seven nodes are present compared to 23 on the web site. The 
nodes listed on the web site but not downloaded to JOSM are the problem. 
Browsing the xml for instance for 
http://www.openstreetmap.org/api/0.5/node/385719 shows a totally blank page.

* How are blank nodes removed?
* Is there a way to find such nodes in the database in the case there 
are more of them out there?

Erik


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

Re: [OSM-talk] Can not remove a way from a relation

[Erik Lundin]

Erik Lundin skrev:
> * How are blank nodes removed?
> * Is there a way to find such nodes in the database in the case there 
> are more of them out there?

I correct myself. It's stated on the web site that the nodes are 
deleted, so that's why they are blank. :-|

But how can the nodes be listed as members of way 29370058 when they are 
deleted?

Erik


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

Re: [OSM-talk] Can not remove a way from a relation

[Erik Lundin]

Thanks!

Erik

Richard Fairhurst skrev:
> Erik Lundin wrote:
>> But how can the nodes be listed as members of way 29370058 when 
>> they are deleted?
> 
> http://lists.openstreetmap.org/pipermail/dev/2009-January/013540.html
> and passim
> 
> cheers
> Richard

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

Re: [OSM-talk] Wiki: chriscf vandalism

[Erik Lundin]

Richard Fairhurst skrev:
> The important thing is that there is no prescription. No rejected. No
> approved. Just easy-to-use documentation of what people are using, why. If
> you feel a need for a particular tag, start using it, and document it. If
> the tag is good, it'll catch on. It's much more akin to OSM-style
> crowdsourcing than the rather Wikipedia-esque procedures we have at the
> moment.
> 
> Then, the people who maintain Map Features can pull out the most popular
> tags and descriptions from here; same goes for Potlatch presets, JOSM
> presets, and any other lists of tags.

It's an interesting idea, and I like when people come up with 
constructive suggestions. Just one thought: what would prevent an edit 
war on the data when people interpret the tags differently? I think that 
there still is a need for some kind of consensus.

Erik

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