Wikipedia:Descriptive image tagging
This is a failed proposal. Consensus for its implementation was not established within a reasonable period of time. If you want to revive discussion, please use the talk page or initiate a thread at the village pump. |
- This proposed policy was initiated by User:Carnildo.
- See also Wikipedia is not censored for the protection of minors, which is official policy and which this proposal seems to contravene.
This is a proposed system for descriptive tagging of image contents. It is meant for allowing people to filter out images they might not want to see while browsing Wikipedia. For example, someone viewing recent changes during their lunch break may want to filter out all disease symptom images to preserve their appetite, or someone may wish to restrict all non-lineart images of sex acts. This system does not attempt to make any judgements of "good" or "bad" images; rather, it is intended to be a strictly objective description of certain types of image contents.
This proposal does not cover any of the technical details of descriptive tagging, but I envision that tagging will be done with either checkboxes accessible from an image's page, or templates added to the image's page, while filtering will be done by a set of checkboxes in the user's preferences.
Related policies and discussions are at
- Wikipedia:Image use policy General image use policy
- Wikipedia:Profanity#Offensive images
- Wikipedia:Image censorship Discussion on how Wikipedia should be censored, if it does get censored
- Wikipedia:Graphic and potentially disturbing images Discussion on dealing with "offensive" images; the discussion on tagging seems to have bogged down from lack of a specific tagging system, and from problems with systems that need to make subjective judgements.
- m:Offensive content Discussion of the mechanics of tagging images; the discussion of specific systems seems to have run into trouble with systems that require subjective judgements.
Suggested tags
editThis list of tags is meant to cover categories of content that large groups of people may find objectionable. It is by no means complete. Category titles are simply a convenient way of grouping individual tags, with no specific meaning attached to them.
Human body
edit- Visible male genitals
- Visible female genitals
- Uncovered buttocks
- Uncovered breasts
- Bare male torso
- Naked male
- Naked female
Sexuality
edit- Sexual act (humans)
- Sexual act (animals)
- Sex toys
- Sex toys in use
Violence/death/medical
edit- Blood/gore
- Dead humans
- Dead animals
- Open wounds
- Disease symptoms
Religion
edit- Religious figure (Christianity)
- Religious figure (Judaism)
- Religious figure (Islam)
- Religious figure (Baha'i)
Drugs
edit- Note: The distinction between medicinal and recreational is whether the drug is approved for treating a non-drug-related medical condition.
- Drug apparatus
- Recreational drugs
- Medicinal drugs
- Cannabis/marijuana
- Tobacco
- Alcohol
Other
edit- Gambling apparatus
- Firearms
- Non-firearm weapons
Image type
edit- Photograph
- 3D computer model
- Sculpture
- Painting
- Lineart
- Colored lineart
Specific examples
edit- Image:Human-woman.png would be tagged as "naked female, lineart"
- Image:Mass Grave Bergen Belsen May 1945.jpg would be tagged as "dead humans, naked male, naked female, photo"
- Image:Sixtynine.png would be tagged as "naked female, sexual act (human), lineart"
- Image:Bahaullah.jpg would be tagged as "religious figure (Baha'i), photo"
Advantages and disadvantages
editAdvantages
edit- This system does not require judgement calls such as "is this suitable for children?" or "is this obscene?". The only decisions needed when tagging images are "does the image contain these subjects?".
- The system lets people decide for themselves what they do and do not want to see.
- The system is internal to Wikipedia, so that, in addition to the options of showing or hiding images, there's the option of turning images into links.
- The tags can be translated into other content-filtering systems.
Disadvantages
edit- The system does not distinguish based on context.
- The system does not work for users who haven't set up their computers for it.