OpenStreetMap

[micahcochran]

OpenStreetMap (OSM) is Google Maps like mapping service that provides open data. OpenStreetMap has a database of spatial data with tags. Currently, it tags amenity=toilets and unisex=yes for unisex toilets. Unisex is OSM's terminology for tagging.

I am working on a proposal for OSM for tagging single-stall toilets and unisex toilets within businesses. If you have some OSM experience and have helpful feedback on the matter, please subscribe to the tagging mailing list and send an email.

Even without my proposed improvements, I think OSM information could be leveraged to located a few more refuge restrooms. OSM has an active community of contributors with 41,000 users contributing to the map worldwide in a 30 day period.

OSM data could be used to get a better lat/long point for a business.

My familiarity with how to develop applications for OSM is non-existent.

If you want to get a sense of the current number of unisex toilets tagged in OpenStreetMap, use overpass turbo tool. The point locations themselves should be very close to the actual locations.

Paste the below code in overpass turbo's query code box, click the Run button. It will show in a given area where the unisex toilets are. Around San Francisco and Oakland there were 34 that popped up (Around Huntsville, Alabama, there are currently none.):

[out:json][timeout:25];
// gather results
(
  // query part for: “amenity=toilets and unisex=yes”
  node["amenity"="toilets"]["unisex"="yes"]({{bbox}});
  way["amenity"="toilets"]["unisex"="yes"]({{bbox}});
  relation["amenity"="toilets"]["unisex"="yes"]({{bbox}});
);
// print results
out body;
>;
out skel qt;

In my opinion, logistics for using OSM data should be discussed. There could be possibilities of using Refuge Restroom to improve OpenStreetMap's data. This post was to introduce that this database is out there of potential restrooms.

Thank you!

[tkwidmer]

we've talked about some of this: https://github.com/RefugeRestrooms/refugerestrooms/issues/169 https://github.com/RefugeRestrooms/refugerestrooms/issues/80

One of the issues is all of our lat/longs are geocoded with google maps, so we would have to re-geocode our addresses using OSM to start with.

In the past we looked into holesale importing our data into OSM, but OSM isn't cool with that. We have some data that is dirty / not verified. Also out lat/longs are Google Maps which is against their TOU. They also were not very friendly when we talked to them about the possibility in the past.

But its been a while ... so who knows?

[tkwidmer]

http://wiki.openstreetmap.org/wiki/Legal_FAQ#2a._Can_I_trace_data_from_Google_Maps.2FNokia_Maps.2F....3F Relevent legal information from OSM

[glassresistor]

@micahcochran I think its great your doing this on OSM i was the person pushing to do that but ran into issues that @tkwidmer mentioned.

Even though importing RRs stuff into OSM is likely not plausible it would be great to import them into RR. Also its possible OSM can take some of the new data from RR if it doesn't have the Google lat/long issues.

I'm researching at ETL from yelps bathroom data as well.

[micahcochran]

@tkwidmer Thank you for sharing RR's history with OSM and the relavant legal info regarding OSM.

One of the issues is all of our lat/longs are geocoded with google maps, so we would have to re-geocode our addresses using OSM to start with.

I'm not so confident of the OSM's ability to nationally geocode addresses. It could be worth testing because some of the addresses will definitely be in there.

In the past we looked into holesale importing our data into OSM, but OSM isn't cool with that. We have some data that is dirty / not verified. Also out lat/longs are Google Maps which is against their TOU. They also were not very friendly when we talked to them about the possibility in the past.

It'd definately take some work to try to import RR data into OSM.

I can understand OSM's reluctance to allow breaking their TOU. Still, there might be a way to work around that. What that is I'm not quite sure of.

Just thinking of the logistics of it, I would say that it might be preferable to use your current data to inform editing of OSM data.

Users are pretty good about moving a pushpin around to get the location that is close enough to the actual location. That's what Google has been allowing users to do for the last few years. It is exactly why Google has pretty good geocoding. Still, it is a painful to have to recreate the wheel.

Even though importing RRs stuff into OSM is likely not plausible it would be great to import them into RR. Also its possible OSM can take some of the new data from RR if it doesn't have the Google lat/long issues.

Good points. Creating data on OSM can take familiarity of a general area or survey of the area.

Another option is for RR to create a policy that would allow/encourage RR data to be used for the creation of OSM data.

I understand exercising caution because you are providing a level information of restroom "experience" that is just not really in line with OSM itself.

Once my OSM proposal gets done, I'd be willing to add RR designated restrooms in the Huntsville, Alabama area into OSM.

They also were not very friendly when we talked to them about the possibility in the past.

But its been a while ... so who knows?

OSM is mainly a group of mainly volunteers just like you donating their time to a project that they are passionate about. They are trying to look out for their project's best interests. In the linked discussion, I didn't see anything that seemed particularly unfriendly. However, I've seen in the past that the fear of legal problems can make a normally cool and collected person downright unfriendly.

My OSM proposal will probably needs 2 months before it will have run its coarse. Hopefully, it gets approved.

[micahcochran]

I have written my proposal for OSM: Proposed_features/Toilet_Tagging_Improvements

It came out a little more passionate than I thought it would.

[micahcochran]

I reported an issue OSM's iD editor for amenity=toilets being a radio box that wouldn't allow you to report multiple "genders". Got some great guidance about a way to address this. https://github.com/openstreetmap/iD/issues/3954

iD editor is the top 1 or 2 editors of OSM and is the one that requires no extra software to be installed (only a Javascript enabled web browser).

There was an earlier issue that made the point that RR only exists because OSM isn't easy enough to use. Great point. I think that is still true. https://github.com/openstreetmap/iD/issues/2422

[DeeDeeG]

Hi all,

I've been in touch with a legal advisor at OpenStreetMap (legal-questions[at]osmfoundation.org) to try to clear up some questions/roadblocks toward using OpenStreetMap instead of Google Maps. Seems promising if e.g. we wanted to start displaying OSM with our dataset. (First reply came in yesterday.) Here's my correspondence so far:

Hello,

I am a volunteer developer helping Refuge Restrooms, a public, free-to-use, crowd-sourced database of restrooms.

https://www.refugerestrooms.org

https://github.com/RefugeRestrooms/refugerestrooms

We are wondering if our existing database of restrooms, having been geocoded (address -> coordinates) with the Google Maps API is compatible with OpenStreetMap.

Specifically, we'd like to 1) Use our existing restrooms database to show restrooms on an OpenStreetMap map, 2) Start using OpenStreetMap-based services as the source of our future geocoding. (address -> coordinates, and coordinates -> address), 3) In any other applicable way, replace use of the Google Maps API with an open (non-proprietary) alternative.

Our specific questions:

How much of this can we do without changing the existing dataset? Can we switch right away without legal issues?

Also, would we be able to transform or discard part of our existing database and then use it legally? For example, could we use only the street addresses from our current database, geocode them with a new, OpenStreetMap-based geocoder, and then use only the OpenStreetMap coordinates going forward?

If you require any further details, I and the rest of the team would be very interested in following up with you to make sure this is okay.

We are a small volunteer team, and this project is intended to remain strictly "free as in beer" to use, "free as in freedom" where possible, and publicly accessible to all. (Full disclosure: we are attempting to get some crowd-funding going, and some funds have been awarded for development, but these are social-benefit-oriented charitable grants, and small crowd-funding donations, not fees for service.)

We hope you will advise us as to how best to use your open service, as opposed to a proprietary and mostly commercial solution, because we feel using an open service is in the best interest of our users.

Thank you for reading.

Sincerely,

[DeeDeeG]

---

[resending as I was on the road at the time and are not sure if the mail actually went out] Hi [DeeDeeG]

Thank you for your interest in OpenStreetMap.

There are numerous tricky questions that you are asking, but fundamentally you first need to determine the legal status of your data with google, not with us. This may be clear from googles terms of use, or you may have to ask google directly if any and what rights they claim in your data.

Whatever the answer, I don't see an issue with simply displaying your existing data on an OSM derived map, that is is in general unproblematic from an OSM pov. As to going further have you crosschecked how much overlap there is between your data and what is already in OSM?

I'm currently on the road and will be back in roughly a weeks time and will be able to have some more time to look at the issues.

Kind regards Simon