Property talk:P2677

Documentation

• Start a query
• Properties using this qualifier (statistics)
• Count

relative position within image
position of a motif within a larger image, defined by x-offset (from left), y-offset (down from top), width w, and height h of the crop region (as a percentage of the whole image) values

Data type	String
Allowed values	pct:(?:(?:100|[1-9]?\d(?:\.\d+)?),){3}(?:100|[1-9]?\d(?:\.\d+)?)
Usage notes	use as a qualifier on P180 statements
Example	Portrait of a Woman with a Squirrel (Q17335769) → pct:82.5,53.7,17.6,16.3
Source	Primarily likely to be user-supplied using a specific tool; but in some cases might possibly be available from external references (note: this information should be moved to a property statement; use property source website for the property (P1896))
Formatter URL	https://wd-image-positions.toolforge.org/iiif_region/$1
Robot and gadget jobs	I would imagine a tool similar to the crop-tool on Commons, challenging the user whether can they find this item as a detail in this image, and to move a rectangle around the detail if they can.
See also	depicts (P180), region within image (P8276)
Lists	<search Commons for files with depicts-statement and this property as qualifier> Most recently created items Items with novalue claims Items with unknown value claims Usage history (total) Usage history (main statement) Chart by item creation date Database reports/Complex constraint violations/P2677 Database reports/Constraint violations/P2677
Proposal discussion	Proposal discussion
Current uses	Total 39,533 Qualifier 39,532 >99.9% of uses Reference 1 <0.1% of uses
Search for values

[create Create a translatable help page (preferably in English) for this property to be included here]

Scope is as qualifier (Q54828449): the property must be used by specified way only (Help)

List of violations of this constraint: Database reports/Constraint violations/P2677#Scope, hourly updated report, SPARQL

Format “pct:(?:(?:100|[1-9]?\d(?:\.\d+)?),){3}(?:100|[1-9]?\d(?:\.\d+)?)”: value must be formatted using this pattern (PCRE syntax). (Help)

List of violations of this constraint: Database reports/Constraint violations/P2677#Format, hourly updated report, SPARQL

2-d position on something that isn't a 2-d visual artwork
This qualifier should only be used with depicts to indicate the location on a 2-d object or image that has a natural co-ordinate system. (Help)
Violations query: SELECT ?item WHERE { ?item p:P180 ?s . ?s pq:P2677 ?v FILTER NOT EXISTS { ?item wdt:P31/wdt:P279* wd:Q110304307 } FILTER NOT EXISTS { ?s pq:P518 ?part } FILTER(STRSTARTS(STR(?v), "pct:")) SERVICE wikibase:label { bd:serviceParam wikibase:language "en" . } } GROUP BY ?item
List of this constraint violations: Database reports/Complex constraint violations/P2677#2-d position on something that isn't a 2-d visual artwork

How to

The value format is based on IIIF standard, pct corresponds to percent (so it it's possible to change the image file with an image at the same ratio but not the same dimensions in pixels) and w,x,y,z numbers delimit a rectangle. Example for Idleness (Q19953492), with depicts (P180) qualifier house cat (Q146) value "pct:65,81,35,15" → https://tools.wmflabs.org/zoomviewer/proxy.php?iiif=Godward_Idleness_1900.jpg/pct:65,81,35,15/,240/0/default.jpg .
A specific tool on Crotos IIIF Cropper can be used to get the adapted values. In the same tool, it's possible to display the result ; for example : depicted items in Ancient Rome (Q14619165). --Shonagon (talk) 01:18, 22 November 2016 (UTC)

@Shonagon: I don't understand, it doesn't work, right? Tubezlob (🙋) 16:16, 6 October 2018 (UTC)

Hello Tubezlob. Zoomviewer is currently broken : Phabricator T206260. Best regards --Shonagon (talk) 09:07, 7 October 2018 (UTC)

Second property needed

I just encountered this use of the property referring to a specific image file. As pointed out in the proposal discussion if we want to provide this functionality we need another property for that. --Marsupium (talk) 15:48, 19 September 2017 (UTC)

User script

The user script: User:Husky/ifff-viewer-link.js turns values of this property into working web links, in the form: https://tools.wmflabs.org/zoomviewer/proxy.php?iiif=Godward_Idleness_1900.jpg/pct:65,81,35,15/full/0/default.jpg Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:14, 29 June 2018 (UTC)

Wikidata Image Positions tool

I wrote a new tool, Wikidata Image Positions, which shows the image areas indicated by this qualifier. You can view the image of a single item, or search for all items with a certain relative position within image (P2677) string, and I’ve set up that second mode as the formatter URL (P1630) for this property now: since relative position within image (P2677) values are usually unique, clicking the value will usually show you the image for the item that you came from. Let me know if you have any feedback, and of course, feel free to revert the formatter URL (P1630) change if you disagree with it :) --Lucas Werkmeister (talk) 17:45, 21 October 2018 (UTC)

List with examples - Salgo60 (talk) 15:13, 23 October 2018 (UTC)

Use on 3-d art

I'm finding cases where editors are using this qualifier on 3-d artwork such as statues. Is this intended? How is it defined? Examples: Q30502944, Charles XII statue (Q15120907). Bovlb (talk) 18:04, 5 February 2019 (UTC)

more examples Bovlb (talk) 19:33, 5 February 2019 (UTC)

I have removed almost all cases. I hesitated in the case of bas reliefs. If the bounding rectangle and orientation are clear, then a co-ordinate frame is arguably well-defined (even if not consistent with that of the reference image), unless it was clear from the values that the editor was working from the reference image instead. Bovlb (talk) 19:08, 1 July 2020 (UTC)

@Bovlb Thanks for raising this issue and trying to help the community to be careful and consistent in the use of relative positions statements. I do think we need some way to contribute annotations about specific portions of sculptural items and other 3D media, at least far as I understand what you are saying, I don't think there's a problem in using the image positions tool to make statements about a 3D object. Of course those statements are going to be relative to the P18 reference image for the sculptural object, to me that seems ok since the "instance of" statements make it clear that the 3D object in question is not the same as the digital image that is appended to it as a statement with P18, and the bounding boxes show up on that appended image to specify which part of the reference image refers to which iconographic element. For instance, in the example of this sculpture depicting Heracles, the imagery is complex and the sculpture is broken in a way that it would be unclear, say, that the figure holds an important attribute (the tip of the broken club is visible on the left)--without an annotation, this feature would not be clear to someone who doesn't know this object well (of this I'm sure--my students struggled with it!). Being able to search sculpture by attribute is important for both art historians and lay audiences. Art historians of antiquity are often interested in things like patterns of how different regions depict a specific god, so it is important to have metadata statements that make a sculpture searchable by the attributes a given sculptural example includes. For instance, I might want to query something like "show me all the Heracles statues from this date to that date where Heracles is depicted with a club" in order to look for patterns in where that particular iconography shows up. I confess I'm not certain what you mean by "if the bounding rectangle and orientation are clear". Why would they not be clear when the tool is used? Happy to continue the conversation, I just need more specifics about your concerns. Ahc84 (talk) 21:44, 11 February 2022 (UTC)

Image area

Could this be widen to image area, more complex identication of the part of the image? --Juandev (talk) 17:21, 7 September 2019 (UTC)

Description slightly misleading

Description says "... width w, and height h ...". This suggests that "w" and "h" are used as parameter names somewhere, and that the syntax might be "...,w=15.0,h=23.4" or something like that. (Note that the "Structured data" UI does not show the example value.) But the correct syntax uses no parameter names, associating arguments solely by position. I looked for a way to fix the description, but can not find any. What am I missing? --RainerBlome (talk) 15:03, 10 May 2022 (UTC)