Axel Boldt wrote:
Toby Bartels wrote:
Axel Boldt wrote:
How about if we give up our current distinction between images and text and require a single confirmation whenever people upload material to Wikipedia:
By uploading material to Wikipedia, you certify that either * the material is in the public domain, or * the copyright owner of the material releases it under GFDL, or * the material can be used under the [[fair use]] doctrine and such fair use will likely extend to all third party users of Wikipedia
(maybe even with three [radio buttons].)
But in the case of images (along the lines of album cover thumbnails), I don't think that we need to worry about future users of Wikipedia (much less try to guess who's /likely/ to be one of these and who isn't). This is because images can be easily removed by the future user.
Granted that it's easier to remove fair-use images than fair-use quotes, but even the former involves going through the articles that link to the removed images and removing the now-broken links.
We should write a utility that will strip them. That can't be hard, and providing it with our work, whether or not it's necessary to comply with the text of the FDL, certainly helps in complying with the spirit.
Not to mention the case-by-case decisions about which fair-use images to keep and which to remove. In effect, this necessary work prevents a quick-and-dirty third-party $10 Wikipedia CD from ever seeing the light of day. And there isn't much room: the Encyclopedia Britannica DVDROM sells for $50.
It's easy to produce a /quick-and-dirty/ CD this way; simply strip out all questionable (3rd option) images. We do need to save the choice made when it was uploaded, of course; but I think that everybody agrees on adding such metadata.
In the end, to get a cheap CD, we would probably have to produce a sanitized version of Wikipedia ourselves.
To get a cheap CD with some of the fair use images, yes. But to decide who the image is fair use for and who not -- that's a difficult decision to make once and for all. I'm certainly not up to judging whether something is fair use for all of the /likely/ third-party future users. How am I to judge that when we argue over an image's status for just us? If a third-party user has a particular usage in mind, then they can judge their circumstances better than we can. I agree that it's nice if we can judge it ahead of time, but I don't agree that we're capable of it.
For quotations, we should be able to ensure that the quotation is fair use for /every/ potential user, so that would be all right.
Even for quotations you need the word "likely" somewhere: your quoting a line of Martin Luther King in an article about him is certainly fair use; my selling t-shirts with that quote on it is certainly not.
Fair enough. But in this case, "likely" is easy to judge.
Thus, there is a practical difference between text and images.
True, but I'd like to maintain the spirit of GNU-free for the whole encyclopedia, not just for its text.
I'd also like to do this, but the only pracitcal way is to not include fair use images in "the whole encyclopedia" at all. Of course, this is what /you/ want (and is consistent with my practice too), but the other Wikipedians aren't likely to come around to that. This leaves us with a difference between text and images -- to with, that we /can/ maintain the spirit for the entire text, but cannot for all of the images.
-- Toby