Re: [Talk-ca] Building Canada 2020 initiative

2017-09-29 Thread James
To answer you question about who would organise the tasking manager, I'm
willing to do so.

On Fri, Sep 29, 2017 at 9:24 AM, john whelan  wrote:

> A couple of comments:
>
> 1. Pierre Beland  has identified
> 95% of contributors only map 6% of the assets.
> https://mobile.twitter.com/pierzen/status/910551645498552321  We would
> need to use those 5% of mappers who do the most mapping and they have their
> own agendas and reasons for mapping. They are more likely to throw their
> weight behind something that looks achievable and I'm not sure this is.
>
> 2. I've put up a sample of buildings being mapped from a mapathon.
> https://www.jatws.org/johnw/building3.jpg as you can see in my opinion
> the quality is not suitable for Stats Canada's use.  Some buildings are
> grouped together with others as a single building, others are mapped the
> wrong shape or size.  Quite often buildings are omitted.  There are better
> examples and there are worse examples but it is not untypical and it was
> this experience that made me suggest the Open Data import route in the
> first place.
>
> So Open Data import is better for quality.  Adding tags to building foot
> prints is less error prone.
>
> 3. There are more than 5,000 municipal governments in Canada source Stats
> Can.  It took five years to get the City of Ottawa to update their Open
> Data license.  Treasury Board still hasn't released their Open Data tool
> kit for the municipalities.  With good will I estimate it will take two
> years to get the Open Data licenses amended.  Kingston might be a good
> target.  With any questions and there will be a number, this figure can be
> expected to drift out to three to four years.  Who is responsible to answer
> questions, in both official languages?  Who will make the requests to
> municipal governments to adopt a usable Open Data license?
>
> In Ottawa we had the right mix of resources.  We had enough local mappers
> to discuss things through which is part of the import process. We had good
> will from the City of Ottawa and they were happy to release building foot
> print data which had not been part of their Open Data so far.  The import
> process is not simple these days, it would need  the steps to follow to be
> documented and then you get the technical side of the import.  I'm a fairly
> experienced mapper and to be honest I wouldn't attempt the sort of complex
> import that was done in Ottawa.  I'm not sure the Ottawa experience is
> repeatable more than five thousand times.
>
> We can approach the OSM LWG for an opinion on existing licenses but they
> are volunteers and for five thousand opinions that would take a
> considerable amount of time and if the license weren't the TB toolkit ones
> I wouldn't even bother.
>
> The community is supposed to be doing this.  Fine but a project manager
> and a project plan might make it run more smoothly.  Data quality will be
> important so how will it be verified?  Who will be responsible for
> organising task manager tiles for the whole country? Who will identify the
> group of mappers who are "local" to a small municipality?  Remember these
> have a critical decision making role to play in the import process.
>
> Have fun.
>
> Cheerio John
>
>
> On 28 September 2017 at 16:48, Alasia, Alessandro (STATCAN) <
> alessandro.ala...@canada.ca> wrote:
>
>> Hello all!
>>
>> Statistics Canada was a partner of the 2017 HOT Summit held in Ottawa,
>> Sept 14-15. In parallel to the summit, Statistics Canada hosted a workshop
>> to discuss the possibility of launching a community-led initiative titled
>> “Building Canada 2020”.
>>
>> The goal and vision of this initiative is simple: *map all buildings in
>> Canada on OSM by the year 2020*. The workshop was well attended. There
>> were about 50 people from various sectors (federal, academic, civic group,
>> and private). This was a preliminary discussion amongst a small group of
>> people, but now that broad interest has been confirmed more stakeholders
>> need to be involved!
>>
>> A short summary of the workshop along with a first draft *Roadmap to
>> implementation* has been posted on the OSM Wiki at:
>> *https://wiki.openstreetmap.org/wiki/WikiProject_Canada/Building_Canada_2020*
>> 
>>
>> Everyone that shares the vision of “mapping all buildings in Canada on
>> OSM by the year 2020” is invited to contribute to the discussion and the
>> roadmap to implementation. In addition, there is still need to discuss how
>> coordination, communication and governance of this initiative can be set in
>> place.
>>
>> One important aspect that needs to be emphasized is that “Building Canada
>> 2020” is not a Statistics Canada project. In fact, it is not properly a
>> project of any sort. It is a vision and an aspirational goal. The hope is
>> that many organizations and contributors working with an open data 

Re: [Talk-ca] Building Canada 2020 initiative

2017-09-29 Thread James
I and others can give access to people to create projects on the tasking
manager. I understand that one person might disapear, i'll try not to die
in the next 2 years ;)

On Fri, Sep 29, 2017 at 9:37 AM, john whelan  wrote:

