Re: [OSM-talk] Map features page on wiki

[Matthijs Melissen]

On Thursday, 1 September 2016, Jochen Topf  wrote:

> On Do, Sep 01, 2016 at 12:42:10 +0200, Matthijs Melissen wrote:
> > We have currently a Map Features page on the wiki:
> > http://wiki.openstreetmap.org/wiki/Map_Features
> >
> > The page also contains definitions for all features. We therefore
> > store the definitions now in two places: in the map features tables
> > and in the infoboxes on the pages themselves.
> >
>
> See https://wiki.openstreetmap.org/wiki/Taginfo/Taglists . This "just"
> needs some people going through every tag list in the wiki and replacing
> it by the special syntax. The problem ist, as you say, the differing
> descriptions and such, which need to be consolidated in the process.


Great, that's exactly what I needed. Will see what we can do with that.

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

Re: [OSM-talk] Analysis of usage of similar tags over time

[Oleksiy Muzalyev]

On 01/09/16 00:15, Matthijs Melissen wrote:

Hi all,

Recently, a new tool was created by Martin Raifer (@tyrasd) to
generate graphs of the usage of tags over time. This is a great tool
that gives us more insight in what drives the choice of tags by
mappers. The tool can be found at http://taghistory.raifer.tech/.

I wrote an OSM diary to discuss some results:
http://www.openstreetmap.org/user/Math1985/diary/39404#comment35887

-- Matthijs

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


I entered

man_made=tower

man_made=mast

amenity=atm

amenity=police

etc.

in this tool and it showed graphic lines in different color. Amazing.


However I have got yet some doubts as one usually does about a new tool. 
Could it be that there are only 104000 ATMs mapped in the whole world? I 
mean, - is it really the data from the OSM database? Is it retrieved via 
the Overpass API or from a tool's own database? If the latter how often 
it is synchronized with the live OSM database?



One more thing: in JOSM when one selects a key there is a readily 
available drop down list of values for this key. And even when one 
starts typing the key there are suggestions based on first letters 
entered. It would be great if Martin Raifer could add a similar 
functionality to the page. It is possible to do with jQuerry 
Autocomplete feature [1]. Otherwise I have to switch to JOSM back and 
forth to look up the exact key&value pairs spelling. Could you please 
forward to Martin this suggestion?



[1] https://jqueryui.com/autocomplete/


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

Re: [OSM-talk] Map features page on wiki

[François Lacombe]

Hi Matthijs,

+1 to keep Map Feature pages up to date.

It would be great to sync definitions with info boxes (with some template
tricks but it more looks like wiki-data than simple wikimedia templates)
TagInfo already got it from wiki

All the best
Francois
Le 1 sept. 2016 12:45 AM, "Matthijs Melissen"  a
écrit :

> Hi,
>
> We have currently a Map Features page on the wiki:
> http://wiki.openstreetmap.org/wiki/Map_Features
>
> The page also contains definitions for all features. We therefore
> store the definitions now in two places: in the map features tables
> and in the infoboxes on the pages themselves.
>
> Duplication of definitions seems not ideal to me. Even though a lot of
> people try to update this, there are still quite a lot differences
> between the definitions on the map feature pages and the definitions
> in the infoboxes.
>
> Do we want to keep the Map Features page? If yes, do we have technical
> means to keep the definitions synchronised? Could we perhaps generate
> it from Taginfo, or somehow include the definition from the Infobox on
> the Map Features page?
>
> -- Matthijs
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Map features page on wiki

[Jochen Topf]

On Do, Sep 01, 2016 at 12:42:10 +0200, Matthijs Melissen wrote:
> We have currently a Map Features page on the wiki:
> http://wiki.openstreetmap.org/wiki/Map_Features
> 
> The page also contains definitions for all features. We therefore
> store the definitions now in two places: in the map features tables
> and in the infoboxes on the pages themselves.
> 
> Duplication of definitions seems not ideal to me. Even though a lot of
> people try to update this, there are still quite a lot differences
> between the definitions on the map feature pages and the definitions
> in the infoboxes.
> 
> Do we want to keep the Map Features page? If yes, do we have technical
> means to keep the definitions synchronised? Could we perhaps generate
> it from Taginfo, or somehow include the definition from the Infobox on
> the Map Features page?

A solution using Taginfo has been available for about a year now, but
nobody took this up. (I guess because not many people know that this
exists, and I didn't have time to promote it really.)

