Hello Mateusz, yes why not ? It s a good idea ; i didnt think about it. I ll try to find some time later today to set it as proposed or you may change it yourself, fine with me.. it s good that other people looks at the data and standardise things.. thanks for your review and proposal !!

"Survey" here means that i personally know this site and i saw what the building is. Yes it's a shop but also and firstly the location where the cider is made. This is why "factory".
Cheers.

247100046 : msg sent to author "schandlers" for his input.

241361362 : ok, for me it's quite clearly visible on BingSat that it's not a "Heather only" area... However, I just sent a mail to the author/Creator for his input.

ok. I got it.
anyway "water" is more generic than "spring". hence cannot be an error.

1 more comment - be sure that I'm going to take into account your point related to the fact of contacting the previous mapper !!

starting to look at the 1st ref. you mentioned...
The very 1st text of place=locality at "place=locality" is "the place=locality tag can be used to name unpopulated place...".
So as you mentioned "natural=locality" makes no sense when "place=locality" has even if less precise than "natural=fell".
i would like also say that as far as I know "place=locality" and "natural=fell" are not incompatible even if in such case "place=locality" becomes unneeded....
so to "weigh" the change - it was from "non sense" to "unprecise" - certainly not enough but better.

ok. I think I understand that we have a different point of view. my mindset is that "heath" is "heath" whatever surface is covered and using 1 value is better than 2 especially when the difference is only the covered surface.
For me this is of the same level than having for exemple natural=small_wood and natural=big_wood for another surface; Both are only wood...
but ok I agree that my view could be not shared - probably some sort of "professional deformation" on my side - I'm in IT ;-).

sure.

agreed - changes reverted back.

ok. I read that when involved in a relation, the attributes should be mainly supported by the relation itself. here the relation itself is a closed object : fine. the specific part we are speaking about is only a part of it and supports polygon "attributes"

because of the vocabulary: "source" is the french word for "spring".

sure. sorry for the inconvenience. a mistake.