Talk:Proposed features/Toilet Tagging Improvements

From OpenStreetMap Wiki
Jump to navigation Jump to search

Comments by Javbw

Here is an example of the layout I would want to be able to replicate (Javbw's comment)

Mapping different kinds of toilets and and the different genders must be taken into account when mapping restrooms.

Here is a sign for a restroom in Japan, and the ability to label and map what toilets are available to men, women, and true elderly, along with types of toilets, stalls, and other amenities ( baby holders, changing tables, kids urinals/toilets, etc, should be considered from the start.

Javbw (talk) 02:06, 31 March 2017 (UTC)

I can easily see the path to add baby holders and changing tables. I think those tags would be extremely useful for parents with babies. The diaper=* key set forth a decent specification for changing tables. Baby holders could follow a similar specification. It seems like that it just needs to be formalized. diaper=* would need to have the d-word happen: discouragement.
Folding Restroom Stall Child Seat/Holder Example (Micahcochran's comment)
I think currently child urinal/toilets is too much detail. My fear that it might slow down the acceptance of tagging other things in this proposal.
What is typically referred to as "gender" of the restroom should be the designated groups for which restrooms are designed to accommodate. That's my small technical quibble.
I was thinking about family toilets that we have in the USA (I assume that Europe might have it too) that are unisex restrooms that have many of those types of amenities for babies, children, "true" elderly, and handicap people who require an assistant. Elderly restrooms sounds like a similar idea for just elderly and handicap people who require and assistant, but I assume that there might be handicap-only designated restrooms that would have similar function. Perhaps a historic building might have just added a handicap-only restroom to comply with disability laws. Elderly restrooms are unisex when it comes to gender designation.
These are a few tagging schemes that might address the restroom group designatation:
  1. Create a new toilets:gender=* tag --- that would be a messy transition. I've seen the gender=* with a couple other tags combos, but it is inaccurate for this use. I'd have to have a really good reason to even propose that. Nope.
  2. Fabricate a toilets:unisex_ref=unisex/family/disabledonly/elderly tag. This will pose some problems when there are multiple unisex restrooms. It gives you the flexibility to create your own designations, which might not be such a good thing. Yuck.
  3. Add new designated group tags that work like male/female/unisex tags. My favorite solution.
    • toilets:family=* -- for a unisex family toilet
    • toilets:disabledonly=* -- for a unisex handicap/disabled only restroom... Does that adequately describe these Japanese restrooms or should this be elderly or senior?
I can see adding the group designation ("gender") at the end of the tags, sorta like parking:lane=*.
You'd only specify it if it differs from the rest of the group designation restrooms.
Perhaps you could tag that picture as a single like so:

(Tagging them separately would be a slightly less cumbersome, but let's show the more complicated example.)

I'm not too sure exactly what baby facilities are provided, so I threw in something. toilets:stall:male=* might be "mult;open" or just "multi".
I have a feeling that this addition of group designations could be a little controversial. -- Micahcochran (talk) 21:47, 31 March 2017 (UTC)

"true" was a mistype, it was supposed to be "the". In Japan, similar to trains, they have a unisex stall colored green for handicapped (the legal definition), elderly, or people with a baby or other people who need something beyond "a normal" toilet (beyond obvious injuries or being pregnant, this is is a self-selecting group, on the honor system). This designation of people (pregnant/disabled/old/injured/assisting) is a common designation. You find special chairs for them in train, you find special parking spots (beyond handicapped), and in the same way that in the US they mandate handicapped restrooms, they mandate these "green" toilets. The *mandate* for these type of *green* toilets are legal and ubiquitous.
Also, the men's toilets in Japan *in public sports parks* *sometimes* have completely visible toilets to deter people from habitating or hiding in them. Most public toilets in anywhere (train stations, publicly else but sports parks have normal restrooms, and the restrooms found in larger public parks *are* private - the person advising that they are "always" open to view is patently wrong.
thanks for fixing my image ^_^ Javbw (talk) 00:00, 1 April 2017 (UTC)
In the US, I'm not too sure how many handicap only restrooms there are. The law does require facilities to provide for some level handicap access for restrooms. I mostly see handicap stalls within the gender designated restrooms.
What you are describing with the *green* toilets fits my idea of a family restroom/toilet. A definition of a family restroom. In the US there isn't a legal definition of the group, but the family restrooms are intended for those groups you've identified. It is available for those self-selected groups who need that type of restroom.
toilet:stall=open is really intended to describe the privacy among users within the toilet. In the diagram you have shown above, it looks like there aren't partitions between urinals, which leads me to tag that as "open". I haven't really figure out a way to tag that it doesn't have a door. As a US citizen, this would be strange and could cause a bit of culture shock. I do not intend to spread misconceptions about Japanese toilets. Please correct me if I am making/spreading any misconceptions.
I looked at taginfo a little more for tagging of child/kid toilets and could not finding any tagging of those features.
Diaper/nappy changing tables also need to include tagging to show if the table only accommodates babies or can accommodate an adult. Thank you for all the good comments. -- Micahcochran (talk) 15:21, 3 April 2017 (UTC)
thanks for your continued work. I am also an American, who lives in Japan. Most men's urinals have no privacy panel, this might be familiar to you if you visit school bathrooms or stadium bathrooms. Some urinals are either sized or placed on the wall for children. Some urineals have bracing bars for old people. These are found in normal restrooms. If you are mapping urinals, then the disabled or kids urinals should be an option.
Because you are mapping stalls, and in Asia there is a distinction between "western" and "squat" toilets, I hope the proposal has a way to map the number of each type available, and the privacy might be different.. This distinction in toilet types is not something that you would find in the US - but it is as distinct as the way we would feel between a urinal and a toilet. The different toilet icons in the (small) map I posted shows the different types.
Finally, there are small toddler toilets, usually a toddler "stall" with a very tiny sink and a small western/squat toilet. These are usually semi-private, as the adult standing nearby can look over the 3-4 foot stall wall. These are found in stand-alone restrooom (kids restrooms adjacent to a nursery room) or inside a dedicated adult restroom. These are what I think of for a "family toilet or a "kids stall". - the green toilet here can be thought of as a unisex handicapped restroom outside the gener separated toilets (which would also have a few handicapped accessible urinals or toilets as well).
If you are going to map restrooms down to the stall detail, then the toilets, stalls, and rooms need different types available to mix and match properly. Trying to map the stalls (or lack thereof) means being able to map the different types of access, amenities, and type of each fixture. The stall is an amenity of the toilet - so mapping the toilet is the center so point to me. If you are looking just to put a stall type on a toilet POI, then iit will be too hard to accurately reflect the real world. This will work for simple toilets - but there are a very large amount of public toilets with many permutatations, it may be difficult to accurately reflect what you are trying to map, especially with the the lower lack of privacy (sometimes) in other cultures, once the genders are divided.
Thank you for your patience explanations. I don't have the goal of entirely replicating something that looks something like the plan you've provided in tagging. I anticipate that level of tagging will get way too complicated. At a certain point, this could also be interior mapping. I am push the envelope a little, not to that extent.
There is a small section in the proposal about repurposing capacity, but it really don't really want to pursue it. It quickly could gets a little too complicated. The difference between 5 toilets and 50 toilets is not a huge concern. Tagging 2 urinals, 1 seated toilet, 2 squats toilets, and 1 child's toilet and then separating those by gender designation will be quite a few tags. At most 15% of toilets are even tagged male/female/unisex.

"If you are looking just to put a stall type on a toilet POI, then iit (sic) will be too hard to accurately reflect the real world [...] it may be difficult to accurately reflect what you are trying to map, especially with the the lower lack of privacy (sometimes) in other cultures, once the genders are divided."

Great points. Based on your points I think it is best to abandon the toilet:stall=* tag.
Since, I'm trying to have a way to distinguish restrooms intended for a single person or multiple people at a time. Instead of tagging the stall configuration perhaps toilets:user=single/multi.
The Australia Government created a database of their public toilets, National Public Toilet Map. Their symbols show more or less the level of detail that--with the right tagging scheme--I think OSM could replicate, but in some cases it goes a little beyond OSM.
The term "Family Restroom" as used in the US is a unisex restroom for a single occupant or a family. It may be a father and his young daughters may go in together. A caregiver (related or unrelated) and an elderly or disabled person who may be of different genders. Because of the different genders it is unisex. There should also be diaper changing facilities. The handicap design is almost always taken into account atleast as much as other restroom and sometime moreso. There is also space for a small group and for maneuvering and moving someone from a wheelchair. The family restroom has a lock, just like a single occupancy restroom would. I'm pretty doubtful there are partitions in family restrooms, but there could be if it large enough. The facilities are signed as "Family Restroom".
Transgender people are using these because they are single occupancy unisex restrooms and they have less to worry about from other occupants and sometimes the law itself.
Here's what a few descriptions that I have on Family Restrooms: (I know that I have read more about this, but most of it has been from news articles.)
I think we are in violent agreement that something needs to be done to designate this group (these groups?). What is the best terminology for classify the Japanese green toilets for a group designation? Thank you! -- Micahcochran (talk) 21:41, 5 April 2017 (UTC)

Changing Tables and Child Holders

Another Example child holder Picture

Javbw mentioned changing tables and child holders (see above). These changes are a bit far from the scope of my original proposal. Therefore, I am deferring action on this.

Changing Tables are currently served by an the in-use key diaper. I added my comments in the Key:diaper discussion page.

My reservations about proposing tagging for "baby holders", "child protection seats", or some sort of "restraint". The fact is that I do not really know the proper term to use. The seat design would probably determine the age cutoff. A simple yes/no value for the existence of this piece of equipment in a restroom could be helpful for parents.

Koala Kare calls it a Child Protection Seat. This model supports 22.7 kg, which is about the weight/mass of a 6 year old. There may be other dimensional constraints that would prevent putting a typical 5 year old in this holder. Other brands may have different design constraints that might lead to it working for different ages.

Here is some text that might be useful for an eventual proposal:

Add toilets:baby_holder=yes/no for a baby holder being present (yes) in a public toilet or absent (no). Use toilets:baby_holder=no for instances when you might expect to have one. For example, a shop=baby_goods would be a place that a customer might expect this equipment to be the business's toilet.

-- Micahcochran (talk) 16:13, 7 April 2017 (UTC)

Why single/muti vs a number ?

Multi could be 2 or 20 .. would it not be more usefull and informative to have a number?



The present toilets:wheelchair=yes/no could also be expanded to toilets:wheelchair=0/1/2/* .

Warin61 (talk) 23:00, 9 April 2017 (UTC)

"Why single/muti vs a number?" My short answer to this question is tagging simplicity.
I am working on a rewrite of the proposal for a second RFC. Based on Javbw's comments, I am replacing stall with occupancy, which I think is similar idea with a little less specificity. It removes the tagging of privacy expectations that vary across the world. So, Template:Toilets:occupacy would identify single-occupancy, multi-occupancy, and perhaps for Family Restrooms I might coin a term "family-occupancy".
The power of toilets:wheelchair=* tag is that it is very simple. There are only three values in use yes/no/unknown, which are 99% of them. There is also "limited", but that has been used 35 times out of nearly 62,000 instances of toilet:wheelchair=*. wheelchair=* itself (with 1,000,000 instance) has the same situation in which 99% of values yes/no/unknown/limited (and designated). I am trying to attain that simplicity with my proposal.
If you think toilets:wheelchair=* could use some sort of count/capacity, please submit a proposal. That is outside of my knowledge and outside of the scope of this proposal.
Yes, "multi" could refer to 100 in a stadium or airport. Most of my arguments against numbering are in the proposal. Also, I'm unlikely to go into a women's restrooms (since I am a male). It is often an okay assumption that the male and female restrooms have the same number of facilities. Note that I am not against reusing the capacity=* tag, nor would I be against someone using it instead of toilet:occupancy=*. Along with toilet:capacity=*, it would be almost necessary to have to add prefix group "gender" designation of for all toilets tags, which I'm not quite sure if I want to do. As an examples take urinals (rarely located in women's restrooms), so you'd tag toilets:position=seated for restrooms and then toilets:position:male=seated;urinal to override the value only in the men's restroom. That change is quite a bit bigger and adds some complexity. If I am going to propose an even bigger change, I need to ensure that this change does not create an even bigger problem than the one it attempts to solve.
Regarding the above-mentioned toilets:stall:male=0/1/2/*;open/closed -- I know only a little about using overpass for querying (being from the GIS world I don't think I'm completely ignorant). Requiring a semi-colon in the value seems like a recipe to make querying a nightmare. Thank you for your comments. -- Micahcochran (talk) 15:38, 10 April 2017 (UTC)

gender implication

In the current version 1460593 it it written:

If toilets:for:male=* is set it should imply toilets:for:female=no and toilets:for:unisex=no. [...]

I dislike this implication, as I usually only tag what I have surveyed, and often don't survey an object completely. When there is for example a (larger) shop or amenity/building/.., maybe with multiple levels, I might only add toilets:for:male=yes as these had been the only toilets I've seen, although there might be very well female and/or unisex toilets in another area or on another level. I would add toilets:for:*=no only when I'm certain there are none. Because of the implication I would need an additional value, e.g. toilets:for:*=unknown to avoid it being interpreted as no.

--Athalis (talk) 22:12, 20 April 2017 (UTC)

If it is a multi-level building or large shopping complex, the mapper should probably be using a separate node (or area) amenity=toilets with that tags level=* and indoor=yes. This tagging is really for a place that just has one location of public restrooms, and tagging it separately would not be easy or create clutter on the map. Let's put that aside and talk about implicit "no" versus explict "no" for gender designations for restrooms.

amenity=toilets does not discuss the default behavior.
Summarizing, this confusion comes from of a misunderstanding of access=* tags (not necessarily my misunderstanding). The wiki pages for male=*, female=*, and unisex=* state that these are access tags.
Here's the Description from male=*:

Denotes that access is restricted to persons of the male sex or gender. This is an access tag following the same rules as those described in access=*, though the presence of this tag with a yes value necessarily implies a no for all other gender-based access tags.

My understanding of access=* tags is that those work by having a default setting for based on the type of highway=*. So, highway=footway, means by default motor_vehicle=no and foot=yes.
I think what is it the Description section from male=*, is where misunderstanding of how access tags came from. There was a revision that added male/female/unisex tags to access=*, the very next revision removed them with the note "gender/sex restrictions follow a different logic than access tags, so they shouldn't be listed on this page". The same problem is on female and unisex. I guess nobody followed editing the other tag's pages (male=*/female=*/unisex=*). That was in 22 December 2010, so I guess the damage has been done.
A minor remedy would be to edit the pages (male=*/female=*/unisex=*) to remove the language that this a type of access=* tag, but that doesn't really address how these tags should behave.
My intent was to have toilets:for:*=* work the same way that male=*/female=*/unisex=* tags work. I can see that it would a good idea to define how these tags should behave in this proposal. -- Micahcochran (talk) 19:29, 25 April 2017 (UTC)
One thing that I did not consider at the time is that entrance=yes is used to show the entrance/door of a restroom along with one of these male=*/female=*/unisex=*. I also don't like the implication that setting male=yes makes female=no by default. Perhaps there are different uses of male=*/female=*/unisex=* that may have different nuanced meanings. -- Micahcochran (talk) 22:07, 10 May 2017 (UTC)

Add prefix tag for gender designations to all toilet tagging: Example #2

In the current version 1460593 under Add prefix tag for gender designations to all toilet tagging there is Example #2:

Example #2 - a separate wheelchair accessible unisex restroom from male/female restrooms:


Is there a typo and should it be toilets:wheelchair=yes ? If not this example needs more explanation.. --Athalis (talk) 22:16, 20 April 2017 (UTC)

That tagging not a typo. Here's more explanation: The unisex restroom is wheelchair accessible, but the male and female restrooms are not wheelchair accessible. -- Micahcochran (talk) 17:41, 21 April 2017 (UTC)
I think this will lead to much confusion, as toilets:wheelchair=no is currently used and interpreted as "there are no toilets available for wheelchairs" and this proposal makes a backwards-incompatible change to "not every toilet is wheelchair accessible". Additionally, virtually every place I know of with toilets for wheelchairs would need to be manually retagged from toilets:wheelchair=yes to toilets:wheelchair=no having toilets:wheelchair:unisex=yes which creates a lot of (maybe unnecessary/avoidable) work. --Athalis (talk) 21:10, 24 April 2017 (UTC)
Very good point, and that makes this tagging simpler. I will have to revise the proposal with this in mind. Example #2 should only be tagged toilets:wheelchair:unisex=yes. Still, there would be some work for people moving from the toilets:wheelchair=yes tag to toilets:wheelchair:unisex=yes. Also, I just notice this should be called a suffix, not a prefix. -- Micahcochran (talk) 19:51, 25 April 2017 (UTC)

Diversity quarterly project 2018Q2 is this sort of thing

For those not in the know, people on the diversity-talk mailing list are trying a new thing, quarterly projects. And this quarter, we're doing toilet genders Diversity Quarterly Project/2018 Q2. So this is very relevant

Interesting to hear about this, I also just noticed your thread on talk. (For the record, I agree with your proposed definition of "unisex", as that's what I had assumed the key to mean anyway.) Are you also going to add other attributes related to inclusiveness, such as single/multi occupancy, during the project? --Tordanik 20:26, 24 April 2018 (UTC)
I, personally, will not be working too much on occupancy/etc. Others are free to map anything they like. If you'd like to keep track of it, be my guest. :) ____Rorym (talk) 10:08, 25 April 2018 (UTC)