See https://wiki.openstreetmap.org/wiki/Taginfo/Taglists . This "just"
needs some people going through every tag list in the wiki and replacing
it by the special syntax. The problem ist, as you say, the differing
descriptions and such, which need to be consolidated in the process.

Jochen
-- 
Jochen Topf  joc...@remote.org  http://www.jochentopf.com/  +49-351-31778688

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

[OSM-talk] Map features page on wiki

[Matthijs Melissen]

Hi,

We have currently a Map Features page on the wiki:
http://wiki.openstreetmap.org/wiki/Map_Features

The page also contains definitions for all features. We therefore
store the definitions now in two places: in the map features tables
and in the infoboxes on the pages themselves.

Duplication of definitions seems not ideal to me. Even though a lot of
people try to update this, there are still quite a lot differences
between the definitions on the map feature pages and the definitions
in the infoboxes.

Do we want to keep the Map Features page? If yes, do we have technical
means to keep the definitions synchronised? Could we perhaps generate
it from Taginfo, or somehow include the definition from the Infobox on
the Map Features page?

-- Matthijs

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

[OSM-talk] Analysis of usage of similar tags over time

[Matthijs Melissen]

Hi all,

Recently, a new tool was created by Martin Raifer (@tyrasd) to
generate graphs of the usage of tags over time. This is a great tool
that gives us more insight in what drives the choice of tags by
mappers. The tool can be found at http://taghistory.raifer.tech/.

I wrote an OSM diary to discuss some results:
http://www.openstreetmap.org/user/Math1985/diary/39404#comment35887

-- Matthijs

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

Re: [OSM-talk] Mailing lists archive not in Google

[Matthijs Melissen]

On 31 August 2016 at 18:32, Andy Townsend  wrote:
> Maybe I'm missing something, but a web search of e.g. "something
> site:https://lists.openstreetmap.org/pipermail/talk/"; finds relevant posts?
> Is there something I'm not understanding here?

You're right, something must have gone wrong on my side.

-- Matthijs

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

Re: [OSM-talk] Mailing lists archive not in Google

[Andy Townsend]

On 31/08/2016 17:23, Matthijs Melissen wrote:

Hi,

The mailing list archives at
https://lists.openstreetmap.org/pipermail/ are currently not indexed
by Google. I suppose this is caused by the  tag that is included in every page of the
archive.


Maybe I'm missing something, but a web search of e.g. "something 
site:https://lists.openstreetmap.org/pipermail/talk/"; finds relevant 
posts?  Is there something I'm not understanding here?


Cheers,
Andy


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

[OSM-talk] Mailing lists archive not in Google

[Matthijs Melissen]

Hi,

The mailing list archives at
https://lists.openstreetmap.org/pipermail/ are currently not indexed
by Google. I suppose this is caused by the  tag that is included in every page of the
archive.

Would it be possible to remove this tag? I think it would be very
useful if the mailing list archives were searchable.

-- Matthijs

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

Re: [OSM-talk] Artwork problems

[Martin Koppenhoefer]

sent from a phone

> Il giorno 31 ago 2016, alle ore 13:30, Oleksiy Muzalyev 
>  ha scritto:
> 
> The artwork icon resembles the Academy Award, or "Oscar" statuette 
> https://en.wikipedia.org/wiki/Academy_Awards . It does not evoke art either 
> inside or outside the museum.


while I agree, this design was chosen by the osm-carto team in the usual 
process and after discussion, so we should accept it for now. There's no good 
icon anyway that can represent all kinds of artwork, from landart to sculpture 
to wall paintings etc. 
You can find other proposals for artwork icons in the github issue of the style 
(my favorite "idea" was a small generic square)

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

Re: [OSM-talk] Artwork problems

[Oleksiy Muzalyev]

On 25.08.2016 22:11, Martin Koppenhoefer wrote:


sent from a phone


Il giorno 25 ago 2016, alle ore 15:28, Daniel Koć  ha scritto:

there are some places, where the indoor artworks just make noise and I still 
don't know how to avoid this clutter in general:

http://www.openstreetmap.org/#map=18/48.86064/2.33631


these look like artwork in a museum, I guess they don't meet the definition for 
the osm tag according to the wiki (public artwork) and should be fixed 
(different tagging)

cheers,
Martin
___

The artwork icon resembles the Academy Award, or "Oscar" statuette 
https://en.wikipedia.org/wiki/Academy_Awards . It does not evoke art 
either inside or outside the museum.



