What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 134 Stack Exchange communities.

I'm not meaning this to be a duplicate of this feature request. Here, I'd really like to find out if the current migration process is a violation of the license under which contributors grant their content to Stack Exchange Inc. and the community.

Currently, when a question or answer is migrated, the post is migrated in its edited state, but the revision history is not preserved. Because the revision history is not preserved, there is no attribution for the editor's content. There are two problems caused by this lack of attribution.

  1. The Stack Exchange Terms of Service states that subscriber content (defined as any material posted by the user) is licensed under CC BY-SA, which requires attribution. This includes edits, and lack of attribution could potentially cause some serious legal issues for Stack Exchange Inc.
  2. Bad edits could be credited to the post's author instead of the editor. (Not the main issue here.)

If, in fact, the lack of attribution is a violation of the ToS as it seems, I see two possible solutions:

  1. Migrate the revision history with the post, so that the "edited by..." link still exists.
  2. Specify in the section of the ToS detailing attribution that attribution is not required for edits and possibly update/replace the blog post linked in the footer. (I don't know this would affect content that already exists.)

TL;DR

We need to do something to prevent legal issues due to lack of attribution when migrating edited posts.

share|improve this question
4  
Whoa, amazing catch. –  Undo the Snowman Nov 25 '13 at 15:36
20  
brb, suing Stack Exchange, Inc. –  Yannis Nov 25 '13 at 15:44
1  
You do know that there is a link back to the original material on the original site, right? –  Robert Harvey Nov 25 '13 at 16:26
2  
Wanting to report the Stack Exchange network itself as per Updated procedure for reporting SCRAPERs, since the sites are both highly ranking in search results and aren't following attribution requirements. –  Joshua Dwire Nov 25 '13 at 16:27
6  
@RobertHarvey I did know that, but aren't the answers immediately deleted and the question deleted after 30 days? Attribution needs to be visible to anyone, not just 10k users. –  Joshua Dwire Nov 25 '13 at 16:28
2  
I think from a legal standpoint, Stack Exchange isn't likely to have any liability for this, since the license you grant per the TOU is broader than cc-wiki, and it seems silly to claim that the site's software is bound by the terms of use itself since it's not a user. This may be a bug, but a legal issue... probably not. –  Wooble Nov 25 '13 at 16:31
12  
It's disturbing that the edit history isn't migrated regardless of the legal implications. –  tvanfosson Nov 25 '13 at 16:34
2  
@Wooble Maybe, but the ToS still require attribution, as does the CC BY-SA license. Nothing specifically excludes edits from attribution requirements. In addition, the SE software/network is controlled/developed by Stack Exchange Inc., so Stack Exchange Inc. would still be the liable entity. If I created a website that scraped SE content and displayed it without attribution, I'd still be liable even though only the software violates the license. –  Joshua Dwire Nov 25 '13 at 16:35
2  
Well, I combed through the most recent 20 questions migrated to Stack Overflow from other sites looking for examples, but could not find a single one that exhibits this problem. If it is a problem, it appears to be a rare one. –  Robert Harvey Nov 25 '13 at 16:37
3  
@RobertHarvey are you saying it's a non-issue because it only happens rarely? I would find that sort of thinking overly optimistic for a programmer. :) –  tvanfosson Nov 25 '13 at 16:41
6  
@Wooble: The site's software has no right to violate contributors' privileges. –  Lightness Races in Orbit Nov 25 '13 at 16:41
    
@Robert If I only get to sue you for £5m rather than £100m, you'd better know I'm still coming for you! –  Lightness Races in Orbit Nov 25 '13 at 16:42
    
@LightnessRacesinOrbit: no, but the site itself has the right to "use, copy, cache, publish, display, distribute, modify, create derivative works and store such Subscriber Content" and isn't itself bound by any of the Use of Content provisions of Section 2. But I'm not your lawyer; feel free to get one and get laughed out of court. –  Wooble Nov 25 '13 at 16:43
2  
@RobertHarvey: I'm sure there's some judge out there who'll accept that as a valid claim! –  ThiefMaster Nov 25 '13 at 16:44
1  
@Wooble Does it have the right to do so whilst removing public attributions as required by the licence under which such works are submitted to the site? Licences protect both parties: you cannot just opt out of it whenever you please because you are a website. The licence is on the content, not on the website. –  Lightness Races in Orbit Nov 25 '13 at 16:45

3 Answers 3

If it was an issue before, it isn't now.

Edits are tricky. The rules in the license are interpretive:

  • Some edits almost certainly don't require attribution, such as those that are then completely rolled back.
  • Some edits are in a gray area, such as incredibly trivial revisions (removing whitespace, formatting, etc.)
  • But some pretty clearly require attribution, and whatever the legal situation may be regarding a history that's only available to some users, the more important issue is this:

We want to err on the side of more attribution - you trust us with your work, and we want to ensure you to get credit wherever possible.

So, as of today, anyone (regardless of rep) can link back to the original site's revision history on migrated questions, even if the question has been deleted on the original site.

enter image description here

If you go to the post pictured above in incognito, you'll find you click the link in the "migrated from" post notice (circled above) and it will take you back to the posts revision history on StackOverflow, even though the post there is deleted, and normally inaccessible to anonymous (or lower-rep) users.

share|improve this answer
6  
What happens if an answer with edits is migrated along with the question? –  Joshua Dwire Jan 15 at 19:56

TL,DR: Yes, it does. Only in a small minority of cases — but Stack Exchange must nonetheless make sure that that small minority of cases is treated correctly (in the same manner that spending 364 days per year without committing a burglary does not make it ok to spend one day a year committing burglary).

Facts

Law

Stack Exchange operates under United States law. Title 17 of the U.S. Code, and in particular chapter 1, applies.

The content of posts is Subscriber Content as per §3 of the Stack Exchange terms of service.

all Subscriber Content that You contribute to the Network is perpetually and irrevocably licensed to Stack Exchange under the Creative Commons Attribution Share Alike license.

Clause 4c of the Creative Commons license recognizes the right of the author to be attributed:

If You Distribute, or Publicly Perform the Work or any Adaptations or Collections, You must (…) keep intact all copyright notices for the Work and provide, reasonable to the medium or means You are utilizing: (i) the name of the Original Author (…); and (iv) , consistent with Ssection 3(b), in the case of an Adaptation, a credit identifying the use of the Work in the Adaptation (…). (…) in the case of a Adaptation or Collection, at a minimum such credit will appear, if a credit for all contributing authors of the Adaptation or Collection appears, then as part of these credits and in a manner at least as prominent as the credits for the other contributing authors.

The license does not define who must be considered an author in cases where a work is the product of a collaboration.

US copyright law applies to “works of authorship”, and defines the rights of “authors”. These terms is not defined in the law, and thus defined by jurisprudence.

17USC §101 does define “derivative work”, stating that “a work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a ‘derivative work’”. 17 USC §103 adds that “the copyright in a compilation or derivative work extends only to the material contributed by the author of such work”, which does not help in defining whether a contributor can be considered an author.

The article “Defining “author” for purposes of copyright by Russ VerSteeg presents the main views regarding how a contributor to a work gains authorship status. Emphasis is mine.

  • “As a general rule, the author is the party who actually creates the work, that is, the person who translates an idea into a fixed, tangible expression entitled to copyright protection.” (Justice Thurgood Marshall, in Community for Creative Non-Violence v. Reid, 490 U.S. 730, 737 (1989).)
  • “A collaborative contribution will not produce a joint work, and a contributor will not obtain a co-ownership interest, unless the contribution represents original expression that could stand on its own as the subject matter of copyright.” (Paul Goldstein, Copyright: Principles, Law, and Practice § 4.2.1.2, at 379 (1989).)

This is the majority view. VerSteeg's article goes on to present minority views (upheld by some US courts) which have a broader definition of authorship, requiring a lesser contribution.

Application

Credit to contributors

A vast majority[citation needed] of edits on Stack Exchange are of an editorial nature and thus do not pass the Marshall-Goldstein test: the contribution of the editor cannot stand on its own. For example, spelling and grammar corrections, improvements in presentation and formatting, even the correction of mistakes or the writing of a new title are not original expressions that could stand on their own.

