In my endeavors to get open data on public toilets in existence for the benefit of, erm, everyone, I regularly consider the potential of OpenStreetMap.
For anyone who doesn’t know, (and for those that do, feel free to skip my ill-informed explanation..) OpenStreetMap is a map of the country that’s created by members of the public. This began because Ordinance Survey data (along with other maps like those used by Google and SatNavs) was private and expensive and couldn’t be enhanced (or in some cases corrected) by you or me.
They created it by walking around with their personal GPS systems, logging data and adding it to the online map. It struck me as remarkably geeky, not to mention slow progress. Yet when I learnt about it in 2007 due to my MA Industrial Design Engineering project (self-promotion), a skeleton map of Britain was emerging.
Now they seem to be using satellite images provided by Yahoo as well, meaning people no longer have to have a GPS and traipse the streets. Instead you can just go online, log-in, and trace, colour and label the satellite images, thus creating a map. Consequently a huge leap forward has been made in the last few years – it’s looking pretty complete and I’m starting to see it used on websites instead of Google Maps. I’m sure you have to, perhaps without realising it.
… and Public Toilets
One of the things that can be added to the map is a public toilet. The map above, centered on London Bridge, has a few examples (you can click on it then zoom in via the next window. I couldn’t get the map to embed directly). As I’m not remotely proficient in OpenStreetMap it’s hard for me to analyse the quality of this data. I’ve no doubt these toilets exist and are accurately located as that’s the entire ethos of the project.
But I suspect in most instances, like here, there’s no additional information (male/female/unisex/baby-changing/payment/opening-hours) even though the interface does allow for this. I assume this is because OpenStreetMappers are not as ‘into’ public toilets as you or me. Their loss :)
Public toilets are added as a ‘node’, and via the OpenStreetMap wiki I downloaded the dataset as an xml. I haven’t got a clue what to do with this, but the significant thing was it has about 20000 nodes (therefore 20000 toilets?). Not sure if this is worldwide. Or partial-worldwide. Hmm… Still, it’s promising.
(ooh – comments 10 & 11 on ‘The Great British Public Toilet Map‘ answers this, and links to a map showing OpenStreetMap public toilets for London! thanks to Gregory Marler)
A better way for me to evaluate the data is to use Elbatrop’s public toilet app Find Toilets, as this was built using OpenStreetMap data. It allows the public to add additional information about public toilets (and, I think, additional toilets) and shares this information back with OpenStreetMap.
As anyone can add public toilet data, and there’s the option to include extra information, why not just add the public toilets of Britain directly to OpenStreetMap instead of creating the Great British Public Toilet Map?
I’m very open to debate on this as I keep reevaluating my arguments for and against. But so far:
Reasons for TGBPTM (hmm, catchy..)
- Public Participation
Anyone can add data to OpenStreetMap but it’s not the most user-friendly of websites. Don’t get me wrong, it’s improved dramatically and by logging on, reading the instructions and clicking some buttons I could draw lines on the London Bridge area map within the hour (then hastily click ‘Do Not Save These Changes’). But I couldn’t email my mum and say ‘add the village toilets to OSM! <link>’. She’d open it. Then close it. End Of.
The other thing I like about TGBPTM is that it will explain the issue of both public toilet information and open data in a way that the public can understand.
The third thing would be the intended ability to ‘take action’ by emailing your council directly. Asking the public to email (or letter-write) is several pegs down the Scary-Scale than asking them to edit an online wiki-map.
- Accurate & complete data
This accuracy point is not a criticism of OSM at all. However some toilet apps allow people to add public toilets unmoderated, which may mean they’re not public toilets; they’re toilets in a business that may or may not be for customer use only. Or they may be in someone’s house. Or entered incorrectly and someone’s forgotten to delete it. Etc.
The second point, complete data, does come back to OSM, which lacks the motivation (currently) to add (and maintain) information on opening hours and facilities. This is information that I still believe is best coming from the council since they’re in control of such things. Naturally councils will differ Wildly in their ability to do this.
- Community Toilet Schemes
This is fast becoming my justification for everything. Community Toilet Schemes are where councils pay local businesses a small yearly grant (I’ve seen between £300-£1000 offered. Of £0 if they’re a megacorp. Or the Town Hall) in exchange for the business allowing anyone, customer or not, to use their toilet during opening hours. They’re growing in popularity because they’re cheap (Generally the council announces the scheme whilst simultaneously closing a few permanent, expensive and ill-used public loos). It’s advertised on the council website, by a sticker in the window of the business (which is sometimes the cause of contention) and occasionally by printed maps, or even permanent fingerposts in the case of Richmond-upon-Thames.
These ways of informing the public of the Community Toilet Scheme toilets I find unsatisfactory for various reasons which I’m determined not to go into now, however the provision could be vastly improved by including these loos in the open data. The CTS data is likely to change much more frequently than permanent buildings when businesses join and leave the schemes, so needs the responsiveness of open data (and council commitment to maintaining it, though it’s in their best interests to do so).
I don’t know how responsive OSM can be to any toilet data down to the level of CTS participants or opening hours when it’s just the OSM community, open though it is, adding it (though compared to the responsiveness of councils?! What am I talking about?!). Apps like the Elbatrop one which have a simple public interface could be a way around it.
And OSM nodes don’t extend to CTS, though I’m sure they’d consider it.
Reasons for OSM
- It already exists.
- And works.
- It doesn’t require me finding funding
- It’s global, oooh!
- All the data is in the same format…
The Great British Public Toilet Map, which currently consists of a folder with 8 datasets in it, has found or received data in 3 different formats, each with different columns and information and data types. It will require some processing to turn each dataset into something useful and plottable on the map. The website plans to include suggested guidelines for toilet data for the idealistic council-types, but doesn’t expect everyone to comply (taking the ‘anything is better than nothing’ approach).
Once the data is on the map it can be exported as cleaned-up data for use in maps/apps. But there’s probably a level of long-term maintenance in transforming the data.
Adding data directly to OSM avoids this.
- I’m still in love with the public participation element of my idea. I don’t believe that the public can engage directly with OSM or could add toilets to OSM.
- There’s still a need, I believe, to engage councils to publish open data. And the best way to do this is by engaging their residents.
- Could data from TGBPTM be added to OSM, Elbatrop-style? Or from the council datasets directly? Is there a point? or are we adding too many middlemen?
I began today with a tweet:
GaillyK: Dear World, I’ve a question, re: #opendata Public Toilet Map bit.ly/eYG4gC. Should I ditch it + push toilet info on OpenStreetMap?
4 hours and 1000 words later I’m no nearer answering it, though I suspect it’s not an Either/Or situation. (I think some of those who read the original tweet may not have read it as Either/Or either, which has saved them A Lot of time).
However OpenStreetMap people have been discussing it, commenting and being generally helpful and enthusiastic (bits of this post are already out of date!)
So clearly this is just the beginning…
(UPDATE: Some of these issues have been addressed in this blog post by an OpenStreetMapper (and recent creator of a London Toilet Map using the existing data), in particular, ideas for how the public can get involved adding public toilets without tackling the OpenStreetMap site)