On the other hand, what an icon can possibly resemble such a statue 
outside this museum 
https://commons.wikimedia.org/wiki/File:Louis_XIV_Bernin_r%C3%A9plique_Cour_Napol%C3%A9on_Louvre.jpg 
?



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

Re: [OSM-talk] Artwork problems

[Janko Mihelić]

čet, 25. kol 2016. u 15:30 Daniel Koć  napisao je:

>
> I've also noticed that the line between artworks and memorials is
> blurred, especially with statues
>

In my opinion, we can use both tags. If something is a sculpture, add
tourism=artwork+artwork_type=sculpture. If it is a sculpture whose main
purpose is to remind us of a person or an event, add historic=memorial.

About the icon, I think we should add a new key, something like
sculpture_shape. It could have countless values, and only the most frequent
would have their unique icon. Values would be: human, child, human_sitting,
human_standing, human_on_horse, bust, horse, dog, abstract. Those account
for about 95% of sculptures. Others get the default icon. Having an
approximate shape of a sculpture on a map helps a lot with orientation in
space.

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

Re: [OSM-talk] Without an address, an Icelandic tourist drew this map of the intended location (Búðardalur) and surroundings on the envelope. The postal service delivered!

[Oleksiy Muzalyev]

On 31/08/16 10:03, Maarten Deen wrote:


How did people manage before the age of smartphones, accurate GPS and 
free maps?


Regards,
Maarten
Felix Dellatre, a German who lives in Managua, told in his presentation 
"Community mapping in Nicaragua" at the OSM US conference in San 
Francisco in 2013, that people do not come after dark because of the 
risk of being lost. They cannot go to the cinema, theater, courses, etc.


He also said that the absence of address system makes it impossible to 
call police or ambulance, and as a result local informal "authorities" 
emerge which try to establish a semblance of an order. He told that 
addresses are given is such a way as: "make four hundred steps to the 
north from the place of a burnt church, and ask Miguel".


Please, note than both Nicaragua and especially Brasil are relatively 
well-to-do countries with stable central authorities. You may look at 
the maps of the cities of many other countries and see that a lot of 
areas do not have street names at all and will never have them. The 
current system of street names and house numbers is not scalable, and it 
is not suitable for the whole world. The same as say, excuse me, a flush 
toilet system, which is common in Europe and North America, and yet it 
is absolutely not applicable for many other places. You may see this 
excellent BBC documentary "The Toilet An unspoken History" about it: 
https://www.youtube.com/watch?v=0ZHm3vkavgM .


I am sorry for this a bit too strong example, but it illustrates well 
that not all things to which we are accustomed are applicable elsewhere, 
especially in a places with scarce resources.


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

Re: [OSM-talk] [flame] Without an address, an Icelandic tourist drew this map of the intended location (Búðardalur) and surroundings on the envelope. The postal service delivered!

[Iván Sánchez Ortega]

El onsdag 31. august 2016 09.39.46 CEST Oleksiy Muzalyev escribió:
> But this approach is not scalable, - even on this map you can see three
> /Rua C/, four /Rua A/ (Rua means Street in Portuguese). Without a
> central authority it will end up in numerous duplicates.

See, Oleksiy, you're once again trying to steer the conversation your way by 
abusing strawman arguments and false dichotomies.

So it goes like this:

You: "Classical addresses suffer from problem A, and it's either perfection or 
a huge problem A, so magical addressing system must be put in place".

Someone else: "You will not achieve perfection with magical addressing system, 
besides, A will be a problem no matter what and there are other alternatives 
to mitigate A."

You: "Those alternatives suffer from problem B, and it's either perfection or 
a huge problem B, so magical addressing system must be put in place".

Someone else: "You will not achieve perfection with magical addressing system, 
besides, B will be a problem no matter what and there are other alternatives 
to mitigate B."

You: "Those alternatives suffer from problem C, and it's either perfection or 
a huge problem C, so magical addressing system must be put in place".

... and so on, and so on, and so on.

After years of seeing flame threads and trolling, detecting this stuff gets 
tiring. You're mixing the use cases of delivering mail with being lost at 
night, creating a strawman out of that, and with a false dichotomy, point out 
that there is one and just one solution.


Postal addressing, land rights, parcel ownership, record management, and 
territories contested between authorities are all difficult and interconnected 
issues, AND THERE IS NO FUCKING MAGIC SILVER BULLET THAT WILL SOLVE ALL OR ANY 
OF THOSE.


So if anybody tries to push a silver bullet solution to any of those, it will 
trigger reactions and problems in the rest.

e.g. A change in the addressing system will impact land ownership management. 
Does a person reachable by an address own the rights to just the addressed 
area, to more, or to less? How to link those?


I know it's very easy to have this kind of tunnel vision where something cool 
will fix a problem, and fixing that problem cascades in fixing all the 
problems. But in the vast majority of cases, the side effects of these "silver 
bullet solutions" are way, way greater than the good of the expected outcome.





P.S. Kudos to the folks at Cadasta, who know the challenges of land ownership 
better than me or anybody on this list.

P.P.S. If you want to give a fuck about magic addressing systems, use 
www.what3fucks.com

-- 
Iván Sánchez Ortega   


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

Re: [OSM-talk] Maritime boundary missing. How to fix?

[Walter Nordmann]

Hi Evan,

this has been done right now. (not by me, but i know and trust that guy).

Mapnik rendering is not yet finished on all levels.

Regards
walter/germany



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

Re: [OSM-talk] Without an address, an Icelandic tourist drew this map of the intended location (Búðardalur) and surroundings on the envelope. The postal service delivered!

[Martin Koppenhoefer]

sent from a phone

> Il giorno 31 ago 2016, alle ore 09:39, Oleksiy Muzalyev 
>  ha scritto:
> 
> But this approach is not scalable, - even on this map you can see three Rua 
> C, four Rua A (Rua means Street in Portuguese).


first it wouldn't harm to get a tiny bit more creative when choosing the street 
names and second this is where post codes come into play 


> Without a central authority it will end up in numerous duplicates.


there is a central authority in Brazil, there are very few places without any 
kind of central authority 


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

Re: [OSM-talk] Without an address, an Icelandic tourist drew this map of the intended location (Búðardalur) and surroundings on the envelope. The postal service delivered!

[Maarten Deen]

On 2016-08-31 09:39, Oleksiy Muzalyev wrote:

On 31/08/16 07:54, Andreas Vilén wrote:



Imo it's better to let these areas choose their own (classic)
addresses and then report these to the authorities.

/Andreas

 It looks like they are trying to do exactly this. Here is a map of a
part of Rio de Janeiro in Brasil: http://osm.org/go/OVcch3Alh- .

But this approach is not scalable, - even on this map you can see
three _Rua C_, four _Rua A_ (Rua means Street in Portuguese). Without
a central authority it will end up in numerous duplicates.


These look to be gated communities. Like here:


Condominio Barao da Taquara

So the address would be something like Rua E., Condominio Barao da 
Taquara, Rio de Janeiro.



I heard that in such areas of a city a smartphone is a basic
necessity, because if one leaves home when it is dark outside it is
difficult to find way back as there are no street signs, and one may
realistically get lost till dawn.


How did people manage before the age of smartphones, accurate GPS and 
free maps?


Regards,
Maarten


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

Re: [OSM-talk] Without an address, an Icelandic tourist drew this map of the intended location (Búðardalur) and surroundings on the envelope. The postal service delivered!

[Oleksiy Muzalyev]

On 31/08/16 07:54, Andreas Vilén wrote:
"frame;lock;door" hmm... or was it "frame;lock;window"? ---> mail gets 
delivered to the other side of the world. If you write "Main street 
15" instead of "Main street 13" it will probably get delivered 
correctly anyway as long as the name is correct.


Also, tell people who are supposed to deliver mail in the favelas of 
Brazil to constantly wear a smartphone around their neck and you will 
see very few people willing to do that work...


Imo it's better to let these areas choose their own (classic) 
addresses and then report these to the authorities.


/Andreas
It looks like they are trying to do exactly this. Here is a map of a 
part of Rio de Janeiro in Brasil: http://osm.org/go/OVcch3Alh- .


But this approach is not scalable, - even on this map you can see three 
/Rua C/, four /Rua A/ (Rua means Street in Portuguese). Without a 
central authority it will end up in numerous duplicates.


I heard that in such areas of a city a smartphone is a basic necessity, 
because if one leaves home when it is dark outside it is difficult to 
find way back as there are no street signs, and one may realistically 
get lost till dawn.
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk