Re: [Tagging] visa offices tags

2018-11-07 Thread John Willis
Is the office full of people who you pay to help you apply? Or are they 
contracted to be the front-otfice of the agency (acceptance & distribution)?  

It seems to be the latter. 

Think of taxes. 

The office where you submit taxes, and an office where a professional helps you 
prepare the forms are two different places. 

Office=tax_preparation and office=tax are different. 

If this is some office (public or private) with an official mandate to accept 
applications and distribute visas (whether approved onsite immidately or 
approved elsewhere and mailed to the office for distribution), that does sound 
like office=visa. 

If it is some private business who you pay to help you prepare the form, and 
you mail it to the government and get your visa directly, it sounds like 
office=visa_preperation or office=immigration_lawyer. 

I am unfamiliar with the situation you described - so please choose the best 
tag that suits your needs. 


> On Nov 8, 2018, at 11:37 AM, Eugene Alvin Villar  wrote:
>> On Thu, Nov 8, 2018 at 9:29 AM Warin <> wrote:
 On Nov 7, 2018, at 7:12 AM, Warin <> wrote:
 1) this is a commercial firm - not a government 
 2) it 'assist' people to obtain a visa
 3) it is not at an airport/seaport/boarder
 4) the visa is obtained before travel commences.
>> It is an office you go to. You present documents, they ask questions, you 
>> answer, you pay a fee, 
>> the office fills out forms using that information provided (and they then 
>> send it off to an embassy/consulate) 
>> and then some time later you get a visa back from the office (but the visa 
>> itself is actually from the embassy/consulate). 
>> In the above situation, what is wrong with office=visa ? You apply to the 
>> office, they (usually) get you a visa. 
> Here, the Japanese consulate never accepts direct visa application and 
> instructs people to only submit visa applications through accredited travel 
> agencies.
> On the other hand, many European consulates here contract a 3rd-party visa 
> processing company such as the aforementioned VFS Global to handle all visa 
> applications. These companies even have equipment to collect biometric data 
> such as photographs and fingerprints that will be forwarded to the consulates 
> together with the visa applications.
> I would think that the first case should be tagged like other travel agencies 
> because visa handling is just one of their services (they also arrange tours 
> and purchase airline tickets). For the second case, they do nothing else 
> besides processing visa applications on behalf of the contracting consulates. 
> So they are not travel agencies. I think they should indeed be tagged with 
> something like office=visa or better yet office=visa_processing so it is 
> clearer.
> ___
> Tagging mailing list
Tagging mailing list

Re: [Tagging] visa offices tags

2018-11-07 Thread John Willis

> On Nov 7, 2018, at 7:12 AM, Warin <> wrote:
> 1) this is a commercial firm - not a government 
> authority/branch/department/etc
> 2) it 'assist' people to obtain a visa
> 3) it is not at an airport/seaport/boarder
> 4) the visa is obtained before travel commences.
> 5) it is not within the country where the visa is used

if they don’t issue visas, they are immigration or travel paralegals/lawyers. 
this sounds like people assisting tourists. An office assisting people trying 
to get *residence* in a country is certainly an immigration lawyer.  

This does sound like something completely different than the 
amenity=immigration I am thinking of. 

I am unsure of a tag - but simply office=visa should be out. office=travel 
assistance or visa assistance or something. 

Tagging mailing list

Re: [Tagging] visa offices tags

2018-11-06 Thread John Willis

Tourist visas are handled by different offices - usually at airports. 

While they may all be "visas" - tourist/guest 90 day visas or an 8h transit 
Visa are treated very differently than any other visa that allows you to reside 
in the country- work, student, spouse, etc visas require office visits to an 
office usually called "immigration" in Japan and the US. 

Usually, there is some office or building that locals know nothing about - and 
foreign residents have to visit. 

Sometimes it is part of a larger building, sometimes not. 

For example, 1/2 of this building is a prison transfer building. The other half 
is a a branch office of the main Tokyo immigration office.

Every few years I have to go there to get my Visa renewed, get, my new foreign 
resident card, and eventually submit my application for permanent residence. 

My passport issues are still handled by the US embassy in Tokyo. 

No Japanese citizen ever has to deal with that office - only immigrants. They 
don't handle guest/tourist visas either - only people living there. 

I assume this amenity exists in most countries, in some form or fashion. 


> On Nov 6, 2018, at 1:34 PM, Warin <> wrote:
> There is no OSMwiki on what government=migration is about. 
> If you simply use the common definition of  'migration' then tourist do not 
> fit. 
> migration is the movement by people from one place to another with the 
> intentions of settling
> So a place that assist visas for tourist and business people would not 'fit'.
> So I'd say .. office=government, government=migration does not handle this. 
> Particularly when the office is a commercial firm, and I think has no 
> government funding. 
>> On 06/11/18 14:11, Allan Mustard wrote:
>> The office=government, government=migration tags already handle this, no?
>> Sent from my iPhone
>> On Nov 6, 2018, at 7:06 AM, Warin <> wrote:
>>> Lots of people apply for tourist visas. They are not immigrants, so 
>>> immigration does not fit all. 
>>> I don't know if that particular office only does immigration visas, tourist 
>>> visas or does any type of visa. 
>>> I would prefer a tag suitable for an office that could do any type of visa, 
>>> rather than having to find out which particular thing they do. 
>>>> On 06/11/18 12:37, John Willis wrote:
>>>> Amenity=immigration 
>>>> They handle visas and passports and other paperwork needs of legal 
>>>> residents. 
>>>> This is not something for guests/tourists, but people (like me) who need 
>>>> to handle paperwork to continue to live in the country. 
>>>> It is the inverse of an embassy. 
>>>> Javbw
>>>> On Nov 3, 2018, at 12:24 PM, Allan Mustard  wrote:
>>>>> Definitely not an embassy, and not a consulate, either!  More like a 
>>>>> specialized travel agency that focuses only on visa applications.
>>>>>> On 11/3/2018 6:22 AM, Warin wrote:
>>>>>> Hi, 
>>>>>> Node: Visalink Germany (4362535595) is tagged as an embassy. 
>>>>>> It is a commercial firm that arranges applications to the German 
>>>>>> Embassy/Consulate for a visa, see 
>>>>>> I think tags could be office=visa, country=DE, 
>>>>>> website= 
>>>>>> but not amenity=embassy, diplomatic=visa ... 
>>>>>> There are some 13 with the tag diplomatic=visa that may fall under this 
>>>>>> same cloud. 
>>>>>> ___ 
>>>>>> Tagging mailing list 
>>>>> ___
>>>>> Tagging mailing list
>>>> ___
>>>> Tagging mailing list
>>> ___
>>> Tagging mailing list
>> ___
>> Tagging mailing list
> ___
> Tagging mailing list
Tagging mailing list

Re: [Tagging] visa offices tags

2018-11-05 Thread John Willis

They handle visas and passports and other paperwork needs of legal residents. 

This is not something for guests/tourists, but people (like me) who need to 
handle paperwork to continue to live in the country. 

It is the inverse of an embassy. 


> On Nov 3, 2018, at 12:24 PM, Allan Mustard  wrote:
> Definitely not an embassy, and not a consulate, either!  More like a 
> specialized travel agency that focuses only on visa applications.
>> On 11/3/2018 6:22 AM, Warin wrote:
>> Hi, 
>> Node: Visalink Germany (4362535595) is tagged as an embassy. 
>> It is a commercial firm that arranges applications to the German 
>> Embassy/Consulate for a visa, see 
>> I think tags could be office=visa, country=DE, 
>> website= 
>> but not amenity=embassy, diplomatic=visa ... 
>> There are some 13 with the tag diplomatic=visa that may fall under this same 
>> cloud. 
>> ___ 
>> Tagging mailing list 
> ___
> Tagging mailing list
Tagging mailing list

Re: [Tagging] Feature proposal - RFC - assembly_point:purpose

2018-10-19 Thread John Willis

> On Oct 20, 2018, at 2:50 AM, bkil  wrote:
> There are lots of bomb shelters, although nobody knows how to tag
> these. 

Sounds like you found something that needs a proposal! 



Tagging mailing list

Re: [Tagging] Feature proposal - RFC - assembly_point:purpose

2018-10-18 Thread John Willis
Hmm... The active shooter discussion brought up some good things to think 

As far as I know, we are not mapping the evacuation plans of individual 
buildings with assembly_point. 

When talking about the schools, we talked about shelters and assembly_points. 

The pitch at a public school is often considered an assembly_point - not just 
for the children, but for the entire neighborhood. It is a government 
designated place for people to go during a large-scale disaster 

Perhaps thinking of those as active shooter safe rooms as "shelters" is wrong, 
and the mere evacuation point for a random private building is not something to 
include in emergency=* 

Perhaps having some evacuation_plan=* key and an accompanying relation can let 
individual buildings and complexes map areas, points, and evacuation routes on 
a micro level (like indoor mapping , ie: the fire evacuation routes and meeting 
point outside for a large hotel),  **but I don't think mapping a place 
designated for an individual building evacuation in case of fire is proper for 
=assembly_point.** They are for the *public* to gather and receive aid and 
possible rescue in a large scale disaster. They are usually designated and 
operated by the government, and mapped and signed by the government, so they 
know where to send rescue personnel. 

The only exception I can see is for tornado shelter or bomb shelter - as their 
physical existence is the "help" - and (I assume) are publically accessible 
assembly_points, even in private commercial buildings, and they blur the line 
between shelter and assembly_point. I don't know how to map those, as I am not 
very familiar with them. 

But Having a bunch of assembly_points coating a downtown area, even with 
access=private, would turn into tag pollution. The 2-3 locations (the school 
ground, the park, and the sports complex) would be lost in a sea of points on 
lawns and parking lots no one would care to be. 

If a large concrete mall near a coastline has a outside, designated, publically 
accessible stairway to the roof and signs telling people to evacuate there in 
case of tsunami, the fact that it is "privately operated" is not as important 
as it is publically accessible for *anyone* looking for Saftey. And the fact 
that any random building just happens to be tall and have stairs is not enough 
- has to publically known and publically accessible. 

Our local elementary school grounds are the designated evacuation point for our 
community in case of a failure of a nearby dam - we received flyers showing the 
hazard map and evacuation points. 

The building evacuation points do not feel like those are in the same category. 

The idea of assembly_point being publically accessible and designated for this 
purpose is the most important point. 

The narrow_definion of assembly_point seems best. 


> On Oct 19, 2018, at 2:42 AM, bkil  wrote:
> The reason is probably to both increase survival rate by taking
> everybody as far as possible from danger and to ease the work of
> firefighters by not gathering a crowd around the building in question.

Tagging mailing list

Re: [Tagging] historic=memorial tagging question

2018-10-18 Thread John Willis
We need to update the wiki then, to add further description. There are plenty 
of tags we use that are uncommon or have incomplete descriptions. 

My local temple made an erected a stele as-pictured (similar to the Takasaki 
one) last year. 

I think it's orgins are ancient, but still "in use" in many parts of the world. 

Besides the comparison to a grave's headstone, I can't think of any steles I 
have seen in the US. I assume there are some - but I can't remember seeing any 
- so it  a simply not a word we are familiar with. 


> On Oct 19, 2018, at 12:09 AM, EthnicFood IsGreat 
>  wrote:
> And I doubt if many people in the US are familiar with that word.

Tagging mailing list

Re: [Tagging] historic=memorial tagging question

2018-10-17 Thread John Willis
Thanks, but marker only has seven uses - stele has about 5000.

Also- Marker, to me, would be something you would find in the ground with a 
number or a pole with a number on it, or something based around a ref number or 
value of some sort (like a mile marker). 

In the US, a "historical landmark" or "marker" is usually a plaque embedded in 
a pedestal or stone, so it is memorial=plaque, with about 10,000 uses. 



> On Oct 18, 2018, at 4:21 AM, EthnicFood IsGreat  
> wrote:
> memorial=marker

Tagging mailing list

Re: [Tagging] historic=memorial tagging question.

2018-10-17 Thread John Willis

One of the wiki examples (a rubbing from a tablet from Takasaki) is from my 
area, and exactly what I was thinking of. 

The example image I saw on OSM was of a pole or something, but the Wikipedia 
was much more broad. 

There like a hundred or more of those in my town alone, so now I can map them. 


> On Oct 17, 2018, at 6:49 PM, Lionel Giard  wrote:
> I would use memorial=stele (as it looks like a stele to me). If you go on 
> wikipedia it looks like some of the "usual" stele : 
> .
>> Le mer. 17 oct. 2018 à 11:22, John Willis  a écrit :
>> How would I map this object? 
>> It is a stone tablet memorializing when a small levee failed and washed away 
>> part of a hamlet. This is in the park built on top of the levee repair.
>> these kinds of stone tablets - always a freestanding carved or etched stone 
>> (not a plaque attached to a rock), and usually between 50cm and 2m in 
>> height, are the most common form of memorial object in Japan (very tall, 
>> thin, cut stone tablets anchored in the ground, covered with carved letters 
>> - not a memorial=stone). this example is a more modern version, with very 
>> little text. older ones are covered with text. though it looks like a 
>> headstone or a grave, this is not a common shape for a grave marker in Japan 
>> - but for memorial tablets. 
>> interestingly, the sub-tag of historic=memorial, memorial=*,  has a lot of 
>> suggested values in iD, including memorial - so memorial=memorial is a 
>> commonly tagged value (295 uses)!
>> to me, this is a tablet (20 uses), but if this object is better defined by 
>> another value, I would be happy to use it. 
>> Javbw
>> ___
>> Tagging mailing list
> ___
> Tagging mailing list
Tagging mailing list

[Tagging] historic=memorial tagging question.

2018-10-17 Thread John Willis
How would I map this object? 

It is a stone tablet memorializing when a small levee failed and washed away 
part of a hamlet. This is in the park built on top of the levee repair.

these kinds of stone tablets - always a freestanding carved or etched stone 
(not a plaque attached to a rock), and usually between 50cm and 2m in height, 
are the most common form of memorial object in Japan (very tall, thin, cut 
stone tablets anchored in the ground, covered with carved letters - not a 
memorial=stone). this example is a more modern version, with very little text. 
older ones are covered with text. though it looks like a headstone or a grave, 
this is not a common shape for a grave marker in Japan - but for memorial 

interestingly, the sub-tag of historic=memorial, memorial=*,  has a lot of 
suggested values in iD, including memorial - so memorial=memorial is a commonly 
tagged value (295 uses)! 

to me, this is a tablet (20 uses), but if this object is better defined by 
another value, I would be happy to use it. 

Tagging mailing list

Re: [Tagging] Feature Proposal - RFC - landuse=governmental

2018-10-15 Thread John Willis

> On Oct 16, 2018, at 7:13 AM, Graeme Fitzpatrick  wrote:
> How about those "public safety" groups that are manned & operated by unpaid 
> volunteers, often with minimum (or no) Govt funding?

A building dedicated to a volunteer fire department - where they keep the truck 
and the hoses - is still land used by a structure dedicated to public Saftey. 

It may not be a giant police station or a multi-floor firehouse - yet all of 
them usually have dedicated land larger than building. 

If the community depends of them for fire protection services - they are the 
"fire department". 

I assume tagging for a fire station will have some kind of tagging to denote 
it's volunteer or irregular staffing situation.

Pulling them out of "landuse=governmental" seems to be a no-brainer, so making 
a tag for them would help more narrowly define landuse=governmental. 

