In short, it has data on more than a hundred hotels whose bathrooms do not offer an expected amount of privacy, using either no door at all or glass doors. People speculate that this may be to make friends book two rooms while they might otherwise share.
Whatever the reason may be, it’s a verifiable property of the hotel rooms that you would want to know before booking. It seems like a good fit for OSM to me, iff this is actually a common thing and not just a fad in one or two countries!
Someone on Hacker News pointed out to me that the listings on this website are not accurate (not that the data has a license usable for OSM anyway):
I looked at a few and found zero evidence of missing bathroom doors in reviews or photos. One even had a review complaining the bathroom door was broken and not closing fully… indicating it is actually there.
So my question to you:
Have you stayed in hotels that had this issue? Which country?
Interesting that few days ago I saw this post on Mastodon about that same thing, hotels with glass doors in bathrooms. It looks like it’s a rising trend everywhere, no idea why they do it. Personally occurred only once, and it was the only thing we really disliked in that hotel. Not only due to just privacy issues, but because there are people who literally can’t go to toilet if there are other people nearby.
I don’t know how much of verifiability would this “feature” have, but for sure it would be something useful to have mapped, as exactly this is something that violates a person’s privacy, even if the people who stay together in that room know each other pretty well.
Getting a bit ahead of myself, but for mapping, I shared this in a Signal OSM chat last night:
[…] this might be a useful general toilet property like toilet:privacy=* with values:
none (no barrier whatsoever)
glass (that potentially soundproofs)
frosted (contours visible through the glass or plastic)
gapped (doors/stalls that aren’t full-height, like in some public places)
full (not sure about this word, but as an example)
But then hotels often have shared toilets near conference rooms, and private toilets as part of a room. We would somehow have to differentiate, like rooms:toilets:privacy=*?? That’s too complicated for my taste, but as a starting point for some ideas :)
There could also be some level just below “full” for if it has a window. Not sure how to handle those yet. Ideas welcome
But you can only verify this one hotel room at a time, and can’t necessarily be sure it’s the same in other rooms.
Not everything verifiable is a good fit for OSM. (Fuel prices are a canonical example - they change too frequently.) And while I agree people might want to know this about a hotel, it’s not clear to me that OSM is the best place to store this information.
Looking at the long list of hotel and room properties that you can filter for on booking sites such as booking.com and airbnb, I don’t think we should attempt to duplicate this in OSM.
right, but this is not what was proposed, the op specifically asked about tagging glass doors in hotel rooms. As long as it is verifiable it can be tagged, and if many mappers do it, we can see it was wanted, and if only a few instances are tagged, it does not create a problem.
Therefore, a tagging schema should take into account that the tag describing the privacy level of a bathroom door can be different among the rooms or types of rooms of the hotel. On the other hand, the tagging schema should not be so detailed that it is too complicated to use. This means a tagging schema should simplify the floor plan of a hotel and summarize the privacy level like "This hotel has rooms with full-level and without privacy.
Looking at the huge pile of different options I would recommend to refrain from tagging such details. Besides “no doors” or “glass doors” for bathrooms there are other variations already and surely more to come.
What I have seen myself is a hotel room where the shower with all around glass walls and door was located in the bedroom and the toilet set in a small compartment with door. Other rooms of the same hotel had normal bathrooms with regular doors. Another hotel had shower and bathtub incorporated in the bedroom without any separation and the toilet set in a small niche, but without door again.
Of course one may say such installations are physical present so we can map them, but I don’t see any way to map such features with a reasonable tagging scheme. Too many options, too complicated, too many tagging mistakes to be expected.
The video that inspired my toot about this indicates that you generally can’t filter for it on the booking sites. I guess unlike F-Droid they don’t go out of their way to point out things users might perceive as anti-features as that might mean they lose out on comission. If OSM did start recording this it might be one of the few databases that makes it searchable.
Ironically I think the thing that would most likely prevent me from tagging this if I encountered it is privacy as it would probably also generate a suspiciously timed one star review on places like Google Maps.
I kind of suspect that the only thing we could reasonably assert is “has rooms with limited privacy”, as in you’ve found at least one room that meets that criterium. I wouldn’t be against introducing a simple tag that models that.
I agree, but that does not indicate yet if this affects only few rooms or all and most users wouldn’t know that is means a glass bathroom door in your room without reading the wiki.
But as far as I can see, once that has been tagged, it cannot be falsified except by visiting each room in a hotel, right? So if a hotel was to rebuild their rooms back to expected levels of privacy (or claim that they did), it would be difficult to prove or disprove.
So what the tag on its own would mean in practice is “there was at least one room with limited privacy when the tag was added”?
And I guess we could add a :check_date=* to indicate when a limited-privacy room was last seen?
It’s not like rooms are generally very different within one hotel; they’d need different replacement parts. If both are present, map both: *=translucent;none
Huh that is a peculiar choice of layout! I wouldn’t propose to map it to such detail indeed, that gets unwieldy.
If you can poop in peace but need to show off showering, the functional result is the same. The tag name should reflect that, so what I had in mind about using the toilet:* namespace is probably not a good idea. Maybe bathroom instead, referring to the whole place (particularly in British English, as the tagging guide recommends). It’s not like public toilets have (frosted) glass separators anyway
I like what @InsertUser proposed since it’s simpler than tagging specific materials. Changing suite:toilet to bathroom also resolves any confusion about it referring to the (conference) guest-accessible toilets, so we get:
I wasn’t sure what “partial” meant, though, so wanted to ask: why not tag it as “none” when it’s “partially none”. After revising that sentence, it occurred to me that it’s probably the layout for which I suggested “gapped” above. My own wording is equally unspecific, so probably not a good idea . If this is that what was meant, e.g. bathroom:privacy=non_full_height would be clearer, but probably too long? Maybe we simply need to use gapped or partial, with clear pictures on the wiki.
Edit: changed check_date to a prefix, as pointed out by InsertUser below
If it varies from room to room we could indicate that with semicolon separated values.[1]
It would be nice to use something that doesn’t require the wiki, but the alternatives I can think of are even more oblique. Suggestions welcome.
A check_date would probably be a good idea on something with this high an entry price. the wiki currently lists it as a prefix rather than a suffix though.
to be clear, that’s one per type present not one per room ↩︎
. If this is that what was meant, e.g.