On Thu, 2007-10-25 at 14:15 +0100, David Gerard wrote:
On 25/10/2007, Padraic Ryan padraic.j.ryan@gmail.com wrote:
It seems to me a great first step to solving this problem is having an option at [[Commons:Upload]] page like "It is a derivative work of a media file already on the Commons" (perhaps also explaining what qualifies as a derivative work). This would allow (1) automatically insert some kind of template designed to indicate and keep track of derivative works, should one be made and more importantly (2) automatically check that the newly uploaded work is under the appropriate license - preventing someone from licensing a derivative of a GFDL file as CC-BY-SA or PD, etc. I have no technical knowledge in this area but that seems possible.
That sounds like precisely what we need. (Optional extension to allow naming more than one source file. Source can be on Commons, on another Wikimedia project, on the web or described in text.) This will also encourage a culture of reuse.
Bugzilla feature request, anyone? Writeup of request for wikitech-l?
I don't see that anyone wrote this up or submitted a feature request, so I'm volunteering to (demonstrating and encouraging reuse is high priority for Creative Commons, eg, that's why we created ccmixter.org, but that's just an aside).
First, let me see if I understand the ideal way to surface this. I imagine one of these:
A)
On http://commons.wikimedia.org/wiki/Commons:Upload a new "Where is this work from?" option "Derived from one or more existing works"
On something like http://commons.wikimedia.org/w/index.php?title=Special:Upload&uselang=de...
a field for the URL of work derived from, with (+) for adding multiple URLs if derived from more than one work.
These fields are added to the upload's article via a template, as licensing info already is.
B)
Nothing new on [[Commons:Upload]], but expandable URL-of-work-derived from field appears on [[Special:Upload]] always(?), much like the licensing drop-down seems to. As above, added to upload's article via template.
Is just the URL of the source work(s) good enough? Is it reasonable to suggest that the upload process attempt to scrape some info from provided URLs? This would presumably be required for part (2) of Padraic's suggestion.