> But that is only a single person for a project of this size you need some
> sort of team approach.  Although buses are fairly safe the odd one gets hit
> by a train and if you happen to be sitting in the front seat you may not be
> available to sort things out.
>
> Cheerio John
>
> On 29 September 2017 at 09:33, James  wrote:
>
>> To answer you question about who would organise the tasking manager, I'm
>> willing to do so.
>>
>> On Fri, Sep 29, 2017 at 9:24 AM, john whelan 
>> wrote:
>>
>>> A couple of comments:
>>>
>>> 1. Pierre Beland  has
>>> identified 95% of contributors only map 6% of the assets.
>>> https://mobile.twitter.com/pierzen/status/910551645498552321  We would
>>> need to use those 5% of mappers who do the most mapping and they have their
>>> own agendas and reasons for mapping. They are more likely to throw their
>>> weight behind something that looks achievable and I'm not sure this is.
>>>
>>> 2. I've put up a sample of buildings being mapped from a mapathon.
>>> https://www.jatws.org/johnw/building3.jpg as you can see in my opinion
>>> the quality is not suitable for Stats Canada's use.  Some buildings are
>>> grouped together with others as a single building, others are mapped the
>>> wrong shape or size.  Quite often buildings are omitted.  There are better
>>> examples and there are worse examples but it is not untypical and it was
>>> this experience that made me suggest the Open Data import route in the
>>> first place.
>>>
>>> So Open Data import is better for quality.  Adding tags to building foot
>>> prints is less error prone.
>>>
>>> 3. There are more than 5,000 municipal governments in Canada source
>>> Stats Can.  It took five years to get the City of Ottawa to update their
>>> Open Data license.  Treasury Board still hasn't released their Open Data
>>> tool kit for the municipalities.  With good will I estimate it will take
>>> two years to get the Open Data licenses amended.  Kingston might be a good
>>> target.  With any questions and there will be a number, this figure can be
>>> expected to drift out to three to four years.  Who is responsible to answer
>>> questions, in both official languages?  Who will make the requests to
>>> municipal governments to adopt a usable Open Data license?
>>>
>>> In Ottawa we had the right mix of resources.  We had enough local
>>> mappers to discuss things through which is part of the import process. We
>>> had good will from the City of Ottawa and they were happy to release
>>> building foot print data which had not been part of their Open Data so
>>> far.  The import process is not simple these days, it would need  the steps
>>> to follow to be documented and then you get the technical side of the
>>> import.  I'm a fairly experienced mapper and to be honest I wouldn't
>>> attempt the sort of complex import that was done in Ottawa.  I'm not sure
>>> the Ottawa experience is repeatable more than five thousand times.
>>>
>>> We can approach the OSM LWG for an opinion on existing licenses but they
>>> are volunteers and for five thousand opinions that would take a
>>> considerable amount of time and if the license weren't the TB toolkit ones
>>> I wouldn't even bother.
>>>
>>> The community is supposed to be doing this.  Fine but a project manager
>>> and a project plan might make it run more smoothly.  Data quality will be
>>> important so how will it be verified?  Who will be responsible for
>>> organising task manager tiles for the whole country? Who will identify the
>>> group of mappers who are "local" to a small municipality?  Remember these
>>> have a critical decision making role to play in the import process.
>>>
>>> Have fun.
>>>
>>> Cheerio John
>>>
>>>
>>> On 28 September 2017 at 16:48, Alasia, Alessandro (STATCAN) <
>>> alessandro.ala...@canada.ca> wrote:
>>>
 Hello all!

 Statistics Canada was a partner of the 2017 HOT Summit held in Ottawa,
 Sept 14-15. In parallel to the summit, Statistics Canada hosted a workshop
 to discuss the possibility of launching a community-led initiative titled
 “Building Canada 2020”.

 The goal and vision of this initiative is simple: *map all buildings
 in Canada on OSM by the year 2020*. The workshop was well attended.
 There were about 50 people from various sectors (federal, academic, civic
 group, and private). This was a preliminary discussion amongst a small
 group of people, but now that broad interest has been confirmed more
 stakeholders need to be involved!

 A short summary of the workshop along with a first draft *Roadmap to
 implementation* has been posted on the OSM Wiki at:
 

Re: [Talk-ca] Stats Canada building project

2017-09-29 Thread Matthew Darwin

Is there anyone on point to drive this activity?


On 2017-09-28 06:58 AM, john whelan wrote:

Looks like we need to talk nicely to Open-Ouvert.

Thanks John

On 27 September 2017 at 21:54, Stewart C. Russell > wrote:


On 2017-09-27 07:00 PM, john whelan wrote:
> No we need to persuade the municipalities to move to the new
standard
> license in the TB kit

Is this initiative published anywhere, John? I virtually
attended the
conference it was supposed to be announced at, and all there is is
Jean-Noé's announcement:
http://open.canada.ca/en/blog/coming-soon-do-it-yourself-open-data-toolkit


I also don't remember any consultation on what it was going to
look like.

