On 2/5/06, Anthony DiPierro wikilegal@inbox.org wrote:
Image tags were originally descriptive, not prescriptive, and frankly I see no reason that should be changed.
One last note on this:
The old templates said things like, "Posters are thought to be fair use on Wikipedia." and "Television show screenshots are thought to be fair use on Wikipedia."
The new ones are mostly written along the lines of "This is a form of X media. It is copyrighted. It is thought that low-resolution and limited numbers of X media are used on Wikipedia in Y way, it is fair use. Any other use of this media could be a copyright violation."
There are two benefits to this. First, it gives users a quick way to see if their invocation of fair use is likely to actually be fair use -- without having to know the first thing about copyright law.
Second, it is not necessarily descriptive at all. Descriptive and misplaced tags are the sort of inaccurate things you are talking about. They make claims about the copyright status that may or may not be true. The sorts of tags I am talking about, when written well, are much more ambiguous. They discuss it in terms of conditions -- if condition Y is met, then X is probably fair use.
When mistagged, they do not necessarily give incorrect information. They point out whether or not the current image is likely to be fair use or not. Sometimes it is not. Now the way to fix this, in my opinion, is to change the tag to say, "This image SHOULD be X type of media." Now, if it is wrong, it is wrong, but it is clear when it is wrong and nobody gets in a snit.
The benefits of this sort of approach are, I think, legion, over the other. It is a way of both encouraging correct use, allowing users without detailed copyright understanding to quickly spot inappropriate use, and encourage a mindset about fair use which is hinged on the conditions of use rather than the form of the media. Personally, I think it has worked for the mostpart. I hadn't thought about changing the labeling wording itself ("this image SHOULD BE" versus "this image IS"), but if people think that is a good idea then that is easy enough to put into place.
The only recurrent arguments I have ever seen against this approahc is, "Copyright tags shouldn't try to instruct users as to their use." I've never really heard a good reason of WHY that should be true, except for a rather vague and hazy notion about separating out functions, and about how copyright law was hard and fair use was really about the four factors. I know it is, that's why I tried to come up with a model which would result in the factors being fairly satisfied in most situations, if the tags were properly followed. Perhaps it isn't perfect, and I'm completely open to suggestions for improvements.
In any case, simplistic and poorly thought out tags lead to simplistic and poorly thought out tagging; the evidence of this is quite clear, or at least it was months ago.
FF