Tagging mailing list

Re: [Tagging] Feature Proposal - RFC - landuse=governmental

2018-10-15 Thread John Willis


> On Oct 16, 2018, at 5:32 AM, SelfishSeahorse  
> wrote:
> No, landuse=governmental is intended for the land used for organising
> a country, state, municipality etc., that is, the 'core functions' of
> a country, state, municipality etc., which are administering it,
> making rules (laws), interpreting and executing them.

One solution might be to split off 

Landuse=Public_safety into its own landuse. 

Many times, those are mostly private buildings with a small public facing 

Landuse=public_safety would be generic enough to be for police stations, 
rangers, lifeguards, snow patrol, highway patrol, harbor patrol, fire stations, 
wildfire stations, and other "official" services offered to the public to 
ensure their Saftey. 
Tagging mailing list

Re: [Tagging] opening_hours value question

2018-10-12 Thread John Willis
Thanks to both Marc & Markus. 



> On Oct 12, 2018, at 9:08 PM, Marc Gemis  wrote:
> A useful tool to verify the expressions is
>  (is listed on the
> page you link)

Tagging mailing list

[Tagging] opening_hours value question

2018-10-12 Thread John Willis
I am really impressed by the key opening hours

However, it is just at my threshold of understanding. I can read the examples 
and adapt it for my needs. 

In Japan, I was surprised to see shops that are "closed Tuesdays" or similar. I 
can tag those. (map the opening hours around the Tuesdays). 

But some shops are open "on the 3rd Saturday of the month, rather than being 
closed. (Sometimes it's the opposite). 

- Can someone type me the necessary tag value for such a monthly calendar 
dependant item? "Open on the 3rd Saturday from 10am-1pm"

I didn't see an example that I could adapt, or perhaps I don't understand that 
it's the one to use.   

- also, is there a way to positively define "closed Tuesdays" or similar? It 
may not be necessary, but it is a common thing to see, hopefully it is easy to 

Thanks in advance

Tagging mailing list

Re: [Tagging] Greengrocer vs grocery vs shop=food?

2018-10-11 Thread John Willis

> On Oct 12, 2018, at 4:20 AM, Mark Wagner  wrote:
> There are no "more knowledgeable mappers" in my area.

Do what you gotta do - but it shouldn't be the method that is reccommended.

Thanks for mapping such remote places, BTW. 


Tagging mailing list

Re: [Tagging] Greengrocer vs grocery vs shop=food?

2018-10-11 Thread John Willis

> On Oct 11, 2018, at 5:29 PM, Mark Wagner  wrote:
> Empire Foods",

The name doesn't help much either. 

A convenience store? 

Food production? (Not a shop) 

Food distribution? (Not a shop) 

Is it a greengrocer?

A butcher?

A supplier of Prepackaged foods sold B2B for convenience stores or fast food 
restaurants (not a shop)? 

The office building for the chain of shops? 

I see this issue in Google maps all the time, where a distribution warehouse is 
labeled as a "supermarket" because it has the chain's name in the POI data. 
They were mapping by name and got it wrong. 

If it was "Weird Al's world of ketchup", then I could see having trouble 
categorizing it, but  tagging a business shop=food is missing the chance to put 

fixme=please update tagging to correct shop type

 on the shop for a more knowledgeable mapper to update. 

We should have a couple more categories of market-type shops (as discussed) 
before lumping all of them into shop=food. 

By tagging shop=food when you do not know how to tag it is wrong. It is a 
missed opportunity not only for the tagged business (it will probably never be 
revisited and properly tagged), but causes tagging confusion when people go to 
tag similar businesses.  

There may be a use for shop=food - but throwing a business in there because it 
has "foods" in the title is bad tagging (imo). 

Tagging mailing list

Re: [Tagging] Greengrocer vs grocery vs shop=food?

2018-10-11 Thread John Willis

> On Oct 11, 2018, at 12:24 PM, Daniel Koć  wrote:
> It's much more precise than shop=yes (which is used a lot) and I know
> what food is, even if I don't know shop type details.

That is true. 

But it is being used to tag things that should have their own tag value. 

it’s like if we used building=people instead of making =house and =apartments 
and other types of residential buildings. 

Yea, it's better than building=yes, but not much better. 

Javbw. ___
Tagging mailing list

Re: [Tagging] Greengrocer vs grocery vs shop=food?

2018-10-10 Thread John Willis
the definition of shop=food is way way way to vague to have meaning. it needs 
to be much narrower. 

it is like shop=goods. we don’t need that either. 

> On Oct 11, 2018, at 1:10 AM, Jmapb  wrote:
> Amen! I wrote a whole spiel about this on the wiki talk page for shop=deli 
> .

"Also, I am clueless as to how to enter a "gourmet food store" to begin with. 
Any help or clarification in this matter would be greatly appreciated”

(from the talk page)

I think this is the interesting bit - and one tagging shop=food is trying to 

By being a catch-all, it has no set value. 

Carve off specialty import/foreign goods into it’s own tag, and perhaps 
shop=gourmet for fancy specialty foods (though perhaps those are shop=deli ?? I 
dont know). 

the continued discussion for take-n-bake shops (still feels like fast food to 
me, but I will defer to the group) also removes a lot of those uses. 

What is left? probably a bunch of mis-tagged shops. 

somewhere out there is Jimbo’s turnip store and Taro’s Mochi shop, but those 
can probably filter into other tags as well (greengrocer, dessert, etc). 

I don’t think we should have shop=food + food=turnipor   shop=food + 

Honey shops? Miso Paste?

we use shop=yes and sells=* for those (as I understand it) 

for the oddball shops, this seems to be the solution to me.

Javbw.  ___
Tagging mailing list

Re: [Tagging] Greengrocer vs grocery vs shop=food?

2018-10-09 Thread John Willis

> On Oct 10, 2018, at 4:39 AM, bkil  wrote:
> Why is shop=convenience not a proper tag for "the only retail building
> in 40 miles radius"?

Usually, the small retail shop in a very remote place is tailored to the daily 
needs of locals and tourists who do activities in that area. they stock goods 
that the locals need for daily life. 

There is one general store I know of

Someone has tagged it as a convenience store. 

A convenience store is is convenient not only because of its proximity, 
relative to the larger supermarkets, but also because of the limited subset of 

A general store is a store of necessity - there is no choice but it. It is the 
only practical choice unless you want to drive 2 hours round trip into "town" 
and get something at a supermarket. 

Perhaps this is bias: I grew up with 7-11s in suburban San Diego and would 
encounter general stores only when out in the mountains or deserts - they carry 
a very different mix of goods. You usually can't buy firewood, snow sleds, or 
video rental at a traditional 7-11. In Japan, you can find convenience stores 
serving very small communities in remote areas - but they carry similar goods 
to the ones in urban centers. They feel like convenience stores. Shop=general 
has some of the same goods as a convenience store - but other goods as well. 
Goods that cater to tourism activities in the area and locals who have no other 
practical choice. The "general store" you see in old western frontier towns is 
a good example. 

Javbw ___
Tagging mailing list

Re: [Tagging] Greengrocer vs grocery vs shop=food?

2018-10-09 Thread John Willis
sounds like there are several different kinds of shops being discussed

- old old “markets”, from before there were super markets or convenience shops. 

- import/foreign foods shops catering to a local minority population or special 
cultural interest

- “markets” in developing countries.   

> On Oct 9, 2018, at 11:56 AM, Joseph Eisenberg  
> wrote:
> What do you think about the need for a shop=grocery tag for small shops in 
> developing countries and specialty grocers in cities?
> Are there still small groceries in Japan which sell non-perishable food 
> items, but would not be properly considerd a shop=convenience, shop=general, 
> shop=greengrocer or shop=supermarket?

I know the shops that you speak of. They were the local “everyday needs” shop - 
the market/grocery shop, very similar to a general store - but in an urban 
area. they were the only shop that had some of everything that wasn't covered 
by the Rice shop, fish shop, the butcher, and the produce stand:  curry mix, 
spices, dish soap, eggs, milk, toilet paper, etc. they would be shop=market, if 
that exists.They still exist in Japan, but are almost gone. The mom-n-pop ones 
are operated by people that live over the shop, and they are still operated for 
the locals to come sit there and gossip - but everyone goes to the supermarket 
3 minutes away. they never look like they sell anything, and most have been 
shuttered, but a few are still there.  the only corner market I knew of was 
there are a few shop=general out in the mountains - but all the “markets” were 
put out of business by supermarkets a long time ago in California. I know of 
only one from personal experience. I hear of the “corner shop” or “bodegas” in 
New York - similar to the little corner market Bullitt buys his frozen dinners 
from in the movie in San Francisco - they seem to be disappearing in developed 

They are the proto-market: the Convenience store is more convenient, they have 
no departments, they are not specific enough to be a greengrocer nor have a 
stock of blankets, bullets, motor oil, and firewood like a general store - they 
are the “daily market”, not a giant supermarket - the corner store. 

a small market for daily living in developing countries feels like it would be 
a shop=general - a general store has a certain feeling when it is the only 
retail building in 40 miles in any direction, perhaps that is similar to the 
developing country shops. 

I think shop=general for the small developing countries’ markets or these 
fading local markets would be a good kludge, but it is not a fit **at all** for 
some specialty shop in a big city.

> Mediterranean groceries or Caribbean foods, as found in some big cities.

This is a great question. there are all kinds of [asian country] markets in San 
Diego, and there are Philippine, Brazilian, and “Halal foods” shops here in my 
area of Japan. There are also chain shops catering to “foreign foods” : 
American snacks, British mints, South American Coffee, Italian pasta, etc. they 
almost always are around food. 

if there is a convenience store, a supermarket, a “halal foods” shop, and a 
butcher shop on the same block - that isn’t 4 “markets” - I think the idea of a 
“foreign foods" market is good - and then choose a theme or country, or 
religion, or similar tag would work.  . I don’t know how that aspect would be 
tagged - but the type of shop - the “import goods from some far off place 
catering to a minorty group that lives in the region” is a very very common 
occurrence, and very very rarely considered by the majority residents to be a 
place to go shopping (they all shop at the supermarket, as their ethnic and 
culturally specific goods are stocked there). I think having a shop=halal and a 
shop=Japanese would be wrong - as the only place they would be used is outside 
those areas, and confusing for people inside those areas. 

If we try to come up with a tag that fits all these uses, it won’t fit. We need 
to create shop=* tags to fit these separately. 


Tagging mailing list

Re: [Tagging] Greengrocer vs grocery vs shop=food?

2018-10-08 Thread John Willis

> On Oct 9, 2018, at 7:50 AM, Joseph Eisenberg  
> wrote:
> shop=food 

> + cuisine=local/polish.


shop=food is called a market. we have that already.  cuisine=* is for 
restaurants and other non-shop places. 

fast_food=take-n-bake (or whatever is decided: take-n-bake=yes/no/only, etc) 


- make a “takaway” or take-n-bake” subtype of amenity=fast_food **for 
freshly-prepared-food items to be taken home, cooked, and served. ** 

- Unless it is the ingredients for making a food, or sells large amounts the 
food in bulk to take home, shop=* is wrong. Unless this pizza shop is selling 
you pizza ingredients or bulk boxes of frozen pizza - I don’t think it is a 

- don’t pollute shop=* with restaurants and fast food. probably most uses of 
shop=ice_cream and shop=bakery are already not proper for the shop key, and 
most uses should be tagged amenity=fast_food + cuisine=* . a take-n-bake pizza 
shop or a shop selling a tray of freshly prepared, yet uncooked perogis or 
Gyoza is very close to a fast_food takeaway shop selling cooked pizza. 

- People should be free to define a business using amenity=restaurant or 
amenity=fast_food using any cuisine=* value - and tags that go against it 
(amenity=ice cream) should be depreciated. 

- being conscious of creating proper tags for “prepared food to be eaten now” 
that are distinct from shops that sell ingredients (supermarket, butcher shop) 
or bulk (shop=ice cream)  will solve both of the above issues. 

- not solving this issue with generic tags that are not conscious of this issue 
will lead to a lot of “reinventing the wheel” for each type of food and immense 
tagging pollution through confusion.  


I think it is better to use an existing amenity=fast_food or similar tag, and 
not to mix shop=* and cuisine  for the job. We have have fast food, restaurant, 
pub, cafe, and probably a couple others. perhaps we need a sub-tag of one of 
those to be “take-n-bake” type shop. 

make a tag solution for takeout / takeaway/ premade meals or items, otherwise 
you get a ball of confusion and disagreement, like the tagging for an ice cream 

"A place that sells ice cream and frozen yoghurt over the counter.”

and now there is

 shop=ice_cream   "For places selling ice cream to take home”


amenity=cafe + cuisine=ice_cream (? I guess these exist somewhere, where ice 
cream drinks are sold?) 


amenity=fast_food + cuisine=ice_cream (proper replacement for amenity=ice_cream)

There should be standardized tags for an ice cream parlor, where people sit 
down for an extended period of time and enjoy custom made ice cream treats at 
booths and tables. 

If this place sounds similar to similar to an ice cream “restaurant” - that is 
because it *is* a type of restaurant.

They are rare, but a proper ice cream parlor is basically an ice cream 

amenity=restaurant + cuisine=ice_cream is an ice cream parlor. I don’t need to 
define a tag with “ice cream parlor” in it, because it is merely the “cuisine” 

Baskin Robbins, Cold Stone, and a stand selling ice cream in cones/cups in a 
mall/storefront are all amenity=fast_food + cuisine=ice_cream (replaces 

a true shop=ice_cream is selling large amounts of pre-packaged or produced ice 
cream for you to take home - or the ingredients you make your own at home (eg: 

all the shops listed there have the primary purpose of selling you 
ingredients to make other things. 

Places where we sell prepared food to sit and eat is a restaurant. the coffee 
shop sells bagged coffee and brewing machines as it’s primary purpose - not 
brewed coffee drinks like a cafe. the bread shops we have in Japan would easily 
be amenity=fast_food + cuisine=bread - they are full of prepared food for quick 
meals or snacks - not shelves full of bread loaves or bags of rolls to make 
dinner with. There are a couple, but it’s not the focus of the place. I 
consider this an error in using the shop=key. 

I do not want this error to spread via a generic shop=food tag. 

*We made the distinction between restaurant and fast food for this very 
reason.* shop=* shouldn't be involved. 

A butcher shop isn't cuisine=meat for this reason - they are shops. 

I bring up ice cream not because I care so much about it - but it is a good 
example of tagging that that cannot be applied to other types, and can be fixed 
and be much more flexible using generic tags. 

We should create tags describe the type of *business*, as this greatly affects 
how the customer views the business. This also makes it impossible to tag 

Re: [Tagging] landuse for government offices ?

2018-10-02 Thread John Willis


> On Oct 2, 2018, at 8:02 AM, Joseph Eisenberg  
> wrote:
> 90% of the land in some western American States will be owned and managed by 
> some level of government

This is not about who is on the operator=* tag on a hundred square miles of 
grazing lands in Colorado. 

It is about mapping the landuse for the *office building* for the land 
administration office in the state capital that handles the applications from 
ranchers wanting to use that land. 

National parks, preserves, grazing lands, military bases, and other "lands" 
controlled by the government are simply empty land with a boundary (which have 
very specific tags),  which would be tagged as they normally should be now. 

"Civic Admin" is about mapping the offices and centers where Civic employees do 
their job and interface with citizens - not where cows interface with grass 

Land literally used for Civic administration -or any land for sale would be 
"landuse=retail" ! 

Tagging mailing list

Re: [Tagging] How to tag a building constructed for a gastronomic purposes?

2018-09-27 Thread John Willis

> On Sep 27, 2018, at 4:20 PM, Warin <> wrote:
> That was the intention. But many mappers are mapping the function of the 
> building, not the appearance/architecture.

TLDR: it's difficult to understand the spirit of building=* as "constructed 
type" when almost all buildings one is mapping are used by their owner for 
their initially built purpose; this leads to a general misunderstanding of 
building=* usage. 


My house in San Diego was built in 1922, and is in the oldest 1% of structures. 
Most were built after 1960. 

In some areas, the structure is valuable, easily repurposable, and/or is not 
expected to be demolished - which would lead to the building=* tag to be 
different than the current use of the occupant. 

This descrepancy is what has led to some people (myself included) as the tag 
being for defining the general purpose the building  as it is currently used 
(building=retail, industrial, house, etc), which is not it's intended purpose. 

I understand the idea of the "constructed type" being defined through 
building=*, but that is very difficult to grasp when you tag mostly buildings 
made after WWII that are still used for the same purpose. 


Japan is a very old place - but everything old is a temple or has rotted away  
- so many buildings: residential, commercial, and industrial are boom-times 
building (1960,1970s) - there is just so little repurposing of structures that 
I see.  Perhaps 1 in 5000 structures (where I am mapping) are repurposed, and 
the ones I know of are storehouse/warehouses from silk production -  there are 
just a handful. Most everything here gets demolished and the lot gravelled when 
the land is sold - people dont buy structures; they buy land. prewar wood and 
bamboo structures exist but are abandoned, boomyear house-shops have the shop 
portion shuttered, post war low-income housing was (& is still being) 
demolished to make apartment blocks that are now themselves being demolished 
for modern apartment complexes.  The only exception is a glut of 1990's 7-11s 
that were abandoned and repurosed into other shops - but almost 100% are still 
retail.  Old blocks of stores are demolished and convenience stores spring from 
the gravelled lots are like weeds here. I counted 7 new ones under construction 
in the past 10 days - they all are new construction replacing demolished older 

I imagine tagging a bakery In a church building would entail using 
building=church - but for some mappers, it a very rare occurrence, leading to 
building=* to feel like  a usage identifier rather than for structure 
construction type, which I guess has led to mis-mapping and tag value 

Tagging mailing list

Re: [Tagging] Topographic Prominence for Peaks

2018-09-27 Thread John Willis


> On Sep 27, 2018, at 2:17 PM, Graeme Fitzpatrick  wrote:
> How do you determine the height of the saddle / peak?

There is a lot of GIS data available for named points. 

Also, there is a lot of topography available as well, so someone manually 
mapping certian areas could create a pretty decent prominence value - from 
where they choose to measure. The issue a lot of people bring up is the "where 
you measure from" is relative to some other point - some people will assume 
different points are the "proper" point. 

In some places this won't be an issue, in other places - a big issue. 

Hopefully there is enough topographic information to programaticaly create 
values to check, rather than everyone choosing their own points. 

Tagging mailing list

Re: [Tagging] Draft Proposal: Default Langauge Format

2018-09-26 Thread John Willis

> On Sep 26, 2018, at 3:36 PM, Frederik Ramm  wrote:
> be amended with an identical name:xx tag just because xx is the language
> spoken in that country!

You are very lucky then to not have to deal with the documented tagging scheme 
of name= , name:en= name:ja= name:ja_rm=, and name:ja_kana= where I am, which 
is the documented way to tag things in Japan. 

Tagging mailing list

Re: [Tagging] Feature Proposal - RFC - Topographic Prominence

2018-09-25 Thread John Willis

> On Sep 26, 2018, at 6:46 AM, Kevin Kenny  wrote:
> I don't actually mind 'natural=peak' for any named local maximum
> elevation.

In so many places, Lumps and bumps are simply not named. But in some places, 
they are. 

People who see Mount Fuji every day have no idea all 8 high points on the rim 
of a caldera are all named points (unmapped in OSM until yesterday). People in 
their daily lives and the OSM dataset don't know of the blizzard of named 
points that could should be mapped. This is true of mountainous areas all over 
the world - people don't realize the level of naming that occurs. 

In the city near me, there is a hill in a park. It is about 15m AGL. But it is 
an historic hill with a name.

It may have an msl elevation of 108m, but my middle school's library is larger 
than this "peak". It is shorter than the trees in the flat part of the park. 

There are places that need some granularity. 

If you want the tag for a top of a mountain, the word is "summit". 

A peak has a different connotation. "Foobar Peak" is usually used to name a 
tall mountain. It is also used to denote the small points on a large or famous 

If people make a peak=* subtag, make =hill tag, or add prominence data, or 
simply add som way denote locally / regionally / nationally / internationally 
famous mountains (which would affect their rendering at zlevels) - whatever - 
something needs to be done to filter out tiny lumps scattered around and small 
named bumps on a larger mountain. 


Tagging mailing list

Re: [Tagging] maxspeed:type vs source:maxspeed // StreetComplete

2018-09-25 Thread John Willis

> On Sep 25, 2018, at 3:09 PM, Florian Lohoff  wrote:
> Sign posted speeds dont are not telling you "this is the speed which is
> safe for 100% of the vehicles" but this is the maximum allowed. 
> You are still required to drive safely.


Even variable speed roads (where the signs change digitally to reflect road 
conditions, such as wind), this still holds true. 

Tagging mailing list

Re: [Tagging] Feature Proposal - RFC - Topographic Prominence

2018-09-24 Thread John Willis

> On Sep 25, 2018, at 12:14 PM, Kevin Kenny  wrote:
> I live in country with long ridges, and almost anything with enough isolation 
> and a little bit of prominence winds up being a named summit.

Yea, long strings of peaks are difficult to deal with.

A caldera relation would handle a single volcano with many named peaks, such as 
Mt Akagi - they look like an alligator floating in a pond.

Sometimes, depending on the angle, they appear as a range, but it when viewing 
from above, their circular nature is present. 

The big blue mountain ridges are circular volcanoes (Mt Haruna) and collaped 
volcanoes (Mt Miyogi).

But making a relation to handle your example is very difficult, similar to the 
many long, layered mountain ranges made by tectonic action in Japan. 

I don't know how to handle such things.

Also, Japan is very young geologically, and the extreme erosion from the rain 
has brought several square KM of soil down into the valleys (along with 
quarantary volcanic eruptions) , making soft hills that stick up everywhere. 
Every lump has a name, and even little lumps on lumps have names.

Each of these situations: 

- calderas and very large peaks with named features 
- ranges and ridges of jumbled random peaks 
- lumpy hills and "mountains" on flat/gently sloping  terrain 

Might require a unique approach. 

=Peak is unsuited to handle all three, and simply using ele or a prominence 
score probably can't either. Both are important components, though. 

I agree it will only be mapped correctly via the opinion of local/regional 
mappers - but having them "tune" or "adjust" the values after a programaticaly 
generated solution might be best - however I have no idea how that might be 

Until then, giving mappers the tools to denote "importance" to mountains the 
same way we do for roads (track to trunk) and waterways (ditch to river) is the 
only viable way forward. 


Tagging mailing list

Re: [Tagging] Emergency=levee_breach_materials

2018-09-24 Thread John Willis

> On Sep 24, 2018, at 6:55 AM, Joseph Eisenberg  
> wrote:
> Piles of sand could also be used for flood control (eg to fill sandbags)

Are there permanent mappable locations for such materials - materials 
purposefully set aside for a single purpose? 

I always think of sandbags as a make-shift solution, and sand as a generic 
building material. 

but if there is some permanent stockpile set aside for flood control in other 
countries, perhaps “flood_control_materials” is a good tag value. 


Tagging mailing list

Re: [Tagging] How to tag a building constructed for a gastronomic purposes?

2018-09-24 Thread John Willis

> On Sep 24, 2018, at 7:10 PM, Martin Koppenhoefer  
> wrote:
> I am not even sure if restaurants are a type of retail

Restaurants sell food. Unless it is an “amenity” that belongs to a larger 
thing, I assume that all restaurants are building=retail. this is especially 
true of fast food restaurants. 

there may be situations where a restaurant is a point inside a larger building 
(a tower with a restaurant on top, a hotel with a restaurant inside, an 
apartment complex with a small restaurant on the first floor, etc ), but a 
purpose-built building along a road with a parking lot is certainly retail. 

we have =office and =hotel,  a type of building=commercial, so if you suggest 
we make building=restaurant, that seems fine to me. But in it’s absence, 
building=retail is the correct value (to me). 


Tagging mailing list

Re: [Tagging] Feature Proposal - RFC - Topographic Prominence

2018-09-24 Thread John Willis

> On Sep 25, 2018, at 9:51 AM, Joseph Eisenberg  
> wrote:
> I don't think it works call each of those local high points on Mt Fuji's 
> crater a "hill", if they are all at >3000m elevation with steep slopes 
> dropping >1000 meters down to the valley or plain below.

I think it does - because the rim  and the drop-off doesn't belong to them 
really —  it is a property of the the caldera / overall mountain. 

It is only their height above the low point of the caldera rim that truly 
matters (most of the time). 

the caldera / crater is the “mountain”  - the little peaks are hills around the 
caldera rim. 

they sit upon it, like an ant on my head. they do not get to inherit my head as 
part of their height *when* deciding if it is important or not. Otherwise, each 
hair on my head would count as a =person - which is what we are doing now, 
which seems counter-productive. 

The highest point may be one of those little peaks, and where everyone goes, 
and it has the highest =ele value of all the little peaks, ***but the =ele 
value is shared with the =volcano tag as well***. 

I know there are mountains where the lesser peaks are important — Mt hoei on 
the side of Mt Fuji is a good example — but for most little points on 
calderas/stratovolcanoes I am familiar with, they are little ant-size “peaks” 
that stand upon the body of the volcano. 

They may have a large MSL elevation, but very little importance unless you are 
on the mountain itself. Very few people could name any of them, whereas Mt Fuji 
itself is internationally famous. I assume the same is true for most mountains 
- local, regional, or national). 

This confusion lead to some horrible mis-tagging of mount fuji I just repaired. 

there was a =volcano tag for Mount Fuji, A =volcano tag for the crater of mount 
fuji, a =peak tag for one of the small peaks *with* a (mount fuji) 
parenthetical, and an incomplete =cliff+ =volcanic_caldera_rim line around the 

=cliff+ =volcanic_caldera_rim line for the rim of the crater/caldera. 

=volcano tag for the body of the volcano, centered in the caldera. 

= peak or =hill for the little points around the edge (whatever solution is 
agreed upon). 

it would be nice if there was a "caldera relation" to connect them all 
together, which would allow the rendering of the named, yet overall unimportant 
=peaks to be reduced. 

This would exclude =peaks on the “arms” of large mountains or further down the 
sides (like Mt Hoei), which might have to be solved by another tagging 

Tagging mailing list

Re: [Tagging] How to tag a building constructed for a gastronomic purposes?

2018-09-24 Thread John Willis

> On Sep 25, 2018, at 2:15 AM, Tobias Zwick  wrote:
> I find it kind of unfitting to tag those as building=retail because the
> kind of building is almost like a residential one (or like a hotel).

the buildings look like a hotel (or was perhaps a hotel in the past) - but if 
it is just a restaurant now, then it is building=retail. 

If it is a place where you can rent a private room to sleep, it is a 
building=hotel with a commercial landuse, a pin for the hotel, and another pin 
for the restaurant  (the lobby restaurant in hotels is usually a separate 
mappable place, as it’s purpose, operating hours, and access to the general 
public is different than the hotel itself. 

there are all kinds of amenities - pubs, restaurants, bed, Ryokans, 
fast food shops, etc,  But if they take up the entire building, almost all of 
them would fall into building=retail or building=hotel.  you are tagging the 
purpose of the building - not it’s design, except in the rarest of cases. 

the building=* tag helps define the rough purpose fo the building - but not the 
exact purpose. The pin or other tags on the building do that. And that building 
looks like it wants to sell food to tourists.  ^_^

I understand the “rest stop” nature of the building - there are similar 
buildings in Japan, some larger complexes registered as official “road 
stations” often using the nickname "Oasis” with the government, and others that 
are merely private businesses that provide a place to sit and relax and enjoy a 
coffee - but mapping the small  private businesses that do this as anything 
other than a “cafe” or “restaurant” or “convenience store” is very very 
difficult without some larger complex with a larger landuse with more 

an official “Road station” on a very narrow road in the mountains. 

A pretty large road station down in the valley. 

they are a collection of several shops and amenities - not a single building 
with a single purpose. 


Tagging mailing list

Re: [Tagging] Feature Proposal - RFC - Topographic Prominence

2018-09-24 Thread John Willis

> On Sep 25, 2018, at 5:08 AM, Andy Townsend  wrote:
>> There is an attempt to document what a hill is and how its separated from a 
>> (natural=)peak by separating them on prominence.

TL;DR - you are dealing with a very high volume of named “sub-peaks / 
prominences / whatever-mountaineering-term-you-want ” on large mountains *and* 
a large volume of small and modest hills in valley floors or otherwise flat 
terrain - Perhaps prominence can handle them, but please remember that this is 
not *just* about defining sub-peaks of very tall mountains - there are a lot of 
little tiny hills that in some cultures would never be named that have 
officially mapped names in others. 


There are landscape types *and* cultural naming traditions that lend themselves 
to using =peak exclusively - places with very large mountains with easily 
identifiable points as the top. 

in California, at least down in San Diego, naming almost all mountains using 
=peak is probably acceptable. tiny lumps and bumps on the mountains and on the 
valleys simply aren't named (at least offically, I have looked at a lot of 
topos). Official maps are not cluttered with mountains names. 

I was amazed when I moved to Japan - the mountains are not only much denser, 
but they are very jagged. And since people lived in hamlets around each of them 
for hundreds and hundreds of years, every lump, bump, and little hill the size 
of a 4 story building is a named “mountain”. 

Google even makes fun of it when they made an online ad for “okay google” about 
a tourist family that confused the 30m tall “Mt Fuji” in my town with the 
iconic Mount Fuji when getting directions. 

I was browsing reddit a couple days ago, and someone posted a map from 1843 
that someone drew of the regions surrounding Mount Fuji. it is roughly 400x300 

every one of those little green bumps is a full =peak.  there are probably 5X 
little bumps and lumps on them that are named. **And then** there are the 
little hills that poke out of a valley floors by 20-50 meters that also get 
officially named - you could hide them behind an average size school building - 
but are named. local natural paper maps have names cluttering up every 
available spot on the paper.   

even modest hills get rendered the same as massive volcanoes.

Also, several caldera volcanoes are nationally or internationally famous, and 
they have named every little point along their asymmetrical rim - the volcano 
tag for them should be rendered z8, and the smaller hills around the top at z16 
- but as it stands there is no way to say “these little points on the rim are 
insignificant compared to the 2000m tall volcano visible to 20 million people 
and namesake to thousands of things and places, and these 6 little points are 
only important to people hiking nearby. “


trying to map all of these as =peaks ***reeks*** of data pollution. 

I hope a good solution is found - EVEN IF it is just the mapper’s intuition. 
enough input should provide consensus. 


Tagging mailing list

Re: [Tagging] Emergency=levee_breach_materials

2018-09-23 Thread John Willis

> On Sep 23, 2018, at 6:33 AM, Anton Klim  wrote:
> not levee_repair? 

Hmm - no one at that location repairs levees. It is a station just to store 

I just want it to be unambiguous. 

Like the difference between car repair and car parking

Even then, levee_repair might be good enough. 

I know there are gravel and salt (bunkers?) Along the sides of roads - I'll 
have to see how they are tagged. 


Tagging mailing list

Re: [Tagging] How to tag a building constructed for a gastronomic purposes?

2018-09-23 Thread John Willis
Building=retail if it is a purpose-built building for selling stuff ( a chain 
restaurant, a McDonald's, etc), as it is purpose-built for retail. 

If it is a rentable location in a mixed use building (shop spaces on the 
bottom, apartments on the 2nd floor) we have no suitable building tag for them 
(so =yes). 

Very large apartment buildings (15 stories tall) often have shops on the first 
floor, I would tag that busing as =apartments and pin the shops - it is 
primarily apartments. 

If it is some outdoor pavilion for seating to eat food, building=roof might be 
good, and small tiny eating areas might even be a picnic shelter. For dedicated 
outdoor seating of a restaurant, I am unsure how to tag a structure it might 

A shop/stall/booth/space/unit/location/suite in a larger building is not a 
mappable building (if it is indoors), a pin suffices. Outdoor stalls made of 
temporary materials (but left in place permanently) are mappable, but I don't 
know what as. If it is purpose-built little stalls in a row that never move and 
are at least permanently present,  building=retiail on the row and pins for the 
individual shops. 

Motorway buildings are usually one large building with many pins for different 
shops and amenities. I forget if there is a unique building type for motorway 
services, but I would still probably go with =retail for a large SA on the 
tollway - it is 95% souvenir shops and fast food. 


> On Sep 23, 2018, at 9:11 PM, Tobias Zwick  wrote:
> Hey there
> Do you have suggestions what would be the appropriate building value for
> a building constructed for gastronomic purposes only, such as a
> restaurant/café/fast food/food courts etc?
> In context of Europe, I am thinking of:
> - inns/taverns (eating only) (de: "Gasthof"), for example in forests,
> national parks or other touristic areas, or that typical one inn in a
> village
> - the fast food place around the corner (de: "Imbiss")
> - restaurants and stalls at rest stops
> - motorway restaurants, food courts
> - staple food buildings (McDonalds etc)
> Tobias
> ___
> Tagging mailing list

Tagging mailing list

Re: [Tagging] Feature proposal - RFC - assembly_point:purpose

2018-09-23 Thread John Willis
Thanks - I added a small note about the tsunami elevation in the examples. 


> On Sep 23, 2018, at 9:14 PM, Daniele Santini  wrote:
> Ok, I updated the existing proposal

Tagging mailing list

Re: [Tagging] Feature proposal - RFC - assembly_point:purpose

2018-09-22 Thread John Willis

> On Sep 22, 2018, at 11:03 PM, Daniele Santini  wrote:
> edit the proposal substituting the old tag and values with the new tags or 
> create a new proposal with the new tags?

If you feel the original proposal (or the discussion around it) is valuable, 
start a new one. If it is of minor or no importance, just rewrite the existing 
page. You can note "originally started as assembly_point:purpose=tsunami, etc., 

Under "rationale", you can use it as an example where assigning a single value 
is not flexible enough to accurately tag the multi-use nature of most 

This will also document the originally proposed tag's role in refining the 
currently proposed one, and show up in OSM searches without leading to a dead 
proposal page. 

Tagging mailing list

Re: [Tagging] Emergency=levee_breach_materials

2018-09-22 Thread John Willis

> On Sep 21, 2018, at 4:09 PM, Anton Klim  wrote:
> Do these have anything to identify them, like a ref?

I found a sign cycling today. 


Okudo (village) "Disaster prevention station" 

The "river" is implied - "river disaster prevention station" is a huge 

The government has a PDF listing of a hundred of them and their addresses. 
These might be the very large stations with a highground and a helipad, not 
just a cache of breakwater blocks and gravel. 

I will investigate further, but levee_breach_materials might be the most 
flexible for uses elsewhere.


Tagging mailing list

Re: [Tagging] landuse for government offices ?

2018-09-21 Thread John Willis

> On Sep 22, 2018, at 6:07 AM, Martin Koppenhoefer  
> wrote:
> the important tags here are “religion” and “denomination”,

Yep, those are useful, but it doesn't state that they are "operated" by a 
specific group (the specific temple down the street) - just what is the 
religion/denomination is practiced there. 

Tagging mailing list

Re: [Tagging] Feature proposal - RFC - assembly_point:purpose

2018-09-21 Thread John Willis

> On Sep 22, 2018, at 5:17 AM, Volker Schmidt  wrote:
> Problem: most assembly points are multi-purpose around here. At least fire 
> and earthquake. And they are not marked with a purpose. 

Very true - I think most people assume an assembly point is "safe", as the 
location is chosen because it is low-risk for many types of disasters. 

Perhaps we need to have a few assembly_point:foobar=yes, in case people want to 
map a specific aspect of one - especially if it is *not* good for one aspect. 

Tsunami (height in M) 
Flood (out of the path of a possible dam breach, levee break, or flash floods. 
Tornado (assumed no, yes has to be explicit)

With certian assembly points, the idea it is "safe" from a tsunami is very 
important. Tornadoes will be basements/bunkers/buried shelters, possibly 
fallout shelters. 

But this would be a very small minority of assembly_points. Most will have no 

Perhaps if we can say :tsunami=25 means it is 25m above sea level (the safe top 
of the structure) or tsunami=yes/no to say at least go/don't go here. Same with 

Many of the assembly points in Japan are chosen specifically because they will 
not be flooded if a nearby dam bursts, to be away from known landslide risks, 
and to have no tall buildings nearby to fall in an earthquake. 

:Purpose=foobar locks you into a certian purpose, Whereas :tsunami=yes just 
means it is "safe" from a tsunami - *if you care to map that*. 

Besides tornado, all are implied yes, so an the assembly point inherits all the 
implied traits. 

Tagging mailing list

Re: [Tagging] Area of Firestations / Area of Ambulancestations

2018-09-21 Thread John Willis

> On Sep 21, 2018, at 7:47 PM, John Willis  wrote:
> Landuse primarily for *services* for the public: libraries, community halls, 
> fire stations, police stations, etc.

I messed up: civic_saftey for all manner of Saftey services. Ambulances, fire, 
police, ranger, lifeguard, and other public safety stations may be privately 
operated, but are a public Saftey service. 
Tagging mailing list

Re: [Tagging] Area of Firestations / Area of Ambulancestations

2018-09-21 Thread John Willis

> On Sep 21, 2018, at 6:29 PM, Colin Smale  wrote:
> There is a parallel discussion going on about landuse=civil_administration 
> which might include smaller areas for individual government offices

In my original landuse=Civic proposal, I included all Civic buildings. 

Later, I split off those kinds of things into landuse=civic_services, as people 
suggested Civic was too broad.

Landuse primarily for *services* for the public: libraries, community halls, 
fire stations, police stations, etc.  

Perhaps that would be of use. 

Javbw. ___
Tagging mailing list

Re: [Tagging] Area of Firestations / Area of Ambulancestations

2018-09-21 Thread John Willis

> On Sep 21, 2018, at 6:17 PM, Anton Klim  wrote:
> I’m not sure I understand why it would be a landuse instead of an amenity tag 
> on the area, or the other way round? Are landuses supposed to be for larger 
> areas?

landuse areas can be as small as the land for housing plot, a factory building, 
or a convenience store. 

look at the landuse polygons there. 

Just like parks or farmland, they can be very big - or very small. 

landuse does *not* show zoning or what is “allowed” it is used to denote the 
general purpose of the area when mapped in large polygons **or** to denote the 
extenant of a single map pin (ie, the retail land a mall sits on, the land used 
by a single factory in a industrial zone. if you are in a mixed area - you 
might have a ton of different landuses. ) 

the landuse is often several times bigger than the building, allowing the 
amenities of the site to be grouped together logically 

older tags (like school) use one tag to define the landuse as well, just train 
station or fire station. But this is a tagging discrepancy (to me). landuse is 
used for the other tags discussed. 

I would prefer that this discrepancy is evenutally resolved (landuse=school) - 
but tags like school are way too entrenched to ever be changed. 

new tags created should follow the landuse=* model, but the older way of using 
area=yes+building=yes/no is still common, and some people might like it better 
- or wonder what landuse=* is for. 

Tagging mailing list

Re: [Tagging] Emergency=levee_breach_materials

2018-09-21 Thread John Willis

> On Sep 21, 2018, at 4:09 PM, Anton Klim  wrote:
> Do these have anything to identify them, like a ref

I cannot find an official name or ref - but I can see their purpose. 

The next time I am on a cycling survey, I will take a picture of any small 
plaques they usually put somewhere on a man_made construction. 

I haven’t been able to find any sign onsite or web site explaining them - they 
are for someone with access to big (military) helicopters, so there is 
little-to-no public facing information at the sites themselves - but somewhere 
someone has a PDF with all the locations marked and ref'ed. 

I imagine it is not important to map them, but since I know what they are, I’d 
like to do it. 

All the signs tell people not to dump garbage, and one that says that the 
public parking lot is an emergency=assembly_point, but that is not present at 
the others - most are just stacks and stacks of blocks fenced in, waiting for a 
helicopter year after year. Some of the areas are very small - 100m2 - but the 
contents are just breakwater blocks surrounded by a fence, and they are stored 
adjacent to the levee - so there is only one purpose they could serve. you 
usually only see these kind of blocks near the ocean, protecting coastlines and 
and stacked into breakwaters.  

all of the permanent levee protection is poured concrete and boulders, so they 
are not normal building materials. 

I have found 15-20 caches of these blocks along the river in my area. Some are 
really old. The more I look, the more I find. I imagine there are several 
hundred Just around Tokyo. 

there are very few things that can stop a levee breach:

Levee_breach_materials is a mouthful, but I cannot think of a shorter way to 
make the tag descriptive enough to avoid mistagging. 


Tagging mailing list

Re: [Tagging] landuse for government offices ?

2018-09-20 Thread John Willis

> On Sep 21, 2018, at 7:28 AM, Clifford Snow  wrote:
> Why should farm be tagged as landuse=religious instead of landuse=farmland 
> just because it is run by monks?

I agree. land used for farming (ie: a big field full of rice, wheat, corn, etc) 
should be tagged as farmland. 

A temple ground, with the various shrines, temples, POWs, etc, is 

Obviously, inside a very large temple ground there are trees, gardens, shrubs, 
and maybe a little 5x5m patch of farmland for vegetables - tag those 
accordingly. they are all encompassed in the larger temple landuse. but the 
landuse ends at the logical boundary of the temple grounds - it’s fence, 
parking lot, driveway, etc. The landuse polygon is not representing what it is 
legally owned by the temple - but that the amenities inside “belong” to the 
building or the point. Sometimes that is difficult to discern - but it is 
usually mappable if you know the place you are mapping.  

Large fields surrounding the temple may be owned or operated by the temple - 
but are landuse=farmland +  operator= [religious org]. 

Similar with religious schools. My buddhist school is operated by a “business” 
representing the temple.  

the school is a school, operator=[temple business]. 

The temple nearby is landuse=religious, with it’s graveyard amenity and various 
statues, POW buildings, and a small building=house that the head monk lives in.

Tagging mailing list

[Tagging] Emergency=levee_breach_materials

2018-09-20 Thread John Willis
I ran into an interesting thing when mapping my local rivers. 

All of the rivers in my area have levees running 100% of their length from the 
mountains to the coast, there are probably over 1000 linear KM of earthen 
levees in the Tokyo region alone - 20mx10m levees (or bigger) along the outer 
banks of all rivers. 

the small and the large rivers in my area have emergency levee repair stations 
- hundreds of 1000kg concrete breakwater blocks (like you would see in the 
ocean), a helicopter pad, and access to dirt/gravel. 

They are situated on high ground, above flooding. 

After I found one, I found other collections of emergency breakwater supplies 
along my river, and there are others along the larger rivers. 

I have never seen one of these purpose-built stations before. 

I am unsure of the landuse of such a station, nor is there a good tag for it. 

I thought up Emergency=levee_breach_materials


Tagging mailing list

Re: [Tagging] landuse for government offices ?

2018-09-20 Thread John Willis

> On Sep 20, 2018, at 8:49 PM, Colin Smale  wrote:
> But this discussion is about land usage

Yes, and the Civic functions of governing a people are very different than the 
usage of educating them, treating their health, selling them goods, and 
manufacturing said goods. We have landuses (or at least area uses) for each. 

It's not about the buildings, nor the grass, but the importance people give to 
the set activities and services performed by Civic representatives and 
servants. That makes the location special - so it is mapped differently. 

We have a special name for the building they occupy, and often have large 
multi-building complexes - with it's own name - that deserves the same level of 
mapping granularity we give to office buildings, elementary schools, malls, 
hospitals, industrial complexes, power stations, and transportation centers. 

Even if all the buildings were *exactly the same* and sat on identical square 
plots of land, owned by the same person, one after another down the street, a 
fence separating each, with only the sign out front to denote their differences 
- each would have it's point(s), it's building, and it's landuse tagged in a 
different manner.

I do not want to shoehorn in one type into another for no discernable reason, 
other than "it looks like an office building" - nor does it represent the 
"ground truth" in thousands of mappable landuses. It may not apply to your 
area, but it sure does to mine and others around the globe - possibly in the 
capital city of every region and country on Earth. 

Just because you don't feel it is necessary in your region, that is zero 
justification for it's disapproval. There are other regions where it is totally 
justified. OSM is flexible enough to handle all use cases if we have enough 
categories of points, buildings, and a enough landuses to mix and match to 
represent _all_ the types of situations, not just what you (or I) am familiar 

Tagging mailing list

Re: [Tagging] landuse for government offices ?

2018-09-20 Thread John Willis

> On Sep 20, 2018, at 5:39 PM, Colin Smale  wrote:
> Maybe it's just me, but I really can't understand why landuse for government 
> functions needs its own tagging. The buildings are often indistinguishable 
> from commercial properties

Why does what the buildings look like matter? 

Hospitals look like office buildings. So do schools. We differentiate their 
tagging based on *what we call it and what is expected (regionally) inside*. We 
wouldn't tag a regional hospital or a university as an office park because the 
"buildings look the same". 

Many parks are indistinguishable from natural=grassland or landuse=farmland, 
but we make the distinction. 

In many places, a city hall is a very different place than an ordinary office 
building - even if we don't care that much about them. 


Tagging mailing list

Re: [Tagging] landuse for government offices ?

2018-09-20 Thread John Willis

> On Sep 20, 2018, at 3:25 PM, Andy Townsend  wrote:
> Certainly here in the UK many formerly "civic" services have been privatised 
> and are run for out-and-out commercial gain

Parliament's landuse isn't landuse=retiail, right? 

You are thinking of *amenities and services* - not buildings and landuses. 
That's what I am talking about. 

Mapping shop pins and civic buildings are different. 

If you live in a town where your city council office is in a mall store front, 
please feel free to drop a pin on a mall building. 

But we aren't talking about that. 

We already have plenty examples of Civic services in commercial / retiail 

For example: A DMV "registration reneweal" office in a strip mall is a point on 
building=retiail, landuse=retail, amenity=shopping_center. 

You can find post offices and other government services in a store front in a 
mall (La Mesa, CA post office in Grossmont Center mall, for example off the top 
of my head). There might be other government admin/services in commercial/ 
retail office complexes - but that is a commercial building, on a commercial 

What about public city hall buildings on dedicated land? Legislative body 
meeting buildings? What landuse do they get? The Same one as a chip shop? 

Similarly, almost every large city hall or regional government building in 
Japan has a convenience store in the basement. They are required to be there, 
so there is a 3rd party cashier for revenue stamps. It is a tenant in a little 
4x5m room in city hall. Our regional city government office has 4 (private) 
restaurants on the 32nd observation floor. Those are pins on the building too. 

The convenience store or restaurant is a pin in the city hall building on a 
civic_admin landuse. 

A standalone public town hall **building with external amenities that belong to 
the building**, let alone town hall complex of many Civic buildings should 
never be on landuse=retiail. 

Yep, never ever ever on landuse=retiail. 

Never. Which is why I am proposing Landuse:Civic_admin. 

Tagging mailing list

Re: [Tagging] landuse for government offices ?

2018-09-19 Thread John Willis

> On Sep 20, 2018, at 1:04 PM, Joseph Eisenberg  
> wrote:
> Would you be willing to revive the proposal and get it voted? 

SelfishSeahorse asked me if he could take over the proposal, and I agreed. 

There is not much else I can offer other than what I typed in the proposal, 
though a few years of writing practice would probably let me rewrite it better. 

I left OSM mapping because of a change in schedule and internet access, but 
another change brought me back a few months ago. 

But I decided just to work on repairing the mapping of cycling roads, river 
channels/levees, and other’s horrible farmland mapping here in Japan. 

There is not many missing or incomplete tagging systems for those features, so 
I have few complaints.

I use civic_admin already in tagging, and will continue to do so. 

I thought I’d throw in my 2 cents in the mailing list for old time’s sake, and 
please sign my name as a yes vote when the proposal comes due (i might miss 

Good luck to you all. 

Tagging mailing list

Re: [Tagging] landuse for government offices ?

2018-09-19 Thread John Willis

> On Sep 20, 2018, at 9:40 AM, Joseph Eisenberg  
> wrote:
> It’s not necessary to have a separate landuse area if the government office 
> is in a single building or shop. In that case  the overarching “landuse” is 
> still retail or commercial.

Do they sell "legislation" at a town hall? Does the DMV "conduct commerce"? 
Nope. Retail is always wrong. Commercial is a crutch. 

As for not needing an area polygon, I disagree. People mapping in a detailed 
way will want a point, a building, and an area. There are areas are available 
for many landuses, but not civic.

Here is a convenience store I mapped. It is surrounded by fields, a restaurant, 
Civic buildings and surge resivoir. 

It is a single building on a single landuse, similar to the restaurant, the 
police station, fire station, and rice fields.  

Because it is a single use area - it doesn't need an area tag to encompass it's 
parking lots? Fences? Driveways? I think it does. 

The same is true with any Civic building. 

Even a single shop or a single single city hall building can and (eventually) 
should be mapped with such details, just as we would map commercial, 
industrial, and retail buildings. Point - building - area. Maybe the point is 
merged with the building or the area, but building *and* area are necessary in 
all but the most urban of environments. 

This is why I proposed land use=Civic, later revised to landuse=civic_admin. 

If a town hall or government office is a single building, you could just map 
the building and drop a pin on it. 

But if you wanted to map the extent of the land, using commercial or retail 
(imo) is *never* acceptable. It is not for commerce nor selling goods. We don't 
use landuse=retail for a hospital or a park for similar reasons. 

I wholeheartedly believe the decision to use commercial for Civic buildings was 
wrong, and came from an adversity to making enough new landuse values at the 
start. I want to correct that. 

I would like all landuse=Foo and building=foo to be similar, and be obvious 
what to use with new mappers. Industrial buildings are mapped on an industrial 
landuse. Building=Civic is mapped on... Commercial? Ugh. 

Landuse=civic_admin not only allows the proper mapping of stand-alone offices, 
but is the *only* way to proper way to map Civic complexes of multiple 
buildings with unique names. 

It is useful for both stand-alone buildings or multi-building complexes: there 
are always parking lots, walkways, and other amenities that "belong" to the 
point/building, yet are outside the building's footprint. An area polygon is 
the only way to map them, and landuse=* is the most versitille and consistent 
tag key to do it with.

Javbw. ___
Tagging mailing list

Re: [Tagging] highway=service // public road?

2018-05-26 Thread John Willis

> On May 26, 2018, at 11:54 PM, Greg Troxel  wrote:
> I use highway=service service=driveway for the entire
> length of the way, even the end part that is within the government-owned
> road parcel.



Tagging mailing list

Re: [Tagging] highway=service // public road?

2018-05-26 Thread John Willis

> On May 26, 2018, at 6:09 AM, Greg Troxel  wrote:
> The fix is to add highway=alley, for things that are too small but
> nevertheless fit the legal definition of a road, and then have service
> be for "place you can drive that is not a legal road".

Sounds great to me. 

If you want to keep service completely exclusive to roads you would find on 
private property or "behind the entrance gate" kind of roads, then having a 
sub-tag value that implies public access (service=alley) is not good. 

This would also solve having a road "lower" than unclassified for some 
particular uses. 


Tagging mailing list

Re: [Tagging] highway=service // public road?

2018-05-24 Thread John Willis

> On May 25, 2018, at 2:29 AM, Florian Lohoff  wrote:
> Interestingly the key:highway wiki page lists unclassified as
> the lowest classification of a road:

That’s weird - since service=alley seems to be the lowest class, being service 
and all - yet "alley” is a public road. 

I personally think that the idea of “alley” (urban or rural) is a great 

When mapping in California, I would rarely encounter a road (urban, suburban, 
or rural) which would be a public maintained asphalt road with severe physical 
width limitations (under 2m) or is used only for local (but not necessarily 
residential) access. most rural roads are unmaintained tracks, and the the 
existing OSM tagging neatly fits. I’ve driven thousands of miles of grade 2-5 
tracks all over California. 

Then I came to Japan. Japan is a spaghetti pile of 1.5m wide alleyways,  2.5-4m 
residential and unclassified roads, and overbuilt tertiary bypasses and trunk 
roads bulldozed through the whole spaghetti pile. 
Japanese maps have to accurately convey *MASSIVE* disparities in width, turn 
radius, shoulders, expected hazards, etc (the road classification), - which is 
why most consumer maps use hand-drawn “area based” maps at z16 and above, 
usually provided by the government or private mappers such as Mapion. It is a 
wholly different mapping Job than mapping anything in Southern California - 
except perhaps adjacent to the ocean, where service=alley is common as well.  
there are simply more levels of road to map, mostly below tertiary. but we only 
have 2 grades until we get into tracks, which is not enough. 

Because of the extensive land management over hundreds of years, there are many 
public paved, maintained, yet very narrow and limited usefulness roads all over 
the urban centers and rural countryside, often used only by farmers to access 
the farming tracks that further subdivide the farming land. I personally use 
them when cycling - and the distinction between a public paved road that cuts 
through a rice field along a levee and a narrow logging track in the mountains 
is huge. 

This leads to rural maps of farmland that are an order of magnitude more 
complicated than most suburban neighborhoods in California. And since no road 
is straight, and everything is spaghetti pile, choosing the best route through 
it via basic classification is a must.

This is where unclassified falls down, and the idea of a publicly accessible, 
yet very restricted and narrow “allley” road comes into great usage. Using 
service=alley allows further classification of the road network down below 
unclassified - because in places with as much variance in the road network 
(like in Japan), this is necessary to properly convey the distinction between 
roads that are useful, yet remote -  and roads that loop off of them to allow 
tractors access to their sets of fields, or roads adjacent or on top of levees 
that are technically passable by public cars but are often too narrow to do so 
regularly. This is not necessary in all places. In addition, there are still 
tracks, private driveways, cycling roads, trails (ugh - no proper way to map 
those either), sidewalks, stairs, and other pedestrian considerations out in 
the middle of nowhere as well. 

In other countries, these would easily just be grade 3 unmaintained tracks. and 
in some places in Japan, they are. There are plenty of grade1 “paved” tracks as 
well. But most of these very narrow and marginally useful roads are mostly 
publicly maintained roads, and part of the road network - but add unnecessary 
noise into a mapping program. map viewers and routers should be directed to a  
proper road 1m wider for normal unclassified car traffic nearby - and looking 
at a pile of spagetti unclassified roads is not useful to data customers, 
routers, nor map users. 

Dumping these tiny roads all into unclassified to be defined by “width” or by 
“smoothness” would not capture the true nature of the roads - they are rural 
alleys. not tracks. Not residential. below unclassified. 

So I map them as such, and will continue to do so, under the idea of “alley”.


Tagging mailing list

Re: [Tagging] access=disabled

2018-05-17 Thread John Willis


> On May 10, 2018, at 9:19 AM, Warin <> wrote:
> Hi,
> I'm tagging a 'disabled parking area' - these are fairly common in my country.

I know I am just jumping in - but this is also something I am interested in. 

I know if we have a big parking lot waiting the a few disabled spots along an 
isle near the store entrance, than capacity:disabled=4 added to a normal 
parking lot is appropriate.

But the instances I am trying to map are large disabled-only lots. They 
(sometimes) have a gate that the security guard opens, allowing anyone with a 
disabled plackerd to enter. It is a separately mappable lot near the normal 
access=customers. Most of them are physically separated from any other parking 
by kerbs and shrubs. 

I really think access=disabled is appropriate for this parking lot. All others 
are denied. 

Having to map space-by-space to just show that this *lot* is disabled-only 
seems weird. 

Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-17 Thread John Willis

> On Apr 12, 2018, at 6:28 PM, Marc Gemis  wrote:
> So what are you going to use/propose ? landcover or man_made ?
> Curious, because I want to correct my mapping.

I Have been thinking about it a lot, and I think we need two separate tags - 
one for the object and one for the “surface” (landcover). 

So I think there is a need for both man_made=flowerbed and landcover=flowers — 
maybe this is overkill, but I want to follow the theme followed by other tags. 

But there is a discrepancy...

Other types of plant-based tags have the ability to map an individual item 
(tree, shrub), as well as a linear line (tree row, barrier=hedge), and an area 
(scrub, wood/forest). 

Flowers are difficult because in some usages they are shrubs - like the 
distinct, named, cataloged, and mappable-from-imagery rose bushes in a rose 
garden; in other usages they are tiny plants in massive quantities that should 
only be mapped by area. 

So, I think it would be good to use the existing natural=shrub with 
man_made=flowerbed when mapping large gardens of carefully managed flowers, and 
to have a general landcover=flowers to use for all other circumstances. but I 
am not sure how to use flower_bed by itself when it contains unnamable “tiny” 
flowers or a field of nameless, unmappable flowers, like nemophila or a named 
breed of a single flower (like tulips). would you combine man_made=flowerbed & 
landcover=flowers in such a situation? similar to a road, park, pitch, or track 
- we map the item *and* what it’s “surface” is separately — is such a 
distinction necessary here? 


Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-12 Thread John Willis

> On Apr 12, 2018, at 4:43 PM, Martin Koppenhoefer  
> wrote:
> landuses don’t refer to features like a house or a supermarket, they refer to 
> a purpose/activity class like retail, residential, industrial or commercial.

Yea, you got me on that one. You are right about the landuse tag being wrong 
for flowerbed. I somehow got mixed up on this one. 

I understand the reasoning against usng landuse now. 

I think we should have some more landuse categories, but I got mixed up 
somehow, equating usage with category. 

Thanks for the continued discussion and comments from everyone. 


Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-12 Thread John Willis

> On Apr 12, 2018, at 4:48 PM, Martin Koppenhoefer  
> wrote:
> I thought you also agreed they could be seen as a kind of garden? It isn’t 
> the only possibility, one might also see them as a kind of park, or maybe 
> even a kind of meadow, but my choice, from what I imagine, would be kind of 
> garden.

I assume a basic flower park is a garden, and a big "spectacle" like a huge 
field of flowers (non-farm, grown and marketed for tourism) is a garden as 
well, but should be defined with a new garden:type=* tag of some kind. 

Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-11 Thread John Willis


> On Apr 11, 2018, at 7:37 PM, Martin Koppenhoefer  
> wrote:
> An area with decorative flowers organized and presented in flowerbeds, with 
> visitors and maybe a fee, will be something, like a flower show, with a name, 
> maybe a website etc.

That is the overall Garden.  Garden:type=flower_field. Or flower_Spectacle. 
Whatever people suggest.  Tag the fees, name, website, etc on the enclosing 
garden tag (along with tourism=attraction). 

I micromap tourist destinations. I tag the walking paths and walls. Hedges and 
driveways. Individual trees. Fences, gates, parking lot refs and individual 
vending machines. Many of the places I personally visit are flower parks. I 
have visited  9 flower parks in the last few years. I map places I visit 
and revisit. Most are large outdoor gardens that fit into existing garden tags. 
A few of them are either a stand-alone flower fields or a garden/park/ with a 
named feature that is a dedicated large flower field (like that hill of blue 
nemophila). But all of them have beds of flowers with areas easily mappable 
from imagery. 

I need another tag that says "this area here visible in imagery is decorative 
flowers". This exact 5m2 Or 100m2. If that is landuse=flower_bed or 
man_made=flowerbed or whatever, it's okay with me, I would like to "approve" it 
so we can get it documented, along with the garden:type=value. 

Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-11 Thread John Willis

> On Apr 10, 2018, at 11:09 PM, Martin Koppenhoefer  
> wrote:
> they are. A flowerbed is about something human made. What you have been 
> posting is a forest


There are many natural spectacles (the fall colors on the mountains, certian 
flowers that grow on wild hills, etc) that is a form of attraction, but is 
neither a subset of garden nor a flowerbed. 

Just like a fountain in a park and a waterfall in the wilderness. 

Just to muddy the waters, several places I take pictures of flowers "in the 
wilderness" are transplanted and cultivated by the locals. Some of the flowers 
are native to other regions, and transplanted to similar climates to recreate 
the natural spectacle, and to diversify the locations (in case a volcano 
explodes and kills all the others in one spot). They care for the plants and 
increase their density to keep the (moneymaking) attraction. But these are 
pretty rare compared the flower spectacles I am talking about. 

There was an eruption on Mt Kusatsu-Shirane a few months ago; it was 400m from 
some mountain flowers they painstakingly transplanted a couple decades prior. I 
hope they all lived.


Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-11 Thread John Willis
Actual flower Farms are landuse=farmland crop=flowers. Yea, they may have a 
viewpoint and a gift shop. But those large commercial farms are not what I'm 
talking about. 

These are about tagging the actual beds of decorative flowers with 
landuse=flowerbed (which I think is totally a landuse - it is land dedicated to 
flowers for display or decoration), and tagging gardens that are "flower 
spectacles" - places that grow flowers primarily as a spectacle (and often 
charge admission) using a garden:type=foobar is the two tags I am asking for 
feedback on. Landuse=grass is crappy - is it for sports? picnicing? Roadside 
shoulder? Landscaping? 

Luckily flowers in a non-farm sense serve a single purpose - to be looked at. 
They are colorful decorations. You don't sleep on them. You don't play sports 
on them. People grow flowers in dedicated land merely to be enjoyed. 

Several places around the world grow tulips and build a Dutch windmill to 
emulate a working landuse=farmland - but just as Space Mountain is neither a 
spaceship nor an actual mountain, these are tourist attractions made to emulate 
the look of a farm for people looking to take pictures. These fall into the 
category of "flower attractions" and I want to tag these as such. 

When I lived in San diego, the only thing I had ever seen like this is the 
Carlsbad flower fields. There are formal botanical gardens and rose gardens - 
but a town or large commercial park just doesn't purposefully grow very large 
fields of flowers in a large field and put out a viewing platform like they do 
in Japan *and* get hundreds, thousands, or tens of thousands of people a week 
that come to just merely view the spectacle  that they purposefully made, year 
after year in the same spot and static configuration. 

Maybe it is common in the rest of the world, but these flower spectacles (and 
their dedicated area just for flowers) seems something that needs precise 


> On Apr 10, 2018, at 2:19 PM, Clifford Snow  wrote:
> In John Wills original post he talked about tulip farms. T

Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-09 Thread John Willis

> On Apr 10, 2018, at 5:47 AM, John Willis <> wrote:
> landuse=flowerbed

Although I searched the wiki and didn’t find the page I was looking for — when 
I googled for it, I found a “defacto” page made for landuse=flowerbed. 

1200 uses. 

I think making this page more fleshed out would serve well. 

Define it as an area used primarily for decorative flowers and other ornamental 
plants (that are not hedges or trees). 

That tag can be used in the decorations seen out front of a mall or school, the 
beds of flowers in a rose garden, large outdoor flower installations, or a 
decorative display found in roadside landscaping.

Let the larger landuse define the purpose (garden, retail, park, median, etc), 
just like we use fence, wall, or parking. 

Tagging mailing list

Re: [Tagging] Flower fields as tourism attraction

2018-04-09 Thread John Willis
During this discussion it is evident That we need a macro "venue" tag *And* A 
micromapping "flower bed" tag together. 

let's take a Rose garden for example. 
The garden itself is one big landuse. The standard "garden" tag should suffice. 

Then there are all the little pieces inside. The paths, fountains, etc. The 
Rose garden is 30-40 smaller flowerbeds bordered by paths and walls. This is 
where we need the new micromapping tag. 

Landuse=flowerbed. (Flower_bed?) 

Next: Tulip "attraction" grounds (not the actual commercial farm). 

First problem - how do you tag the entire venue? It is a single landuse with 
parking, shops, etc. Some new value for the existing garden:type=* subtag 
should probably be created for these outdoor large-scale "spectacles". 

Then map the paths and tracks and driveways and the windmill as normal. 

Then the tulip fields themselves. Theey could be mapped as one big field 
between the service roads. My venue would use 2-3. 
But the tulips are mapped to their color and variety (they each have a name) 
and we are now back to having 30-40 small landuse=flowerbed. 

Lastly:  a large single "spectacle" my flowers on the hill, the sunflower 
displays for tourism (as Dave mentioned, happens here too). These often 
re-occur in the exact same spot year after year and a small toursit spot is 
open for the few weeks it is beautiful; part of the late summer feastivals. 

Same new garden:type=* tag. Paths and parking. The viewpoint, if present. 

Then a large landuse=flowerbed for the big field of sunflowers or nemophila or 


> On Apr 9, 2018, at 9:15 PM, Martin Koppenhoefer  
> wrote:
> maybe it could be tagged as garden with fee and a (new) subtype? For reference

Tagging mailing list

[Tagging] Flower fields as tourism attraction

2018-04-08 Thread John Willis
TL:DR - we need a “flower field” tag or a variant of flowerbed.  tat is not 
related to farming, but more to garden or tourism. This is a tag for the field 
itself, not the entire location, which may need a garden:type=spectacle value 
to define it. 
 (yay! more 
landuse= values!) 

I visited a tulip field In Japan yesterday; thousands of people coming to take 
photographs of a field of flowers and the Dutch windmill they built. 

This is very similar to the flower fields (and windmill) in Calrlsbad, 
California - the tulips and other flowers grown are a spectacle to attract 

It is not a commercial farm which happens to be pretty.  it is a carefully 
planned place to attract tourists to see the flowers as a spectacle - like a 
garden or park. 

I understand that in Holland, those are commercial farms (landuse=farmland 
crop=flowers), the beauty is incidental. But people recreate the spectacle on a 
smaller scale as purely a tourist attraction.

Similarly, Hitachinaka Park has a large hill which they rotate the crop of 
flowers (Nemophila is the most popular) and tens of thousands of people come to 
see the flowers every season.

I tagged the flower area on the hill a year or so ago - but was unsure of how 
to tag the flowered ground. I left it untagged. 

Visible here 
Spring) . 

There are also large fields of colorful shibazakura that are planted as a 
tourist attraction. in many large flower grounds. 
In all these instances, the flower fields are a tourist attraction. their 
primary purpose is similar to a botanical garden - they are grown to be viewed 
- but in a garden/park like atmosphere. 

These images are very popular online, especially the large field near Mt Fuji. 


People may buy some flowers, but it is incidental - people primarily come to 
see them. It’s not flower art - there is no “image” grown into the flowers - 
the flowers themselves are the attraction.  

having it tagged as farmland seems wrong. It’s not a crop. It’s not a plant 
nursery. It’s not a botanical garden.

 It’s not a “park” -  the entire location may be a form of “garden” tagged as a 
tourist attraction, but the **field itself** needs a tag. 

Flowerbed? seems a little weird to tag 3000m2 as a flowerbed. But if it is 
approved I will use it. 



Tagging mailing list

Re: [Tagging] Area of loading docks = ?

2018-03-27 Thread John Willis

> On Mar 28, 2018, at 9:53 AM, Graeme Fitzpatrick  wrote:
> How about loading / unloading / freight _area, still with access=no?

so we would have to create a tag - amenity=freight_area ?

Tagging mailing list

[Tagging] Area of loading docks = ?

2018-03-27 Thread John Willis
I was looking how to tag the area used for loading & unloading freight, usually 
a “private” area of an otherwise public retail building.

I would call this area a loading dock, but “loading dock” 
 is used to tag 
the individual docks that trucks use at this larger area. 

Mapping this area is not an issue at a commercial cargo warehouse (it is the 
entire building & area), but many commercial retail buildings have a private 
freight loading dock for good deliveries that take up 10-20% of the area around 
the building.

example areas: 
  Maebashi-Minami Costco 
 Ota Aeon Mall (one of 4, but the 
only outdoor one). 
 Costco Santee 
 Home Depot Santee 

For some buildings, this is merely a driveway and an entrance to a totally 
indoor area (common at shopping malls), but for a lot of retail places this is 
often an outdoor area, with different private amenities (freight storage, 
palettes, etc), delivery truck parking, etc. that are hidden away from 
public/customer access. Tagging that this is a “delivery” or “freight” area, 
and is access=private, (as opposed to access=customers) would be nice. many 
larger facilities (a warehouse store, a Shopping Mall, etc) have a separate 
entrance and area dedicated to these, so mapping them is easy, but other 
facilities are much more ambiguous, and being able to separate the areas would 
show what areas are not for customer access - and where a delivery truck driver 
should go. These often have names/refs for a very large facility that may have 
more than 1 (like myAeon Mall example), and then the loading docks would have 
individual refs and tags. 

I assume amenity=loading_dock is used on nodes where this area meets the 

What would this larger (outdoor) area be? calling it amenity=parking seems to 
be wrong. 

searching the wiki doesn't offer any clues. 

How would you go about mapping these areas? 

Tagging mailing list

Re: [Tagging] Feature proposal - RFC - Aviation obstacle light

2018-03-08 Thread John Willis

> On Mar 8, 2018, at 6:27 AM, Warin <> wrote:
> airmark:light I think is better. 

there are already approach lights, PAPI lights, etc. I think saying that these 
are obstacle / hazard warning lights is appropriate. 

Beyond mapping runways and terminal buildings, The airmark stuff is very 
specific because people flying planes all use extremely specific language for 
things to avoid confusion. 

Javbw. ___
Tagging mailing list

Re: [Tagging] wayside shrine tag definition page

2018-02-21 Thread John Willis

> On Feb 21, 2018, at 9:18 PM, Martin Koppenhoefer  
> wrote:
> I would not consider a freestanding statue or several of them (not in a niche 
> or covered or inside some other construction) , a “shrine”

They are usually the same objects inside other wayside shrines (as I understand 
it).  The other “statues” are simply luckier to have a roof. 
I understand that mapping objects - small wooden or similar structures too 
small for people to enter vs a statue seems like a good distinction to make, 
especially for “duck” mapping. 

but similar to how we would map bicycle parking as covered=yes/no, the parking 
is the major concern, rather than then the rain cover. 

Maybe it is proper to separate the freestanding jiso statues into a separate 
category (wayside_statue), but Tomoya is approaching it as they are both 
wayside statues of jizo. 

I believe these are the same statues. They serve the same purpose. they both 
get little clothes made for them and prayed to. one group has a roof, the other 
doesn’t.  this is a way_side shrine on a larger POW ground, along a walking 
path to the main building. They are pretty common too. 

Because they are basically the same thing, Tomoya chooses to view them like 
bicycle parking (one category, covered=yes/no), while you see them as separate 
wayside_statue / wayside_shrine. 

It might be better to let wayside_shrine be kind of a catchall, and make a 
wayside_shrine=* subtage with extenable values, rather than making a new tag 
for every object: column, painting, statue, etc to keep them grouped together - 
but still allow proper definition. 


Tagging mailing list

Re: [Tagging] reviving hollow way

2018-02-19 Thread John Willis

> On Feb 19, 2018, at 8:32 PM, Colin Smale  wrote:
> cutting=sunken_lane feels good.

this feels right to me too. 


Tagging mailing list

Re: [Tagging] Cycling "service area"

2018-02-18 Thread John Willis
TL;DR - highway=wayside_station  ? 


> On Feb 19, 2018, at 12:40 AM, Volker Schmidt <> wrote:
> At present the wiki says that highway=services is mainly for motorways and 
> similar roads, but we have many places now that look like highway=services, 
> but are on single carriageway roads and offer similar  services, but are 
> tagged as amenity=fuel.

Japan has these too, growing from the “Road Station” idea. there are over 1000 
registered road stations. many of them are on heavily trafficked trunk roads, 
so in urban areas, it is probably a massive 40m wide road, but they are also 
found on narrow single carriageway roads high in the mountains, for people 
coming to or going through the mountains to popular destinations. 

All of them are registered with the government, which I bring up not to 
validate them - but one of the requirements to me of a “Service Area” - a 
singularly named place that says “I am a Service Area”. 

The Japanese road Stations are all called “Michi-no-eki”  literally a "station 
of the road". often written as “Michi-no-eki Foobar” or "Road Station Foobar” 
in english.  People know that it is a place to stop. While they may not have 
gasoline (trunk roads have gasoline stands along the road), they basically 
offer the other amenities of a service area in a singularly named location. 

the creator of the english wikipedia page says Japan “officially” translates 
them as “roadside stations"

We got into (well, I got into) a long discussion about how to tag road_stations 
last year, with:

Tomoya saying they are simply named landuse=retail, since they usually do not 
offer gasoline, making them ineligible to be highway=services.
I said they need their own subtag, highway=services & services=road_station, as 
they are a variant to me. 
Paul saying highway=services retail=road_station might work. 

I quoted the entire last email at the end. 

Perhaps  we should make highway=wayside_station to cover all of the 
transportation methods. this would avoid the trouble with ‘Service_area” 
confusion. you can apply it to a landuse=retail or whatever environment your 
wayside_station happens to be.  

his would also give a proper tag to non-motorway stops that fit into the same 
traveller focused area that is way adjacent, using landuse=retail to show it’s 
operating many shops under a single POI, with separate communal public 
restrooms and eating area (shelter) denoted by the highway=wayside_station 

This “communal amenity use” distinction is present in other tags - a road lined 
with individual shops is different than a mall, an amusement park is a 
collection of attractions.  It is a distinction I don’t want to overlook when 
tagging these kind of places. 


Quoted email From Paul, Jan 27th 2017

Re: [Tagging] highway=services & "Road Stations" - subtag requested.

> On Wed, Jan 25, 2017 at 7:34 PM, John Willis < 
> <>> wrote:
>>> On Jan 26, 2017, at 12:45 AM, Paul Johnson < 
>>> <>> wrote:
>>>  Usually just a vending machine in a cage and some overused, 
>>> under-maintained toilets and lots of parking.
>> yea, i’m more familiar with the view points (Interstate 5 near camp 
>> Pendleton) and that style of crap rest area near Castaic lake. Maybe the map 
>> stop on interstate 8 as you come into san diego. I have driven 300,000 miles 
>> in California, and those are the 3 I can name.  they are usually barren 
>> dirty places you stop at only if you have to.
> I wouldn't say that's typical of CalTrans rest areas to be that horrible.  
> Randolph Collier 
> <>
>  seems like it may have been a small state park with campground at one point, 
> it's quite nice.  Solano Westbound 
> <>
>  is a little bit of a dirty trick, though.  If you're in the middle of a 
> transcontinental drive and heavily loaded, you're going to have to stop there 
> because you probably need to let the engine cool off a bit after the long 
> climb up from the California Valley to the east; good taco truck, 
> understandably rekt restrooms, in no small part of being in the middle of the 
> suburbs of an unbelievably spread out metropolis.  You can see all the way to 
> the Pacific Ocean from there, so worth stopping if you haven't made the trip, 
> even if you don't need

Re: [Tagging] Cycling "service area"

2018-02-18 Thread John Willis


> On Feb 18, 2018, at 10:44 PM, Volker Schmidt  wrote:
> what's the difference between a service area on a road that is tagged with 
> highway=services and one that is tagged amenity=fuel

I'll write a full response tomorrow, but highway=services (SA) is a singularly 
named place with a collection of specific amenities with dedicated access along 
a motorway. Fuel is one of those amenities. It is important to defining a SA, 
but it is not an SA itself. 

Similar to how a mall is a singularly named location full of shops and 
amenities for shoppers, it is the collection of amenities  together for a 
particular usage in a singularly named location that makes a Service Area, 
along with specialized access. 

Tagging mailing list

Re: [Tagging] Cycling "service area"

2018-02-18 Thread John Willis


> On Feb 18, 2018, at 10:11 AM, Warin <> wrote:
> service _area:hgv=yes

Isn't a highway=services already for HGV ? 

The issue I ran into is highway=services that are not on a motorway (a Japanese 
road station) - but how does a facility "cater to" HGV and cyclists at the same 
time? I know I can get rest and food at a highway=services  along a trunk road, 
but it is a place not tailored to cyclists. The Cycle_service_area that led me 
to think of this tag doesn't even have a single designated car parking spot! 

The other issue is that highway=service_area already has an automatic redirect 
page to highway=services on the wiki. Creating this tag value is probably a bad 

Considering motor_vehecles have highway=services, creating 3 tags: 
cycle_service_area, foot_service_area and bridle_service_area would match the 
three types of current non-motor ways.

I am unsure that is is the best idea, but the idea of matching the existing 
_path types seems proper. 

Tagging mailing list

Re: [Tagging] Cycling "service area"

2018-02-17 Thread John Willis
Thanks for all the comments everyone. 


> On Feb 18, 2018, at 5:43 AM, Warin <> wrote:
> highway=service _area
> service _area=bicycle, hgv, horse,  etc 

Warin's suggestion is probably best, that way the wheel doesn't have to be 
reinvented for each type, service_area can just get another subtag value. 

If we were starting from scratch, I would use highway=services and then 
services=foobar to define type. 

But highway=services is explicitly for motorway service areas now; I think 
trying to extend that tag will just lead to confusion. 

I think hgv is already covered by motorway=services. Am I wrong? 

Please reply back to tell me specific services offered for a type (beyond 
food/shelter/toilets) That are at facilities which are: 

Self-identified, way-adjacent, type-specific facilities offering specific and 
general amenities to that type of transient traveller (kayak? Cycle? Horse? 
Hiker?) That isn't a park, sports complex, convenience store, or a retail shop.

- self-identify as service_area for a specific type. There may they be a nice 
convenience store along the route with a bike pump and a parking rack - but it 
is just a 7/11. 

- Must not be confused with a facility dedicated to that activity - an 
equestrian center, a MBX off-road park, etc. Service_area is a facility for 
transient visitors, just passing through.  

- Must be a facility along a dedicated route (bridlepath for horses, for 
example) or busy area for the specific type.  This is not a shop in a mall. 

- Must offer amenities specific to the type (bicycle maintaince, watering spot 
for horses). Additional non-specific amenities are usually present as well 
(food, shelter, toilets), that could be used by anyone visiting there, but that 
is not a defining feature. A park could have a picnic shelter, a toilet, a 
cycle rack and a vending machine. It is a park. 

- Must be more than a souvioir shop / [type] rental / torestaurant/ [type] 
shop/ or a convenience store with parking. It may have a restaurant or such, 
but the pesence of such does not define it as a service_area. 

With feedback, I can include some common types in the draft page and use some 
amenities offered as defining features, and present what existing tags go with 
such a facility. 


Tagging mailing list

Re: [Tagging] Cycling "service area"

2018-02-17 Thread John Willis


> On Feb 15, 2018, at 11:54 AM, Dave Swarthout  wrote:
> think the highway=cycle_service_area tag is both useful and specific enough.

I will make a draft page for it and start using it for the very rare places 
here. I assume the rest area in the OSM question thread is a sun protection or 
picnic shelter with name=Cycle Rest Area. 


Tagging mailing list

Re: [Tagging] wayside shrine tag definition page

2018-02-17 Thread John Willis

Tomoya - Thanks for the translation help! I wasn't expecting that. 

 This particular building is 300m into the forest, at the top of a small hill, 
accessed only by a walking trail. I assumed it was a shrine for the hill it is 
on. Would a Mikoshi be stored so far away from the road? 

I know of several portable shrine storehouses (two story tall "garages" for the 
large wagon-style ones used in Omama and Kiryu), but I do not know a lot about 
the smaller mikoshi carried by a few people. 

Is "building=garage" standard tagging for mikoshi storage buildings of any 


> On Feb 16, 2018, at 9:49 PM, tomoya muramoto  wrote:
> >
> * 1st and 2nd: Shinto shrine?
> I cannot deside its a Shinto shrine or not. It could be a Shinto 
> garage(Mikoshi is placed in it).
> If so, I would tag it as building=garage(or yes) + religion=shinto (without 
> amenity=place_of_worship).
Tagging mailing list

Re: [Tagging] wayside shrine tag definition page

2018-02-15 Thread John Willis

> On Feb 15, 2018, at 9:41 AM, Martin Koppenhoefer  
> wrote:
> +1 to require „enshrining“ for a shrine.

For some context, here are 4 pictures.

The first two is a remote Shinto shrine (building=shrine) located on a hilltop. 

It is a building the size of a garage.

The third is a small wayside shrine for a natural spring, visible as white 
bubbles on the left. A much larger temple (?) Is 30m away.

The last one is a mystery to me. There are often carved tablets or writing 
carved into the stone near certian natural features. These are on enoshima 
island near the cliffs and tidal flats, with a view of Mt Fuji. I see these 
kind of carvings and tablets - but I have no idea of their origin or purpose. 

They are below this building.

Similar to this carving in Kyoto in a temple complex

Or this tablet in a rural mountain forest.

They might all have different meanings and completely different form one 
another - but represent the kind of "wayside objects" I see that are in the 
smaller side. 

Tagging mailing list

Re: [Tagging] wayside shrine tag definition page

2018-02-15 Thread John Willis

> On Feb 15, 2018, at 10:53 PM, tomoya muramoto  
> wrote:
> I added some photo example to wiki, 

Thanks for telling us about things going on in the Japanese discussions and 
adding examples. Please correct any of my statements if I get something wrong. 

I was surprised to find man_made=torii when I was napping near Mt Haruna. I 
assume there are changes to the Japanese tagging that happen that I don't 


Javbw ___
Tagging mailing list

Re: [Tagging] wayside shrine tag definition page

2018-02-15 Thread John Willis


> On Feb 15, 2018, at 5:44 PM, Martin Koppenhoefer  
> wrote:
> This seems to imply that wayside_cross is a subclass of wayside shrine (all 
> instances of crosses are also shrines)?

I think it is really difficult to tease apart "memorial" from a lot of these 
religious objects. I imagine the wayside cross and shrine in many cultures are 
put there to memorialize someone who died or some event that occured. I imagine 
(without fully understanding) that this is true in Japan for some of them. 
Others, particularly the small ones near some natural feature (tiny waterfall, 
small hill) are for that feature, not as a memorial to someone - they are for 
the natural spirit of of that feature. 

There is also so much decorative Christian iconography in some places that it 
may be hard. I can see the second example (statue in the wall) as possibly 
being a wayside something. 

Wayside crosses are usually very personal to the person that set them (as I 
understand them being for someone who died), similar to the more secular 
memorial plaque, whereas man_made=cross or =torii are usually very large 
symbols for symbolism's sake, or a part of a landuse=religious complex. 

But figuring out what is a wayside shrine, a memorial, or just some painting on 
a wall could be really hard for someone not very familiar with some of the edge 

Tagging mailing list

Re: [Tagging] wayside shrine tag definition page

2018-02-14 Thread John Willis


> On Feb 15, 2018, at 9:41 AM, Martin Koppenhoefer  
> wrote:
> What do you think about the column, does it meet your expectations for 
> wayside shrine?

If the object is religious in nature, I think it is. 

Very very very old wayside shrines in Japan are usually destroyed. The weather 
demolishes everything. The small stone figure (of indigenous gods?) Carved into 
a stone is still on a rock along the road or sitting on a wall. In urban areas, 
they get small little platforms or shelters, like the one shown in the thread. 
They are sometimes POIs in local maps. 

This column seems to be like that - a religious object dedicated to a 
worshipped spirit right there, nearby, or a place for locals to leave small 
offerings to the dedicated spirit/dieity/god. Can be a small case for holding 
very small objects or accepting donations as well. 
Tagging mailing list

Re: [Tagging] wayside shrine tag definition page

2018-02-14 Thread John Willis
There are tons and tons of little singular Budda statues on a pedestal and 
perhaps a little roof,  and people leave coins, sake, and dress them in 
jackets. There are also very tiny Shinto "shrines" on the top of mountains 
(like the size of a shoebox or microwave) that have a spot for offerings and 
candles and whatnot. 

There are probably several million of these in Japan. Most are not mapped.

To me these are wayside_shrines. 

To me, a building=shrine / temple "enshrines" an object or offers a place for 
worship of a statue or object. 

My School's temple has a gigantic Budda in the main temple. You come pray in 
front of it. You can shrink down the building and the statue a lot and still 
have it be "a temple" using these 2 rules: 

1 - There is (somehow) room for someone to go inside. Basically the smallest 
building=temple or building=shrine is about the size of a garden shed. It has a 
door, a person or 3 people can go inside,

2 - the building itself is recognizable as an object itself - unlike the 
way_side shrine or roadside cross or similar - just a little cover or roof over 
a statue and an offering box. 

There are many of these small "garden shed" POWs - but they are mapped and 
considered "shrines" or "temples" by people. 


> On Feb 12, 2018, at 10:06 PM, Nelson A. de Oliveira  wrote:
> For example, this Buddhist shrine doesn't seem to have any space or
> cavity (the Buddha is placed on an altar, it seems).

Tagging mailing list

[Tagging] Cycling "service area"

2018-02-14 Thread John Willis
I have come across an interesting mappable feature when out cycling last 

The ~60 km long cycling roads along the Rivers here in Japan occasionally have 
city operated facilities, and similar to the centuries old “road stations” and 
the more recent motorway “Service areas”, I found a “Cycle Center”, best 
described at a cycling service area. Someone has mapped it as a building. 

the Japanese phrase サイクリングセンター “Cycle Center” is used for these buildings. this 
is a play off of the “road station” name, but tagging as 
highway=cycle_service_area seems proper. This is where an east-west and 
north-south cycling route cross, and the cycle center is a public facility. It 
has public toilets, cycle parking, outdoor sinks, air pumps, small information 
office, a room for getting out of the weather and sitting and eating lunch, 
showers, a few vending machines, a nurse’s first aid room,  and a (seasonal) 
cycle rental service. 

this sounds a lot like a motorway Service Area to me.

I know the features are mappable by themselves, and that this kind of 
purpose-built facility is rare, but they seem to exist. The feature is a named 
and singular mappable object with multiple amenities offered. 

someone else asked this question through OSM questions.

his “rest area” is a picnic shelter. There are no real amenities offered there 
beyond the shade and a bench. I am not talking about a bicycle parking spot 
with a drinking fountain. This is why I think “bicycle_rest_area” should be 
avoided - as service area implies there are actual “services" to the cyclist. 

This is a purpose-built facility along the route, catering to the travellers 
and their vehicles along this route, similar to a British/Japanese Parking 
Area/Service area.

Saitama Prefecture (near Tokyo) operates two of these “Cycle Centers”, with 
images viewable here. 

Additional searches lead to other prefectures operating 1 or 2, so I assume 
there are about 20 in Japan, when you filter out all the cycle shops using 
“cycle center” in the name. 
I think we should have some way of tagging a cycle center that isn’t a bike 
shop or a Park, even though it is rare. 

Tagging mailing list

Re: [Tagging] Produce tags for orchards

2017-08-02 Thread John Willis


> On Jul 29, 2017, at 11:22 PM, Tobias Zwick  wrote:
> I find to tag the crop / produce of farmland is rather ineffective, as
> the actual crop being cultivated can and will change over the years or
> even multiple times a year. (See "crop rotation")

Most of the fields I tag with crop= haven’t changed much in a couple hundred 
years. The “orchards” of mulberry trees have disappeared as silk production 
dwindled, but the rice paddies that remain have been around for a very long 
time and grow exclusively rice for as long as the town has existed - the 
temples built in the 1500s and 1600s mean the town existed then, and they were 
completely dependent on rice. 

AFAIK, Rice paddies in Asia are usually not rotated nor used for anything 
besides growing rice. 

Even the corn fields here near my house in Japan produce feed corn and nothing 
else. They may be fallow for a year, but when they grow anything, it is corn. 

The greenhouses near my house grow nothing but strawberries. Only strawberries. 

The greenhouses up the street grow eggplants every year. For the ten years I 
have lived here, that land grows eggplants and nothing else.  

The farms I visit in Nagano grow Wasabi. They are purpose built to grow wasabi. 
It is extremely difficult to grow. Nothing else could ever grow in the prepared 
gravel beds for them. 


There will always be examples of why a tag is bad - but that means you don’t 
use it. 

***but that doesn’t mean it is not useful to someone in another region or 
situation you are unfamiliar with*** 

It may be that we need to modify the tag to suit the needs of everyone, but 
killing a useful tag or removing a use case because your region or experience 
says it is not useful is not a great idea - unless you propose a new tag that 
the old tag would conflict with. 

Tagging mailing list

Re: [Tagging] Solar shop

2017-08-01 Thread John Willis

> On Aug 2, 2017, at 2:45 AM, Dave Swarthout  wrote:
> That would leave us with:
> sells:photovoltaic_devices=yes/no
> sells:wind_turbines=yes/no

Seems like a good fit. I assume storage battery systems (like the Tesla 
powerpacks) will become something sold as well, so having a flexible frame to 
hold all those types and combinations is nice. I think most of these shops will 
have 2-4 sells:*, so it isn’t too big a deal. 

This also allows for stores that are not predominantly power shops to be tagged 
that they do sell them if the mapper thinks it is worth mapping.  It is rare 
for the big-box “electronics” shops here to sell them, but smaller “electric” 
shops (the old mom franchised outlets from a single manufacturer) offer 
weird mixes of stuff, including full solar installations. 
Tagging mailing list

Re: [Tagging] Solar shop

2017-08-01 Thread John Willis

> On Aug 2, 2017, at 12:06 AM, SwiftFast  wrote:
> shop=power_equipment
> power_equpiment="(photovoltaic, wind, battery, etc)"
> (NOT power=

Yea, you guys are right. My mistake. 


Tagging mailing list

Re: [Tagging] Solar shop

2017-08-01 Thread John Willis

> On Aug 1, 2017, at 1:37 AM, SwiftFast  wrote:
> shop=renewable_energy_equipment

This is the right idea, but I would actually go more broadly. There is gonna be 
different methods and combinations going forward. 

I would go with shop=power_generators and power=*, so any type of power 
generation can be specified (photovoltaic, wind, battery, etc), and can be made 
more specific as needed without making a new shop=* value. 

If you want to stick with a single value just for solar panels, 
shop=solar_generators or similar is good. 

Solar is a common value for the power keys. We shouldn’t deviate from that 
because somewhere it means diesel fuel. It’s like the paraffin wax/fuel 
language problem, and OSM already uses “solar” to define power from the sun. 

Tagging mailing list

Re: [Tagging] Recommendation for building tags

2017-07-27 Thread John Willis

> On Jul 27, 2017, at 3:52 PM, Vozniuk Taras  wrote:
> So non-permanent is a building structure constructed for a defined short 
> period of time. Like Circus tent for example. It stays there for a defined 1 
> month. 

Unless it is for some international event (like the olympics, World Cup, etc) I 
would not map buildings that would be there for a month, especially something 
as local as a circus. 

If you want to map such local and temporary structures, please set a personal 
reminder in your calendar app on your phone to go remove it. Otherwise it may 
stay there for years. 

There are some “temporary” structures that are permanent: a Bus is used as an 
office for a nearby driving school because no permanent structures are allowed 
to be built in the area. That bus has been parked in the same spot for over 10 
years, (moved only during typhoons), so I consider it permanent and mappable - 
a much larger erection (like a circus tent) only there for a month is much more 
“temporary” than that bus. 

Tagging mailing list

Re: [Tagging] Beautified JSON presets for natural=tree

2017-07-22 Thread John Willis

> On Jul 22, 2017, at 9:50 PM, Adam Snape  wrote:
> Removing the name key from the JOSM preset wouldn't stop somebody adding a 
> name tag in the few cases where a tree really was named. Nor would it remove 
> name tags from existing trees. 

Gotcha. I assume the vast majority of mapped trees are unnamed. Javbw. 
Tagging mailing list

Re: [Tagging] Beautified JSON presets for natural=tree

2017-07-22 Thread John Willis


> On Jul 22, 2017, at 7:23 PM, Pander  wrote:
> 5. remove key `name` as it is rare for trees

As a person who has tagged a named tree, please don’t remove it. 

A tree was named and planted near my house 300 years ago, and it is a local 
tourist spot.  There is no park, no other kind of infrastructure. Just a tree. 

Tagging mailing list

Re: [Tagging] Formally informal sidewalks

2017-07-15 Thread John Willis

> On Jul 15, 2017, at 7:04 PM, Svavar Kjarrval  wrote:
> Just to be clear: Is it valid, in your opinion, to connect the end of a
> footway along a street, directly to the street itself?

If the street becomes the route, I say yes, especially if there is no 
reasonable barrier to prevent pedestrians from continuing. 

In some cases, there are barriers to stop to stop any kind or traffic. In other 
cases, one side ends and the other continues, meaning peds would reasonably 
cross there to access the other sidewalk (providing there is no reasonable 
zebra crossing nearby, a situation the occurs in my suburban-rural area 
frequently), which would the. Require an unmarked crossing. 

 But If you are walking on the sidewalk and it abruptly ends (especially if it 
makes some affordance to transition you onto the shoulder, like a taper or a 
break in the curb to allow cyclists free passage) - then by all means! The road 
is now the way you are traveling on, not the sidewalk - which up until 3 meters 
before, could have been a curb, hedge, and fence separated way! 

Tagging mailing list

Re: [Tagging] Formally informal sidewalks

2017-07-15 Thread John Willis

> On Jul 15, 2017, at 4:18 PM, Nick Bolten  wrote:
> sort (not unlike a *_link for roads)

This was my reasoning for highway= footway_link earlier, perhaps 
highway=footway_routing might be a more accurate tag. 


Tagging mailing list

Re: [Tagging] Formally informal sidewalks

2017-07-15 Thread John Willis

> On Jul 15, 2017, at 3:13 PM, Marc Gemis  wrote:
> My neighbour's driveway is longer than
> mine (it's a company) and now OsmAnd insists on taking his, because it
> comes closer to my house.

Admittedly, I don’t map a lot of residential driveways (because most 
residential in Suburban Japan has no sidewalks), AND I do not use OSM for 
routing, but wouldn’t a footway linking your building to your driveway solve 

When I map retail buildings and I know where the doors are, I map the entrance 
and add the mappable footway to the entrance node from the parking driveway. 

Would doing the same for your location solve a routing problem? 
Tagging mailing list

Re: [Tagging] Formally informal sidewalks

2017-07-14 Thread John Willis

> On Jul 14, 2017, at 11:32 PM, Nick Bolten  wrote:
> > --> need to add all driveways?
> This is generally a good idea - and to make sure they share a node.

To me, if you are considering adding sidewalks, you’ve already committed to 
adding the service roads/tracks/etc. 

Adding the hidden crossing (crossing=unmarked) IMHO is the thing to be 

An example of this issue is where a road with no sidewalks meets another road 
with sidewalks, but does not cross it (and is not in an urban environ, so there 
is no real paint to show a crossing=zebra) . Do you add a crossing=unmarked 
that goes from the sidewalk to the node of the road’s T junction? People on the 
sidewalk far side of the T junction will expect to be able to cross the street 
there and continue on the road. 

In my region, even on major national roads sidewalks abruptly stop, let alone 
on tertiary roads. Usually a road is being brought up to a modern standard 
section by section, but the surrounding roads are not. A building project 
forced the adjacent roads to be upgraded, but the beginning and end of those 
roads are still the older narrow versions, such as this tertiary road here:

Because where the sidewalk abruptly ends is dangerous for peds, I put in an 
unmarked crossing to the other side, and linked the sidewalk to the road where 
it ends to the east. 

I am committed to mapping all sidewalks as separate ways, because they often 
have routing completely separate from roads in Japan, and the nature of them 
appearing and disappearing be mapped with a separate way is the best way to 
show this - but how peds “rejoin” the road when it does end is what is not 
documented in OSM.

Javbw ___
Tagging mailing list

Re: [Tagging] Formally informal sidewalks

2017-07-13 Thread John Willis

> On Jul 13, 2017, at 10:28 PM, Svavar Kjarrval  wrote:
> when the "common sense" approach would be to "just go
> across the street".

This is a question I have too, and I’m wondering if this is something you solve 
at the tagging or engine level. 

Afaik, this is why the roads have the sidewalk=left/right/both tag - so the 
sidewalk is considered “part of the road” and routing engines do not have to 
think about what side of the road you are on. 

Places with complicated and separate footpaths need Highway=footway, but this 
is the downside to that. 

I wonder if adding foot=yes tag to the roads without sidewalks and and foot=no 
to ways where it is dangerous would help. 

In places complicated enough to warrant separate footpaths, then assuming they 
*cannot* cross the street wherever they want (and forced to go to crosswalks or 
signals) is by far the best choice. But where this complicated sidewalk tagging 
ends, and the minor, residential, and service roads without sidewalks begin 
interests me greatly. Is there a “footway_link” ? Not a traditional _link road, 
but a logical link to when sidewalks end - do they need some kind of “link” to 
the adjacent road so Routing continues on? 

Currently, in some situations I link them to the adjacent road segment or 
across an intersection with crossing+crossing=unmarked, but I am unsure if this 
is necessary or proper. 


Tagging mailing list

Re: [Tagging] key trees

2017-07-10 Thread John Willis
I might be wrong, but there seemed to be a big push a while back to make 
“landuse=farmland” take over any land where stuff is grown and have crop=* 
handle what is physically there and the produces=* tag handle what is 
harvested. A lot of the grapes grown in my area are on horizontal steel 
trellis’ in plastic greenhouses (landuse=greenhouse_horticulture) - grown for 
sale in bunches (they are really expensive), and don’t look anything like a 
California nor Japanese “vineyard” with grapes in big long rows outdoors on 
vertical supports - nor a basic dirt farm field. 

Rice fields are flooded, are named and special here in Japan (tanbo: 田んぼ), with 
the extremely common kanji “ta” 田 representing a rice field - I suppose this is 
true elsewhere in Asia.  Perhaps some cultures would have a special word or 
crop that is treated differently than all others That is “just another crop” 
elsewhere, like rice is here. Making everything that grows plants for food or 
processed plant products (cotton) has its merits. Being able to tag everything 
under the same landuse idea (farmland) and then add additional tags to say what 
is on the land, what grows on it, and what it makes: greenhouse=yes, trellis=* 
netting=* crop=* produces=* is nice. Most commercial apple and pear orchards 
here have retractable netting that goes over an entire orchard, it would be 
nice to tag it somehow. 

But this runs counter to tagging things in a way that makes it fit into English 
words for new tagger friendliness (orchard / vineyard /  etc) and the idea of 
“duck tagging (tag an orchard as an orchard), so I think both systems are in 
use at the same time and adaptable enough for any situation - I wouldn’t call 
the few greenhouses trapped between a trunk road and a train road a “vineyard” 
- and the large swaths of California or North of Mt Fuji are most certainly 
“vineyards”, regardless of what the grapes are used for. 

With increased uses of tagging presets in iD, this might be come a moot point. 


> On Jul 11, 2017, at 8:43 AM, Warin <> wrote:
> Hi,
> According to the OSMwiki the key tree;
> 1) is to be used for plants ... that would include shrubs and grasses.
> 2) is to be used to define what is growing in an orchard.
> 3) has been 'approved'?
> I would suggest it would be better
> If this is only for use with the key landuse=orchard then a sub key such as 
> orchard=berry_tree, * would be better?
> ___
> Tagging mailing list

Tagging mailing list

Re: [Tagging] definition of the key "office"

2017-07-09 Thread John Willis

> On Jul 9, 2017, at 8:30 PM, John Willis <> wrote:
> civic

It also matches building=civic and the (unapproved) landuse=civic_admin


Tagging mailing list

Re: [Tagging] definition of the key "office"

2017-07-09 Thread John Willis

> On Jul 9, 2017, at 7:29 PM, Warin <> wrote:
> The word civic ... conveys what

Something for the public good; a civil service. 


Tagging mailing list

Re: [Tagging] definition of the key "office"

2017-07-08 Thread John Willis

> On Jul 8, 2017, at 11:11 PM, Martin Koppenhoefer  
> wrote:
> Why is it shop=hairdresser or travel_agency, but office=estate_agent?


Shop: a retail space selling foods, goods, or services to customers, and is not 
a restaurant. 

Office: an commercial space where professionals at desks perform their job in 
private or for customers. Alternatively, it is the “home desk” for a 
professional who does their work predominantly “in the field”. 


I think it is the line between “blue collar” and “white collar” jobs. Service 
jobs where you wear are similarly not “white collar” jobs. Blue collar jobs are 
in a retail or industrial space of some kind (car repair or car factory) or are 
in the service or retail industry (people wearing uniforms at McDonald’s or an 
electronic shop). Many travel agencies look like retail shops, I I think that 
is a good example of the edge of this explanation. 

People who predominantly sit at a desk in a commercial space and move paper, 
information, or money are in an office, like a lawyer or an architect. These 
are commercial spaces. 

This is why banks refer to their public branches as “retail banking” - they are 
“shops” full of customer service agents talking to walk-in customers dealing 
with their checking account. Posters everywhere asking if you want a car loan. 
Their large corporate offices full of brokers, traders, IT people, loan 
officers and managers are commercial offices that handle the management of all 
of the paperwork and money generated by their retail division. 

An office is also a “home base” for a professional who does their work in the 
field. An estate agent meets all of these: 

-commercial setting
-moves paper and money
- office is a “home base” for their field activities. 

People who need a shop full of advertising to prospective clients (travel 
agency)  or specialty equipment to perform their duties in that location (hair 
dresser) is a shop. 

An estate agent spends their time at other locations. Their office is their 
“home desk.” hair dressers perform a service on a person in a building with 
specialty equipment (nail salons, dry cleaners, Etc) and a travel agency is 
full of millions of brochures and posters, waiting for prospective clients to 
call or come in. 

My friend is an electrical engineer. He is an industrial inspector and damage 
investigator. When a factory or a supermarket has an electrical issue (which 
happens due to age, animals, and lightning), he is under contract with the 
business to come out immediately 24/7 and inspect the damage and tell them how 
to repair the system. He also performs government required yearly tests of the 
breakers and earth-leakage and inspects the private transformers (the pole is 
the power company’s, but the transformer on the ground belongs to the factory) 
and write all the reports and submit the government forms. He has a tiny little 
office with a desk and a computer for writing his reports, and spends his free 
time watching sumo wrestling on TV there. Maybe he has a meeting with his 
client there to sign a contract (very rare) - but his job takes place 
elsewhere.  His little office is his “home base” for his professional job that 
predominantly takes place “in the field”.

My other friend has a auto glass repair shop. It is a small garage full of 
boxes of windshields and tubes of glue. People bring their cars there and he 
removes and replaces damaged glass. He wears a 1-piece mechanic’s jumpsuit.

Traditionally, any person who does field work or inspection involving reports 
and paperwork has an “office” where they take care of that - estate agents, 
soil & other field engineers, building inspectors, and the myriad of 
consultants out there. They are treated the same as lawyers and attorneys and 
other paper-pushers in having “an office”. 

Tagging mailing list

Re: [Tagging] definition of the key "office"

2017-07-08 Thread John Willis

> On Jul 8, 2017, at 7:01 AM, Warin <> wrote:
> How then are 'offices' that do 'sell a service' to the public distinguished? 
> The 'access' key?

That is the tough question. Apple HQ has visitors, but 99% of their offices 
(and they have a hundred or so office *buildings* in close proximity to their 2 
campuses) are totally 100% access=private - but that might be interpreted as 
access to the building is private, not that the purpose of the office is 

> Or are these now to be removed from the key 'office'?

I think it is reasonable to keep private offices in office=*  

Perhaps office:access=customers / visitors / private / destination might be a 
way to differentiate between the different kinds of business that are in 
existence, with the default being office:access=customers . 

Perhaps a simple office:public=no switch would work, or perhaps I am 
overthinking it. 

> There are government offices of both descriptions.
> Can the 'service' key to be used to describe the type of service offered 
> (public or private)?

The easy way is to go office=government and let the name=* take care of the 
local assumptions. An American has an assumptions of services at a city hall or 
US federal building, and a Japanese person has similar assumptions of what can 
be done at the City office or the (giant) Prefectural Office. The same would be 
true for the myriad of other offices as well . Just naming them and identifying 
them as “government” would allow different renderings and icons, so people 
could find them in a sea of strip malls and convenience stores. 

If we wanted more granularity, I would go office=government (or civic) and then 
a colon separated subkey with many values. These can be private or public 
facing - because the government has a ton of buildings - especially in bloated 
government countries (like Japan). 

Civic: [insert OSM regional system]tax=yes

Buildings or offices handle many different things, it would be good to be able 
to define what they are without depending on a name field, and in this way, can 
also be added to non-office buildings for exceptions (some shops let you pay 
your DMV registration in the US, in Japan you go the city office for domestic 
Passport services, while you go to a post office or Fedral building in the US - 
so the mix and assumptions of splits of services must vary wildly around The 


Tagging mailing list

Re: [Tagging] definition of the key "office"

2017-07-07 Thread John Willis

> On Jul 7, 2017, at 8:45 PM, Martin Koppenhoefer  
> wrote:
> +1, let's get rid of "like a shop", and of "selling services" as only 
> requirement.

I think we are in agreement in almost all respects. 

I know healthcare proposals are huge and have been adopted (?) - I assume there 
is some schema for handling clinics and treatment offices. I assume admin-only 
buildings are for insurance - private or public.

We should have office=government + government=* or something similar - I think 
a lot of people have “office” linked to “government FooBar office” pretty hard. 

If we keep pounding on this, I’m sure we can think it through. 

Tagging mailing list

Re: [Tagging] definition of the key "office"

2017-07-07 Thread John Willis

> On Jul 7, 2017, at 9:10 AM, Martin Koppenhoefer  
> wrote:
> charity office, a government office, the office of an ngo, the office of an 
> association, etc., they don't sell a service,

I think in he strictest of terms, they do - the embassy offers passport and 
visa services, an NGO offers aid coordination, an association handles logistics 
for members - but a better definition might be office where services are 
*performed* and offices where a business or service is administered or 

To me personally, people confuse:

- the office as a physical structure 
- the office as a “white collar” shop: a lawyer or clerk or consultant who 
offered his services.
- the administration section of a larger company: Apple’s HQ in Silicon Valley 
is where they manage all aspects of their business if manufacturing their 
devices in China. 
- a (government) service that is “public facing” but not represented by the 
narrow tagging system: we have court house and DMV, but not “federal building” 
or pension_office or other civic_admin offices. 

My landuse Civic_Admin is (surprisingly) in de facto use because people want to 
separate government services from commercial ones (a public pension office from 
a private lawyer’s office), but we need to separate out “customer facing 
service offices” from “private managerial offices” - Offices of a newspaper, a 
car manufacturer, etc - as well. 

I know the government part is easy where to draw the line, but the line 
separating “service” from “management” is difficult to see, but it might be 
necessary to make some arbitrary decisions to aid in making a more 
understandable and narrow office=* key to more accurately tag things. 


Tagging mailing list

  1   2   3   4   5   6   7   >