 Stewart

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





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


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


Re: [Talk-ca] Building Canada 2020 initiative

2017-09-29 Thread john whelan
But that is only a single person for a project of this size you need some
sort of team approach.  Although buses are fairly safe the odd one gets hit
by a train and if you happen to be sitting in the front seat you may not be
available to sort things out.

Cheerio John

On 29 September 2017 at 09:33, James  wrote:

> To answer you question about who would organise the tasking manager, I'm
> willing to do so.
>
> On Fri, Sep 29, 2017 at 9:24 AM, john whelan 
> wrote:
>
>> A couple of comments:
>>
>> 1. Pierre Beland  has identified
>> 95% of contributors only map 6% of the assets.
>> https://mobile.twitter.com/pierzen/status/910551645498552321  We would
>> need to use those 5% of mappers who do the most mapping and they have their
>> own agendas and reasons for mapping. They are more likely to throw their
>> weight behind something that looks achievable and I'm not sure this is.
>>
>> 2. I've put up a sample of buildings being mapped from a mapathon.
>> https://www.jatws.org/johnw/building3.jpg as you can see in my opinion
>> the quality is not suitable for Stats Canada's use.  Some buildings are
>> grouped together with others as a single building, others are mapped the
>> wrong shape or size.  Quite often buildings are omitted.  There are better
>> examples and there are worse examples but it is not untypical and it was
>> this experience that made me suggest the Open Data import route in the
>> first place.
>>
>> So Open Data import is better for quality.  Adding tags to building foot
>> prints is less error prone.
>>
>> 3. There are more than 5,000 municipal governments in Canada source Stats
>> Can.  It took five years to get the City of Ottawa to update their Open
>> Data license.  Treasury Board still hasn't released their Open Data tool
>> kit for the municipalities.  With good will I estimate it will take two
>> years to get the Open Data licenses amended.  Kingston might be a good
>> target.  With any questions and there will be a number, this figure can be
>> expected to drift out to three to four years.  Who is responsible to answer
>> questions, in both official languages?  Who will make the requests to
>> municipal governments to adopt a usable Open Data license?
>>
>> In Ottawa we had the right mix of resources.  We had enough local mappers
>> to discuss things through which is part of the import process. We had good
>> will from the City of Ottawa and they were happy to release building foot
>> print data which had not been part of their Open Data so far.  The import
>> process is not simple these days, it would need  the steps to follow to be
>> documented and then you get the technical side of the import.  I'm a fairly
>> experienced mapper and to be honest I wouldn't attempt the sort of complex
>> import that was done in Ottawa.  I'm not sure the Ottawa experience is
>> repeatable more than five thousand times.
>>
>> We can approach the OSM LWG for an opinion on existing licenses but they
>> are volunteers and for five thousand opinions that would take a
>> considerable amount of time and if the license weren't the TB toolkit ones
>> I wouldn't even bother.
>>
>> The community is supposed to be doing this.  Fine but a project manager
>> and a project plan might make it run more smoothly.  Data quality will be
>> important so how will it be verified?  Who will be responsible for
>> organising task manager tiles for the whole country? Who will identify the
>> group of mappers who are "local" to a small municipality?  Remember these
>> have a critical decision making role to play in the import process.
>>
>> Have fun.
>>
>> Cheerio John
>>
>>
>> On 28 September 2017 at 16:48, Alasia, Alessandro (STATCAN) <
>> alessandro.ala...@canada.ca> wrote:
>>
>>> Hello all!
>>>
>>> Statistics Canada was a partner of the 2017 HOT Summit held in Ottawa,
>>> Sept 14-15. In parallel to the summit, Statistics Canada hosted a workshop
>>> to discuss the possibility of launching a community-led initiative titled
>>> “Building Canada 2020”.
>>>
>>> The goal and vision of this initiative is simple: *map all buildings in
>>> Canada on OSM by the year 2020*. The workshop was well attended. There
>>> were about 50 people from various sectors (federal, academic, civic group,
>>> and private). This was a preliminary discussion amongst a small group of
>>> people, but now that broad interest has been confirmed more stakeholders
>>> need to be involved!
>>>
>>> A short summary of the workshop along with a first draft *Roadmap to
>>> implementation* has been posted on the OSM Wiki at:
>>> *https://wiki.openstreetmap.org/wiki/WikiProject_Canada/Building_Canada_2020*
>>> 
>>>
>>> Everyone that shares the vision of “mapping all buildings in Canada on
>>> OSM by the year 2020” is invited to contribute to the discussion and
>>> the roadmap to implementation. In 

Re: [Talk-ca] Building Canada 2020 initiative

2017-09-29 Thread john whelan
A couple of comments:

1. Pierre Beland  has identified
95% of contributors only map 6% of the assets.
https://mobile.twitter.com/pierzen/status/910551645498552321  We would need
to use those 5% of mappers who do the most mapping and they have their own
agendas and reasons for mapping. They are more likely to throw their weight
behind something that looks achievable and I'm not sure this is.

