Andreas Kolbe wrote:
Again, I think you are being too philosophical, and lack pragmatism.
We already have bad image lists like
http://en.wikipedia.org/wiki/MediaWiki:Bad_image_list
If you remain wedded to an abstract philosophical approach, such lists are not neutral. But they answer a real need.
Apart from the name (which the MediaWiki developers inexplicably refused to change), the bad image list is entirely compliant with the principle of neutrality (barring abuse by a particular project, which I haven't observed). It's used to prevent a type of vandalism, which typically involves the insertion of images among the most likely to offend/disgust large numbers of people. But if the need arises (for example, in the case of a "let's post harmless pictures of x everywhere" meme), it can be applied to *any* image (including one that practically no one regards as inherently objectionable).
I would invite you to think some more about this, and view it from a different angle. You said earlier,
A reputable publication might include textual documentation of a subject, omitting useful illustrations to avoid upsetting its readers. That's non-neutral.
You assume here that there is any kind of neutrality in Wikipedia that is not defined by reliable sources.
There isn't.
Again, you're conflating two separate concepts.
In most cases, we can objectively determine, based on information from reliable sources, that an image depicts x. This is comparable to confirming written facts via the same sources.
If a reliable source declines to include an image because it's considered "offensive," that's analogous to censoring a word (e.g. replacing "fuck" with "f**k") for the same reason.
"Include suitably licensed images illustrating their subjects" is a neutral pursuit. (Debates arise regarding images' utility — just as they do regarding text — but the goal is neutral.) "Include suitably licensed images illustrating their subjects, provided that they aren't upsetting" is *not* neutral, just as "include properly sourced information, provided that it isn't upsetting" is not.
If I go along with your statement that reliable sources avoid upsetting their readers, why would we be more "neutral" by deciding to depart from reliable sources' judgment, and consciously upsetting our readers in a way reliable sources do not?
"This is an image of x" (corroborated by information from reliable sources) is a neutral statement. "This image is upsetting" is not.
In an earlier reply, I cited ultra-Orthodox Jewish newspapers and magazines that refuse to publish photographs of women. If this were a mainstream policy, would that make it "neutral"?
As I noted previously, to emulate such a standard would be to adopt a value judgement as our own (analogous to stating as fact that "x is bad" because most reputable sources agree that it is).
Likewise, if most publications decide that it would be bad to publish illustrations alongside their coverage of a subject (on the basis that such images are likely to offend), we might address this determination via prose, but won't adopt it as *our* position.
That exact same argument could be made about text as well:
"Likewise, if most publications decide that it would be bad to publish that X is a scoundrel (on the basis that it would be likely to offend), we might address this determination via prose, but won't adopt it as *our* position."
So then we would have articles saying, "No newspaper has reported that X is a scoundrel, but he is, because –."
"X is a scoundrel" is a statement of opinion. "X is a photograph of y" (corroborated by information from reliable sources) is a statement of fact.
And as noted earlier, this is tangential to the image filter discussion.
David Levy