Founding Agreement comments by Sean Barrett on 1-20-2005 at 6:33 CST
I (Sean--note that starting an essay with "I" and signing at the bottom just makes everybody scroll) am a little surprised to see nick respond as if everything about the license is obviously hunky-dory and act like DW is the only one to have ever expressed these concerns, since, lo and behold, I made many similar points on ifMUD while we awaited DW's written reply.
The CC-SA license is a complex piece of legal writing, and it is not always particularly clear how things are meant to apply to a wiki-like scenario. (Please forgiving my lack of clever formatting here, but the 'editing help' popup is empty.) For example, DW asks: "It's not clear to me whether the proposal plans to have IFWiki as a whole covered by one license, or if each and every page of IFWiki is to get its own individual copy of the license." Nick replies, "We don't need to have any copies of the license on the site, just links. We just change a line in the config file for the wiki. I'm proposing that whole wiki be covered." I would guess DW's point was, in terms of the CC-SA license, is the "Work" to which it refers supposed to be the whole wiki, or the individual contribution. The answer is (I assume) that each editor licenses each individual contribution ("to the wiki", in some sense) as its own Work under the terms of the CC-SA. The entire Wiki is a "Collective Work" under the terms of the CC-SA.
Note that authors don't actually license anything to the wiki; they license their content in general, and the wiki is reproduced under the terms of the license.
The big issue is attribution and how this interacts with being edited. Jon says (and people said this on mud), "It seems to me that the history mechanism in the wiki provides a valid reference for individual contributions to articles", and it's great that it *seems* so to him, but does it really mean that? Do you want to take a chance on that interpretation and get burned later? (Some of these issues might be helped by writing a "what the license means to the wiki" article that documents all of these sorts of assmuptions. This might have no true legal standing, but it might actually help in that in some hypothetical legal case, someone says, "no, our interpretation of the license was _blah_", and you can point at the interpretation article and say the person contributed after reading it. This is something like how legal decisions sometimes make reference to the congressional record to understand what the intent of the law was.)
Nick produces the big, central, troubling quote from the license, commenting, "The license itself says you only have to convey the name of the original author if it is supplied:"
You must keep intact all copyright notices for the Work and give the Original Author credit reasonable to the medium or means You are utilizing by conveying the name (or pseudonym if applicable) of the Original Author 'if supplied;' the title of the Work if supplied; to the extent reasonably practicable, the Uniform Resource Identifier, if any, that Licensor specifies to be associated with the Work, unless such URI does not refer to the copyright notice or licensing information for the Work; and in the case of a Derivative Work, a credit identifying the use of the Work in the Derivative Work (e.g., "French translation of the Work by Original Author," or "Screenplay based on original Work by Original Author").
"I don't know what other valid interpretations there are of this," he adds.
This presents a number of problems for the process of continuous editing and refactoring that goes on a living, healthy wiki. First is the problem of copyright notices; second is the problem of attribution; third is the problem of 'crediting the use of the Work in the Derivative Work'.
There is subtle context in applying these to a wiki. At any instantaneous moment, the wiki is a collection of static pages that can be browsed. Over time, the wiki is an editable document in which one editor can edit the documents of another. So the question is how the three items above apply in these two cases.
Well, to understand that, we need the clause nick dropped from the front of that paragraph: "If you distribute, publicly display, publicly perform, or publicly digitally perform the Work or any Derivative Works or Collective Works," [insert above]
So, the first case of this ("distribute [...] or [...] perform the Work") isn't about editing, it's just about displaying. If the Work (an article submission or article update by an editor) contains copyright notices or other attributions, then these must be shown when distributing / publically displaying the work. Well, that's trivial; the wiki shows everything the author submitted. Done.
Now, what are the cases about editing? Well, the wiki hands an editor a page to be edited. The editor creates a Derivative Work of that Work, and hands it back to the wiki. The wiki will then publically display it. But the important constraint is on the editor while handing it back to the wiki, which is presumably the 'distribute' case of it. So, an editor of an existing page, pressing 'save page', has the following obligations to the Work on that page that he is creating a Derivative Work of:
1&2. "keep intact all copyright notices for the Work and give the Original Author credit reasonable to the medium or means You are utilizing by conveying the name (or pseudonym if applicable) of the Original Author if supplied": so you have to preserve copyright notices and bylines that are in the actual text (unless you delete all of the text to which the copyright or byline applied, presumably, although this isn't entirely clear)
3. "in the case of a Derivative Work, a credit identifying the use of the Work in the Derivative Work (e.g., "French translation of the Work by Original Author," or "Screenplay based on original Work by Original Author")": this would seem to require you to write something like "based on the previous wiki page". People are assuming that the infrastructure of changelogs (or simply the fact that it's a wiki) means that you don't have to do this, but I think a close, literal reading of the license, there is no other "valid interpretation", as nick said. When you press "save page", surely you're submitting _the contents of the text box_ as a (Derivative) Work, not the entire constructed page that the wiki will create out of it that provides the context and changelogs? The text box contents are all that you personally are "distributing", so surely that submission itself is the Derivative Work, and you personally are the person who bears a responsibility to make your Derivative Work meet the terms of the license.