2. I've put up a sample of buildings being mapped from a mapathon.
https://www.jatws.org/johnw/building3.jpg as you can see in my opinion the
quality is not suitable for Stats Canada's use.  Some buildings are grouped
together with others as a single building, others are mapped the wrong
shape or size.  Quite often buildings are omitted.  There are better
examples and there are worse examples but it is not untypical and it was
this experience that made me suggest the Open Data import route in the
first place.

So Open Data import is better for quality.  Adding tags to building foot
prints is less error prone.

3. There are more than 5,000 municipal governments in Canada source Stats
Can.  It took five years to get the City of Ottawa to update their Open
Data license.  Treasury Board still hasn't released their Open Data tool
kit for the municipalities.  With good will I estimate it will take two
years to get the Open Data licenses amended.  Kingston might be a good
target.  With any questions and there will be a number, this figure can be
expected to drift out to three to four years.  Who is responsible to answer
questions, in both official languages?  Who will make the requests to
municipal governments to adopt a usable Open Data license?

In Ottawa we had the right mix of resources.  We had enough local mappers
to discuss things through which is part of the import process. We had good
will from the City of Ottawa and they were happy to release building foot
print data which had not been part of their Open Data so far.  The import
process is not simple these days, it would need  the steps to follow to be
documented and then you get the technical side of the import.  I'm a fairly
experienced mapper and to be honest I wouldn't attempt the sort of complex
import that was done in Ottawa.  I'm not sure the Ottawa experience is
repeatable more than five thousand times.

We can approach the OSM LWG for an opinion on existing licenses but they
are volunteers and for five thousand opinions that would take a
considerable amount of time and if the license weren't the TB toolkit ones
I wouldn't even bother.

The community is supposed to be doing this.  Fine but a project manager and
a project plan might make it run more smoothly.  Data quality will be
important so how will it be verified?  Who will be responsible for
organising task manager tiles for the whole country? Who will identify the
group of mappers who are "local" to a small municipality?  Remember these
have a critical decision making role to play in the import process.

Have fun.

Cheerio John


On 28 September 2017 at 16:48, Alasia, Alessandro (STATCAN) <
alessandro.ala...@canada.ca> wrote:

> Hello all!
>
> Statistics Canada was a partner of the 2017 HOT Summit held in Ottawa,
> Sept 14-15. In parallel to the summit, Statistics Canada hosted a workshop
> to discuss the possibility of launching a community-led initiative titled
> “Building Canada 2020”.
>
> The goal and vision of this initiative is simple: *map all buildings in
> Canada on OSM by the year 2020*. The workshop was well attended. There
> were about 50 people from various sectors (federal, academic, civic group,
> and private). This was a preliminary discussion amongst a small group of
> people, but now that broad interest has been confirmed more stakeholders
> need to be involved!
>
> A short summary of the workshop along with a first draft *Roadmap to
> implementation* has been posted on the OSM Wiki at:
> *https://wiki.openstreetmap.org/wiki/WikiProject_Canada/Building_Canada_2020*
> 
>
> Everyone that shares the vision of “mapping all buildings in Canada on OSM
> by the year 2020” is invited to contribute to the discussion and the
> roadmap to implementation. In addition, there is still need to discuss how
> coordination, communication and governance of this initiative can be set in
> place.
>
> One important aspect that needs to be emphasized is that “Building Canada
> 2020” is not a Statistics Canada project. In fact, it is not properly a
> project of any sort. It is a vision and an aspirational goal. The hope is
> that many organizations and contributors working with an open data resource
> (OSM) can coordinate their efforts through a multitude of projects,
> initiatives, and activities towards a common goal that would benefit
> society at large.
>
> My team at Statistics Canada (DEIL) has been working on a pilot project
> with OSM (which was presented at the HOT Summit). We are 

Re: [Talk-ca] Building Canada 2020 initiative

2017-09-29 Thread Matthew Darwin
Another good reason to create an organization that holds the keys to 
key OSM resources... then it can out-survive all of us. :-)



On 2017-09-29 09:41 AM, James wrote:
I and others can give access to people to create projects on the 
tasking manager. I understand that one person might disapear, i'll 
try not to die in the next 2 years ;)


On Fri, Sep 29, 2017 at 9:37 AM, john whelan > wrote:


But that is only a single person for a project of this size you
need some sort of team approach.  Although buses are fairly safe
the odd one gets hit by a train and if you happen to be sitting
in the front seat you may not be available to sort things out.

Cheerio John

On 29 September 2017 at 09:33, James > wrote:

To answer you question about who would organise the tasking
manager, I'm willing to do so.

On Fri, Sep 29, 2017 at 9:24 AM, john whelan
> wrote:

A couple of comments:

1. Pierre Beland**
 has identified
95% of contributors only map 6% of the assets.
https://mobile.twitter.com/pierzen/status/910551645498552321

We would need to use those 5% of mappers who do the most
mapping and they have their own agendas and reasons for
mapping. They are more likely to throw their weight
behind something that looks achievable and I'm not sure
this is.

2. I've put up a sample of buildings being mapped from a
mapathon. https://www.jatws.org/johnw/building3.jpg
 as you can
see in my opinion the quality is not suitable for Stats
Canada's use.  Some buildings are grouped together with
others as a single building, others are mapped the wrong
shape or size.  Quite often buildings are omitted. 
There are better examples and there are worse examples

but it is not untypical and it was this experience that
made me suggest the Open Data import route in the first
place.

So Open Data import is better for quality.  Adding tags
to building foot prints is less error prone.

3. There are more than 5,000 municipal governments in
Canada source Stats Can.  It took five years to get the
City of Ottawa to update their Open Data license. 
Treasury Board still hasn't released their Open Data

tool kit for the municipalities.  With good will I
estimate it will take two years to get the Open Data
licenses amended.  Kingston might be a good target. 
With any questions and there will be a number, this

figure can be expected to drift out to three to four
years.  Who is responsible to answer questions, in both
official languages?  Who will make the requests to
municipal governments to adopt a usable Open Data license?

In Ottawa we had the right mix of resources.  We had
enough local mappers to discuss things through which is
part of the import process. We had good will from the
City of Ottawa and they were happy to release building
foot print data which had not been part of their Open
Data so far.  The import process is not simple these
days, it would need  the steps to follow to be
documented and then you get the technical side of the
import.  I'm a fairly experienced mapper and to be
honest I wouldn't attempt the sort of complex import
that was done in Ottawa.  I'm not sure the Ottawa
experience is repeatable more than five thousand times.

We can approach the OSM LWG for an opinion on existing
licenses but they are volunteers and for five thousand
opinions that would take a considerable amount of time
and if the license weren't the TB toolkit ones I
wouldn't even bother.

The community is supposed to be doing this.  Fine but a
project manager and a project plan might make it run
more smoothly.  Data quality will be important so how
will it be verified?  Who will be responsible for
organising task manager tiles for the whole country? Who
will identify the group of mappers who are "local" to a
small municipality?  Remember these have a critical
decision making role to play in the import process.

Have fun.

Cheerio John


On 28 

Re: [Talk-ca] Building Canada 2020 initiative

2017-09-29 Thread Pierre Béland
J'étais au SOTM France en juin. On voit là une organisation forte, très 
développée avec des membres très engagés dans les différentes régions de 
France. Ils ont développé des collaborations avec les autorités locales, ils 
ont des serveurs, offrent différents services tels Osmose, tuiles pour le style 
Humanitaire OSM, serveur Overpass, etc. Dans un tel cas, cela a du sens d'avoir 
une organisation plus structurée. Et des membres capables de s'impliquer et 
être représentatifs des différentes régions.
Mais au Canada, quelle est la situation exactement? Si on se fie aux 
discussions sur talk-ca, on constate peu d'activités. Au Québec entre autres, 
les collaborateurs de Montréal et Québec préfèrent discuter sur leur propre 
groupe de discussion local.
Il me semble prématuré d'incorporer une ONG qui se prétendra pan-canadienne. 
Sinon quelle sera la représentativité de cette ONG, qui la contrôlera ? 

Je penses qu'il faut d'abord développer des organisations locales. Celles-ci 
pourraient ensuite voir l'opportunité de se fédérer au niveau canadien.
Quelques individus ne peuvent pas s'incorporer en utilisant le signe OSM et 
prétendre représentation la communauté OSM du Canada, récolter des fonds au nom 
de la communauté.
Statistique Canada fait un bel énoncé de principe, très évocateur, «Tracer tous 
les immeubles d'ici 2020». Par contre, StatCan n'est pas prêt à s'engager 
malgré ses moyens financiers et expertise. 

Je suis d'accord avec John que c'est un projet irréaliste, que l'on ne pourra 
pas rapidement négocier avec les municipalités pour les licences de données 
ouvertes. Et en avons-nous les moyens?
Et il faudrait plus que cela pour justifier de créer une ONG pan-canadienne.  

 Pierre 


  De : Matthew Darwin 
 À : talk-ca@openstreetmap.org 
 Envoyé le : vendredi 29 Septembre 2017 10h06
 Objet : Re: [Talk-ca] Building Canada 2020 initiative
   
 Another good reason to create an organization that holds the keys to key OSM 
resources... then it can out-survive all of us. :-) 
   ___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca


Re: [Talk-ca] Mapping of bilingual destination signs

2017-09-29 Thread john whelan
Whilst I think about it Ottawa is an amalgam of smaller municipalities so
is slowly changing street names to avoid duplicates.  I seem to recall an
employee in the street naming bit is adjusting street names in OSM.  So
please do not change a street name to match a photo that might have been
taken some time ago.

In Quebec I understand province wide the standard for names on maps is "Rue
xyz" in Ontario it is left to the municipality whether to capitalise the
first letter or not so you need to know the rules for each municipality.

Have fun

Cheerio John

On 29 Sep 2017 4:20 pm, "john whelan"  wrote:

> Ottawa is one of the few places that has bilingual street names.
>
> On the same street I've seen just the name, name street and rue name
> street signs.
>
> In Ottawa the majority are Slater street in name then rue Slater in
> name:french.
>
> Anything else means it is difficult to search for the name electronically.
>  "rue Slater Street"  is not easy to enter.
>
> Note for Ottawa it is rue Slater not Rue Slater.  Other places such as
> Quebec may have different rules.
>
> Cheerio John
> .
>
> On 29 Sep 2017 4:10 pm, "Martijn van Exel"  wrote:
>
>> Hi all,
>>
>> How do you map bilingual signposts? Ones that say for example 'Rue Regent
>> St'?
>> My thought would be destination:street=[name in primary language for the
>> province] and destination:street:en / destination:street:fr for the name in
>> the other language. But I've also seen just 'destination:street:Rue Regent
>> St'.
>>
>> My team would like to help make this consistent if you're up for that,
>> but what should be the convention? From a machine parsing perspective,
>> separating out the languages in separate tags is preferable.
>>
>> We have a ticket for this question as well, https://github.com/Telen
>> avMapping/mapping-projects/issues/27
>>
>> Thanks / Merci
>> Martijn
>>
>> ___
>> Talk-ca mailing list
>> Talk-ca@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk-ca
>>
>>
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca


Re: [Talk-ca] Mapping of bilingual destination signs

2017-09-29 Thread john whelan
OSMand will show street names in Ottawa in either English or French as will
Maperitive but I don't recall seeing destination:street=Rue Regent

Cheerio John

On 29 Sep 2017 4:23 pm, "James"  wrote:

Also do cosumers use destination:street:[lang code]? Is it a standard or
just invented?

That seems to be in the Montreal area(guessing) so i'd put
destination:street=Rue Regent
destination:street:en=Regent Street

just like street names.
Name=default language and then you can addon name:fr or name:en


On Sep 29, 2017 4:17 PM, "James"  wrote:

Rue Regent St is incorrect. Rue Regent is french and Regent Street would be
English. Just the fact that Street is shorten to St is wrong as osm fully
expands names

On Sep 29, 2017 4:10 PM, "Martijn van Exel"  wrote:

> Hi all,
>
> How do you map bilingual signposts? Ones that say for example 'Rue Regent
> St'?
> My thought would be destination:street=[name in primary language for the
> province] and destination:street:en / destination:street:fr for the name in
> the other language. But I've also seen just 'destination:street:Rue Regent
> St'.
>
> My team would like to help make this consistent if you're up for that, but
> what should be the convention? From a machine parsing perspective,
> separating out the languages in separate tags is preferable.
>
> We have a ticket for this question as well, https://github.com/Telen
> avMapping/mapping-projects/issues/27
>
> Thanks / Merci
> Martijn
>
> ___
> Talk-ca mailing list
> Talk-ca@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-ca
>
>

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


[Talk-ca] Brève analyse, Statistiques des contributeurs OSM

2017-09-29 Thread Pierre Béland
Dans la chronique [Talk-ca] Building Canada 2020 initiative, John Whelan 
faisait mention ce matin de mon analyse twitter des contributeurs OSM. Tout 
cela en 140 caractères!
voir https://twitter.com/pascal_n/status/893573996364673024

Ce ne sont là que des données préliminaires. Je suis à compléter l'analyse et 
publierai ensuite sur OSM.
 Cette brève analyse était en réponse à un tweet de Pascal Neis sur le nombre 
de contributeurs depuis 2004 selon le nombre de changesets créés (sessions 
d'édition). On fait souvent référence à ces grands chiffres cumulant le nombre 
de contributeurs depuis 2004 avoisinant le million. On les retrouve aussi sur 
la page wiki des statistiques OSM. 

On mentionne aussi qu'il y a de grands flux d'entrée et de sortie et j'ai voulu 
montrer cette autre perspective, faire ressortir le grands flux d'entrée et de 
sortie des contributeurs OSM. 

Deux problèmes lorsque l'on veut examiner les contributions OSM avec ces 
données cumulatives. Si on compte le nombre de personnes qui ont eu un emploi 
au Canada depuis 2004, peux-t-on dire Il y xxx millions de travailleurs au 
Canada. Ensuite le profil des activités. On  compte ici des contributeurs à 
temps plein, à temps partiel et très pontuels (ie une heure, une journée maxi). 

D'un côté, il y a des contributeurs «plus permanents» qui traitent de grands 
volumes de données, d'autres qui contribuent régulièrement «à temps partiel» 
pour leur activité préférée tel vélo, randonnée, et d'autres qui ont une 
intervention ponctuelle et ajoutent leur restaurant préféré ou participent un 
seul soir à un mapathon et ne reviennent pas par la suite.
Si on  regarde les «gros chiffres» de contributeurs de plus près, on constate 
de très grands flux de nouveaux contributeurs, la très grande majorité de 
participant que quelques jours ou n'ajoutent que quelques données. 

De 2004 à août 2017,on compte 944,000 contributeurs qui ont édité au moins une 
node. Parmi eux, 895,000 ont créé moins de 100 changesets. Il y a des nouveaux 
qui sauvegardent après chaque immeuble ou POI créé. C'est donc très peu. Ils 
représentent 95% des contributeurs mais uniquement 6% des objets édités. En 
moyenne, ils ont chacun édités, durant tout leur séjour comme contributeur OSM, 
100 objets (ie. node, chemin ou relation).  Cela représente 500 POI ou encore 
100 maisons. En comparaison, les 41,500 contributeurs avcec 100 à  1,500 
changesets (Casual Mapper) ont édité en moyenne 140,000 objets.  

Puis ensuite les très gros contributeurs, les BOTS, les comptes d'importation 
avec eux de très gros chiffres. Un compte Import US pour les données Tiger a 
édité à lui seul en 2007 147,642,297 d'objets (ie. node, chemin ou relation) en 
159 jours. Un bot du DWG a ensuite enlevé une bonne partie de ces objets par la 
suite.  

Depuis 2004, seuls 50,000 contributeurs à travers la planète ont créé plus de 
100 changesets.  Le graphique des contributeurs par mois sur la page wiki 
https://wiki.openstreetmap.org/wiki/Stats#Active_contributors_per_month nous 
indique que de 30,000 à 40,000 contributeurs sont actifs à chaque mois.  Encore 
ici, une analyse des flux montre qu'une grande majorité de ces contributions 
sont ponctuelles.. La forte hausse en 2016 était dû en bonne partie à des 
comptes Maps.Me où des usagers de l'application ajoutaient leur maison 
personnelle etc
Et tout comme Statistique Canada, il nous faudrait affiner nos statistiques 
mensuelles de participation et distinguer selon le niveau d'activité (ie temps 
plein, temps partiel,  contribution ponctuelle, etc). 
Pierre 
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca


Re: [Talk-ca] Mapping of bilingual destination signs

2017-09-29 Thread Pierre Béland
Les différentes provinces ou états ont souvent un organisme responsable de 
faire l'inventaire des noms officiels. Au Québec,  c'est la Commission de 
toponymie qui est responsable.http://www.toponymie.gouv.qc.ca/ct/accueil.aspx

Sur leur site, on retrouve des listes de noms et les règles qui s'appliquent 
pour les noms au Québec. 
Pour les règles, voir 
http://www.toponymie.gouv.qc.ca/ct/normes-procedures/regles-ecriture/

Les noms affichés sur Geobase.ca correspondent souvent à ces règles puisque les 
données de Ressources naturelles Canada sont fournies par les provinces. Par 
contre, il peut y avoir un certain retard lors de modifications de noms. Dans 
la section Fournisseurs d'image de JOSM, on retrouve un lien vers la couche RRN 
de Geobase. Les données sont aussi disponibles par province en 
shapefile.http://ouvert.canada.ca/data/fr/dataset/3d282116-e556-400c-9306-ca1a3cada77f

cordialement 
Pierre 


  De : john whelan 
 À : Martijn van Exel  
Cc : Talk-CA OpenStreetMap 
 Envoyé le : vendredi 29 Septembre 2017 16h52
 Objet : Re: [Talk-ca] Mapping of bilingual destination signs
   
Whilst I think about it Ottawa is an amalgam of smaller municipalities so is 
slowly changing street names to avoid duplicates.  I seem to recall an employee 
in the street naming bit is adjusting street names in OSM.  So please do not 
change a street name to match a photo that might have been taken some time ago.
In Quebec I understand province wide the standard for names on maps is "Rue 
xyz" in Ontario it is left to the municipality whether to capitalise the first 
letter or not so you need to know the rules for each municipality.
Have fun
Cheerio John
On 29 Sep 2017 4:20 pm, "john whelan"  wrote:

Ottawa is one of the few places that has bilingual street names.
On the same street I've seen just the name, name street and rue name street 
signs.
In Ottawa the majority are Slater street in name then rue Slater in 
name:french. 
Anything else means it is difficult to search for the name electronically.  
"rue Slater Street"  is not easy to enter.
Note for Ottawa it is rue Slater not Rue Slater.  Other places such as Quebec 
may have different rules.
Cheerio John    .
On 29 Sep 2017 4:10 pm, "Martijn van Exel"  wrote:

Hi all, 
How do you map bilingual signposts? Ones that say for example 'Rue Regent 
St'?My thought would be destination:street=[name in primary language for the 
province] and destination:street:en / destination:street:fr for the name in the 
other language. But I've also seen just 'destination:street:Rue Regent St'.
My team would like to help make this consistent if you're up for that, but what 
should be the convention? From a machine parsing perspective, separating out 
the languages in separate tags is preferable.
We have a ticket for this question as well, https://github.com/Telen 
avMapping/mapping-projects/ issues/27
Thanks / MerciMartijn
__ _
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.or g/listinfo/talk-ca



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


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


[Talk-ca] Mapping of bilingual destination signs

2017-09-29 Thread Martijn van Exel
Hi all,

How do you map bilingual signposts? Ones that say for example 'Rue Regent
St'?
My thought would be destination:street=[name in primary language for the
province] and destination:street:en / destination:street:fr for the name in
the other language. But I've also seen just 'destination:street:Rue Regent
St'.

My team would like to help make this consistent if you're up for that, but
what should be the convention? From a machine parsing perspective,
separating out the languages in separate tags is preferable.

We have a ticket for this question as well,
https://github.com/TelenavMapping/mapping-projects/issues/27

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


Re: [Talk-ca] Mapping of bilingual destination signs

2017-09-29 Thread James
Also do cosumers use destination:street:[lang code]? Is it a standard or
just invented?

That seems to be in the Montreal area(guessing) so i'd put
destination:street=Rue Regent
destination:street:en=Regent Street

just like street names.
Name=default language and then you can addon name:fr or name:en


On Sep 29, 2017 4:17 PM, "James"  wrote:

Rue Regent St is incorrect. Rue Regent is french and Regent Street would be
English. Just the fact that Street is shorten to St is wrong as osm fully
expands names

On Sep 29, 2017 4:10 PM, "Martijn van Exel"  wrote:

> Hi all,
>
> How do you map bilingual signposts? Ones that say for example 'Rue Regent
> St'?
> My thought would be destination:street=[name in primary language for the
> province] and destination:street:en / destination:street:fr for the name in
> the other language. But I've also seen just 'destination:street:Rue Regent
> St'.
>
> My team would like to help make this consistent if you're up for that, but
> what should be the convention? From a machine parsing perspective,
> separating out the languages in separate tags is preferable.
>
> We have a ticket for this question as well, https://github.com/Telen
> avMapping/mapping-projects/issues/27
>
> Thanks / Merci
> Martijn
>
> ___
> Talk-ca mailing list
> Talk-ca@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-ca
>
>
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca


Re: [Talk-ca] Mapping of bilingual destination signs

2017-09-29 Thread john whelan
Ottawa is one of the few places that has bilingual street names.

On the same street I've seen just the name, name street and rue name street
signs.

In Ottawa the majority are Slater street in name then rue Slater in
name:french.

Anything else means it is difficult to search for the name electronically.
 "rue Slater Street"  is not easy to enter.

Note for Ottawa it is rue Slater not Rue Slater.  Other places such as
Quebec may have different rules.

Cheerio John
.

On 29 Sep 2017 4:10 pm, "Martijn van Exel"  wrote:

> Hi all,
>
> How do you map bilingual signposts? Ones that say for example 'Rue Regent
> St'?
> My thought would be destination:street=[name in primary language for the
> province] and destination:street:en / destination:street:fr for the name in
> the other language. But I've also seen just 'destination:street:Rue Regent
> St'.
>
> My team would like to help make this consistent if you're up for that, but
> what should be the convention? From a machine parsing perspective,
> separating out the languages in separate tags is preferable.
>
> We have a ticket for this question as well, https://github.com/
> TelenavMapping/mapping-projects/issues/27
>
> Thanks / Merci
> Martijn
>
> ___
> Talk-ca mailing list
> Talk-ca@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-ca
>
>
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca


Re: [Talk-ca] Mapping of bilingual destination signs

2017-09-29 Thread James
Rue Regent St is incorrect. Rue Regent is french and Regent Street would be
English. Just the fact that Street is shorten to St is wrong as osm fully
expands names

On Sep 29, 2017 4:10 PM, "Martijn van Exel"  wrote:

> Hi all,
>
> How do you map bilingual signposts? Ones that say for example 'Rue Regent
> St'?
> My thought would be destination:street=[name in primary language for the
> province] and destination:street:en / destination:street:fr for the name in
> the other language. But I've also seen just 'destination:street:Rue Regent
> St'.
>
> My team would like to help make this consistent if you're up for that, but
> what should be the convention? From a machine parsing perspective,
> separating out the languages in separate tags is preferable.
>
> We have a ticket for this question as well, https://github.com/
> TelenavMapping/mapping-projects/issues/27
>
> Thanks / Merci
> Martijn
>
> ___
> Talk-ca mailing list
> Talk-ca@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-ca
>
>
___
Talk-ca mailing list
Talk-ca@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-ca