So, there is a certain amount of obvious irony to the revelation that Chris Anderson’s book Free contains significant unattributed copying from Wikipedia. The entire piece from Virginia Quarterly Review‘s Waldo Jaquith is worth a read (and I leave it to you to decide if Anderson’s response is appropriate), but I just want to fixate on one small detail from Jaquith’s article:
Though reproducing words or original ideas from any uncredited source is widely defined as plagiarism, using text from Wikipedia presents an even more significant problem than reproducing traditional copyrighted text. Under Wikipedia’s Creative Commons Attribution-ShareAlike 3.0 license, Anderson would be required to credit all contributors to the quoted passages, license his modifications under the Creative Commons Attribution-ShareAlike license, note that the original work has been modified, and provide the text of or a link to the Creative Commons Attribution-ShareAlike license. Anderson has not done any of these things in Free. [Bold text added.]
Now I don’t think “all contributors” implies that each editor to the relevant passage in Wikipedia would need to named individually. I dare suggest such a requirement would make it nearly impossible to accurately cite a passage (especially given the way MediaWiki tracks edits). The handy-dandy “Cite this page” link which fixes to a specific Wikipedia edit (example) doesn’t deal with individual editors. And the page on Citing Wikipedia says nothing about these Creative Commons rights. (Interesting to note that Anderson claims the absence of a good citation format was largely why there aren’t any citations.) But then why does Jaquith, clearly a very clever Waldo, suggest that quoting Wikipedia is “an even more significant problem” than a copyrighted text?
In any event, this passage in a widely-read online article sows some real confusion. And it concerns a process that needs to be absolutely confusion-free if it is going to take flight — especially in a domain like academia.
I know that around v2.0 of the CC licenses language was added that specifically allowed attribution by link and this change was explicitly designed to allow for attributing wiki pages without having to list every author. I’m sure the CC site and/or mailing lists have much discussion (I remember some of it). In fact, I seem to remember quite a bit of discussion on the topic when Wikipedia admins first started discussing the possibility of moving to a CC license.
Maybe Jaquith is referring to the SA part of the clause. With traditional copyright, assuming I am within the parameters of fair use, I have no obligation to offer my derived work back to the community. With SA, I do. That potentially does make quotation more of a problem… except, since CC licenses don’t abrogate fair use, the argument could be made that as long as one is within those parameters (and I haven’t read the article, so this may be moot) then the SA doesn’t apply.
I look forward to seeing what some of the CC people and/or Larry Lessig have to say.
It does not sounds like to me (and I am an optimist) that Anderson deliberately lifted stuff like a desperate student writing a paper the night before it is due. I’d like to think there might be a day when attributing is an every day reflex, as much as I hope as more online writers pervasively hyperlink. THINK ATTRIBUTION cause it comes back around to ya. To me, a link to the source makes total credit to anything- the point is to connect back to the source, not to create a movie credit list of everyone who might have tossed a sentence into a wikipedia article. The notion of authorship in wikis is fuzzy.
I do find that applying Creative Commons to text is a bit complicated. I have a CC stamp on the bottom of my blog, but really, writing really seems like always a process of rewriting bits that has been written before. Who is really going to use a blog post in its full text besides a spammer?
I get the part about giving credit to written idea, but text is not really the same in terms of being a coherent whole as an image, sound file, video– the relationship of part to whole is not nearly the same.
The way you’ve framed this points out one the glaring conceptual failings of the Creative Commons: it’s not a commons, because the “BY” part, the attribution requirement, is inescapable. Getting rid of the attribution, by CC standards, puts you pretty close to the Public Domain. So if not a real commons, CC is perhaps better understood as “some rights reserved.”
People achieve commons-like effects through pseudonyms, or through self-effacement. A better job of achieving a commons is the GNU GPL, because althrough the copyright holder is named ‘technically,’ the license terms essentially make the actual holder irrelevant; the work becomes community property, and attribution is less of a problem. By putting the attribution at the centre, CC hobbles its ability to ever really achieve this.
Not that any of this excuses Anderson; if he submitted Free as academic work, it would be plainly called plagiarism—a very different thing than what CC and fair use are talking about.
Chris Kelty, in his excellent book Two Bits: The Cultural Significance of Free Software points out that the law can only set general rules; it’s up to an actual community of practice to figure out the social and ethical norms that govern what people actually do. The CC muddies this distinction.
Pingback: The Difficulties of Open Content Attribution « Open Education News
Pingback: So many thoughts, so little time — Internet Time Blog