Marco-
In my humble opinion we should think about how we can solve the problems we already have with the quotes and not make our situation worse by allowing "fair use" images which clearly violate the GFDL.
Let's be clear on one point: On en:, we are *already* allowing fair use of images. You would like to see this status *changed*. The legal case for doing so is no stronger than the case for prohibiting fair use of quotes or sounds (of which there are plenty). In fact, I will now demonstrate that it is weaker; that we need to worry more about quotes than about images.
The FDL was developed primarily for books and other printed works. As such, it does not contain any reference to "linking" of any kind. The relevant sections to our discussion are the following:
----------------------------
5. COMBINING DOCUMENTS
You may combine the Document with other documents released under this License, under the terms defined in section 4 above for modified versions, provided that you include in the combination all of the Invariant Sections of all of the original documents, unmodified, and list them all as Invariant Sections of your combined work in its license notice, and that you preserve all their Warranty Disclaimers.
....
7. AGGREGATION WITH INDEPENDENT WORKS
A compilation of the Document or its derivatives with other separate and independent documents or works, in or on a volume of a storage or distribution medium, is called an "aggregate" if the copyright resulting from the compilation is not used to limit the legal rights of the compilation's users beyond what the individual works permit. When the Document is included in an aggregate, this License does not apply to the other works in the aggregate which are not themselves derivative works of the Document.
----------------------------
What is now, in the context of Wikipedia, the work that is licensed under the FDL? It is not what the server generates, it is not what the web browser renders, for neither of these entities have the legal rights to claim a copyright. It is the document entered by the user in the Wikipedia article submission form, in its original, "transparent" wikitext source code form.
This document may contain a line like
[[Image:Britney_Spears.jpg]]
This line of text is licensed just like all other text in the document. It is implicitly understood by the user that this line will produce an image that is rendered in-line, but the copyright status of that image is not confirmed in the article -- it is confirmed during the upload of the image. The image data is stored in a separate row in the table, and can be referenced by *several pages*. The image pages even have their own history, both for the image and the image description. The Britney image is, for all intents and purposes, a separate work.
When the server creates the HTML page, it turns the above line into something like
<img src="http://www.wikipedia.org/upload/2/2e/Britney_Spears.jpg">
The user's web browser then produces the *aggregate work* consisting of the text and the image by fetching both from the server, from different locations.
The author did not create a "combined work" by combining the article with the line [[Image:Britney_Spears.jpg]], any more than he would by combining the article with the link [[Christina Aguilera]] or [http://www.britneyspears.com Britney official homepage].
Now what happens if I insert a Fair Use quote?
We do so in the text [[Martin Luther King]] under the heading "Views on anti-Zionism", for example. This quote is indented and inserted in the source text. Martin Luther King's estate holds the copyright on all his works. The resulting article is, for all intents and purposes, a *combined work*. This means that the work as a whole has to be FDL-licensed -- which we cannot do because we do not own the King copyrights.
Of course, you might argue that the quote is not copyrightable, but that is a very difficult stance to take given that even cell phone ringtones are considered "intellectual property".
Directly included quotes are a much more serious legal issue than client- side included images. As I previously suggested, that might be addressed by supporting transclusion of some type, so you could say
[[>Martin Luther King on Zionism]]
which would automatically fetch the quote from a separate namespace (which does not have to be specified because transclusion is only possible from that namespace -- hey, even my hypothetical features have to be user friendly) and insert it in the proper place. This would put quotes on the same legal level as image pages (with the distinction that the transclusion would happen server-side), and allow us to use the same quote in multiple places comfortably. Editing would get a bit trickier, though.
But in my understanding, it's not image pages that we have to worry about with regard to the FDL. Of course *we* do have to worry about what fair use means in each individual instance, but the same applies to text.
Regards,
Erik