Wikidata:Property proposal/forest cover

forest cover edit

Originally proposed at Wikidata:Property proposal/Place

Not done

Description	percentage of land area covered by forests or the forest canopy or open woodland. Use qualifiers "determination method" and "point in time" to state how and for which year it's computed.
Represents	forest cover (Q49001692)
Data type	Quantity
Allowed values	>=0,<=100
Allowed units	percent (Q11229) (same as P2927)
Example	Maine (Q724) → 89% Nevada (Q1227) → 0.5%
Planned use	import w:Forest cover by state in the United States
See also	water as percent of area (P2927)

Motivation

@Amadalvarez, Jey, Thayts, Thierry Caro, B.Zsolt:@Triggerhippie4, Cavaliere grande: this is similar to water as percent of area (P2927) you already used.
--- Jura 12:10, 6 May 2018 (UTC)[reply]

Discussion

@Jura1: Conceptually, it's similar. To me, the difference is that water as percent of area (P2927) talks about a geographical characteristic (more or less stable), but this new property could by related with a set of country developement indicators as the World Bank (Q7164) handle in its opendata series, like this one: http://wdi.worldbank.org/table/3.4 Amadalvarez (talk) 12:59, 6 May 2018 (UTC)[reply]
- Let's say it's less clear-cut. I don't think it says much about Nevada's development, but the more sources are available, the better.
  --- Jura 13:05, 6 May 2018 (UTC)[reply]
Support David (talk) 14:45, 6 May 2018 (UTC)[reply]
Oppose Better to have a generic "Land use" property, qualified with a percentage or area (i.e. Land use=forest; percentage =89%; Land use=industry, percentage=5%, etc.). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:40, 6 May 2018 (UTC)[reply]
Oppose per Andy --Pasleim (talk) 21:29, 6 May 2018 (UTC)[reply]
Oppose I agree with Andy and Pasleim. However, Will "Land use" accepts physical/geographical criteria like water, desert or forest.. and economical/development values as industry, agriculture, residential, etc. ?. Or do we have two differents properties for this two kind of classification ?. In any case, the future of water as percent of area (P2927) is affected by the new "land use" property in order to avoid overlapping concepts. Thanks Amadalvarez (talk) 06:07, 8 May 2018 (UTC)[reply]
- @Amadalvarez: do we have a sample where that approach actually works? I know you edit in the field, so I suppose your suggestion isn't just a theoretical one. Wikidata is somewhat limited by fact that qualifiers on qualifiers aren't possible and properties with multiple values are hard to aggregate.
  --- Jura 06:18, 8 May 2018 (UTC)[reply]
  - @Jura1: I don't work too much. I used water as percent of area (P2927) just in infoboxes and the manual parameters the infoboxes had since the beginning of the WPs were exactly this and no one else. So, I did not figure out any other value at that moment. However, cause of this proposal, I wonder if "% of forest would be the last proposal" or it will open a new world of classification, then I searched and found the above reference of worldbank. I don't bet to have a close list of values and, I understand, the Andy's proposal is an open list. However, any classification should have excluding values (among them) to avoid ambiguities or overlap, and must add 100% or less. Probably, someone needs to describe "the territori" (water, desert, forest, tundra, etc.) and it should not be mixed with an "economical point of view" ( industry, agriculture, residential, etc.), because it will most likely overlap with the other criteria because they may not be excluding concepts. I appologise to have a common sense (but theoretical) approach. Now, I found a good document showing that "land use classification" is not an easy simple matter. Sorry.Amadalvarez (talk) 17:35, 8 May 2018 (UTC)[reply]
    - @Amadalvarez: Thanks for your feedback. I thought I'd better ask you as you already added quantity properties and would have some practical experience with it. It does seem complicated to add more data about land, beyond simple ones like water or forest. As there doesn't seem to be any demand for the others and international organizations don't combine them either, maybe we should try to focus on the task at hand. If ever needed, a parent property could be added to the existing P2927 and this one. How should we go about it?
      --- Jura 05:54, 9 May 2018 (UTC)[reply]
      - @jura1: I like the Andy's proposal and my first comment just added some complexity seeing there are not a clear (and few) classifications method. So, in my opinion, we can approve this proposal "as is" hopping nobody else will ask for any other single and specific value, or else, suggest to the proponent change this proposal in the way Andy said (Land use). In fact, it means change to a "more open and versatile version" that covers perfectly the original needs manifested in the proposal. If David, Andy Mabbett, Pasleim and You agree with this second option, we can try to re-focus and vote again. Thanks, Amadalvarez (talk) 16:40, 9 May 2018 (UTC)[reply]
        You could easily create several properties if ever needed. Properties can be subproperties of another one. The question is if there is any demand or practical use for the alternate proposal. I understand it might appear tempting, but Wikidata isn't exactly a table based database. Maybe there are working samples we could compare with.
        --- Jura 08:39, 13 May 2018 (UTC)[reply]
Comment Very good idea, but something like french "classes d'habitats" (see [1]) that is to say habitat (Q52105) would be better, talking about ecology, and perhaps "land use" as bove for human activity. --El Caro (talk) 11:16, 13 May 2018 (UTC)[reply]
Not done given the lack of consensus. A more generic property could be proposed though. − Pintoch (talk) 11:26, 16 June 2018 (UTC)[reply]