I have some questions about heritage=* and associated tags.
First, I’ve edited the wiki page for heritage https://wiki.openstreetmap.org/wiki/Key:heritage to add United Kingdom, Wales to the table of Heritage Registration Organizations towards the end of the page. I’ve used the “abbreviation” cadw (the name of the organization, pronounced “kadu”) because I did see some discussion suggesting that abbreviations were a bad idea and because others in the table use full names.
I would have added Cadw to the list of organizations near the top of the page, but none of the UK organizations appear there although they appear in the table. Also the list and table are almost, but not quite, alphabetical and I’m wondering if it would be a good idea to alphabetize them properly. So, was the UK left out for a reason and should things be alphabetized?
The thing that really puzzles me is the use of “criteria” for listed building/scheduled monument/shipwreck/conservation area/whatever, which I’d describe as types, with criteria being the reasons why the object was listed. It makes me wonder if I’m interpreting something wrong, but if that’s what’s been used then so be it. However, the way to tag the criteria looks the wrong way around: cadw:criteria=* rather than criteria:cadw=* (or even something like heritage:criteria:cadw=*). So does cadw:criteria conform to current best practise?
It gets even worse, because it seems I could, and should, use cadw:inscription_date for the date it was registered. Not only is inscription_date bad English language for registration_date (which is what it appears to mean) but we already have start_date for artworks which could possibly be borrowed, in the same way that opening_hours apply to more than just shops these days.
And then there appears to be no way of indicating category/grade (e.g., I, II*, II, or A, B, C) other than a partial overlap between criteria and exception.
Any time I see such a big mismatch between what I read and what I expect, I suspect I am looking at it in completely the wrong way. Or is it that mappers have invented criteria and exception on the fly?
Any guidance would be appreciated.