Looking at the existing tagging, I think I will take a ride down the slippery slope you mention. For example: an "aed" is an emergency medical device people are mapping. Learning the locations of these could be important in an emergency: checking the mapping makes sense.
So how about: lastcheck=date (the only required tag) lastcheck:source=* (a description of your mapping party or expedition) lastcheck:status=(operating/needs_maintenance/restricted/closed) lastcheck:note=* (notes on the above e.g. 'sprays water on feet') operator=* (the operator of the facility, who is responsible for attending to the facility) This could apply to any node or way, but it would be silly on most. I've seen a few lastcheck= on meadows, but would never add one myself. I would however check drinking water, aed's, toilets, and maybe opening hours. Thus we'd have: - lastcheck/lastcheck:status=* *for facilities that are intended to remain in operation, even if they are inoperable when observed.* - disused=yes *for facilities that have ceased operation, but still exist on the ground* - disused:oldtag=oldvalue *for facilities that have ceased to exist on the ground (e.g. building is a vacant lot).* On Sat, Jun 29, 2013 at 6:43 AM, Frederik Ramm <[email protected]> wrote: > I have no strong opinion on the general idea but one thing sticks out: > "operational_status=operating"**. > > Obviously, this should be the default assumption. If there's a toilet > mapped then I assume that it is working. > > I understand how you'd like mappers to reaffirm that it is *indeed* > working by setting "operational_status=operating"**, but this takes you > way into the terrain of "verification mapping" which is a larger topic - > what you're suggesting really only makes sense if one can assume that a > mapper passing by a working toilet would make sure it is tagged > operational_status=operating and then update the operational_status:date to > the current date. (If you do not expect this behaviour then you wouldn't > need operational_status:date because it would simply be the date when the > tag has been last changed.) > > This leads to a situation where a mapper is expected to, as he or she > walks the streets, update every object in the database with "yep, this is > still there, I walked past it right now". Because just as a toilet could > fall into disrepair, a shop could close or a house vanish, and what we > currently do is we map this when we see it but we don't map "yep the house > was still there last Sunday". Attempting to do this would change the > typical mapper workflow and the structure of our data drastically. > > I know it's a slippery slope argument, and you're only proposing to do > this for a narrow subset of things - I just wanted to point out that > "verification mapping" is not something we do currently. > > Bye > Frederik
_______________________________________________ Tagging mailing list [email protected] http://lists.openstreetmap.org/listinfo/tagging
