Wikidata:Property proposal/angle

angle from vertical / slopeEdit

Originally proposed at Wikidata:Property proposal/Generic

Descriptionangle from vertical: angle the principle axis of an item makes with the vertical
Representsangle (Q11352)
Data typeQuantity
Allowed unitsdegree (Q28390), minute of arc (Q209426), arcsecond (Q829073)
ExampleLeaning Tower of Pisa (Q39054) -> 3.97±0.01 degree

Originally proposed at Wikidata:Property proposal/Generic

   Done: slope (P4184) (Talk and documentation)
Descriptionslope: angle the surface of an item makes with the horizontal
Representsslope (Q275447)
Data typeQuantity
Template parameter"slope" in en:template:infobox pyramid
Allowed unitsdegree (Q28390), minute of arc (Q209426), arcsecond (Q829073)
ExamplePyramid of Unas (Q1478801) -> 56.3 degree

I was surprised to discover that we don't have a decent property to represent the angle of something. We currently have the odd situation that the angle of Leaning Tower of Pisa (Q39054) is represented by angular distance (P2212), which is an astronomical concept for the size of something on the sky! In Pyramid of Unas (Q1478801) the best I could find to represent the angle of the slope of the pyramid was average gradient (P2198), but that's not right (it's supposed to be a value between 0 and 1 to start with!). Ideally we could do with something that displays like half of a coordinate (DMS notation), but just allowing degrees/arcmins/arcsecs should be good enough for now. Mike Peel (talk) 11:20, 11 August 2017 (UTC)

Note I removed the example Pyramid of Unas (Q1478801) -> 56.3 degree as this doesn't match the current label. I also updated the description text. Not sure this is quite right yet though. ArthurPSmith (talk) 19:56, 17 August 2017 (UTC)
  •   Support as something like this is needed, though I wonder if we need to be more specific - an angle is between two directions; the natural axis of the item would presumably be one of them, but is the reference direction horizontal or vertical? ArthurPSmith (talk) 18:03, 11 August 2017 (UTC)
    @ArthurPSmith: Good point. Unfortunately, I think the normal reference direction for the tower of Pisa is vertical, while for pyramids it's horizontal... Maybe this need a qualifier (as is currently the case in Leaning Tower of Pisa (Q39054)), or we just pick one and enforce that? Thanks. Mike Peel (talk) 01:05, 12 August 2017 (UTC)
Primarily, angle (Q11352) is a angle (Q11352) with unit 'DMS' or otherwise. Therefor, as a property it should not be limited into a specialised (limited) definition (this will run into trouble for the rest of its property life).
Best solution would be to add (require) a qualifier by QID (Pisa, pyramid). A tilted pyramid should have both. ArthurPSmith (talkcontribslogs), is that possible (in WD design)? -DePiep (talk) 09:40, 12 August 2017 (UTC)
@DePiep, ArthurPSmith: I don't know if requiring qualifiers is possible. Converting between horizontal and vertical is trivial (one is 90 minus the other), so I'm not sure both values need to be recorded. I'm going to suggest we go with 'angle from vertical' for now, but let's keep the discussion going. Thanks. Mike Peel (talk) 22:44, 13 August 2017 (UTC)
It's not trivial in a property definition. You cannot leave it to the (automated) reader to do extra calculations to get it right (and left wrong if not done). This would be an example of Wikidata not taking responsibility of providing incomplete data BTW. See also my opposition below. -DePiep (talk) 19:20, 16 August 2017 (UTC)
  •   Support in principle. Mahir256 (talk) 02:50, 13 August 2017 (UTC)
  •   Comment I would prefer 3 angles from six degrees of freedom (Q1071507) d1g (talk) 18:47, 13 August 2017 (UTC)
    • @D1gggg: Those three angles look like angles of motion, rather than static angles, so I think that is a different issue - maybe we should also have roll, yaw and pitch as properties as well, although perhaps those would need to be 'maximum' values. How does the change to "angle from vertical" look to you, or would you prefer something else? Thanks. Mike Peel (talk) 22:44, 13 August 2017 (UTC)
      • we could name them "delta in yaw" and so on.
      • Motion is when we add time on top of this. We can do this if we really want, e.g. 2 degree rotation per 2 years :-) d1g (talk) 01:48, 14 August 2017 (UTC) @Mike Peel: d1g (talk) 01:50, 14 August 2017 (UTC)
  •   Comment natural objects should use different property "with respect to earth" or local surface. d1g (talk) 19:07, 13 August 2017 (UTC)
  •   Oppose. An angle in itself cannot be a property. It always needs a specification: angle between what? For Leaning Tower of Pisa (Q39054) it would be "Angle with the vertical", for Pyramid of Unas (Q1478801) it is "angle with the horizon (slope)" etc. So either we need two properties (for these), or this one must require a specifier.
IOW: angle (Q11352) is a quantity (think measured), but it is useless without specification. Similar: a property "speed" would not be helpful (and so: dangerously wrong) without specifiers like "max", "at take-off", etc. -DePiep (talk) 19:20, 16 August 2017 (UTC)
    • OK, it looks like we need two properties here then, so I've added the second one above. These can be split to different pages if needed, but it might be simpler to keep the conversation about them here given how closely they are linked. I'm not sure that requiring a qualifier would work. Thanks. Mike Peel (talk) 20:13, 17 August 2017 (UTC)
      • for the pyramids, the axis is pretty much vertical, it's the surface that has an interesting slope. I modified the second proposal description, does this make sense?
        • Thanks, the change looks good to me. Thanks. Mike Peel (talk) 23:26, 18 August 2017 (UTC)
  •   support slope natural definition and used in documents; another angle is something rare d1g (talk) 18:41, 28 August 2017 (UTC)
  • @Mike Peel, ArthurPSmith, Ita140188, D1gggg, DePiep, Mahir256:   Done. Please make good use of it.
    --- Jura 18:50, 28 August 2017 (UTC)