Re: [OSM-talk] Changes in the rendering

[Mateusz Konieczny]

I opened a ticket:
https://github.com/gravitystorm/openstreetmap-carto/issues/934

2014-09-09 23:35 GMT+02:00 Markus Lindholm :

> Thanks for fixing it, but there's one problem remaining. There's a
> quite visible gap in rendering when moving from a highway+railway to a
> highway+railway+bridge. E.g.
>
> http://www.openstreetmap.org/?mlat=59.33140&mlon=18.09368#map=19/59.33140/18.09368
>
> /Markus
>
> On 10 August 2014 14:33, Matthijs Melissen 
> wrote:
> > On 10 August 2014 10:09, Markus Lindholm 
> wrote:
> >> - highway=platform is no longer rendered
> >> - railway=tram is no longer rendered if it shares a way with highway=*
> >>
> >> Are these bugs or a deliberate change?
> >
> > These are bugs, thanks for reporting. I created issues on Github for
> them:
> > - https://github.com/gravitystorm/openstreetmap-carto/issues/871
> > - https://github.com/gravitystorm/openstreetmap-carto/issues/874
> >
> > In general, if you want a quick response, it is easier to create the
> > issues directly on Github as not all stylesheet maintainers follow the
> > talk mailing list.
> >
> > -- Matthijs
> >
> > ___
> > talk mailing list
> > talk@openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/talk
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] 1.61 Gigapixel density map of the POI in Europe

[Jóhannes Birgir Jensson]

Looks very interesting for *most* of Europe.

Signed,
Iceland

It is very nice though, showing the density in Central Europe, both due to 
population and very active OSM communities.



 Upphafleg skilaboð Frá: Badita Florin 
 Dagsetning:10/09/2014  02:48  (GMT+00:00) 
Til: talk@openstreetmap.org Efni: [OSM-talk] 1.61 
Gigapixel density map of the POI in Europe 
If somebody find this interesting, i made a visualization of the POI in 
Europe

https://www.openstreetmap.org/user/baditaflorin/diary/23729

have a good day,
Florin Badita
https://www.facebook.com/OsmRomania
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

[OSM-talk] 1.61 Gigapixel density map of the POI in Europe

[Badita Florin]

If somebody find this interesting, i made a visualization of the POI in
Europe

https://www.openstreetmap.org/user/baditaflorin/diary/23729


have a good day,
Florin Badita
https://www.facebook.com/OsmRomania
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Changes in the rendering

[Markus Lindholm]

Thanks for fixing it, but there's one problem remaining. There's a
quite visible gap in rendering when moving from a highway+railway to a
highway+railway+bridge. E.g.
http://www.openstreetmap.org/?mlat=59.33140&mlon=18.09368#map=19/59.33140/18.09368

/Markus

On 10 August 2014 14:33, Matthijs Melissen  wrote:
> On 10 August 2014 10:09, Markus Lindholm  wrote:
>> - highway=platform is no longer rendered
>> - railway=tram is no longer rendered if it shares a way with highway=*
>>
>> Are these bugs or a deliberate change?
>
> These are bugs, thanks for reporting. I created issues on Github for them:
> - https://github.com/gravitystorm/openstreetmap-carto/issues/871
> - https://github.com/gravitystorm/openstreetmap-carto/issues/874
>
> In general, if you want a quick response, it is easier to create the
> issues directly on Github as not all stylesheet maintainers follow the
> talk mailing list.
>
> -- Matthijs
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk

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

Re: [OSM-talk] tag with value lists Was: Proposed mechanical edit to convert alt_name tags

[Andreas Goss]

Isnt the semicolon the list seperator typically used in OSM? My
intuitive answer would have been alt_name=a;b;c;d


Isn't this sometimes going to cause issues when it comes to the number 
of characters?

__
openstreetmap.org/user/AndiG88
wiki.openstreetmap.org/wiki/User:AndiG88‎


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

Re: [OSM-talk] Visually detect missing roads

[SomeoneElse]

On 09/09/2014 08:11, Stephan Knauss wrote:


I have created a map which visually diffs our data against Google Maps.


I can see how Google could find this useful locally to me - it would 
enable them to remove some of the roads on their map that don't exist, 
like the one to a coal mine that closed in 1957 :-)


Cheers,

Andy


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

Re: [OSM-talk] Visually detect missing roads

[Mateusz Konieczny]

It would be useful to allow switching between diff, OSM only and google
only. Currently in my area results are too confusing to be useful.

2014-09-09 9:11 GMT+02:00 Stephan Knauss :

> I did announce this on the German list last week. As the load did not
> cause the server to catch fire I'm now announcing it to a wider audience.
>
> I have created a map which visually diffs our data against Google Maps.
> Currently it compares major highways (unclassified and higher) and water
> features.
>
> The data is styled to show up in bright colors. If there is "matching"
> data in OSM it would hide the Google data.
>
> So in a "perfect" area the map would be grey. Differences stay visible.
>
> You can try it here: http://compare.osm-tools.org/
>
> It has the possibility to directly load the visible area into the editors.
>
> More details can be found here: http://www.osm-tools.org/compare.html
>
> or if you're able to read German in my blog post at
> http://www.technologyblog.de/2014/08/wo-fehlen-bei-
> openstreetmap-noch-daten/
>
> In well-mapped areas the differences are usually caused by OSM data not
> being tagged as a major highway. If you're looking for areas where roads
> are actually missing and can be drawn from aerials head over to Asia.
>
> Hope this helps all people interested in arm-chair mapping to focus on the
> major missing parts of OSM-data.
>
> If you pan the map fast you'll see the original Google data. This is
> caused by the way technical way the images are layered. When using Google
> API instead of OpenLayers it would not do this but caching of tiles is
> worse with Google. That's why I decided to use OL.
>
> Enjoy mapping!
>
> Stephan
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] tag with value lists Was: Proposed mechanical edit to convert alt_name tags

[Jochen Topf]

On Mon, Sep 08, 2014 at 05:52:57PM -0500, Andrew Buck wrote:
> On 09/08/2014 05:44 PM, Eugene Alvin Villar wrote:
> > On Tue, Sep 9, 2014 at 3:00 AM, Florian Lohoff  wrote:
> > 
> >> Isnt the semicolon the list seperator typically used in OSM? My 
> >> intuitive answer would have been alt_name=a;b;c;d
> >> 
> > 
> > +1
> > 
> > I think using a semicolon-delimited list is better than a
> > potentially open-ended set of keys such as "alt_name_x", and
> > already has precedent. While it is true that semicolon as a
> > multi-value separator is not written as a "law" of OSM tagging, it
> > is quite frequent enough to be a de facto tagging guideline.
> > 
> 
> Both systems are in use and were in use before we started working on
> this GNS stuff.  The tiger tags are one example, but there were
> alt_name_N tags before we started as well.  If you want to have a
> discussion about changing all of these worldwide, that is fine, but
> this thread is about fixing the alt_name:2 ones.
> 
> Please either weigh in on that, or keep silent.  I do not have time to
> have a general discussion about broader topics of how multi-valued
> keys should be handled.  There have been dozens of threads that
> discussed that and no one has ever come up with a good solution, so I
> am not going to have this thread get dragged into the same discussion
> that has been had many times before.
> 
> There has been one person who has said it makes sense to change them,
> and all of the other posts have been about discussion of other topics.
>  So unless someone has some concrete reason _not_ to convert the few
> mistaken ones, I will go ahead and do that.  We can have these other
> discussions some other time when people are not waiting on the results.

Sorry, but this is not how OSM works. You have to allow discussion and
you have to listen what people have to say and engage with them in the
discussion. If you can not do that you can not do any automated edits.

I understand your frustration with the discussion culture in OSM that
often seems to go in endless rounds and leads nowhere, but for better
or worse, this is what we have and what has built OSM as we know it. So
somehow it can't be all bad. There are a lot of people on this mailing
list with long experience in OSM and the software using its data. Chances
are there are a few who know something you don't. It might appear tedious,
but it actually leads somewhere. And if you want to help the discussion
along, you can read all the answers and suggestions and pull them together
into one coherent pro/contro-type document with the different proposals.

Your argument that others are discussing a different topic is bogus. If
you want to change alt_name:2 into alt_name_2 and people don't agree that
this is the right tag, how can that be a different discussion?

Btw I think you have a good chance that this discussion will lead somewhere,
because you seem to be actually using those tags. Most endless discussions
revolve around lofty issues where nobody actually uses any of the variants
in question. So make your case, but keep an open mind, collect information
about the solutions impartially and you will get somewhere. If you think
you don't have the time, maybe you should think about the time of all the
others involved here, too.

Jochen
-- 
Jochen Topf  joc...@remote.org  http://www.jochentopf.com/  +49-721-388298

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

[OSM-talk] Visually detect missing roads

[Stephan Knauss]

I did announce this on the German list last week. As the load did not 
cause the server to catch fire I'm now announcing it to a wider audience.


I have created a map which visually diffs our data against Google Maps.
Currently it compares major highways (unclassified and higher) and water 
features.


The data is styled to show up in bright colors. If there is "matching" 
data in OSM it would hide the Google data.


So in a "perfect" area the map would be grey. Differences stay visible.

You can try it here: http://compare.osm-tools.org/

It has the possibility to directly load the visible area into the editors.

More details can be found here: http://www.osm-tools.org/compare.html

or if you're able to read German in my blog post at 
http://www.technologyblog.de/2014/08/wo-fehlen-bei-openstreetmap-noch-daten/


In well-mapped areas the differences are usually caused by OSM data not 
being tagged as a major highway. If you're looking for areas where roads 
are actually missing and can be drawn from aerials head over to Asia.


Hope this helps all people interested in arm-chair mapping to focus on 
the major missing parts of OSM-data.


If you pan the map fast you'll see the original Google data. This is 
caused by the way technical way the images are layered. When using 
Google API instead of OpenLayers it would not do this but caching of 
tiles is worse with Google. That's why I decided to use OL.


Enjoy mapping!

Stephan


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