Re: [OSM-talk] The movie Eye in the Sky credits OpenStreetMap

[Pine W]

Hi Simon,

Not quite. Per "https://www.openstreetmap.org/copyright;, licensees "must
also make it clear that the data is available under the Open Database
License, and if using our map tiles, that the cartography is licensed as CC
BY-SA. You may do this by linking to this copyright page
."

So, while noting “© OpenStreetMap contributors”, is a good start, it's
insufficient to fulfill the terms of the license.

I would expect that a project with a cost of about $75 million would have
plenty of resources with which to do legal research and completely fulfill
the terms of the license.

Pine

On Sat, Oct 22, 2016 at 1:08 AM, Simon Poole  wrote:

> “© OpenStreetMap contributors” has been the suggested attribtion text
> since day one post licence change (see http://www.openstreetmap.org/
> copyright and http://wiki.osmfoundation.org/wiki/Licence/Licence_and_
> Legal_FAQ ). While you could argue that the OSMF is wrong headed to ask
> for that, you defintely can't complain about the productions.
> Simon
>
>
> Am 22.10.2016 um 07:00 schrieb Pine W:
>
> With a budget of about $75 million [1], productions like this certainly
> have the resources to get the credits and licensing right, so there is no
> excuse for being sloppy. Does the OSM Foundation have lawyers or
> communications staff who would be in a good position to address this issue
> with the responsible parties?
>
> Pine
>
> [1] https://variety.com/2016/film/news/box-office-tom-hanks-
> inferno-international-1201890418/
>
> On Oct 21, 2016 9:06 PM, "Eugene Alvin Villar"  wrote:
>
>> I saw another mainstream film that credits OSM in the end credits and
>> this is the film adaptation of Dan Brown's novel Inferno. The credit
>> went “© OpenStreetMap contributors”. There was no mention of any license
>> again.
>>
>> I've decided to create a new wiki page[1] to list these films, linked
>> from the "Press" wiki page[2].
>>
>> [1] https://wiki.openstreetmap.org/wiki/Films
>> [2] https://wiki.openstreetmap.org/wiki/Portal:Press
>>
>> ~Eugene
>>
>>
>> On Fri, Jul 22, 2016 at 3:12 PM, Eugene Alvin Villar 
>> wrote:
>>
>>> Hi all,
>>>
>>> I just saw the film Eye in the Sky, starring Helen Mirren and Alan
>>> Rickman (the movie is really good), and I was pleasantly surprised to
>>> see OpenStreetMap and its contributors get a credit in the end
>>> credits. The surprise is partly because I do not recall seeing any map
>>> in the film that looked like it came from OSM.
>>>
>>> IIRC, the credit went like “OpenStreetMap © OpenStreetMap
>>> contributors”. No mention of any license though.
>>>
>>> Now I'm wondering whether there are any other mainstream films that
>>> also use and credit OpenStreetMap as well. This is the first time I've
>>> seen such a credit and I don't recall any mention as well on this
>>> mailing list.
>>>
>>> ~Eugene
>>>
>>
>>
>> ___
>> talk mailing list
>> talk@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk
>>
>>
>
> ___
> talk mailing 
> listtalk@openstreetmap.orghttps://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

[talk-ph] Fwd: [OSM-talk] weeklyOSM #326 10/11/2016-10/17/2016

[Eugene Alvin Villar]

-- Forwarded message --
From: weeklyteam 
Date: Sat, Oct 22, 2016 at 7:04 PM
Subject: [OSM-talk] weeklyOSM #326 10/11/2016-10/17/2016
To: t...@openstreetmap.org


The weekly round-up of OSM news, issue # 326,
is now available online in English, giving as always a summary of all
things happening in the openstreetmap world:

http://www.weeklyosm.eu/en/archives/8214/

Enjoy!

weeklyOSM is brought to you by ... https://wiki.openstreetmap.
org/wiki/WeeklyOSM#Available_Languages
___
talk mailing list
t...@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk
___
talk-ph mailing list
talk-ph@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ph

Re: [Talk-ca] [Imports] [Import] Ottawa Buildings & Addresses [Statistics Canada project]

[Begin Daniel]

Well, only very large buildings can be found as polygons in the Canvec product.
Furthermore, NRCan did not update the Canvec buildings layer for more than 20 
years (the oldest is 1944), with only a few exceptions…

Daniel

From: James [mailto:james2...@gmail.com]
Sent: Saturday, 22 October, 2016 10:54
To: Stewart C. Russell
Cc: Talk-CA OpenStreetMap
Subject: Re: [Talk-ca] [Imports] [Import] Ottawa Buildings & Addresses 
[Statistics Canada project]

http://ftp.geogratis.gc.ca/pub/nrcan_rncan/vector/canvec/shp/ManMade/
50K man made just seems to be nodes of where buildings are located, instead of 
outlines.
Judging by :
http://atlas.gc.ca/toporama/en/index.html
They have a few buildings in CanVEC, but not all of them. Seems like massive 
buildings like schools and malls, residential buildings are out.

On Sat, Oct 22, 2016 at 10:00 AM, Stewart C. Russell 
> wrote:
On 2016-10-21 11:41 PM, James wrote:
> Sounds like it, but the data handed to us didnt have sidewalks and
> roads, driveways etc. Ottawa may have exported data from this file

Yes, for sure.

I've now had more of a chance to look at the data (thanks, Ottawa, for
providing no docs at all ...). I'm pretty sure that the data at
http://data.ottawa.ca/dataset/cad-topographic-data is the source of what
the Ottawa group were given.

In the 31 gigabytes of converted files, about 8-10 of the 177 total
layers might be of interest. But:

* The files are in some kind of MTM projection, but I don't know the
datum. Some munis still love their NAD27, so getting this right is crucial.

* These were digitized 2010-2011 at the latest. Since municipalities
share data with NRCan, aren't these outlines already available in a
recent iteration of CanVec in a much more useful (i.e., anything but
DWG) format?

My notes on the files, so far:
https://gist.github.com/scruss/e7f85da2e7943cb1a1d13772fbe144d3#file-ottawabfomapdata-md

(feel free to use/modify/etc)

If anyone wants 31 GB of converted DXFs, let me know. It took Teigha
several hours on a quad core with SSDs to convert this, so I'm not going
to delete it lightly.

cheers,
 Stewart


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



--
外に遊びに行こう！
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca

Re: [OSM-talk-fr] Tag pour constructeur de maisons ?

[pepilepi...@ovh.fr]

  
  
Le 22/10/2016 à 21:35, Romain MEHUT a
  écrit :


  

  Bonsoir,

  
  Celui-ci http://wiki.openstreetmap.org/wiki/Tag:craft%3Dbuilder
  pourrait faire l'affaire ?
  

Romain
  

Pas exactement... Mais c'est vrai que j'avais pas pensé au tag
  "craft". En fait le "constructeur de maison" auquel je fais
  référence est plus du bureau d'études que de la réalisation.
Merci,
JP


  
Le 22 octobre 2016 à 15:34, pepilepi...@ovh.fr
  
  a écrit :
  

  Bonjour,
  Après mon angoisse sur les kinés je
  suis aujourd'hui confronté à un autre problème :
  comment taggue-t-on un constructeur de maisons ? À
  savoir le gars à qui vous expliquez ce que vous voulez
  et qui vous fait les plans, organise et coordonne tous
  les corps de métiers et vous livre votre petite maison
  terminée. Pas vraiment, d'ailleurs, un promoteur
  immobilier, puisque lui achète en gros et revend au
  détail.

  Dans le wiki j'ai bien vu un
  "office=estate_agent", mais le "estate agent" se
  traduit parfaitement en français par "agent
  immobilier" et ce n'est pas ça qui corespond à ce que
  je veux indiquer.
  Question annexe : quand on ne
  trouve pas dans la nomenclature officielle ce dont on
  a besoin précisément y a-t-il un moyen de préciser une
  approximation ? Par exemple ici mettre office=estate_agent et un fr:=constructeur de maisons individuelles ?
  Merci,
  Jean-Pierre



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

  


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




  


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

Re: [OSM-talk-fr] données maritimes pour Osmose ?

[Frédéric Rodrigo]

Salut,

Pas sûr que ça puisse servir pour Osmose.
Pour pouvoir valider les mers et océans, il faut :
- des extracts de ces régions,
- un polygone exact par zone,
- des idées de règles à que l'on cherche à valider.

Frédéric.


Le 21/10/2016 à 21:53, osm.sanspourr...@spamgourmet.com a écrit :


Bonjour,

MarineRegions vient de sortir sa nouvelle version des ZEE.

Attention, ils prennent non la définition légale mais celle des zones 
de pêches.


Elles vont donc jusqu'à la côte.

Ils ont par le passé envisagé de mettre leurs données dans OSM, elles 
ne sont pas en ODbL pour le moment mais si d'autres voix se joignent à 
la demande...


À noter aussi un gazeetter intéressant pour nommer les objets en mer.

Jean-Yvon


*From:*users-ow...@marineregions.org 
[mailto:users-ow...@marineregions.org] *On Behalf Of *Marine Regions 
users

*Sent:* Friday, October 21, 2016 1:24 PM
*To:* us...@marineregions.org
*Subject:* [Marine Regions] New version Maritime Boundaries on 
MarineRegions.org


*Dear Marine Regions User,*

**

*MarineRegions.org  is proud to launch 
the new version of the Maritime Boundaries Geodatabase. **In this new 
release, Marine Regions updates the global Exclusive Economic Zones 
(EEZ) (version 9), and **launches global datasets for **Territorial 
Seas (TS), Contiguous Zones (CZ), Internal Waters (IW) and 
Archipelagic Waters (AW). The global baseline used in this new 
version is a combination of the publically available straight and 
archipelagic baselines and the coastline as a proxy for the low-water 
line. The source for the straight baselines is the United Nations 
repository on maritime claims, together with national publications on 
maritime delimitations and agreements. The ESRI Countries 2014 
database has been used as the primary source for the Maritime 
Boundaries coastline. The coastline was further updated with reefs 
data extracted from the Coral Reef Distribution database from UNEP 
for the countries where reefs were fundamental for the correct 
calculation of the maritime areas. *


**

*With the new release, we hope to further serve the global marine 
data user community working on geospatial analysis. As Marine Regions 
depends on data and knowledge sharing from global, European, regional 
and national data providers and relevant experts, we aim to set up 
Collaboration Agreements with our users and data providers. An 
example template of a Collaboration Agreement can be found online, 
please contact i...@marineregions.org  
if your organization is interested to explore this collaboration.*


**

*All datasets from marineregions.org’s Maritime Boundaries 
Geodatabase continue to be the only freely-available global datasets 
on the World’s Maritime Boundaries.*


**

*With Kind Regards*

**

*The Marine Regions Team*

*Follow us on Twitter @marineregions *



Jean-Yvon

*From:*users-ow...@marineregions.org *On Behalf Of *Marine Regions users
*Sent:* Friday, October 21, 2016 1:24 PM
*To:* us...@marineregions.org
*Subject:* [Marine Regions] New version Maritime Boundaries on 
MarineRegions.org


*Dear Marine Regions User,*

**

*MarineRegions.org  is proud to launch 
the new version of the Maritime Boundaries Geodatabase. **In this new 
release, Marine Regions updates the global Exclusive Economic Zones 
(EEZ) (version 9), and **launches global datasets for **Territorial 
Seas (TS), Contiguous Zones (CZ), Internal Waters (IW) and 
Archipelagic Waters (AW). The global baseline used in this new version 
is a combination of the publically available straight and archipelagic 
baselines and the coastline as a proxy for the low-water line. The 
source for the straight baselines is the United Nations repository on 
maritime claims, together with national publications on maritime 
delimitations and agreements. The ESRI Countries 2014 database has 
been used as the primary source for the Maritime Boundaries coastline. 
The coastline was further updated with reefs data extracted from the 
Coral Reef Distribution database from UNEP for the countries where 
reefs were fundamental for the correct calculation of the maritime 
areas. *


**

*With the new release, we hope to further serve the global marine data 
user community working on geospatial analysis. As Marine Regions 
depends on data and knowledge sharing from global, European, regional 
and national data providers and relevant experts, we aim to set up 
Collaboration Agreements with our users and data providers. An example 
template of a Collaboration Agreement can be found online, please 
contact i...@marineregions.org  if your 
organization is interested to explore this collaboration.*


**

*All datasets from marineregions.org’s Maritime Boundaries Geodatabase 
continue to be the only freely-available global datasets on the 
World’s Maritime Boundaries.*


**

*With Kind 

Re: [Talk-ca] Light rail mapping questions

[Michael Reichert]

Hi Mike,

Am 2016-10-22 um 20:43 schrieb Mike Boos:
> I don't know if I'd characterize this as a "Karlsuhe model" train-tram -
> the system is entirely within a single urban area (even if it does span two
> adjacent cities), it just uses an existing rail corridor in some places.
> Unlike the Karlsuhe system, the vehicles have only a single operating
> voltage.

There are lines in Karlsruhe which use only the tram current (750 Volt
DC), e.g. line S1 from Hochstetten via Karlsruhe to Bad Herrenalb. This
line shares tracks with freight trains between Lepoldshafen Frankfurter
Straße and Welschneureuter Straße. That's similar to the train track in
Waterloo.

>> Tag the tracks as they look like. Sections where tracks share space with
>> cars [1] are railway=tram. Where the trams are physically separated from
>> the traffic [2], it's a railway=light_rail. That's how tagging works in
>> cities which only have *one* tram/light rail system. If the city has two
>> or three (low-floor tram and high-floor light rail; some German cities),
>> it becomes more difficult because we also try to get the systems
>> distinguishable (there are use cases). But that is not important now and
>> the reason why Germans discuss correct tagging of trams, light rails and
>> subways at their OSM Forum over multiple pages and threads. :-)
>>
> 
> There are not really spaces shared with cars, (thank goodness) so the only
> appropriate tag along roads is light_rail.

It does not share the space but according to your photos at Twitter the
trackbed seems to be paved and cars could use it. A proper light rail
has either ballast or grass between the tracks, i.e. car drivers will
recognize that the cannot use this space.

https://www.mapillary.com/map/im/jx4Z1BdL_nhENDjJMxHI2w
http://www.regum.de/en/rail_products/lawn_track

There is no sharp border between trams and light rails.

Best regards

Michael


-- 
Per E-Mail kommuniziere ich bevorzugt GPG-verschlüsselt. (Mailinglisten
ausgenommen)
I prefer GPG encryption of emails. (does not apply on mailing lists)



signature.asc
Description: OpenPGP digital signature
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca

Re: [Talk-ca] Light rail mapping questions

[Michael Reichert]

Hi

Am 2016-10-22 um 21:11 schrieb James:
> Could lanes work?
> https://wiki.openstreetmap.org/wiki/Lanes
> 
> Example:
> https://wiki.openstreetmap.org/wiki/Lanes#Two_driving_directions
> 
> What ever the train tag would be:
> train:lanes:forward=no|yes
> train:lanes:backwards=no|yes
> Then for passenger cars would be opposite?
> car:lanes:forward=yes|no
> car:lanes:backwards=yes|no
>
> I'm not sure about this maybe Micheal has more insight on this

I would use different tags but the ideas is similar.

The right/back of the photo:
lanes = 4
lanes:forward = 2
lanes:backward = 2
access:lanes:forward=no|yes
access:lanes:backward=no|yes
tram:lanes:forward = yes|no
tram:lanes:backward = yes|no

The left of the photo:
lanes = 4
lanes:forward = 2
lanes:backward = 1
turn:lanes:forward=
tram:lanes:backward = no|yes
access:lanes:backward = yes |no

Differences to James' suggestion:
My suggestion is compatible with the widely used lanes tagging scheme. I
used "backward", not "backwards" (see Taginfo). I used "tram", not "train".

Best regards

Michael


-- 
Per E-Mail kommuniziere ich bevorzugt GPG-verschlüsselt. (Mailinglisten
ausgenommen)
I prefer GPG encryption of emails. (does not apply on mailing lists)



signature.asc
Description: OpenPGP digital signature
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca

Re: [OSM-talk-fr] Tag pour constructeur de maisons ?

[Romain MEHUT]

Bonsoir,

Celui-ci http://wiki.openstreetmap.org/wiki/Tag:craft%3Dbuilder pourrait
faire l'affaire ?

Romain

Le 22 octobre 2016 à 15:34, pepilepi...@ovh.fr  a écrit
:

> Bonjour,
>
> Après mon angoisse sur les kinés
> 
> je suis aujourd'hui confronté à un autre problème : comment taggue-t-on un
> constructeur de maisons ? À savoir le gars à qui vous expliquez ce que vous
> voulez et qui vous fait les plans, organise et coordonne tous les corps de
> métiers et vous livre votre petite maison terminée. Pas vraiment,
> d'ailleurs, un promoteur immobilier, puisque lui achète en gros et revend
> au détail.
>
> Dans le wiki j'ai bien vu un "office=estate_agent", mais le "estate agent"
> se traduit parfaitement en français par "agent immobilier" et ce n'est pas
> ça qui corespond à ce que je veux indiquer.
>
> Question annexe : quand on ne trouve pas dans la nomenclature officielle
> ce dont on a besoin précisément y a-t-il un moyen de préciser une
> approximation ? Par exemple ici mettre office=estate_agent et un fr: sais pas>=constructeur de maisons individuelles ?
>
> Merci,
>
> Jean-Pierre
>
> ___
> Talk-fr mailing list
> Talk-fr@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-fr
>
>
___
Talk-fr mailing list
Talk-fr@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-fr

[OSM-talk-fr] Map4Haiti Ouragan Matthew Une premiere - 15km2 images UAV disponibles pour évaluation des dommages

[Pierre Béland]

Envois aux listes hot-francophone, talk-ht, et talk-fr


 
--
Deux semaines après l'arrivée de l'ouragan Matthew en Haiti, plus de 1.4 
millions de personnes ont besoin d'aide et la situation est encore très 
précaire. De nombreux villages n'ont pas encore reçu d'aide. Les services de 
santé déjà précaires avant le séisme ont été fortement touchés. MSF rapporte 
que 23 centres de santé ont été endommagés ou partiellement détruits. Le risque 
d'épidémie de choléra inquiète les organisations de santé. De nouvelles 
dépressions tropicales sont arrivées sur Haiti causant encore des inondations 
et augmentant les risques de glissement de terrain.
Les organismes internationaux qui planifient la reconstruction ont besoin 
d'évaluer l'ampleur des dégâts et estimer l'effort financier nécessaire.
Images UAV couvrant 15km2
Une première avec les réponses humanitaires, des images UAV de grande précision 
couvrant déjà plus de 15km2 et une dizaine de sites sont disponbiles. Elles 
sont hébergées sur les serveurs TMS et WMS de OSM-fr. Ces images ciblent des 
localités fortement affectées par l'ouragan. Elles viennent compléter 
l'information fournie par les images satellite post-désastre. La carte 
UAVCompare, en ligne depuis hier, avec des panneaux Avant / Après permet 
d'observer l'ampleur du désastre. Les images pré-désastre prises par avion sont 
mises à disposition par le CNIGS, l'agence geospatiale du gouvernement haitien. 
http://pierzen.dev.openstreetmap.org/swipe/UAVCompare-Haiti-Matthews-before-after.php

Les images UAV fournissent des images souvent jusqu'à 10 fois plus précises que 
les images satellite. Zoomez sur ces images, et vous verrez quels détails elles 
fournissent sur l'état des structures dans les diverses localités et comment 
elles peuvent être utilisées pour l'évaluation des dommages. Pour rappel, les 
contributeurs expérimentés sont invités à nous contacter pour contribuer à 
l'évaluation des dommages.Voir http://taches.francophonelibre.org/

Les missions UAV sur le terrain ont été réalisées par deux équipes, travaillant 
dans des conditions difficiles et des moyens matériels réduits.  Nos collègues 
Fred Moine et Presler Jean de la ONG haitienne Potentiel3.0 et Nicolas Clarens, 
consultant pour la Banque inter-américaine de développement, ont organisé les 
missions sur le terrain. 

Les plans de vol, les missions de collecte d'images sur le terrain, le 
traitement et le transfert de données sont des éléments essentiels pour assurer 
le succès de telles opérations. Malgré les difficultés matérielles, les 
problèmes d'électricité et de transfert internet, Fred, Presler et Xavier ont 
fait le traitement à partir d'ordinateurs personnels, et fait le transfert de 
données vers le serveur OSM-France. Jean-Guilhem Cailton a alors pris le 
relais, converti ces images lorsque c'était nécessaire, et les a intégrées aux 
serveurs WMS et TMS de OSM-Fr.
.Avec les pluies abondantes et le pont endommagé sur la route principale, Fred 
et Presler ont de la difficulté à rejoindre les localités sinistrées. Ils 
poursuivent néanmoins leurs efforts et vont organiser des missions de vols UAV 
notamment au dessus de zones à risque de glissement de terrain.
Chapeau à tous ces collègues qui ont réussi à travailler en flux tendus avec 
très peu de moyens et livrer des images saisissantes.  
Pierre twitter pierzen
   ___
Talk-fr mailing list
Talk-fr@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-fr

Re: [Talk-ca] Light rail mapping questions

[James]

https://wiki.openstreetmap.org/wiki/Key:train

Train does exist as a vehicle type

As for cars:
I think it would be motor_vehicle:

https://wiki.openstreetmap.org/wiki/Key:motor_vehicle

I got the vehicle types from:
https://wiki.openstreetmap.org/wiki/Key:access#Transport_mode_restrictions

On Oct 22, 2016 3:10 PM, wrote:

> Could lanes work?
> https://wiki.openstreetmap.org/wiki/Lanes
>
> Example:
> https://wiki.openstreetmap.org/wiki/Lanes#Two_driving_directions
>
> What ever the train tag would be:
> train:lanes:forward=no|yes
> train:lanes:backwards=no|yes
> Then for passenger cars would be opposite?
> car:lanes:forward=yes|no
> car:lanes:backwards=yes|no
>
> I'm not sure about this maybe Micheal has more insight on this
>
> On Oct 22, 2016 2:45 PM, "Mike Boos"  wrote:
>
> Thanks Michael for your reply
>
>
>> Am 18.10.2016 um 03:52 schrieb Mike Boos:
>> > Along on-road sections, the dedicated rail right-of-way moves from
>> > centre-running to the outsides of the street at certain intersections.
>> (A
>> > by-product of some of the political compromises in route choices.) Does
>> > anyone know of any examples of tracks going from the centre to the side
>> of
>> > the road with traffic lanes in OSM? I expect these are going to look
>> messy.
>>
>> Look at any German, Austrian or Swiss city of your choice where every
>> tram track is mapped as a single way in OSM (i.e. no tracks=2). I need
>> more details (show us photos) to give a useful answer.
>>
>
> Here's an example of an intersection that's largely complete:
> https://twitter.com/mikeboos/status/789892670453723136
> The tracks on the right/behind of most of the photos are centre-running
> between the two directions of car travel. Across the intersection, on the
> left of the photos, one track changes sides with the travel lane so that it
> lies between the sidewalk and the roadway. (The other track branches off
> onto the perpendicular road.)
>
> > There are also portions of the line that will share track with a freight
>> > corridor. From what I can tell, convention appears to be to tag it with
>> the
>> > heavier mode, i.e. railway=rail instead of railway=light_rail. However,
>> the
>> > use of the track for freight is quite small - at most one freight train
>> > to/from Elmira uses the track at night, when light rail service won't be
>> > operating. Should the track still be marked as 'rail' instead of
>> > 'light_rail,' or should we attempt to have the tags represent the
>> dominant
>> > use? (At present, some of these are tagged as railway=construction, even
>> > though the freight train has been consistently using it overnight. This
>> > section is also largely complete.)
>>
>> Yes. If the track is still usable for freight trains (even if limited to
>> certain hours), it is a normal railway track and therefore gets
>> railway=rail. What you describe is called "Karlsuhe model" – don't
>> confuse it with our tagging scheme at OSM. ;-)
>>
>> I assume, that some people of Grand River Transit have visited the
>> German cities Karlsruhe and/or Kassel. :-) The first one has been
>> operating a tram-train system for more than 40 years.
>>
>> https://en.wikipedia.org/wiki/Karlsruhe_Stadtbahn
>> https://en.wikipedia.org/wiki/Kassel_RegioTram
>>
>
> I don't know if I'd characterize this as a "Karlsuhe model" train-tram -
> the system is entirely within a single urban area (even if it does span two
> adjacent cities), it just uses an existing rail corridor in some places.
> Unlike the Karlsuhe system, the vehicles have only a single operating
> voltage.
>
>
>> Tag the tracks as they look like. Sections where tracks share space with
>> cars [1] are railway=tram. Where the trams are physically separated from
>> the traffic [2], it's a railway=light_rail. That's how tagging works in
>> cities which only have *one* tram/light rail system. If the city has two
>> or three (low-floor tram and high-floor light rail; some German cities),
>> it becomes more difficult because we also try to get the systems
>> distinguishable (there are use cases). But that is not important now and
>> the reason why Germans discuss correct tagging of trams, light rails and
>> subways at their OSM Forum over multiple pages and threads. :-)
>>
>
> There are not really spaces shared with cars, (thank goodness) so the only
> appropriate tag along roads is light_rail.
>
> > Further, there is gauntlet track to allow freight trains to pass station
>> > platforms. Do we tag the track closest to the platform as
>> > railway=light_rail and the outer track as railway=rail? There's some
>> > discussion here on gauntlet tracks here that suggests this is the case
>> in
>> > Europe: http://forum.openstreetmap.org/viewtopic.php?id=29131
>>
>> It is the case in Kaufungen near the city of Kassel which has a
>> Karlsruhe-like tram-train system ("Regiotram").
>>
>> https://commons.wikimedia.org/wiki/File:Haltestelle_Niederka
>> ufungen_Mitte_02.JPG
>>
>> Yes, the track for heavy 

Re: [Talk-ca] Light rail mapping questions

[James]

Could lanes work?
https://wiki.openstreetmap.org/wiki/Lanes

Example:
https://wiki.openstreetmap.org/wiki/Lanes#Two_driving_directions

What ever the train tag would be:
train:lanes:forward=no|yes
train:lanes:backwards=no|yes
Then for passenger cars would be opposite?
car:lanes:forward=yes|no
car:lanes:backwards=yes|no

I'm not sure about this maybe Micheal has more insight on this

On Oct 22, 2016 2:45 PM, "Mike Boos"  wrote:

Thanks Michael for your reply


> Am 18.10.2016 um 03:52 schrieb Mike Boos:
> > Along on-road sections, the dedicated rail right-of-way moves from
> > centre-running to the outsides of the street at certain intersections. (A
> > by-product of some of the political compromises in route choices.) Does
> > anyone know of any examples of tracks going from the centre to the side
> of
> > the road with traffic lanes in OSM? I expect these are going to look
> messy.
>
> Look at any German, Austrian or Swiss city of your choice where every
> tram track is mapped as a single way in OSM (i.e. no tracks=2). I need
> more details (show us photos) to give a useful answer.
>

Here's an example of an intersection that's largely complete:
https://twitter.com/mikeboos/status/789892670453723136
The tracks on the right/behind of most of the photos are centre-running
between the two directions of car travel. Across the intersection, on the
left of the photos, one track changes sides with the travel lane so that it
lies between the sidewalk and the roadway. (The other track branches off
onto the perpendicular road.)

> There are also portions of the line that will share track with a freight
> > corridor. From what I can tell, convention appears to be to tag it with
> the
> > heavier mode, i.e. railway=rail instead of railway=light_rail. However,
> the
> > use of the track for freight is quite small - at most one freight train
> > to/from Elmira uses the track at night, when light rail service won't be
> > operating. Should the track still be marked as 'rail' instead of
> > 'light_rail,' or should we attempt to have the tags represent the
> dominant
> > use? (At present, some of these are tagged as railway=construction, even
> > though the freight train has been consistently using it overnight. This
> > section is also largely complete.)
>
> Yes. If the track is still usable for freight trains (even if limited to
> certain hours), it is a normal railway track and therefore gets
> railway=rail. What you describe is called "Karlsuhe model" – don't
> confuse it with our tagging scheme at OSM. ;-)
>
> I assume, that some people of Grand River Transit have visited the
> German cities Karlsruhe and/or Kassel. :-) The first one has been
> operating a tram-train system for more than 40 years.
>
> https://en.wikipedia.org/wiki/Karlsruhe_Stadtbahn
> https://en.wikipedia.org/wiki/Kassel_RegioTram
>

I don't know if I'd characterize this as a "Karlsuhe model" train-tram -
the system is entirely within a single urban area (even if it does span two
adjacent cities), it just uses an existing rail corridor in some places.
Unlike the Karlsuhe system, the vehicles have only a single operating
voltage.


> Tag the tracks as they look like. Sections where tracks share space with
> cars [1] are railway=tram. Where the trams are physically separated from
> the traffic [2], it's a railway=light_rail. That's how tagging works in
> cities which only have *one* tram/light rail system. If the city has two
> or three (low-floor tram and high-floor light rail; some German cities),
> it becomes more difficult because we also try to get the systems
> distinguishable (there are use cases). But that is not important now and
> the reason why Germans discuss correct tagging of trams, light rails and
> subways at their OSM Forum over multiple pages and threads. :-)
>

There are not really spaces shared with cars, (thank goodness) so the only
appropriate tag along roads is light_rail.

> Further, there is gauntlet track to allow freight trains to pass station
> > platforms. Do we tag the track closest to the platform as
> > railway=light_rail and the outer track as railway=rail? There's some
> > discussion here on gauntlet tracks here that suggests this is the case in
> > Europe: http://forum.openstreetmap.org/viewtopic.php?id=29131
>
> It is the case in Kaufungen near the city of Kassel which has a
> Karlsruhe-like tram-train system ("Regiotram").
>
> https://commons.wikimedia.org/wiki/File:Haltestelle_Niederka
> ufungen_Mitte_02.JPG
>
> Yes, the track for heavy trains is a normal train track (railway=rail)
> while the outer ones can only be used by light rail vehicles due to the
> smaller structure gauge. Therefore the light rail track gets
> railway=light_rail. Because we map one way per track at the centerline
> of the track, there are two (in Kaufungen three) parallel tracks and all
> get railway:interlaced=yes. This is useful for routing engines.
>
> If there were up to date Mapillary photos, I could give more 

Re: [Talk-ca] Light rail mapping questions

[Mike Boos]

Thanks Michael for your reply


> Am 18.10.2016 um 03:52 schrieb Mike Boos:
> > Along on-road sections, the dedicated rail right-of-way moves from
> > centre-running to the outsides of the street at certain intersections. (A
> > by-product of some of the political compromises in route choices.) Does
> > anyone know of any examples of tracks going from the centre to the side
> of
> > the road with traffic lanes in OSM? I expect these are going to look
> messy.
>
> Look at any German, Austrian or Swiss city of your choice where every
> tram track is mapped as a single way in OSM (i.e. no tracks=2). I need
> more details (show us photos) to give a useful answer.
>

Here's an example of an intersection that's largely complete:
https://twitter.com/mikeboos/status/789892670453723136
The tracks on the right/behind of most of the photos are centre-running
between the two directions of car travel. Across the intersection, on the
left of the photos, one track changes sides with the travel lane so that it
lies between the sidewalk and the roadway. (The other track branches off
onto the perpendicular road.)

> There are also portions of the line that will share track with a freight
> > corridor. From what I can tell, convention appears to be to tag it with
> the
> > heavier mode, i.e. railway=rail instead of railway=light_rail. However,
> the
> > use of the track for freight is quite small - at most one freight train
> > to/from Elmira uses the track at night, when light rail service won't be
> > operating. Should the track still be marked as 'rail' instead of
> > 'light_rail,' or should we attempt to have the tags represent the
> dominant
> > use? (At present, some of these are tagged as railway=construction, even
> > though the freight train has been consistently using it overnight. This
> > section is also largely complete.)
>
> Yes. If the track is still usable for freight trains (even if limited to
> certain hours), it is a normal railway track and therefore gets
> railway=rail. What you describe is called "Karlsuhe model" – don't
> confuse it with our tagging scheme at OSM. ;-)
>
> I assume, that some people of Grand River Transit have visited the
> German cities Karlsruhe and/or Kassel. :-) The first one has been
> operating a tram-train system for more than 40 years.
>
> https://en.wikipedia.org/wiki/Karlsruhe_Stadtbahn
> https://en.wikipedia.org/wiki/Kassel_RegioTram


I don't know if I'd characterize this as a "Karlsuhe model" train-tram -
the system is entirely within a single urban area (even if it does span two
adjacent cities), it just uses an existing rail corridor in some places.
Unlike the Karlsuhe system, the vehicles have only a single operating
voltage.


> Tag the tracks as they look like. Sections where tracks share space with
> cars [1] are railway=tram. Where the trams are physically separated from
> the traffic [2], it's a railway=light_rail. That's how tagging works in
> cities which only have *one* tram/light rail system. If the city has two
> or three (low-floor tram and high-floor light rail; some German cities),
> it becomes more difficult because we also try to get the systems
> distinguishable (there are use cases). But that is not important now and
> the reason why Germans discuss correct tagging of trams, light rails and
> subways at their OSM Forum over multiple pages and threads. :-)
>

There are not really spaces shared with cars, (thank goodness) so the only
appropriate tag along roads is light_rail.

> Further, there is gauntlet track to allow freight trains to pass station
> > platforms. Do we tag the track closest to the platform as
> > railway=light_rail and the outer track as railway=rail? There's some
> > discussion here on gauntlet tracks here that suggests this is the case in
> > Europe: http://forum.openstreetmap.org/viewtopic.php?id=29131
>
> It is the case in Kaufungen near the city of Kassel which has a
> Karlsruhe-like tram-train system ("Regiotram").
>
> https://commons.wikimedia.org/wiki/File:Haltestelle_Niederka
> ufungen_Mitte_02.JPG
>
> Yes, the track for heavy trains is a normal train track (railway=rail)
> while the outer ones can only be used by light rail vehicles due to the
> smaller structure gauge. Therefore the light rail track gets
> railway=light_rail. Because we map one way per track at the centerline
> of the track, there are two (in Kaufungen three) parallel tracks and all
> get railway:interlaced=yes. This is useful for routing engines.
>
> If there were up to date Mapillary photos, I could give more and better
> advice. (Mapillary photos by pedestrians are better because are located
> on the sidewalk)
>

I think that makes sense. Here's an example of the gauntlet track at one of
the stations:
https://twitter.com/Canardiain/status/768589656581509122/photo/1


>
> Greetings from Karlsruhe
>



-- 
Mike Boos, MASc.
mike.b...@gmail.com
519-580-5804
http://real.uwaterloo.ca/~mboos
___
Talk-ca mailing list

Re: [OSM-talk] Forum admin(s) wanted

[Daniel Koć]

W dniu 16.08.2016 11:45, Jóhannes Birgir Jensson napisał(a):


As for "no panic" then we've got several projects cooking in Iceland
after a beneficial annual meeting but we've not launched them yet
because we don't want to fragment any more than strictly needed so the
forum issue has been plaguing us. We could set up our own forum and go
from there but that is counter-productive in the long run.


Jóhannes, "Users: Iceland" subforum is finally created:

https://forum.openstreetmap.org/viewforum.php?id=90

What's more important, the whole forum is now migrated to a server under 
OSMF control, so it's pretty safe now:


https://forum.openstreetmap.org/viewtopic.php?pid=613559#p613559

--
"To co ludzie zwą marskością wątroby/ Tak naprawdę jest śmiercią z 
tęsknoty" [Afro Kolektyw]


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

Re: [Talk-lt] Talk-lt Digest, Vol 92, Issue 6

[zjonaitis]

Pritariu Aido komentarams dėl žymėjimo. Sveikas prisijungęs, Lukai! 

Žilvinas


From: Ričardas
Sent: 22 October 2016 19:38
To: talk-lt@openstreetmap.org
Subject: Re: [Talk-lt] Talk-lt Digest, Vol 92, Issue 6

Sveiki,

Pritariu Aido kriterijams. Gal tik norėčiau patikslinti, kad mano nuomone, 
supermarket žymi vieną objektą - parduotuvę kaip atskirą vienetą. Jei viename 
pastate yra didelis prekybos centras su keletu mažų papildomų parduotuvių, tai 
jį žymiu supermarket plius visas kitas parduotuves atskirai. Pažymėjus vien tik 
bendrinį mall maža naudos, nes nesužymėjus atskirai kiekvienos parduotuvės 
(bent jau svarbiausių) nebus galima paieškoje jų rasti.
Galbūt galima diskutuoti, ar tuo atveju, kai pastate yra vienas didesnis 
prekybos centras su keletu parduotuvėlių tame pastate, supermarket geriau 
žymėti visam pastatui ar tame pastate kaip atskirą tašką. Mano nuomone, didelio 
skirtumo nėra ir teisingi abu variantai.
Ričardas

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

Re: [Talk-it] mappa osmand Italia

[girarsi_liste]

Il 19/10/2016 23:16, MAx1234 Ita ha scritto:
> Per chi fosse interessato, ieri mattina il mio PC ha finito di generare una
> nuova edizione delle mappe per OsmAnd: le ho condivise su GDrive e potete
> scaricarle dai link qui sotto:
> 
> 
>- Address Map (tutta Italia, solo indirizzi, per facilitare le ricerche
>se si usano mappe regionali): https://goo.gl/ySmjRD
> 
> 
> 
>- Mappa completa (Tutta Italia con Strade, PDI, Indirizzi, Trasporto):
>https://goo.gl/4PfAko
> 
> 
> Spero vi siano utili! :-)
> 

Grazie, l'ho scaricata qualche giorno fà, ma ti ringrazio adesso, visto
l'impegno non da poco, spero non sia incompreso.




-- 
Simone Girardelli
_|_|_|_|_|_|_|_|_|_
|_|_|_|_|_|_|_|_|_|_|



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

Re: [Talk-lt] Talk-lt Digest, Vol 92, Issue 6

[Ričardas]

Sveiki,


Pritariu Aido kriterijams. Gal tik norėčiau patikslinti, kad mano nuomone,
supermarket žymi vieną objektą - parduotuvę kaip atskirą vienetą. Jei
viename pastate yra didelis prekybos centras su keletu mažų papildomų
parduotuvių, tai jį žymiu supermarket plius visas kitas parduotuves
atskirai. Pažymėjus vien tik bendrinį mall maža naudos, nes nesužymėjus
atskirai kiekvienos parduotuvės (bent jau svarbiausių) nebus galima
paieškoje jų rasti.
Galbūt galima diskutuoti, ar tuo atveju, kai pastate yra vienas didesnis
prekybos centras su keletu parduotuvėlių tame pastate, supermarket geriau
žymėti visam pastatui ar tame pastate kaip atskirą tašką. Mano nuomone,
didelio skirtumo nėra ir teisingi abu variantai.

Ričardas
___
Talk-lt mailing list
Talk-lt@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-lt

Re: [OSM-talk-fr] Appli mobile pour mapper les arrêts de bus

[Philippe Verdy]

Déjà les "bus_stop" (ancienne version) devraient correspondent point par
point aux nouveaux objets "public_transport:platform" s'ils ne sont pas sur
le chemin mais doivent servir à indiquer les abris, les bancs ou
l'accessibilité et la zone d'attente (ou le poteau indicateur)

S'ils sont sur le chemin (donc sans indication du côté) ce sont des
"public_transport:stop_position" (et il n'y a pas d'attributs pour les
bancs, abris et l'accessibilité).

Selon que c'est un type ou l'autre tu en déduis le rôle qu'il faut mettre
dans la relation route. Idéalement on devrait avoir deux rôles pour chaque
arrêt: stop et platform

Les "route segments" (proposés) sont une autre problématique liée à la
topologie des chemins, et la multiplicité des variantes qui empruntent des
sections communes. Mais ils sont surtout nécessaire pour lever les
ambiguités de parcours sur les chemins quand ils ne forment pas une ligne
continue sans intersection, car dans l'immédiat on est amené à devoir
mettre tous les ways membres dans l'ordre, y compris avec des doublons pour
les segments communs, et il n'est pas facile de déterminer un ordre des
segments quand il y a des intersections, même sur une seule route et après
avoir séparé chaque direction et chaque variante de la ligne dans des
relations "route" séparées mais regroupées dans un "route_master"

Quand on a compris tout ça, on peut faire une appli qui saura distinguer
les vrais "stop_position" des "plateform" (c'est facile: le "stop" est sur
le chemin, mais pas la "plateform"), ce que "bus_stop" ne sait pas classer
correctement. Juste faire ça fera beaucoup avancer les choses.



Pour aller au delà avec les "stop_area" par exemple pour regrouper
différents arrêts et plateformes de différentes lignes destinées à la
correspondance directe ou même pour revenir en arrière sur la même ligne
mais par une autre route), c'est un autre travail à faire: celui de
vérifier et assurer les correspondances.

De même que les objets "station" qui sont des regroupements plus large
comprenant plusieurs "stop_area", des batiments, des accès (escaliers,
portes, ascenseurs...) où la correspondance est plus complexe (et où on
peut avoir des services annexes (comme la vente des billets et abonnements)
et peut grouper des réseaux de nature différente et différents opérateurs
pour l'intermodalité (exemple: les gares routières et gares ferroviaires)


Le 22 octobre 2016 à 12:10, Florian LAINEZ  a écrit :

> Si on propose ce que l'on sait déjà de l'arrêt et qu'on propose de
>> confirmer/infirmer/ne pas se prononcer, on gagne des possibilités de
>> réponse rapide.
>
> tout à fait, je pense à de l'autocomplétion avec la meilleure proposition
> mise en avant le cas échéant.
>
> Peut-être que si on laisse la personne dessiner le trajet (juste en
>> chaînant les arrêts) on a pas mal de matière.
>> Ensuite un moteur de calcul de trajet (favorisant les rues larges et
>> droites) peut être utilisé côté serveur pour avoir un trajet vraisemblable.
>> Trajet à valider bien-sûr.
>
>  C'est à peu près ça que j'ai en tête. Avec la possibilité de drag le
> chemin pour le replacer (du style de ce que propose OSRM sur osm.org)
> sans avoir à gérer des way un par un.
>
> @philippe tout ceci est passionnant mais ce sont des problématiques
> complexes, or mon appli est vraiment focalisée sur la simplicité. D'où le
> parti pris de ne gérer que les highway=bus_stop (ou son équivalent
> public_transport=platform).
>
> Mon idée est bien de pouvoir gérer les deux modèles, ancien et nouveau, de
> manière transparente pour l'utilisateur.
> Charge aux mappeurs expérimentés de compléter le travail avec des outils
> plus classiques.
>
>
> Le 21 octobre 2016 à 22:26, Philippe Verdy  a écrit :
>
>> Les tags "from=*", "via=*" et "to=*" sont plutôt destinés à un
>> utilisateur humain qu'à subir un vrai rapprochement avec un arrêt.
>>
>> Dans le shéma actuel ("public_transport:version=2" dans les relations
>> "route"), ce sont les noeuds membres (posés sur un les ways membres) ayant
>> un rôle "stop" qui servent à ça, le premier noeud (from) devrait être de
>> rôle "stop_enter_only", le dernier de rôle "stop_exit_only". La direction
>> de la ligne devrait s'en déduire L'ennui c'est qu'il existe parfois des
>> stations intermédiaires qui n'autorisent que la descente ou que la montée,
>> ayant ces même rôles (c'est rare...). On n'a en fait rien de clair pour
>> indiquer réellement quels arrêts (noeuds de rôle "stop") sont les deux
>> terminus (l'idéal serait d'ajouter un suffixe ":start" ou ":end" à ces
>> rôles)
>>
>> On peut ajouter aussi les plateformes (noeuds, lines ou polygones) en
>> membres de rôle "platform" (avec "platform_enter_only",
>> "platform_exit_only" pour le premier et le dernier arrêt, mais cela me
>> semble inutile si on a identifié clairement les deux noeuds "stop" de
>> départ et d'arrivée: le rôle "platform" suffit à lui seul; ces rôles
>> variantes sont plutôt destinés à 

semanárioOSM Nº 326 11/10/2016-17/10/2016

[weeklyteam]

Bom dia,

O semanárioOSM Nº 326, o resumo de tudo o que acontece no mundo OpenStreetMap, 
está publicado em português:

http://www.weeklyosm.eu/pb/archives/8214/

Aproveite!

O semanárioOSM em português é produzido por: 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-br mailing list
Talk-br@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-br

semanárioOSM Nº 326 11/10/2016-17/10/2016

[weeklyteam]

Bom dia,

O semanárioOSM Nº 326, o resumo de tudo o que acontece no mundo OpenStreetMap, 
está publicado em português:

http://www.weeklyosm.eu/pb/archives/8214/

Aproveite!

O semanárioOSM em português é produzido por: 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-pt mailing list
Talk-pt@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-pt

Re: [Talk-it] Utente che modifica confini

[Marcello]

On 21/10/2016 20:11, Martin Koppenhoefer wrote:
>
> sent from a phone
>
>> Il giorno 21 ott 2016, alle ore 19:30, Marcello  ha 
>> scritto:
>>
>> Convinto che avere un dato migliore dalla stessa fonte fosse già motivo 
>> sufficiente per aggiornarlo in OSM ho provveduto ad aggiornare i confini 
>> dei comuni della provincia di Perugia, mantenendo inalterati solo quelli 
>> coincidenti al confine di regione, non credevo ci fosse bisogno di 
>> referendum o altro.
>
> si, quando hai un dato migliore lo puoi inserire senza referendum. Il 
> discorso sarebbe l'importazione: se non hai autoprodotto il dato devi seguire 
> le import guidelines (se per un determinato dato è già avviato una procedura 
> ti puoi associare)
>
> ciao,
> Martin 
>
Martin,
chiaramente il dato non l'ho autoprodotto, ma non so se può essere
definito un import. Ho modificato la geometria delle way già presenti,
mantenendo quindi lo stesso id, controllandole una ad una, ho fatto un
changeset per comune mettendo nel source del changeset "ISTAT 2014 data
(CC-BY 3.0) - http://www.istat.it/it/archivio/124086; e aggiornando il
valore source delle way in "Based on ISTAT data - 2001 Italian
Census-Updated with 2014 ISTAT data". In effetti probabilmente era più
corretto lasciare il source originale, che è riportato alla pagina
dell'import, e aggiungere ad esempio un source:note o qualcosa di simile
per specificare che il dato era stato aggiornato.

Guardando i confini della provincia di Terni vedo che non sono più
quelli dell'import con dati Istat 2001, gli originali sono stati
cancellati e le way attuali riportano nel campo source "Based on ISTAT
data - Italian Census 2011", per questo sono coincidenti a quelli del
2014. Nella pagina delle potenziali fonti di dati in effetti per i
confini amministrativi riporta i dati Istat del 2011, verificando con
una query Overpass (http://overpass-turbo.eu/s/jy1) vedo che da molte
parti quelli del 2001 sono stati sostituiti con quelli del 2011.

Ciao
Marcello

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

Re: [Talk-ca] [Imports] [Import] Ottawa Buildings & Addresses [Statistics Canada project]

[Stewart C. Russell]

On 2016-10-22 12:12 AM, James wrote:
> Converting, DWG to DXF to open in qgis should be fun

nah, all that pointenclicky would do me no good at all. This is what OGR
is for:

ogr2ogr -f GeoJSON -s_srs epsg:2951 -t_srs epsg:4326 \
   -dialect SQLite \
   -sql "select * from entities where LAYER='house'" \
   glebe-houses-4326.geojson 368029.dxf

Assuming I picked the right datum (I've been burned before) here are
some houses from in and around The Glebe:

https://gist.github.com/scruss/5a3f469c47df5d27fdba28258c273b45

I see that building outlines get sliced on tile boundaries dammit.

 Stewart

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

Re: [Talk-ca] [Imports] [Import] Ottawa Buildings & Addresses [Statistics Canada project]

[James]

http://ftp.geogratis.gc.ca/pub/nrcan_rncan/vector/canvec/shp/ManMade/
50K man made just seems to be nodes of where buildings are located, instead
of outlines.

Judging by :
http://atlas.gc.ca/toporama/en/index.html
They have a few buildings in CanVEC, but not all of them. Seems like
massive buildings like schools and malls, residential buildings are out.

On Sat, Oct 22, 2016 at 10:00 AM, Stewart C. Russell 
wrote:

> On 2016-10-21 11:41 PM, James wrote:
> > Sounds like it, but the data handed to us didnt have sidewalks and
> > roads, driveways etc. Ottawa may have exported data from this file
>
> Yes, for sure.
>
> I've now had more of a chance to look at the data (thanks, Ottawa, for
> providing no docs at all ...). I'm pretty sure that the data at
> http://data.ottawa.ca/dataset/cad-topographic-data is the source of what
> the Ottawa group were given.
>
> In the 31 gigabytes of converted files, about 8-10 of the 177 total
> layers might be of interest. But:
>
> * The files are in some kind of MTM projection, but I don't know the
> datum. Some munis still love their NAD27, so getting this right is crucial.
>
> * These were digitized 2010-2011 at the latest. Since municipalities
> share data with NRCan, aren't these outlines already available in a
> recent iteration of CanVec in a much more useful (i.e., anything but
> DWG) format?
>
> My notes on the files, so far:
> https://gist.github.com/scruss/e7f85da2e7943cb1a1d13772fbe144
> d3#file-ottawabfomapdata-md
>
> (feel free to use/modify/etc)
>
> If anyone wants 31 GB of converted DXFs, let me know. It took Teigha
> several hours on a quad core with SSDs to convert this, so I'm not going
> to delete it lightly.
>
> cheers,
>  Stewart
>
>
> ___
> Talk-ca mailing list
> Talk-ca@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-ca
>



-- 
外に遊びに行こう！
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca

Re: [Talk-it] Ricerca affluenti di un dato torrente/fiume in Josm.

[girarsi_liste]

Il 22/10/2016 14:25, Samuele Battarra ha scritto:
> In JOSM se installi il plugin utilsplugin2 ti ritrovi un menù "selezione" 
> dove 
> ci trovi le voci "Tutti i percorsi connessi" che dovrebbe fare quello che 
> chiedi.
> https://josm.openstreetmap.de/wiki/Help/Action/SelectConnectedWays
> 

Diciamo che funziona a metà, seleziona una parte dei percorsi, a patto
che tutte le ramificazioni siano isolate, se c'è un ford attaccato,
seleziona anche le strade, a me questo non interessa.

Comunque per quanto di mio interesse su livello a parte, mi va bene.

Grazie per la dritta.


-- 
Simone Girardelli
_|_|_|_|_|_|_|_|_|_
|_|_|_|_|_|_|_|_|_|_|



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

Re: [Talk-de] FOSSGIS als deutsches OSMF-Chapter

[Michael Kugelmann]

Am 21.10.2016 um 21:52 schrieb Frederik Ramm:

2. Meinungsbild der deutschen OSM-Community einholen - das tue ich hiermit!

+1


Michael.


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

Re: [Talk-ca] [Imports] [Import] Ottawa Buildings & Addresses [Statistics Canada project]

[Stewart C. Russell]

On 2016-10-21 11:41 PM, James wrote:
> Sounds like it, but the data handed to us didnt have sidewalks and
> roads, driveways etc. Ottawa may have exported data from this file

Yes, for sure.

I've now had more of a chance to look at the data (thanks, Ottawa, for
providing no docs at all ...). I'm pretty sure that the data at
http://data.ottawa.ca/dataset/cad-topographic-data is the source of what
the Ottawa group were given.

In the 31 gigabytes of converted files, about 8-10 of the 177 total
layers might be of interest. But:

* The files are in some kind of MTM projection, but I don't know the
datum. Some munis still love their NAD27, so getting this right is crucial.

* These were digitized 2010-2011 at the latest. Since municipalities
share data with NRCan, aren't these outlines already available in a
recent iteration of CanVec in a much more useful (i.e., anything but
DWG) format?

My notes on the files, so far:
https://gist.github.com/scruss/e7f85da2e7943cb1a1d13772fbe144d3#file-ottawabfomapdata-md

(feel free to use/modify/etc)

If anyone wants 31 GB of converted DXFs, let me know. It took Teigha
several hours on a quad core with SSDs to convert this, so I'm not going
to delete it lightly.

cheers,
 Stewart


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

[OSM-talk-fr] Tag pour constructeur de maisons ?

[pepilepi...@ovh.fr]

  
  
Bonjour,
Après mon
  angoisse sur les kinés je suis aujourd'hui confronté à un
autre problème : comment taggue-t-on un constructeur de maisons
? À savoir le gars à qui vous expliquez ce que vous voulez et
qui vous fait les plans, organise et coordonne tous les corps de
métiers et vous livre votre petite maison terminée. Pas vraiment,
d'ailleurs, un promoteur immobilier, puisque lui achète en gros
et revend au détail.
  
Dans le wiki j'ai bien vu un "office=estate_agent",
mais le "estate agent" se traduit parfaitement en français par
"agent immobilier" et ce n'est pas ça qui corespond à ce que je
veux indiquer.
Question annexe : quand on ne trouve pas
dans la nomenclature officielle ce dont on a besoin précisément
y a-t-il un moyen de préciser une approximation ? Par exemple
ici mettre office=estate_agent et un
fr:=constructeur de maisons individuelles ?
Merci,
Jean-Pierre
  
  


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

[Talk-it] Notiziario Settimanale OSM #326 11/10/2016-17/10/2016

[Stefano]

La raccolta settimanale delle notizie OSM, edizione #326, è adesso
disponibile online in Italiano, come sempre fornisce un riassunto di tutte
le cose che accadono nel mondo Openstreetmap:

*http://www.weeklyosm.eu/it/archives/8214
*

Godetevelo!

Il notiziario settimanale OSM vi è fornito da 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Languages
___
Talk-it mailing list
Talk-it@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-it

Re: [Talk-it] Ricerca affluenti di un dato torrente/fiume in Josm.

[Samuele Battarra]

In data mercoledì 19 ottobre 2016 20:45:01 CEST, girarsi_liste ha scritto:
> Quale stringa devo scrivere per permettere la ricerca di affluenti di un
> dato waterway:stream con name:* ?
> 
> Sapete darmi qualche indicazione di dove guardare sulla wiki OSM o Josm?
> 
> Finora non ho trovato nulla che mi dia l'idea per formulare la corretta
> sintassi.
> 
> Grazie a chi mi risponderà.

In JOSM se installi il plugin utilsplugin2 ti ritrovi un menù "selezione" dove 
ci trovi le voci "Tutti i percorsi connessi" che dovrebbe fare quello che 
chiedi.
https://josm.openstreetmap.de/wiki/Help/Action/SelectConnectedWays

-- 
Samuele Battarra
batta...@email.it

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

Re: [Talk-de] FOSSGIS als deutsches OSMF-Chapter

[Christoph Hormann]

On Friday 21 October 2016, Frederik Ramm wrote:
>
>der FOSSGIS e.V. würde sich gern bei der OSMF als offizielles
> "deutsches Local Chapter" bewerben. Ein Local Chapter ist eine
> meistens auf ein Land beschränkte regionale Organisation, die
> ähnliche Ziele wie die OSMF verfolgt.
>
> [...]

Ein paar Überlegungen dazu:

- der FOSSGIS hat ja zwei inhaltliche Schwerpunkte: Freie GIS-Software 
und freie Geodaten.  Im Allgemeinen ergänzen sich diese Themen auch 
recht gut und profitieren im Rahmen des Vereins voneinander.  
Allerdings muss man auch klar sagen, dass strukturell erhebliche 
Unterschiede bestehen.  Insbesondere ist OSM in Deutschland von Leuten 
geprägt, die das als Hobby betreiben, während der Bereich freier 
GIS-Software ein hohes Maß an Professionalisierung zeigt.  Es gibt vor 
diesem Hintergrund durchaus auch Konfliktlinien, die sich auftun 
könnten.  So ist die FOSSGIS-Konferenz mit freiem Eintritt für 
OSM-Aktive sehr Hobby- und OSM-freundlich, andere Aktivitäten des 
Vereins, wie zum Beispiel die diesjährige FOSS4G, sind da jedoch 
deutlich anderer Natur.  Meines Erachtens ist das OK und eine große 
Bandbreite von Aktivitäten hilft allen Seiten, nicht zu sehr im eigenen 
Saft zu schmoren und auch mal andere Teile der Welt wahrzunehmen.  
Allerdings sollte man klar sehen, dass in einem solchen Verein mit 
breiter Zielsetzung eben nicht alle am selben Strang in die selbe 
Richtung ziehen.  In diesem Sinne denke ich, dass die Anforderung 
des "Local Chapter Agreement", dass die Ziele des local chapter mit 
denen der OSMF im Einklang stehen müssen, nicht nur eine leere Phrase 
sein sollte, sondern vom FOSSGIS auch bewusst als Anforderung 
aufgenommen werden muss.

Ich könnte mir zum Beispiel vorstellen, dass es gut wäre, wenn der 
FOSSGIS in seiner Satzung über das allgemeine Ziel der Förderung freier 
Daten hinaus konkreter auch freie Daten, die wie bei OSM in einem frei 
zugänglichen und offenen Prozess produziert und gepflegt werden, mit 
aufnimmt.

- ein weiterer Punkt ist, dass im FOSSGIS viele Prozesse vereinsintern 
sind. Daran ist im Grunde nichts auszusetzen und es schafft einen 
Anreiz zu Mitgliedschaft.  Traditionell wird aber in der OSM-Community 
viel Wert auf Öffentlichkeit der Prozesse gelegt.  Da gäbe es eine 
Menge zu verbessern - ich sehe im FOSSGIS-Wiki zum Beispiel nicht mal 
alle Protokolle der Jahreshauptversammlungen.

- Der Entwurf für die Vereinbarung sieht ja vor, dass das local chapter 
öffentlich als "OpenStreetMap Deutschland" auftritt.  Wenngleich das so 
generell sicher problematisch wäre, würde ich die Idee, dass der 
OSM-Teil des FOSSGIS sich auch dediziert öffentlich präsentiert nicht 
so ohne weiteres abtun.  Wenn OSM formell immer nur als eines von ggf. 
hunderten vom FOSSGIS geförderten freien Software- und Daten-Projekten 
dargestellt wird, würde das der Bedeutung des Projektes nicht wirklich 
gerecht.

Meine Meinung ist also grundsätzlich positiv, aber nicht nur einfach 
irgendwie machen, sondern bewusst schauen, was man ändern kann und 
sollte, um der Idee einer Vertretung der deutschen OSM-Community 
gerecht zu werden.

-- 
Christoph Hormann
http://www.imagico.de/

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

Re: [Talk-dk] Oplevelsesstien i Ringsted Kommune

[Peter Leth]

Hej

Jeg har sendt nogle folk afsted her til eftermiddag med Mapillary - håber
det giver lidt data. Jeg cykler selv på ruten mandag.

/ Peter

Den 22. oktober 2016 kl. 13.00 skrev Soren Johannessen <
soren.johannes...@gmail.com>:

> Hej Peter
>
> Jeg ser at en ny del åbner 1 november. Er det ikke muligt at nogen fra
> Oplevelsesstiens venner  (http://oplevelsesstien.dk/) eller
> skoleklassen cykler turen igennem og laver et GPX spor fra en GPS app?
> (eller et GPX spor fra en fritids GPS'er a.la. Garmin) - gerne flere
> GPX fra forskellige - da vi så får lettere ved at se hvordan ruten
> kører gennem landskabet, ved at have flere GPX spor at arbejde ud fra.
>
> Der er selvfølgelig også Mapillary fotograferingsmulighed, så vi OSM
> ad den vej kan få området ind.
>
> Vh
> Søren Johannessen
>
> 2016-10-21 21:37 GMT+02:00 Peter Leth :
> > Kære alle
> >
> > Undskyld jeg forstyrrer - men jeg skal hjælpe en folk unger på en skole i
> > Ringsted på mandag med at lave et informationsmateriale om deres område
> og
> > det store projekt kaldet Oplevelsesstien.
> >
> > Området er desværre ret mangelfuldt indtegnet for tiden, og trænger
> derfor
> > til en omgang, for at være velegnet.
> >
> > Så hvis man havde en ledig stund hen over weekenden, så ville det være
> fedt
> > hvis I ville hjælpe med at tegne lidt, så stien kunne bruge OpenStreetMap
> > til kortdelen.
> >
> > Projektet er rimeligt stort for deres område, og støttet med 14 millioner
> > vil det få en del opmærksomhed, også derfor tænker jeg at det kunne være
> > fedt at vi lister os ind i projektet.
> > http://oplevelsesstien.dk/
> >
> > Eleverne som jeg skal hjælpe skal lave nogle VR-guides med 360-graders
> > billeder med interaktive elementer af tekst, billede og filmklip.
> > Kladdehæftet anno 2016.
> >
> > Undskyld forstyrrelsen.
> >
> >
> >
> >
> > --
> > Med venlig hilsen
> >
> > Peter Leth
> > pe...@pluk.dk
> > l...@creativecommons.dk
> >
> > T.
> > 51 52 23 87
> >
> > Skype. peter.leth1
> >
> >
> > ___
> > Talk-dk mailing list
> > Talk-dk@openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/talk-dk
> >
>
> ___
> Talk-dk mailing list
> Talk-dk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-dk
>



-- 
Med venlig hilsen

Peter Leth
pe...@pluk.dk
l...@creativecommons.dk

T. 9749 1180
Skype. peter.leth1
___
Talk-dk mailing list
Talk-dk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-dk

Re: [Talk-de] FOSSGIS als deutsches OSMF-Chapter

[Toni Erdmann]

On 21.10.2016 21:52, Frederik Ramm wrote:



Was sagt ihr dazu?



Ich bin dafür.

Gruß,
Toni


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

Re: [Talk-de] Fwd: [OSM-talk] OSM Forum move / disruption

[chris66]

Am 22.10.2016 um 11:41 schrieb chris66:
> Am 20.10.2016 um 23:03 schrieb Michael Kugelmann:
>> Forum Server down am Samstag wegen "Umzug"
> 
> Ja, wurde soeben abgeschaltet.

Und schon fertig, Danke.

Chris



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

semanarioOSM Nº 326 11/10/2016-17/10/2016

[weeklyteam]

Hola, el semanario Nº 326, el sumario de todo lo que está ocurriendo en el 
mundo de openstreetmap está en línea en español:

http://www.weeklyosm.eu/es/archives/8214/

¡Disfruta!

weeklyOSM en Español está producido por: 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-es mailing list
Talk-es@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-es

semanarioOSM Nº 326 11/10/2016-17/10/2016

[weeklyteam]

Hola, el semanario Nº 326, el sumario de todo lo que está ocurriendo en el 
mundo de openstreetmap está en línea en español:

http://www.weeklyosm.eu/es/archives/8214/

¡Disfruta!

weeklyOSM en Español está producido por: 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-cu mailing list
Talk-cu@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-cu

semanarioOSM Nº 326 11/10/2016-17/10/2016

[weeklyteam]

Hola, el semanario Nº 326, el sumario de todo lo que está ocurriendo en el 
mundo de openstreetmap está en línea en español:

http://www.weeklyosm.eu/es/archives/8214/

¡Disfruta!

weeklyOSM en Español está producido por: 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-cl mailing list
Talk-cl@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-cl

semanarioOSM Nº 326 11/10/2016-17/10/2016

[weeklyteam]

Hola, el semanario Nº 326, el sumario de todo lo que está ocurriendo en el 
mundo de openstreetmap está en línea en español:

http://www.weeklyosm.eu/es/archives/8214/

¡Disfruta!

weeklyOSM en Español está producido por: 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-co mailing list
Talk-co@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-co

[Talk-GB] weeklyOSM #326 10/11/2016-10/17/2016

[weeklyteam]

The weekly round-up of OSM news, issue # 326,
is now available online in English, giving as always a summary of all things 
happening in the openstreetmap world:

http://www.weeklyosm.eu/en/archives/8214/

Enjoy!

weeklyOSM is brought to you by ... 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

[Talk-in] weeklyOSM #326 10/11/2016-10/17/2016

[weeklyteam]

The weekly round-up of OSM news, issue # 326,
is now available online in English, giving as always a summary of all things 
happening in the openstreetmap world:

http://www.weeklyosm.eu/en/archives/8214/

Enjoy!

weeklyOSM is brought to you by ... 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-in mailing list
Talk-in@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-in

[Talk-ca] weeklyOSM #326 10/11/2016-10/17/2016

[weeklyteam]

The weekly round-up of OSM news, issue # 326,
is now available online in English, giving as always a summary of all things 
happening in the openstreetmap world:

http://www.weeklyosm.eu/en/archives/8214/

Enjoy!

weeklyOSM is brought to you by ... 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca

[Talk-us] weeklyOSM #326 10/11/2016-10/17/2016

[weeklyteam]

The weekly round-up of OSM news, issue # 326,
is now available online in English, giving as always a summary of all things 
happening in the openstreetmap world:

http://www.weeklyosm.eu/en/archives/8214/

Enjoy!

weeklyOSM is brought to you by ... 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

[OSM-talk] weeklyOSM #326 10/11/2016-10/17/2016

[weeklyteam]

The weekly round-up of OSM news, issue # 326,
is now available online in English, giving as always a summary of all things 
happening in the openstreetmap world:

http://www.weeklyosm.eu/en/archives/8214/

Enjoy!

weeklyOSM is brought to you by ... 
https://wiki.openstreetmap.org/wiki/WeeklyOSM#Available_Languages
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [Talk-dk] Oplevelsesstien i Ringsted Kommune

[Soren Johannessen]

Hej Peter

Jeg ser at en ny del åbner 1 november. Er det ikke muligt at nogen fra
Oplevelsesstiens venner  (http://oplevelsesstien.dk/) eller
skoleklassen cykler turen igennem og laver et GPX spor fra en GPS app?
(eller et GPX spor fra en fritids GPS'er a.la. Garmin) - gerne flere
GPX fra forskellige - da vi så får lettere ved at se hvordan ruten
kører gennem landskabet, ved at have flere GPX spor at arbejde ud fra.

Der er selvfølgelig også Mapillary fotograferingsmulighed, så vi OSM
ad den vej kan få området ind.

Vh
Søren Johannessen

2016-10-21 21:37 GMT+02:00 Peter Leth :
> Kære alle
>
> Undskyld jeg forstyrrer - men jeg skal hjælpe en folk unger på en skole i
> Ringsted på mandag med at lave et informationsmateriale om deres område og
> det store projekt kaldet Oplevelsesstien.
>
> Området er desværre ret mangelfuldt indtegnet for tiden, og trænger derfor
> til en omgang, for at være velegnet.
>
> Så hvis man havde en ledig stund hen over weekenden, så ville det være fedt
> hvis I ville hjælpe med at tegne lidt, så stien kunne bruge OpenStreetMap
> til kortdelen.
>
> Projektet er rimeligt stort for deres område, og støttet med 14 millioner
> vil det få en del opmærksomhed, også derfor tænker jeg at det kunne være
> fedt at vi lister os ind i projektet.
> http://oplevelsesstien.dk/
>
> Eleverne som jeg skal hjælpe skal lave nogle VR-guides med 360-graders
> billeder med interaktive elementer af tekst, billede og filmklip.
> Kladdehæftet anno 2016.
>
> Undskyld forstyrrelsen.
>
>
>
>
> --
> Med venlig hilsen
>
> Peter Leth
> pe...@pluk.dk
> l...@creativecommons.dk
>
> T.
> 51 52 23 87
>
> Skype. peter.leth1
>
>
> ___
> Talk-dk mailing list
> Talk-dk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-dk
>

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

Re: [OSM-talk-fr] Appli mobile pour mapper les arrêts de bus

[Florian LAINEZ]

>
> Si on propose ce que l'on sait déjà de l'arrêt et qu'on propose de
> confirmer/infirmer/ne pas se prononcer, on gagne des possibilités de
> réponse rapide.

tout à fait, je pense à de l'autocomplétion avec la meilleure proposition
mise en avant le cas échéant.

Peut-être que si on laisse la personne dessiner le trajet (juste en
> chaînant les arrêts) on a pas mal de matière.
> Ensuite un moteur de calcul de trajet (favorisant les rues larges et
> droites) peut être utilisé côté serveur pour avoir un trajet vraisemblable.
> Trajet à valider bien-sûr.

 C'est à peu près ça que j'ai en tête. Avec la possibilité de drag le
chemin pour le replacer (du style de ce que propose OSRM sur osm.org) sans
avoir à gérer des way un par un.

@philippe tout ceci est passionnant mais ce sont des problématiques
complexes, or mon appli est vraiment focalisée sur la simplicité. D'où le
parti pris de ne gérer que les highway=bus_stop (ou son équivalent
public_transport=platform).

Mon idée est bien de pouvoir gérer les deux modèles, ancien et nouveau, de
manière transparente pour l'utilisateur.
Charge aux mappeurs expérimentés de compléter le travail avec des outils
plus classiques.


Le 21 octobre 2016 à 22:26, Philippe Verdy  a écrit :

> Les tags "from=*", "via=*" et "to=*" sont plutôt destinés à un utilisateur
> humain qu'à subir un vrai rapprochement avec un arrêt.
>
> Dans le shéma actuel ("public_transport:version=2" dans les relations
> "route"), ce sont les noeuds membres (posés sur un les ways membres) ayant
> un rôle "stop" qui servent à ça, le premier noeud (from) devrait être de
> rôle "stop_enter_only", le dernier de rôle "stop_exit_only". La direction
> de la ligne devrait s'en déduire L'ennui c'est qu'il existe parfois des
> stations intermédiaires qui n'autorisent que la descente ou que la montée,
> ayant ces même rôles (c'est rare...). On n'a en fait rien de clair pour
> indiquer réellement quels arrêts (noeuds de rôle "stop") sont les deux
> terminus (l'idéal serait d'ajouter un suffixe ":start" ou ":end" à ces
> rôles)
>
> On peut ajouter aussi les plateformes (noeuds, lines ou polygones) en
> membres de rôle "platform" (avec "platform_enter_only",
> "platform_exit_only" pour le premier et le dernier arrêt, mais cela me
> semble inutile si on a identifié clairement les deux noeuds "stop" de
> départ et d'arrivée: le rôle "platform" suffit à lui seul; ces rôles
> variantes sont plutôt destinés à combler des données manquantes quand il
> manque des noeuds "stop", mais les plateformees sont pas faciles du tout à
> gérer et associer avec le parcours de la ligne sans un noeud "stop"
> correspondant).
>
> Il reste cependant le cas des lignes circulaires dont le départ et
> l'arrivée sont au même point: si ce point n'est pas sur un way en sens
> unique (par exemple une petite voie de service latérale à la chaussée), là
> on a bsoin que ce premier chemin ait un rôle "forward" ou "backward" (je ne
> vois pas comment faire autrement) pour indiquer le sens de la ligne.
> L'autre solution serait d'utiliser deux noeuds très proches mais séparés
> pour distinguer le départ et l'arrivée (mais sans mettre alors dans la
> "route" le segment qui les relie ! Ces noeuds étant cependant associés à la
> même plateforme.
>
> Les chemins sont ensuite parcours dans le sens indiqués par leur jonction,
> mais il reste la difficulté des lignes dont l'itinéraire repasse plusieurs
> fois par le même noeud (voire même par un même chemin en cas de détour, et
> pas non plus forcément en sens opposé si ce détour fait une boucle): là
> encore les chemins doivent être orientés pour réduire (avec backward ou
> forward) le nombre de possibilités de succession des chemins, mais il peut
> rester des ambiguités.
>
> Pour éliminer totalement ces ambiguïtés, il a été proposé de créer des
> relation "route" avec un tag "segment=yes", où les chemins forment une
> ligne ininterrompue et ne repassant jamais par un même noeud ou chemin,
> puis d'inclure ces segments de routes en tant que membres d'une relation
> "route" normale. Mais ce n'est pas encore mis en oeuvre.
>
> En attendant, on a encore besoin que les relations "route" mentionnent la
> liste des arrêts ("stop") dans l'ordre exact. Et il est impératif
> d'utiliser des noeuds rôle "stop" (tagués "public_transport=stop_position"
> + "bus=yes") sur les chemins. Ces noeuds n'ont aucun tag pour les abris,
> poubelles, accès handicap, etc. qui sont à mettre plutôt sur les objets
> "public_transport=platform".
>
> Le nouveau schéma recommande même de placer dans l'ordre dans la relation
> les noeuds "stop" suivis immédiatement de l'objet "platform" auquel il se
> réfère; la liste des chemins se met plutôt ensuite après (avec des chemins
> en doublons parfois, faute de prise en charge pour l'instant des "segments"
> de route).
>
> Noter que les chemins peuvent aller commencer et aller plus loin que le
> premier et le dernier arrêt (généralement on inclue dans une route aussi

Re: [Talk-de] Fwd: [OSM-talk] OSM Forum move / disruption

[chris66]

Am 20.10.2016 um 23:03 schrieb Michael Kugelmann:
> Forum Server down am Samstag wegen "Umzug"

Ja, wurde soeben abgeschaltet.

Chris





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

Re: [Talk-cz] Dotaz na mapovani silnicnich pruhu

[jzvc]

Dne 11.10.2016 v 15:35 majka napsal(a):

Mě osobně hlavně přijde merge_to_left na 1-lane trochu zbytečný, protože
při jízdě "jinam" nemůžu. U vícepruhových úseků je to dobrá informace o
tom, že jedu v pruhu který skončí. Naopak, použila bych to případně až
na ten připojovací pruh - tam platí že chci vědět, že krasojízda za
chvilku skončí.
Ale vcelku to podle mě ničemu nevadí, jen je to tag navíc.


Cus,

https://www.openstreetmap.org/way/255619335
https://openstreetmap.cz/#map=19/50.04018/14.48773=oO

V tomhle pripade, je spravny tagovani nasledujici:

1) na tom oznacenem useku (a kus pred nim) by mel by lanes = 2 a to az 
do bodu, kde je to slouceno do toho jednopruhu.
2) na useku, kde jsou slucovaci sipky (a prave jen tam) by melo byt 
merge_to_right

3) pak nasleduje kratky usek lanes = 1
4) ten kraticky pripojovak by pak mohl byt znacen, ale uz na te napojene 
silnici, jako lanes = 3

5) a prave a pouze sem patri prave to merge_to_left

Tak jak je to ted je to jednoznacne spatne. A pochopitelne plati, ze 
merge_to ... a lanes=1 nemuze byt zaroven.




Ty both_ways se u nás v praxi vyskytovat můžou - hned v prvním případě
 toho
původního odkazu, tedy obousměrné odbočení doleva přes čtyř- a vícepruh
mapovaný jako dvě cesty. Samozřejmě je variantou to rozdělit v
prostředku a mapovat každý směr zvlášť.

2016-10-11 15:11 GMT+02:00 Jan Martinec >:

Merge_to_left mi připadá logický i pro 1-lane (zejména v situacích,
kde ta lane přejde v připojovací pruh a skončí).

Both_ways je blbě - to je vyloženě pro pruhy se
střídavým/obousměrným provozem; to snad u nás ani není, nebo je?
Každopádně U Kabla je to klasická silnice "jeden tam, jeden zpět,"
tam to nepatří.

HPM


Dne 11. 10. 2016 14:20 napsal uživatel "Martin Hamrle"
>:

Zdravim,

nedavno jsem zjitil, ze osmose pridal kontrolu mapovani jizdnich
pruhu
viz.

http://osmose.openstreetmap.fr/en/map/#zoom=13=50.0668=14.4731=Mapnik=T=3160=1%2C2%2C3=highway=


Pri oprave nalezenych problemu jsem narazil na dva caste
problemy. Nez
to zacnu opravovat chtel bych se zeptat, jestli je zde shoda, jak by
to melo byt.

Prvni je caste pouziti turn:lanes=merge_to_left u jednopruhovych
najezdu napriklad zde
https://www.openstreetmap.org/way/255619335
 Na wiki je popis
merge_to_left trochu nejednoznacny ale na vsech prikladech se
pouziva
merge_to_left pouze u vicepruhovych silnic, kde lze zmenit pruh v
ramci jedne silnice. Zmapovano je to jako jako jedna way s lanes>=2.
Ostatne takto to kontroluje i osmose.

Druhy casty problem, na ktery narazim, je v pouziti lanes:both_ways.
Napriklad
https://www.openstreetmap.org/way/4779348

https://www.openstreetmap.org/way/4052868

Na wiki je pouze proposal pro tento tag
https://wiki.openstreetmap.org/wiki/Proposed_features/Suffix_both_ways

Z uvedenych prikladu se zda ale ze to s proposal nema nic
spolecneho.
Spis je to pouzito jako zkratka pro lanes:forward and lanes:backward
se shodnyma hodnotama. V techto pripadech bych rad tag nahradil
standartizovanou variantou lanes:forward a lanes:backward.

Diky za pripominky,

Martin

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



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





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




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

Re: [OSM-talk] The movie Eye in the Sky credits OpenStreetMap

[Simon Poole]

“© OpenStreetMap contributors” has been the suggested attribtion text
since day one post licence change (see
http://www.openstreetmap.org/copyright and
http://wiki.osmfoundation.org/wiki/Licence/Licence_and_Legal_FAQ ).
While you could argue that the OSMF is wrong headed to ask for that, you
defintely can't complain about the productions.

Simon

Am 22.10.2016 um 07:00 schrieb Pine W:
>
> With a budget of about $75 million [1], productions like this
> certainly have the resources to get the credits and licensing right,
> so there is no excuse for being sloppy. Does the OSM Foundation have
> lawyers or communications staff who would be in a good position to
> address this issue with the responsible parties?
>
> Pine
>
> [1]
> https://variety.com/2016/film/news/box-office-tom-hanks-inferno-international-1201890418/
>
>
> On Oct 21, 2016 9:06 PM, "Eugene Alvin Villar"  > wrote:
>
> I saw another mainstream film that credits OSM in the end credits
> and this is the film adaptation of Dan Brown's novel Inferno. The
> credit went “© OpenStreetMap contributors”. There was no mention
> of any license again.
>
> I've decided to create a new wiki page[1] to list these films,
> linked from the "Press" wiki page[2].
>
> [1] https://wiki.openstreetmap.org/wiki/Films
> 
> [2] https://wiki.openstreetmap.org/wiki/Portal:Press
> 
>
> ~Eugene
>
>
> On Fri, Jul 22, 2016 at 3:12 PM, Eugene Alvin Villar
> > wrote:
>
> Hi all,
>
> I just saw the film Eye in the Sky, starring Helen Mirren and Alan
> Rickman (the movie is really good), and I was pleasantly
> surprised to
> see OpenStreetMap and its contributors get a credit in the end
> credits. The surprise is partly because I do not recall seeing
> any map
> in the film that looked like it came from OSM.
>
> IIRC, the credit went like “OpenStreetMap © OpenStreetMap
> contributors”. No mention of any license though.
>
> Now I'm wondering whether there are any other mainstream films
> that
> also use and credit OpenStreetMap as well. This is the first
> time I've
> seen such a credit and I don't recall any mention as well on this
> mailing list.
>
> ~Eugene
>
>
>
> ___
> talk mailing list
> talk@openstreetmap.org 
> https://lists.openstreetmap.org/listinfo/talk
> 
>
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk



signature.asc
Description: OpenPGP digital signature
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk