I would like to mark Tag:leisure=swing as deprecated. I have written a note on the Wiki page that it is used 6 times vs 62000 times usage of playground=swing and that the later page explicitly states that it is also meant for standalone swings outside of playgrounds.
Basically my question is, if the standard formal deprecation process is necessary, or if there is a shorter path for such clear (let’s see below if that is even true) cases.
Do we need to follow the official deprecation proposal route as stated in Wiki: Deprecation process, or is there some kind of shortcut?
I imagine that we will find way too many cases like this one, which would kind of spam the list of deprecation proposals.
Should I instead abstain from the quest of requesting a deprecation and simply note on the Wiki page and Data item, that a more common tagging exist?
Feel free to point me to such meta-discussions, I have not been successful searching this forum for previous discussions of this broader issue.
With 6 uses worldwide, it seems basically unused to me… I’d contact the person who created that wiki page, as well as persons who used that tag, to check if they’re fine with deprecating it.
I agree that usage numbers are very clear, but there is an additional issue: the tag structure is abnormal compared to most tags. Usually, playground would describe a kind of playground, not a part of a playground.
Mammi71
(One feature, Six mappers and still More ways to map it)
5
With this objection, you are almost 16 years and over 300,000 uses too late. The proposal was unanimously accepted in 2010, including your approval.
The key playground=* describes devices and installations on a playground. It should be tagged to separate objects within the area of a playground leisure=playground .
I think I approved it back then because of the “theme” values, and I guess in 2010 it wasn’t yet completely clear which direction tagging will take, so using playground=* for something different than specifying a kind of playground maybe was less of an outlier as it appears now. (tag used since 2010). Anyway, 16 years are a long time and I now have a different opinion on this one.
There are also very few other tags with the same issue, namely golf (2.7M uses, wiki since 12/2012, used since 2008) and cemetery (182k uses, wiki sind 2/2015, used since 2008).
Hundreds of thousands of uses do not save tags from deprecation, if there is general agreement, in recent years it was done with site_type retagged to archaeological_site (168k uses), which didn’t even clash with anything.
On the other hand, every deprecation is disruptive, particularly for well used tags, so depending on how much weight you give each of these arguments, the issue may or may not warrant the hazzle of retagging.
For example if the wiki page exists I can quickly add random instances of the tag and have iD automatically start recommending it. Not to mention that if the page hadn’t been created the whole discussion would have never happened.
It is completely ridiculous that we are even having this discussion about 6 instances of minor, even debatable, mistagging, everbody should have something better to do.
Thank you all for taking the time of discussing this issue of a tag with 6 instances. The issue for me is actually the existence of the Wiki page and not the unusual tagging. And even more it’s about learning what to do in such cases, rather then the concrete question of leisure=swing.
Deletion vs. warning on the page
I tend to side with @SimonPoole . It would be a bettter state to not have a Wiki page and a Data Item for almost unused non-standard taggings.
This would be an interesting aspect for me as well. I will maybe have a look into the code if iD to understand this issue, and what would be solutions for such an undesired automatism.
I will go forward and suggest to the Wiki administrators to delete the pages.
If playground=swing for standalone swings isn’t problem, let’s declare leisure=swing deprecated, although I think that playground=value for something that isn’t part of a playground is slightly unintuitive.
Yeah, but it is not a tagging mistake? It is explicitly documented tag (for whatever [likely misleaded, granted] reason). Tagging mistake (as I understand it) would be someone unintentionally tagging playground=swingg or some other typo. The very act of creating the wiki page for ATYL tagging seems to indicate is not a random mistake.
That sound reasonable to me. P12 wikidata concept is different from P7 wikidata equivalent, and explicitly states “this key or tag represents a concept described by the given Wikidata item more precisely than any other key or tag”, which seems to hold water for me? (or is there any OSM tag more closely related to wikidata’s swing?)
Makes sense to not consider it a mistake, especially since playground=swing may be meant for children only.
Thanks for pointing out that P12 does not give authority to the Wikidata item over the meaning of an OSM tag. This is a crucial distinction. We currently seem to have no way of expressing that a Wikidata item is exactly what we mean by a tag, as P7 states “For keys or tags, use P12 instead.”
From what you are saying I now understand that the Key:playground is just not as detailed as playground=swing in explaining that it is an item for kids.
This means that my hole idea of getting rid of leisure=swing was probably a bit stupid, as there is no other correct way of tagging swings which are for adults, such as this one node/13251555342.