So I don't think this is even close to cut & dried trivial the way nick thinks it is. Sure you can make a case for interpreting it as he suggests, but there are lots of other cases you can make. They may be dumb, stupid, cases, but this is a legal document, and the whole point of that is supposed to be to nail things down concretely and specifically and as unambiguously as possible.
Now, I don't think this legal nit-picking is really all that helpful or useful--I just think it's disturbing to pretend that the complexity of this legal document isn't there.
The most important point, which I think people agree on, is that it requires preservation of copyright notices and bylines. This makes editing a wiki really a chore. How much of the content does a given copyright notice or byline apply to? This matters if you decide to split a page up into multiple pages.
I think people may have missed DW's points about the essays: his point wasn't that he disagreed with the contents of the essay, but that the essay format is a problematic use of the wiki for exactly the same sorts of reasons. It doesn't invite collaborative editing. Now, there's nothing wrong with using a wiki to collaboratively edit a collection of documents that are themselves not collaboratively edited; it's just not his vision of a wiki. The question is, is that really other people's vision of it? Because if so, then worries about how you refactor and split pages and deal with copyright notices and bylines are irrelevant. But if you think a wiki page is supposed to be a flexible and fluid thing, they seem to become a serious hassle.
Now, one solution, as nick says, would be to have rules against bylines and just delete content involving them. But then it's not clear what the attribution part of the license requirement buys you, and you might as well have a non-attributive license. (The only thing it buys you is attribution of the whole thing to IFwiki, and that leads to DW's conspiracy-theory self-aggrandizement concerns. I don't see it that way myself, but I can see why he'd draw that conclusion given no other evidence about the reason for attribution.) The thing is, there aren't many non-attributive licenses. Wikipedia's requires attribution to Wikipedia. In fact, the only really 'allow derivative works, no attribution, share alike' license is putting things in the public domain.
I'm not actually saying you _want_ to put this project in the public domain (although I'm not sure where else DW could have been going), but you should probably enumerate the reasons why that would be unsatisfactory. Here are some possible reasons it might be unsatisfactory:
- 1. "People can take my essay and turn it into their own essay and not credit me!" -- you'd have the same problem if you disallowed bylines
- 2. "People can reproduce our wiki and not give us credit" -- is that so bad? this reason opens you to DW's complaint ("Are they looking forward to seeing their names or "IFWiki" plastered onto all things IF")
- 3. "It prevents us from putting copyrighted materials from non-wiki sources onto the wiki" -- so will _any_ license unless the original material was already licensed under a compatible license; e.g. if IFwiki uses CC-SA, then this argument applies to any source material not CC-SA. How much of that material is there really? How much would you be giving up?
- 4. "It may discourage people from contributing if they don't want their material in the public domain." How so? What would those people actually be losing, compared to a no-bylines-no-copyright rule with CC-SA? (This is just #1 in more ambiguous guyse)
- 5. "People can make money off our stuff." That's already true under CC-SA; you want "by-nc-sa" if you want a non-commercial license.
- 6. "It's not share-alike: people don't have to public-domain-ify things they make out if it." Well, yeah, that's the main reason to prefer Copyleft. I don't know why there's no non-attributive but share-alike CC license. Aha, there is one, but it's an "old" one and it doesn't show up in their chooser. I'm not sure why they deprecated it. Hmm, looks like they used to have a full suite of non-attributive ones and deprecated them all. No explanation, though they still link to them all, so I assume they didn't conclude that they were unenforceable.
I presume there are some other reasons I haven't thought of. Of the above, the only one that seems a serious issue to me is #3, and I think that's just evidence of why trying to make IFwiki be all things to all people is going to be problematic for licensing reasons. You don't really want to absorb all other content into a wiki under one big license.
I don't think it's wrong to call CC-SA viral. The "SA" part _is_ viral, pretty much by definition. (Just to point out what a non-viral thing looks like, consider public domain; it doesn't put any restrictions on derivative works. You can copyright a work derived from the public domain. Yes, CC-SA gives you more rights than no license, but that doesn't make it non-viral.) It may well be that it's perfectly fine to have a viral license, but why pretend it's not?
However, DW gets one thing wrong about the viral nature of the copyright notices in CC-SA. He says, "Never to be erased or expunged because a Creative Commons license won't let you revoke it once applied." This is not true. First off, you can of course relicense anything you created under some other, less restrictive license (e.g. one that doesn't require the copyright notice). Second, CC-SA has an explicit clause allowing you to DEMAND that people remove attribution to you (so you can disassociate yourself from derivative work you don't like), and it seems to me you could apply this to strip copyright notices in the wild for other purposes. (At least, legally; the technological/sociological problem is still there.)
I don't know under what terms people uploaded material to Dennis' IF Glossary wiki. Perhaps it was the "we're all friends here" license. If there was something more explicit, and the relicensing to IFwiki is incompatible and yet people are undertaking it lightly, that's a bit disturbing. Otherwise, though, it doesn't seem like that big a deal as an analogy to the future of IFwiki, since the whole point of this exercise is to make clear under what terms such a transfer could ever happen in the future.
I hereby place this section in the public domain (so I don't need to sign a founder's agreement). --Sean 06:33, 20 Jan 2005 (Pacific Standard Time)