However it does sometimes happen that an editor makes an original contribution that could stand on its own. For example, it does happen[citation needed] that someone adds a figure or original photograph to illustrate a point, or a code sample, or adds updated information to follow technological evolution, or clarifies a point where the original author's explanations were unclear. All of these cases are both legitimate edits according to Stack Exchange's editing policy (“include additional information only found in comments”, “add updates as the post ages”, “add related resources”, “clarify the meaning of the post”) — and in the case of Community Wiki posts, edits could even legitimately go beyond that.

Such contributors must be attributed as such as per the license that the Subscriber Content is under.

Misattribution of edited content

The user name indicated at the bottom of a post indicates the author of a post. If a post lacks a revision history, this indicates that the author is the sole author of the post. In case the post had other contributors, this is a misattribution which is potentially punishable under slander law. In some cases, this may also fall afoul of clause 4d of the Creative Commons license:

You must not distort, mutilate, modify or take other derogatory action in relation to the Work which would be prejudicial to the Original Author's honor or reputation

Attribution when the original poster has no account

Even attribution to the original poster is not always done correctly. If the author of a migrated post does not have an account on the target site, then only the author's account's display name is mentioned in the migrated post[citation needed]. I believe that this does comply with the letter of the Creative Commons license, even though it is arguable that authors on Stack Exchange are not defined solely by their display name (which is not unique) but also with their profile (which the display name always hyperlinks to in the interface, as long as the author has not removed their profile, except in the case of migrated posts). However, this violates the attribution requirements, which mandate to “hyperlink each author name”. (I am not sure whether these attribution requirements are opposable to Stack Exchange — the wording in the terms of service is unclear on that point.)

Conclusion

If a migrated post had more than one author, then a list of the names of all co-authors is sometimes legally required. A simple way to achieve this objective is to migrate the revision history of every post.

P.S. Since US law was mentioned… it should be abundantly clear, but I'll state it explicitly anyway. I am not a lawyer. I am not a judge. I am not a law professor. I am some guy on the Internet. So this post is not a legal opinion or legal advice, it's the opinion of some guy on the Internet. (I've never been convicted for burglary though, that must count for something?)

share|improve this answer
    
This is the best answer so far and draws from official sources, but I'm really hoping for a response from the SE team, since only they can change anything. I'm trying to decide whether to accept this and make a new post asking for action, edit this post into a feature request, or just wait for the SE team. Currently, I'm planning to wait for a response from the SE team. If by the time your bounty on this feature request expires we don't hear anything from the SE team, I'll contact them directly. –  Joshua Dwire Nov 25 '13 at 21:33
1  
Great answer. Just one thing to add (though it's always unpopular) is that this is almost certainly not a massive issue because any suit brought against SE for misattribution of edit content will almost certainly fail to demonstrate any damages (let alone damages high enough to make up for the time/expense of filing and arguing the suit). Disclaimer: I am not a lawyer either, just another guy on the internet. I just find issues like this incredibly minor and made in good faith on the part of the SE team without any long-lasting consequences. Fixing it is good, but I won't lose sleep otherwise. –  jmac Nov 25 '13 at 23:25
    
@JoshuaDwire: Did you get a response? –  Ricky Demer Dec 7 '13 at 2:12
    
@RickyDemer I contacted them on Monday or Tuesday, but I haven't received a response yet. I wonder if multiple people contacting them would get a faster response or would just be annoying. Probably the latter. –  Joshua Dwire Dec 7 '13 at 13:30
    
@Joshua I emailed them in 2011 and only got a response showing they didn't care, so don't hold your breath. It was Jeff at the time though. –  Gilles Dec 7 '13 at 18:04
2  
@gilles, 1. Yes, we care. 2. No, this isn't clear. Or simple. Many edits likely don't require attribution, but some probably do. We're looking into both the appropriate interpretation of the license, and better ways to err on the side of more attribution. We have a couple of ideas, but it's going to take some time to figure out which match what's needed. –  Jaydles Dec 16 '13 at 21:56

The real issue here is: do (or I guess, should) editors get the same attribution rights as an author? To answer the question, you have to look at several aspects of how edits work and how content gets attributed.

First and foremost, editing from other users is not meant for users to be adding huge chunks of content, and shouldn't be a radical change. The post should not change in meaning. Editing is only meant to clarify the post by fixing spelling, grammar, and other typos as well as simple things that the author didn't catch (like referencing the wrong link). By editing, the editor has to realize that the post will still appear as the author's post. If the editor expects to be attributed for their additions, then they really should be posting a new answer themselves.

Secondly, you have to think about how saying "edits have to be attributed" impacts the use of content on other sites. You're basically saying that by using content on the site elsewhere, I actually have to go through the revision history and determine who the real author of that specific content was, and attribute multiple users wherever I post that content elsewhere. If I referenced enough content, I'd essentially have to copy the entire revision history over to my site too, which I certainly don't want to do.

Since I'm not actually a lawyer, I can only say that editors should not be included under the attribution license, because in theory they're not contributing anything worth attributing. Although, the theory falls apart for community wikis - I have no idea how attribution is supposed to work on those.

share|improve this answer
    
This seems like a valid theory, but as the ToS stands, edits are posted by the subscriber and fit the description of "materials displayed or performed on the Network, including, but not limited to text, graphics, logos, tools, photographs, images, illustrations, software or source code, audio and video, and animations". Therefore, they are entitled to the same license and attribution requirements in Section 3. Either edits should be attributed after migration or the ToS should be updated to clarify that attribution isn't required for edits. –  Joshua Dwire Nov 25 '13 at 17:00
1  
I'm inclined to agree with Animuson. It's a matter of degree; failing to attribute some mundane spelling and grammatical errors is very different from scraping one or more SE sites and republishing that content without giving anyone credit. Ultimately, common sense prevails, and at the end of the day I don't think anyone really cares about their minor edits being attributed. –  Robert Harvey Nov 25 '13 at 17:02
1  
Possibly the ToS/Attribution requirements can be updated to clarify that no attribution is required for edits and that attribution for CW consists of linking just to the post and attributing the origin site. –  Joshua Dwire Nov 25 '13 at 17:03
2  
"First and foremost, editing from other users is not meant for users to be adding huge chunks of content, and shouldn't be a radical change." — nonsense! Perhaps you should read help/edit. It says: "If you see something that needs improvement, click edit!", not "Try to improve without adding huge chunks of content". Collaborative editing is the whole point of CW posts such as the FAQs maintained here. –  Lorem Ipsum Nov 25 '13 at 17:39
    
@yoda Those are community wiki, though, which I specifically mentioned at the end of the post. –  animuson Nov 25 '13 at 17:41
2  
@animuson but your premise is that editors are not contributing anything worth attributing. That is incorrect. I have a few non-CW answers of mine which were improved by other tag users volunteering to add content (example). Your argument is that it becomes fuzzy whom to cite (outside SE) for a particular revision, so let's just ignore all the contributions made by editors during a migration. That's not right. At the very least, SE has to ensure that it is following its own rules for attribution and the ToS for cases where it is not fuzzy. –  Lorem Ipsum Nov 25 '13 at 17:46
    
@yoda An edit like that is pretty rare, and if it showed up in the suggested edits queue would probably be rejected as a radical change. Something like that certainly falls under the "if he really wanted attribution, he should have posted a separate answer" criteria. He chose to edit it into your post, and now attribution of it should fall onto you. Otherwise you get back into my second point, of having to attribute multiple people when you reference material elsewhere. –  animuson Nov 25 '13 at 17:57
1  
@animuson I don't think it is all that rare. It is unfortunate that these get rejected as "radical change", but I understand that with SO's breadth of scope, it's near impossible for an edit in a small tag to find a reviewer from the same tag (or knowledgable). However, that's a flaw of the system, not the license or the users. There's nothing preventing the devs from allowing the OPs to override rejected edits and apply them. The devs chose not to, because it might have minimal gains, but the absence of a feature isn't grounds to pooh pooh the ToS and license terms. –  Lorem Ipsum Nov 25 '13 at 18:21
3  
I already acknowledged that some people might find it confusing on how to cite a post with multiple editors, but I don't find it that hard to comprehend. People publish co-authored articles in newspapers/journals/magazines all the time, so I see no reason why whoever wants to cite the post cannot do the same (or make a judgment call on the ratio of authorship and attribute accordingly). But attributing an edit to its original author as in the revisions is trivial, not confusing and already recorded. By ignoring it and making it appear as if the OP wrote it, SE is willfully misattributing. –  Lorem Ipsum Nov 25 '13 at 18:24

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .