Wikipedia talk:Non-free content/Archive 46

The logo discussion above reminded me of something else that I was considering regarding logos and the Threshold of Originality (TOO).

We know that this is a very vague concept in legal terms; there's certainly bright line cases where a logo passes the TOO (eg Apple Inc.), and cases where it clearly does not (eg Microsoft). We're left with a huge gap in the middle where "a combination of text and basic shapes" could go either way.

Because our goal is to be free content, and that we cannot easily second guess the TOO, I propose the following (language is flexible, I'm looking at the ideas):

• Logos, unless otherwise stated from sources, should only be considered to fail the TOO, and thus be considered as a free-content image, if they consist of:
  • A single line of text in a single font without any "word art" effects applied (curved text, borders, drop-shadows, etc.)
  • A single simple geometric shape (line, circle/oval/arc, triangle, rectangle, trapezoid, or any regular 5-sided or higher polygon) used in conjunction with that text or by itself.

Any other logo art which uses more details, even if they can be easily replicated in an SVG image or the like, contain originality that we cannot assure fails the TOO, and for purposes of being a free content work, need to be considered as non-free. (This does not bar their use, just requires us to follow our NFC policy on them).

• Word art is all subjective style and thus may be considered original
• Multiple lines of text means there's a placement issue of subsequent lines of text that can be considered as originality and thus should be avoided (eg New York Life Insurance Company)
• Once multiple geometric shapes are in place, their positioning, color, and the like relative to the text again all are artistic aspects that lead to originality.

This does mean that we are probably being overbearing on images that legally would not pass the TOO (eg remain free) if tried in a court of law in the US. But consider that this is the same logic we have an NFC policy over fair use: exactly when something is no longer fair use is only determinable in court, and thus a stricter policy that meets the free content mission is a better metric. Since it is basically impossible to otherwise assess the TOO-ness of an image, we should draw a bright line for ourselves, well beyond the fuzziness of the legal requirement, to avoid having problems in the future. --MASEM (t) 16:13, 10 May 2010 (UTC)

• I think this makes perfect sense. We are deliberately more restrictive, and this parallels this. Also, Feist v. Rural (1991) case before the U.S. Supreme Court, "To be sure, the requisite level of creativity is extremely low; even a slight amount will suffice. The vast majority of works make the grade quite easily, as they possess some creative spark" (emphasis mine). I think the standard Masem proposes gives us a considerably clearer picture of where the threshold, for our purposes, exists. Right now, we have a hodge-podge of opinions, and a bunch of lay people enforcing/tagging based on their own interpretations of law. Also, putting this standard in place doesn't remove content; it just says we'll tag less immediately obvious cases as non-free pending further discussion. --Hammersoft (talk) 17:12, 10 May 2010 (UTC)
  • I think I agree. On Commons not too long ago, the Klingon flag was kept as PD-ineligible, which struck me as absurd. I think Commons should consider adopting a similar standard as that proposed above. (Although I admit I'm not convinced on the multi-line criterion.) Powers ^T 21:27, 10 May 2010 (UTC)
    • Well, take the New York Life logo. Instead of spacing between the lines, there is purposely no spacing, allowing the text to run into itself, and creating a certain effect. Other examples include:
      • The Home Depot; the multiline text is used to "fill" the orange box in a certain way that's artistic. (Note this is a commons image)
      • Reader's Digest both old and new logo involve position of one line with the other to enhance an artistic effect. (Note, also commons)
    • I have a feeling that we are going to have to engage commons on this and gain that consensus there too. --MASEM (t) 21:48, 10 May 2010 (UTC)
      • • Oh, I understand; I just disagree. But then, that's your point; we should limit it to what everyone can agree on. I think you're among the first people I've seen suggest that multi-line text isn't PD-ineligible, though. Powers ^T 01:47, 11 May 2010 (UTC)
          • Ok, maybe I need to be a bit more exact: if it is multiple lines of text that is the same font and size and weight and color, arranged in a standard left-, right-, or full-justified or centered format, that likely would fail TOO and thus be considered free. But when you start having lines with different sizes or colors or fonts, that becomes sufficiently artistic. A case in point, the Weather Channel's current logo would fail the TOO, and thus should be considered free, since it is a single font of the same size and weight, and coupled with only one simply geometric figure. Alternatively, the Home Depot's logo does pass this lower TOO bound due to the differing font sizes. --MASEM (t) 17:57, 11 May 2010 (UTC)
• This is something on which we ought to take advice from Mike Godwin. That's one of the reasons we have a general counsel, to give us counsel. On the one hand, we have a responsibility to minimise any legal threat to the site. On the other hand we have an agenda to increase the amount of clearly declared free content in the world. We therefore have incentives both ways; so we should seek a professional view, rather than trying to act as our own lawyers. Jheald (talk) 17:45, 11 May 2010 (UTC)
  • We already know what Mike Godwin's opinion is: there is nothing legally threatening the encyclopedia by our current use of logos. This is not a legal issue, but again a non-free content vs free mission one. Because of the vagueness of TOO, the only clear cases when a logo fails the TOO and thus can be considered free content are based on the guidelines above (or unless specifically sourced to a court ruling as to the nature of TOO-worthiness). Thus, just like we do with regular content - making the presumption of non-free unless shown otherwise, we should do the same with logos and other images that qualify under TOO considerations. --MASEM (t) 17:51, 11 May 2010 (UTC)
    • I don't want Mike's opinion of whether this would be legally permissible as fair use or not. What I want to hear is Mike's considered opinion on what is legally free and what is not. I repeat what I wrote above: we have a fundamental agenda to increase the amount of clearly declared free content in the world. If we give the impression that properly free content is in any way not free, then we strike badly at our agenda. That is why we assertively stand up to copyfraud, and we should be very reluctant to label images as non-free unless it is absolutely necessary. Jheald (talk) 18:04, 11 May 2010 (UTC)
      • You're not going to get that answer from Mike. He is not here to guide us on the distinction between free and non-free because it is not a legal issue. He has stated that we're presently in no danger of overuse of logos from a legal fair use standpoint, but fair use and the free/non-free distinction are two separate things. It is up to the projects to determine how much beyond the Foundation's resolution to adhere to the free content mission. --MASEM (t) 18:34, 11 May 2010 (UTC)
        • I disagree about this not being a legal issue. If we are going to be marking branding that companies paid thousands and thousands of dollars for as PD (even if we add the TM template), we need to be sure we are accurately and legally stating that these logos are not eligible for copyright. Clearly, I think setting guidelines is a great idea, and I think we'd be entirely in the clear with your guidelines, but seeking counsel's opinion about such guidelines wouldn't hurt, eh? -Andrew c ^[talk] 19:58, 11 May 2010 (UTC)
          • Ok, I should state that it's not a legal "we're about to be sued" issue that Godwin would get involved in, but exactly what you're stating: WP should not be in the position of determining where logos are public domain due to TOO save in the simpliest of cases that are well established. --MASEM (t) 20:16, 11 May 2010 (UTC)

While I think the idea of providing a definition to guide the licensing of logos is a good idea, I'm concerned that the wording above reflects what is here a pervasive misunderstanding of copyright in regards to text. Text is not public domain by virtue of being simple (i.e. failing the TOO), but by virtue of being a useful article. TOO, therefore, is irrelevant for text (the issue of shapes is another matter). Consider the headlight on this car; it's a work of authorship fixed in a tangible medium and is likely sufficiently original (given the aforementioned very low threshold). It is, however, free by virtue of being a useful article. Similarly, Eltra Corp. v. Ringer and the Code of Federal Regulations, among others, have found typeface to be a useful article, which makes the level of originality irrelevant (subject, of course, to the "separate recognition" caveat thereto). We should indeed endeavor to err on the side of caution when choosing between free and non-free, but the "effects" condition above (curved text, borders, drop-shadows, etc.) is far too strict, at least my virtue of being overly broad, imprecise and based on a misunderstanding of the underlying "legal" logic. Эlcobbola _talk 19:19, 11 May 2010 (UTC)

Again, its not a absolute "wikipedia-is-about-to-be-sued" legal issue here; we're already talking about the use of logos within the bounds of fair use. The problem is that people are assuming certain logos fall into the public domain due to TOO but with a questionable amount of creativity. (This is different from the idea of the inability to copyright a utilitarian design such as the car or font issue). Now, I am not against having a line that is farther out, but the important part is to have a line that is reasonably clear to us on WP where there is no guesswork involved and that we avoid the blurry area where TOO is not clear and would require a legal challenge to confirm. --MASEM (t) 20:16, 11 May 2010 (UTC)

I have zero concern about Wikipedia being sued and, accordingly, made no mention of it. Potential litigants would have an obligation to mitigate their damages by sending a cease and desist letter. As evidenced, for example, by the removal of the logo for the Sochi Olympics at request the Olympic Committee, such offending images are promptly dealt with. This is to say nothing of the fact that probably all images on Wikipedia qualify for fair use "in real life", independent of our propriety tagging and policies. That said, my only point was the concern about how typeface (text) and only text appears to be being addressed. Logos that are more than mere text or have text with original creative elements that can be recognized/exist outside of the utilitarian function are not covered by my comment. The discussion seemed to have been neglecting that TOO only becomes relevant if a work is not a useful article. Эlcobbola _talk 20:32, 11 May 2010 (UTC)

I understand now. What I would say though is that while text is utilitarian, and thus the outright display of a word or two text in a single font cannot be eligible for copyright, the embellishments once you move away from a straight line of text all add a creative element to the image, diluting its utility in favor of expression. The only reason I would want to exclude word-art is that it is a brighter line than if it was included: take Google's logo, for example where we have both color and drop-shadows; it may very well fail TOO, but it is difficult to have language that assuredly affirms the Google logo as PD while excluding something like Ebay's. Being stricter is not going to harm us. --MASEM (t) 21:50, 11 May 2010 (UTC)

Being too strict does harm us -- or rather it harms what we set out to do: it marks material as non-free that actually is free. That is something we should get counsel's opinion before we embark on. Jheald (talk) 23:32, 11 May 2010 (UTC)

• It's two different viewpoints. The way I view it, and (correct me if I'm wrong Masem) I think Masem views it is we should not be accepting 'free' content unless we are certain it is free. If we're not certain it's free, we need to treat it with caution. This suggestion provides a line by which the lay person can judge something as free without skirting anywhere near the edge of the law, as opposed to random users with their own interpretations of law making decisions, and then coming in conflict over those decisions. By having this line, we provide a point at which we can say "we need discussion and consensus on this one". Whereas now, none exists. --Hammersoft (talk) 12:51, 12 May 2010 (UTC)

• No, that's right, and that's a good way to put it; we guarentee much less harm - legally and from intellectual property laws - by presuming non-free unless proven otherwise (which may generate a lot of false positives) than by presuming free unless shown non-free (which misses a lot of true negatives). We cannot decide when something is free outside the bright line bounds set by law (eg we can take into account dates of publication, etc., but we should not be judging on the fuzzy middle). --MASEM (t) 13:09, 12 May 2010 (UTC)
• Also as to the "harm" on educational aspect of WP, I don't see it. Non-free content on WP is not prevented, just restricted. For logos, there is typically only one article that uses them, the page of the entity the logo represents, and that's a reasonable rationale if the logo ends up non-free. Compared to the potential harm to the legal nature of the work, this is not a significant issues. --MASEM (t) 13:13, 12 May 2010 (UTC)

• I don't have a problem with taking a precautionary stance. I do have a problem with taking such a stance, if we haven't got qualified counsel's opinion on just how precautionary is necessary.
• Our vision, remember, is to maximise people's freedom. We directly reduce people's freedom if we persuade them that a series of images (and by extension, other similar material that they may come across away from WP), if we persuade them that such images are not free, if in fact they are free. I'm surprised you don't seem to see just how harmful that is to what we stand for. The issue is not whether or not we could still use it on WP. Freedom is about more than just what we can legally use on WP. Jheald (talk) 13:50, 13 May 2010 (UTC)

• "Our" vision is to write the best encyclopaedia we can. ("Our" in quotes as you can count me out now that there's more "NFCC enforcement" and politics going on around here than constructive encyclopaedia writing). --kingboyk (talk) 11:06, 2 June 2010 (UTC)

• Consider the Foundation's resolution; points #1 and #2 basically have the Foundation leave it up to the individual Wikiprojects to decide what EDP to take. In at least one case (de.wiki), the project has opted for zero non-free images even though they are still operating under US's fair use laws and could use them. Thus, this should be taken as the ability for us to take whatever stances are necessary to distinguish between free and non-free. That said, because this change impacts images on commons, I think it needs to be discussed there and within reason brought to the Foundation's attention.
• As the harm, it is the harm of misapplying licenses to IP that we need to be concerned with. IP moves from being non-free to free (or can start off free) ,but cannot move from free back to non-free; once the cat's out of the bag, it's impossible to put back into it. Presently, there may be logos marked here or on commons as free, that, if they were taken to court today and challenged, may be found to pass the TOO test and thus technically be non-free. Sure, I doubt that WP is going to be sued over that, but that hurts the integrity of the free-content mission, our first priority to adhere too. We should only be marking content as free if we are absolutely sure that is the case, and thus it is better to be wrong by mis-labeling a free image as non-free; that presents zero legal issues that WP has to worry about in the future at the cost of simply maintaining appropriate use of these "mislabeled" non-free images. We're still not removing the ability to use logos on pages of entities they represent, only that now a rationale is required (but logos generally have boilerplate text for that). --MASEM (t) 14:17, 13 May 2010 (UTC)
• (@JHeald) I don't think the point is marking things free to make things freer to use. That doesn't make things freer. It leaves the decision among lay people to decide, with purely subjective guidelines, whether something is free or not. It's easy enough to say "published in 1907, so pre-1923, so free. Cool, I'll mark it free". That's objective. When you get into subjective areas, people operating on their personal opinions of whether its free or not are going to make mistakes. I see Masem's suggestion as providing an easy to understand mostly objective line; if you fall to the free side of that line, mark it as free. If you don't, bring it up for discussion so that multiple people can discuss it, rather than just a single person operating on their own opinion. Encouraging discussion about something that might or might not be free is a good thing, not a bad thing. It helps protect the project itself, the original rights holders, and downstream users of our content. If instead we retain the current situation of independent users operating on their own subjective opinions, downstream users have to face the reality that the graphical content they are trying to use might or might not be free. That defeats our purpose here. --Hammersoft (talk) 14:20, 13 May 2010 (UTC)

• You still neither of you seem to be seeing my point. You're not qualified U.S. copyright lawyers, neither am I. Mike Godwin is, and a specialist in the area. If we give out an impression that the law is stricter than it actually is, that directly gives our imprimatur diminishing freedom. That is not harmless. I am not against adding an objective line to distinguish "certainly free" from "possibly free". I am merely pointing out that there is a real harm we can do if we give a misleading impression of where that line is -- something I hope you can see, because I thought you cared about freedom and the size of the commons. It's good that Masem has come up with a suggestion, but we should run it past the lawyer because this is something where we really ought to be advised by a qualified specialist. Jheald (talk) 18:21, 13 May 2010 (UTC)

• But it's not a "legal" issue in the sense that WP's presently in trouble. Our current TOO bar, however subjective that is, is not set at a point that is causing legal problems for WP. (*note, I am saying this based on the last few things we've gotten from Godwin w.r.t. to images, in particular about tv logo galleries, in which he said "we're not in any trouble". One can certainly ascertain through Godwin if we are presently in any trouble, but generally, we're the first to know if there's a true legal problem). What the guidelines do here is now place down a second bar that includes less matter under the "free" qualifier. We are, from a "legal" standpoint, below the first bar, so we're not changing any legal issues for the Foundation. Instead, that bar is designed less about being legal and more about approaching the free content mission, only marking things free that we know for sure are free. Mislabeling free works as non-free is going to bring zero legal issues, thus requiring no input from Godwin specifically. The overall Foundation and other projects and editors, yes we do, but its just not this type of pressing legal issue. --MASEM (t) 18:33, 13 May 2010 (UTC)

@Jheald; I think I do get the point. We're just looking at it from a different viewpoint. But let's work with the point we seem to agree on; having objective line. You seem to agree that would be a good idea? If Masem's line isn't good, what is your suggestion? Also, I don't think Godwin will be useful in any respect here. As noted, Godwin feels we're already ok. Personally, this isn't to me about being ok with respect to the law. It's about making sure we're not mistagging things as free that shouldn't be. Placing those decisions in the random personal opinions of lay people who come and go here seems very misdirected. --Hammersoft (talk) 18:42, 13 May 2010 (UTC)

• • But we also shouldn't be mistagging things as not free if they are free, because that directly misleads people about their freedoms. This is a legal issue -- not in the sense that WP is likely to be sued perhaps, but in the sense that it is squarely a question about U.S. Copyright law. That is why we should seek qualified professional legal input. Jheald (talk) 20:03, 13 May 2010 (UTC)
    • Except the case of a legally free image being labeled as non-free has no harm of any sort to any party involved, which can be considered and rectified as an honest mistake in being too careful with our free content mission once the error is pointed out (whether from existing or new court proceedings that define the logo as failing TOO). The countercase is not true in the larger scheme of things, because we have no legal right to mark things free that are not free to begin with, and while we could say that was also an honest mistake, there can be legal repercussions of that. --MASEM (t) 20:22, 13 May 2010 (UTC)
      • Systematically diminishing the commons, making people think that free content is actually non-free, does harm. Just like giving in to copyfraud. It has a chilling effect on what people think they can freely use. That's why we should seek expert advice on our guidelines. Jheald (talk) 20:47, 13 May 2010 (UTC)
        • But do you know for certain if something like File:TheHomeDepot.svg is free? Again, I point to the fact that for every other type of image, we presume non-free until proven otherwise, which may qualify some free works as non-free but that's never hurt us before. We do have reasonable ability to judge from known cases of TOO where to draw a line that is far from being in the blurry area of TOO without seeking legal advice. And, again, I don't think we'd get any info from Godwin even if you asked him on this because it's not a pending legal issue. --MASEM (t) 20:57, 13 May 2010 (UTC)
          • Any time we identify some free work as non-free that hurts us. It hurts our users, it hurts the world in general, and it hurts what we stand for. I'm not quite sure what you mean by "not a pending legal issue" - it's an issue that is manifestly legal, and it's pending right here. If mean it is "not an immediate existential legal threat" to WP, that may be true; but Counsel is not just there for immediate existential legal threats. When we have a legal question like this, which will significantly affect some of the material we present to the world, we are crazy not to at least to seek to draw on the expert we have in-house. If Godwin says he is too busy and can't look at it, well fine. But we should at least consult as widely as we can, and try to get the best most qualified legal opinion we can. Is "The Home Depot" definitely free? Our existing practice and guidelines say yes. But I would be reassured, rather than it being just our say-so, if I knew that we could present that judgement as following from the assessment of a named specialist in relevant U.S. law. Jheald (talk) 13:11, 14 May 2010 (UTC)
    • @Jheald; We don't have qualified legal help to aid us in determining if a given image is free or not. That's the point of this; to provide a line on one side of which we need to discuss it, so lay people can agree together, as opposed to separately, what is and is not free. This isn't about making a sweeping change to the project. It's just about providing a reasonable line that is readily understandable to encourage discussion when cases become less clear than objective. --Hammersoft (talk) 20:33, 13 May 2010 (UTC)
      • But we should get qualified legal help in setting that line. Jheald (talk) 20:47, 13 May 2010 (UTC)
        • I don't see how Godwin can be of any greater service than the rest of us. This is an issue of identifying a clear, objective line that everyone can understand. Not a line that a lawyer says is ok/can understand. --Hammersoft (talk) 21:03, 13 May 2010 (UTC)
          • It's an issue of setting the right clear, objective line. Of not unnecessarily mislabelling any content that is free, and whose freedom should be upheld and actively asserted. Finding out where we can draw the line is precisely a legal question, on which we should try to get the best legal input we can. Jheald (talk) 12:51, 14 May 2010 (UTC)
            • Well, we're not making any headway here. You want a legal definition. We want a definition people can understand. I was hoping for some middle ground. Apparently, it doesn't exist. --Hammersoft (talk) 12:59, 14 May 2010 (UTC)
              • • I want a definition that people can understand, that reflects accurate legal advice. I don't see the problem with at least trying to ask for that legal advice. Jheald (talk) 13:18, 14 May 2010 (UTC)
              • What we would need (if we had to go that way) is a lawyer that had dealt with TOO issues and can outline the bar we set, but my understanding is that Mike Godwin doesn't have ability (he's WP general counsel, and nothing in his past suggests it; while an IP lawyer, it doesn't appear to necessarily deal directly with copyrights). --MASEM (t) 13:06, 14 May 2010 (UTC)
                • But then he might well know who to approach who could give the advice -- perhaps a relevant specialist academic lawyer in a university department who might give WP an opinion pro bono. We're crazy, on an issue of legal substance like this, if we don't at least try to ask our own legal counsel for advice. Jheald (talk) 13:18, 14 May 2010 (UTC)
                  • Well, I've done my best to try to convey why this isn't a legal issue and failed. So, further words from me on that point will do nothing to aid the discussion. This proposal isn't going to go forward. Or at least, it's highly doubtful. As with just about every other attempt to clear the murky waters of non-free content on this project, this is firmly stuck in the mud. Sigh. --Hammersoft (talk) 15:22, 14 May 2010 (UTC)

This discussion now flagged up here at Commons:Village pump, and here on the Commons template talk page. Jheald (talk) 13:53, 14 May 2010 (UTC)

Arbitrary Break

There appears to be a misconception on Wikipedia (and to a significantly lesser extent on the Commons) that only standard/publicly-available/etc typefaces (e.g. Times New Roman font, Verdana font, etc.) and/or "simple" typefaces/simple arrangements of text are ineligible for copyright protection. This simply isn't true under US copyright and trademark laws. If text falls within the definition of a typeface, it is not considered eligible for copyright; no consideration is given to prevalence or dispensation.

"Typeface" is a term defined by the House Report of the 1976 revision of the Copyright Act as follows:

"...a set of letters, numbers, or other symbolic characters, whose forms are related by repeating design elements consistently applied in a notational system and are intended to be embodied in articles, whose intrinsic utilitarian function is for use in composing text or other cognizable combinations of characters."^[1]

It should be noted that "articles" in this case means "any medium in which it is used".

In Eltra Corp. v. Ringer US courts ruled:

"Under Regulation 202.10(c) it is patent that typeface is an industrial design in which the design cannot exist independently and separately as a work of art. Because of this, typeface has never been considered entitled to copyright under the provisions of §5(g)."^[2]

The United States Copyright Office states:

"...mere variations of typographic ornamentation, lettering, or coloring" ... [are generally not eligible for federal copyright protection]. Likewise, the [mere] arrangement of type...cannot support a copyright claim"^[3]

So typographic ornamentation is not protected and neither is arrangement. While I agree that the line between what is merely text and what is not not text is not always black and white (Does this logo have ornamentation or a tail?, but there are also some things that are not just "simple" shapes that are not eligible for copyright. For example, the US Copyright office states unequivocally that a Fleur-de-lis is not eligible for copyright.

I'm not interested in expanding or contracting the definitions on what is acceptable on WP; making something more constrictive that what is legal is fine, but it should be based on sound reasoning and should be as concrete as possible. Much of the debate on WP could be solved by a concrete policy/guideline.

Given the above referenced material, I think the litmus test should cover the following three scenarios:

1. Is this logo exclusively text whose "utilitarian function" is to be letters?

   If a logo consists entirely of letters, it is important to note the "intrinsic, utilitarian function" of those letters. As noted to the right, the New York Yankees logo consists of an "N" and "Y" whose intrinsic, utilitarian function is to be an "N" and "Y". No matter how ornate and no matter the arrangement, they are still letters and therefore ineligible for copyright and, hence, they form a PD image. This same logic does not hold for an image created from something like ASCII art. In such an instance, letters are arranged to form pictures and the intrinsic utilitarian value of the letters are used in a creative way amd in a manner inconsistent with its use as a letter. In short, ask yourself, "Are these letters, no matter how decorative, intended for use as letters, or do they have a purpose beyond the meaning of the letter like in ASCII art." 999,999 times out of 1,000,000 they are going to simply be letters and ineligible for copyright protection.

2. Does this logo exclusively consist of simple/uncopyrightable shapes that do not form a larger picture?

   This is an example of when it gets significantly more murky. How many shapes does it take? Which shapes are eligible for copyright protection? Simple geometric shapes do not attain any copyright. Some are more complex, but are still PD nonetheless. What happens when you get two? or more? My point is that it doesn't really matter. They are still PD shapes. I think we can help ourselves to stay out of hot water by limiting it to simple geometric (or otherwise PD) shapes in arrangements that are limited to arrangements in a simple geometric shape. For example: three squares in a straight line we will label as PD. Eight squares that form another larger square would be labeled the same. Five squares arranged in the shape of a trapezoid are also the same. Like the example in #1 though, this has limits. If you arrange hundreds or thousands of squares and form something like a Bitmap image, this gets beyond the "simple" realm.

3. Does this logo consist of text and simple shapes that separately meet the criteria of #1 and #2?

   If so, the image is PD. Example: The NY logo above is a blue square and text. Therefore, it is PD.

If it does not meet these criteria, then the image has the possibility of attaining copyright. In the interests of erring on the side of caution, for now we should label them as copyrighted even if they are only labeled as a registered trademark and not copyrighted. — BQZip01 — ^talk 00:11, 12 May 2010 (UTC)

References

1. [1]

   U.S. Code Congr. & Admn. News, 94th Congress, 2d Sess. (1976) at 5668
2. [2]

   Eltra Corp. v. Ringer, 579 F.2d 294, 298 (4th Cir. 1978)
3. [3]

   United States Copyright Office: What Is Not Protected by Copyright?

• There's a very large difference in copyright law between typeface (which is well established to be ineligible for copyright) and layout, again reflected when I mention the multiline text. For example, you're presuming the Yankees logo is ineligible for copyright because it is based on two characters from a font and a box. If it was just one character and a box, I'd easily agree, but the arrangement of those two character in the box is meant to create a certain look. We can't be sure reading the minds of the copyright office or a legal team, so we should take the stance of "presume non-free unless shown otherwise". Of course, with the Yankees' logo, it's moot because it's pre1923, so that's the type of evidence otherwise we can use to show it free. Same with the fleur-de-lis. Even though its a complex shape, it is a shape that dates back to the 1300rds and clearly cannot be considered original nowadays. So the proposed language is not meant to override where there is clear evidence of free-ness of an image. --MASEM (t) 13:09, 12 May 2010 (UTC)

  I again refer you to the basic question in #1. What is the "intrinsic utilitarian function" of the N and Y? If it "is for use in composing text", then it meets the verbatim vriteria of a typeface; the word "typeface" and "font" are largely used interchangeably. It may be to create a certain look, but the utilitarian function remains the same; the same is not true for something like ASCII art. Perhaps a better way to look at it would be to consider is it art with a letter in it or an artistic letter. If we can distinguish one from the other, I think we have our "black line".

  And I'm not "reading the minds" of the US Copyright Office, I'm using what they use: the law and their publications. Why shouldn't we do that? I'm not reading anything into it, I am applying black letter law.

  There are no policies or guidelines on Wikipedia which state that we should "presume non-free unless shown otherwise". If you have a link, please prove me wrong, but I am growing tired of people pushing such an agenda as fact/consensus when, in fact, none exists.

  Oh, and a fleur-de-lis isn't so clear. Lawsuits were threatened by the New Orleans Saints for copyright infringement of their design when they reached the Super Bowl. It was, of course, quickly shot down, but even in the "real world" people make the same mistakes. — BQZip01 — ^talk 23:12, 12 May 2010 (UTC)

  • We presume non-free until proven otherwise because WP has a free content mission and follows an exception doctrine police as per the Foundation's request. We require all media to be tagged with a license, even to demonstrate if it is free; if an image lacks this license, it is deleted presumed non-free per the Foundation's resolution.
  • Also, there is not an intrinsic utilitarian function for the layout of the two letters of N and Y in the Yankees logo. If they were laid out next to each other, as in normal text, then certainly that's utilitarian, but as soon as they are placed in that manner that is non-standard from any other printed text, it has an element of artisticness that, were the logo not out of copyright, may be sufficient to pass the TOO and make it non-free. The problem is we have no metric or means of second-guessing how that logo would be ruled non-free without a previous legal challenge to review. Thus, just like we presume non-free unless otherwise shown, we should be doing the same with logos even if that results in what may be free logos being tagged non-free. --MASEM (t) 23:51, 12 May 2010 (UTC)

    I disagree. I think you, and others, believe "presume non-free unless otherwise proven" is a policy when, in fact, it isn't. Obviously images need to be tagged appropriately as to their copyright status, but saying we presume it is nonfree until a label is placed on it is absurd. We may treat it under certain criteria (as in how to delete it), but we shouldn't assume something is nonfree just because someone missed a step in our abhorent upload process.

    No there is not an intrinsic utilitarian function for the layout of the two letters, but layout is something explicitly stated that is not copyrightable. If so, that same method of overlapping letters could be a copyright violation for every other team/organization that does the same thing with their logo.

    "..as soon as they are placed in that manner that is non-standard from any other printed text, it has an element of artisticness that, were the logo not out of copyright, may be sufficient to pass the TOO and make it non-free." As shown above (and below), that simply isn't the case.

    I agree that "[t]he problem is we have no metric or means [to determine] how [a] logo would be ruled [on]", but that is what I'd like to develop. We already have our WP:NFCC which we developed without a legal precedent. Why not have something like that to more clearly deliniate logos?

    I agree with Lingberg below when he says we shouldn't just label things nonfree because, "meh. I don't want to learn any more." — BQZip01 — ^talk 17:21, 14 May 2010 (UTC)

    • We have to assume that an image is non-free until proven otherwise by not only the points above by US copyright law: an image is presumed to be copyrighted as long as its been published. Now if you can positively show proven of publication before 1923, or other langauage that puts the image into the public domain (such as images that arise from the US Gov't), then that's our evidence that the image is free. Failing that, we must assume someone owns the copyright on that. Now, in terms of "fair use", we're not required to id the original copyright holder and thus show evidence either way of the copyright nature of the work, but we have decided that we need that for our NFC policy - images without such info are to be deleted. --MASEM (t) 21:12, 15 May 2010 (UTC)
      • Masem, again, no we do not have to assume that an image is non-free until proven otherwise. This is not policy, period! There is NO default position. You have to assess each image on its merits, not assume the person who uploaded it is committing a copyright violation unless you personally are convinced. A better standard would be to make a determination as to what is copyrightable in the first place and form guidelines that apply to all images. For example, photographs are copyrightable with the following exceptions, those that are faithful reproductions of PD 2D works of art. Beyond that, a PD source, a date of publication, or an uploader-created and licensed work are basically the only things that can bypass this. There is no reason we cannot set up similar criteria for trademarks. Something that may be useful for registered trademarks is that they are all categorized by their creators under various criteria to include "text only", "text and animals", "text and simple shapes", etc. We could easily use this as one method to show whether or not a registered trademark is copyrightable. Thoughts? — BQZip01 — ^talk 16:31, 16 May 2010 (UTC)
        • No, we do assume an image that lacks a license statement (even if it is a PD image) is non-free and inappropriate and delete it; that's policy. Now, on the other side, we don't have a means of reviewing every image by hand and affirming the license is correct, so if a user, intentionally or not, uploaded a clearly in-copyright image but tags it PD, it will likely sit there until noticed and reviewed and corrected. But it is again important: without any license information, we presume non-free and delete. Of course, this is by human hands and if they can tell its PD or the like, they can tag appropriately too. --MASEM (t) 16:47, 16 May 2010 (UTC)

          1. That is a far cry from "we assume images are non-free unless proven otherwise".
          2. It still isn't policy to delete an image on sight (or nominate for deletion) just because it lacks an image tag. I invite you to prove to me otherwise: quote/link to the policy which states this.

          — BQZip01 — ^talk 04:57, 17 May 2010 (UTC)

          • • "A file in use in an article and uploaded after 13 July 2006 that does not comply with this policy 48 hours after notification to the uploading editor will be deleted." from the end of WP:NFCC. As the policy requires a license tag (#10b), this is what you're asking about. --MASEM (t) 05:09, 17 May 2010 (UTC)
              • That would be for non-free content only as this policy reflects. This rule does not apply to free content. Furthermore, it isn't instantaneous (there is a 48-hr delay). It still does not assume that an image is non-free until proven otherwise. — BQZip01 — ^talk 16:30, 17 May 2010 (UTC)
                • Same applies to any image if its not labeled otherwise. See WP:IUP and WP:GID. We have to go by the fact that in today's age, copyright exists for anything unless proven otherwise in the US (where our servers are hosted). And no, it's not necessarily instantaneously, but the point is there: free content must be explicitly defined as free content to be treated as free content. Admins reviewing and deleting images of course should apply common sense (fixing a typo, identifying something obviously pre1923, etc.) but if the case is questionable and the submitted isn't fixing it, we have to remove it as non-free. --MASEM (t) 17:05, 17 May 2010 (UTC)
                  • • I have read both of the pages you discuss and neither one advocates what you are saying. Copyright does not exist "for anything unless proven otherwise" and the "today's age" has nothing to do with it, only law. 1. there is no standard for "proof" 2. there are many instances where copyright is claimed, but not applicable; copyright applies to a lot of things automatically, but it doesn't apply to many others. I agree that free content should be labeled as such to be treated as such within Wikipedia. Admins should not delete something just because the case is questionable and the submitted user isn't fixing it. Admins should consider if the page can be improved and take the time to fix it (if possible). If it isn't possible to fix (i.e. clearly copyrighted and no possible FUR could exist), by all means delete it post haste. But if the image can be fixed, we should attempt to fix it: "If the page can be improved, this should be solved through regular editing, rather than deletion". Deletion because of an upload error, simple mistake, or simply not knowing what to do is ridiculous. — BQZip01 — ^talk 18:23, 17 May 2010 (UTC)
                      • Of course admins should try to fix it if it is obvious. But that's "should" and not "are required to". It is the onus of those putting information (and by extension, non-text content) to do the necessary steps to ensure it meets the requirements to be kept. If the case is not a clear instance of which way it should go, such as a monotone photo that may or may not be taken before 1923, we can't assume its free, and unless the admin just happens to know the source and confirm it either way, it is not worth their time to try to track it down and fix it and instead delete it. If that does end up a mistake, it can be reuploaded or undeleted. But you're mistaken to think that we can presume something is free without any other evidence to support that. --MASEM (t) 19:13, 17 May 2010 (UTC)
                        • I've never said we should "presume something is free without any other evidence to support" (I believe your interpretation that I disagree with your opinion and therefore believe what you believe to be the exact opposite may be part of the misunderstanding), but I am saying that there is not a default position period. Each image should be considered within our criteria. If eligible for copyright and there is no tag stating otherwise, it should be labeled as copyrighted and treated accordingly with simple FURs added as necessary (I'm tired of seeing company logos up for deletion when all that is missing is a FUR). If it is not eligible for copyright, it should be labeled as such and treated accordingly. If there is a question, it should be submitted to WP:PUF. If it is incorrectly labeled, that label should be fixed and the above rules apply. — BQZip01 — ^talk 20:29, 17 May 2010 (UTC)
                          • If is is copyrighted but not labeled as such, it is not the admin's responsibility to make the FUR, and in some cases, they can't make the FUR because they can't second guess the intent of the image. In some cases, like a company logo that's only used on the company page as the infobox logo, they can figure that out, but again, we do not require them to do this, it is an onus on the uploader. But now we're back to the "ineligible for copyright" works like logos. Unless there is information listed where this can be clearly made (such as a link to a photo on flickr that has the free licenses to reused, or text that says "scanned from a book published in 1903", admins cannot guess that the image is free, and thus must treat it as non-free, and as per our policy and the Foundation Resolution, delete it if its been lacking as such for X days. The same "treat as non-free by default" is the same thing with logos and the TOO. Unless it is a clear case of failing the TOO (as suggested by the points I've made), it is inappropriate for an admit to second guess whether the logo is free or not and should be marked non-free, though as I've suggested before, taking it to a larger consensus to affirm that or not. Unless there is absolutely clear information that asserts that an image may be free, we should never make that call for ourselves. --MASEM (t) 20:46, 17 May 2010 (UTC)

I have not read everything in here in detail, as someone just put a pointer here from Commons, but there do seem to be a couple misconceptions. Going by Copyright Office guidelines, either there need to be copyrightable shapes in a work, or the arrangement of non-copyrightable shapes must be creative. The big issue is "aesthetic attractiveness"; a simple arrangement that leads to something that simply looks really neat or creates a nice negative space effect doesn't qualify, I don't think. The Home Depot or Yankees logos are not close, I don't think. [The Yankees logo predates 1923, so there is no doubt either way, but it is two letters on top of each other -- very simple arrangement.] It is sort of like short phrases or slogans -- no matter how novel or creative they seem, they are not copyrightable. The Copyright Compendium states: In determining the registrability of a print, the copyright claim cannot be based solely upon mere variations of typographic ornamentation, lettering, or coloring. Likewise, the arrangement of type on a printed page cannot support a copyright claim. However, if the type is so arranged as to produce a pictorial representation, the resulting image is registrable. Thus, an advertisement which utilized lettering to achieve a pictorial representation of a person can be registered. The basic arrangement of type on a printed page is not copyrightable (for that reason, they also reject copyrights in book layout, per the same document), so it would seems as though the arrangement itself needs to be non-simple if it is going to be copyrightable -- "non-standard" (i.e. something slightly different than a line of text) doesn't necessarily qualify by itself. We apparently need to ignore the visual effect the arrangement produces, and decide if the arrangement is creative in itself (i.e. would be it copyrightable pretty much no matter the shapes used). I'll post most of some content I wrote in a recent discussion on Commons.:

The U.S. Copyright Office distinguishes "aesthetic appeal" from copyrightability. In their Copyright Compendium, they state Quality, aesthetic merit, ingenuity, and uniqueness are not considered in determining the copyrightability of a work. By that, letters or text in a simple combination are not eligible, regardless of the effect they create (which is aesthetically pleasing, but not necessarily artistically creative in the copyright sense). From 37 CFR 202.10, In order to be acceptable as a pictorial, graphic, or sculptural work, the work must embody some creative authorship in its delineation or form. Letters are still letters in their form, so the only possible copyright claim would be in the arrangement, which has to be non-simple itself. From one of their rulings, It is true that some combinations of common or standard shapes or other unprotectible elements can embody sufficient creativity with respect to how the elements are combined or arranged to support a copyright. [...] However, merely combining non-protectible elements does not automatically establish creativity where the combination or arrangement itself is simplistic. Also in that ruling is that the Office does not consider a work's aesthetics, attractiveness, symbolism, visual effect, uniqueness, commercial success or design choices, nor the time and effort expended in creating the work. To me, the placement results in a nice visual effect, but as mentioned that does not add to its copyrightability. There is a collection of U.S. Copyright Appeals Board decisions here (a few things which were accepted on appeal, and many which were not); it can make for some interesting reading if you want to get in depth. Some of the rejected examples include the Best Western logo, the DUB logo (DUB Magazine), the decision already linked which was about a S-shape in a box, a rosette beaded design, the Nikken USA logo, a Bruce Lee Core Symbol, and a Heartagram symbol. All of this is specific to the U.S. -- other countries can be entirely different. Some countries will have a lower threshold (protecting graphic design in general through copyright, which the U.S. does not -- they have design patents), and some countries will be higher (Germany for example has a really high threshold -- they don't let trademark and copyright overlap much).

Another famous case was one involving a New York Arrows (soccer team) logo; it went through the courts who also denied copyrightability (seen in this PDF, page 45). I would probably read through some of those decisions to get a better feel of the "PD-ineligible" line the Copyright Office uses (which is guided by decades of court decisions we likely aren't aware of). They can give a much better idea, and shows PD-ineligible goes beyond simple "text" -- there are many PD-ineligible shapes, and PD-ineligible arrangements. Carl Lindberg (talk) 14:48, 14 May 2010 (UTC)

• The point here isn't to ascertain a legal bright line to determine what is free and is not free. The written law and case law supporting this area is deliberately vague, as all possible scenarios can not be envisaged, but must be considered on a case by case basis. I don't think Masem, and certainly not myself, are trying to create a definitive legal line in the sand here. The government itself hasn't created one, and for us to do so would be dramatically overstepping our rights and abilities. The point here is create a line beyond which we should discourage solo, personal interpretations of the law and encourage discussion. --Hammersoft (talk) 15:27, 14 May 2010 (UTC)
  • Sure -- but the U.S. Copyright Office has more detailed guidelines, trying to answer the exact same question posed here (what is ineligible), trying for a balanced approach, based on law and many court decisions, and made with infinitely more experience and knowledge than we have. It seems like common sense to use their guidelines too. Those are not solo nor personal interpretations. Carl Lindberg (talk) 15:45, 14 May 2010 (UTC)
    • Of course not. But, do you expect the average person to read through all of that to come to a reasonable decision about the status of an image? What Masem proposes is a simple, easy to understand line beyond which discussion should be encouraged. --Hammersoft (talk) 15:47, 14 May 2010 (UTC)
      • Frankly, yes, when it comes to actual deletion (nominating if unsure is different). While we respect copyright, we should also respect the public domain with equal force, and should look for a balancing line. Deletions based on "I don't feel like learning more" aren't all that great, in my opinion. Also, a lot of the above seems to be about uncertainty surrounding aesthetic attractiveness, which the Copyright Office has stated very very plainly is not a consideration for copyrightability, so a guideline which hints that it may be is not a good guideline, I don't think. This stuff is, as you say, very complex with a lots of uncertainty, and there is plenty of room for debate -- but I don't like creating uncertainty in people's minds where the Copyright Office is quite certain on the matter. Trying to come up with a separate dividing "100% sure" line is just as difficult, in my opinion, and probably more -- it will almost always be possible to create uncertainty, and some people may well take such a line as the deletion threshold, which would be rather bad. A better approach may be to pull together some of explicit language on the matter, together with actual examples which were ruled uncopyrightable (I gave some above) and also some which were ruled copyrightable if we can find them (use on such a page would be fair use, I would think, if we have the graphics already for other reasons), to give people a better idea of the line without trawling through all the above material themselves. In other words, educate :-) I don't think the proposed guideline has any grounding in court decisions -- it appears to be based on personal uncertainty, which is understandable, but without (apparently) having read or being aware of much of the above. Carl Lindberg (talk) 16:17, 14 May 2010 (UTC)
        • This isn't a suggestion to encourage deletion of images. Perhaps that hasn't been explained well enough. --Hammersoft (talk) 16:27, 14 May 2010 (UTC)
          • Perhaps, but "failing the threshold of originality" implies that directly. And there are also several statements, "Word art is all subjective style and thus may be considered original", "Multiple lines of text means there's a placement issue of subsequent lines of text that can be considered as originality and thus should be avoided", and "Once multiple geometric shapes are in place, their positioning, color, and the like relative to the text again all are artistic aspects that lead to originality", all of which are directly discredited by the Copyright Office documents. Color and basic placement of text are generally not copyrightable elements. While there is certainly room for argumentation, the proposed guidelines expands that way beyond (sourced, authoritative) boundaries. To me, the proposed guideline is based around misconceptions -- are there any sources which support those original statements? Carl Lindberg (talk) 16:39, 14 May 2010 (UTC)
            • Meh. What was a simple proposal now has the flight capabilities of a lead duck. It's amazing to me how non-free issues always get stuck in the mud. Always. Thanks for your input Clindberg. I think we're done here. Anyone want to put a resolved tag on this? --Hammersoft (talk) 16:47, 14 May 2010 (UTC)
              • No. We shouldn't put a "resolved" tag on this. We definitely ought to try to come up with some summary guidance notes as to what we believe is PD and what isn't; what sources that view is based on; and where people can go to for a more detailed presentation of the issue. That guidance might even include areas where we decide that flagging an image "Could be free, but we're not going to chance it" is the best course. But above all the guidance needs to be based on accurate legal assessment. Jheald (talk) 17:10, 14 May 2010 (UTC)
                • Its apparent from the dialogue here that such a meeting of minds isn't going to happen. For example, we still disagree on the legal aspects. As with many other non-free proposals, this isn't going to fly. I'll exit the conversation now. If (hopefully) the remainder of you can figure something out, I'll be quite impressed. --Hammersoft (talk) 17:38, 14 May 2010 (UTC)
              • [e/c] Why? Nothing is resolved at all. Why not work towards a consensus here or on a separate page? Lindberg, thanks for the updated information. What are your thoughts about using my 3 guidelines stated above? Too much? Not enough? — BQZip01 — ^talk 17:21, 14 May 2010 (UTC)
                • Sorry for not getting back to this. As for your rules... closer, but I'm not sure they are correct. Letters aren't just utilitarian, they are a method of communication (they have denied copyright on a completely original alphabet under those grounds), and are also age-old symbols where typeface is considered minor variations thereof (note one of the cases where a S-like shape was refused, even though it wasn't necessarily meant as a letter). From looking at decisions, the Copyright Office will look to see if any of the elements are copyrightable in themselves, and if not, if the selection and arrangement is creative in itself. For the first part, copyrightable shapes, 37 CFR 202.10(a) says In order to be acceptable as a pictorial, graphic, or sculptural work, the work must embody some creative authorship in its delineation or form. That means that color choice etc. does not matter -- they are looking solely at the outlines. For the second part, selection and arrangement... one court case (Satava v. Lowery) said that not just any combination automatically qualifies for copyright protection, but rather "a combination of unprotectable elements is eligible for copyright protection only if those elements are numerous enough and their selection and arrangement original enough that their combination constitutes an original work of authorship". The Compendium states (§ 307.01: A compilation is registrable if its selection, coordination, or arrangement as a whole constitutes an original work of authorship. The greater the amount of material from which to select, coordinate, or order, the more likely it is that the compilation will be registrable. Where the compilation lacks a certain minimum amount of original authorship, registration will be refused. For one example which *was* ruled copyrightable, see here: A magazine copied Reader Digest's main page layout, using similar fonts with similar layout, along with other similar elements, and the totality of the whole was considered to be a copyright violation (though since the second magazine only did two issues that way, total awards from the case were $500). The Compendium in section 625 goes into other (non-graphic) examples; a compilation of 15 songs may be given its own copyright on the order alone (i.e. a completely different ordering of the same 15 songs would be a separate copyright). But, ordering 2-3 songs isn't enough, and ordering 15 songs the same way they were played at a concert also wouldn't qualify (the ordering was not original authorship). This may be my own take on things, but for arrangement itself, maybe think if the arrangement would still be copyrightable if all the elements were squares -- it may look ugly, but that is an aesthetic judgement which has no copyright relevance. That in itself isn't quite enough, because if there are a large number of elements, then font or color selection can enter into it as well. Usually with logos though, there are just a handful of elements at most, and quite often they are not creatively arranged. Carl Lindberg (talk) 17:33, 20 May 2010 (UTC)

I'm inviting comments on my image removal from Joran van der Sloot.—Kww(talk) 18:26, 10 June 2010 (UTC)

I think it was justified, due to very substantial noncompliance with NFCC, which the proponents persisted in, without really making an argument based on the policy. In fact, given that this article is presently getting tens of thousands of page views a day, Kww's actions were necessary and appropriate.--Wehwalt (talk) 18:30, 10 June 2010 (UTC)

Completely fair. Clearly not a free image, and thus considering NFCC, as it doesn't show any faces, it has no NFCC#8 bearing (completely replaceable by text). --MASEM (t) 18:35, 10 June 2010 (UTC)

• Concur with removal. The fair use rationale is very weak, and does not overcome WP:NFCC #8. --Hammersoft (talk) 18:49, 10 June 2010 (UTC)
• This is a question about whether a file passes NFCC. The uploader is required to fill in certain information, which they did, so the image could not be deleted on technical grounds - it's fair use or lack thereof could only be established by consensus. At the time Kww removed the file the best reading of the talkpage was that there was "no consensus", and while a discussion is taking place backed by a file deletion review and there is as yet no consensus it seems better to wait until there is consensus before acting unilaterally. Since that time more editors have offered their opinion, and consensus has moved in a definite direction, so now there is something to act on. I personally see how NFCC could be met in this case, but it is everyones responsibility to respect community consensus, so I fully support the fact that the file is no longer used in the article. Weakopedia (talk) 09:31, 11 June 2010 (UTC)

• These are two distinct things: any file deletion review would be answering the question "is there any possible use for this file?", which is a distinct question from "Is this particular use of the file in accordance with WP:NFCC#8." The answer to the second one was a resounding "no" from all editors that directly addressed WP:NFCC#8.—Kww(talk) 15:14, 11 June 2010 (UTC)

Can editors familiar with our fair use policy please join the discussion at Talk:Gaza flotilla raid#Fair use images? There are several news agency photos being used under a fair use justification that may be flimsy. Fences&Windows 13:22, 13 June 2010 (UTC)

Note that at least some of the images being used have already been listed at WP:NFCR#Gaza flotilla raid images. VernoWhitney (talk) 15:02, 13 June 2010 (UTC)

The BP article has a logo on the infobox. Is this really fair use? There are many photos of petrol stations with the BP sign. These are free use. Isn't free use preferred? Suomi Finland 2009 (talk) 20:18, 13 June 2010 (UTC)

We use clearly identifiable logos , free or otherwise, to identify companies. While someone could take a picture of a BP station sign, our use to identify the logo from that sign would not fall under normal fair use allowance (the purpose being to show the logo, not the station; that would be equivalent to a photograph of a copyrighted painting). --MASEM (t) 20:32, 13 June 2010 (UTC)

In WP space (i.e., a Signpost article on published research about WP), can I reproduce a scattergraph from an academic research paper freely downloadable from the Internet? The scattergraph concerns support and oppose votes in RfAs on en.WP. I see a "Copyright 2010" followed by the publisher's name and "All rights reserved". It's the seventh in this google ranking page. Tony (talk) 09:14, 13 June 2010 (UTC) PS And to what extent can I reproduce and discuss their research findings, i.e., expose their detailed data, without breaching copyright? Tony (talk) 09:38, 13 June 2010 (UTC)

Data are not subject to copyright; text and images are. So you could take data from the article to reproduce your own graph, if that's possible. You could contact the authors and get the original dataset. Fair use for images from scholarly articles is a grey area:. You can discuss their research findings to your heart's content, just don't plagiarise their presentation of them. Fences&Windows 13:22, 13 June 2010 (UTC)

Thanks. Hmmm ... a grey area. It can't be fair use, of course, in WP space. Tony (talk) 14:27, 13 June 2010 (UTC)

I think we agree that that is best and safest practice. On the other hand, there is a considerable question as to whether the scatterplot represents any original artistic expression, or on the other hand whether it is merely a simply functional representation of the underlying data. If there is not any original artistic expression in the scatterplot, then there is nothing in it that would attract copyright, and so it would be public domain, and so free for use anywhere on Wikipedia. Jheald (talk) 18:00, 13 June 2010 (UTC)

This is intriguing. The scatterplot (indeed all of the graphical representations) required very basic skills on a program like Excel. There's nothing special about them, and the colours chosen—well, they were arbitrary. Graphing such as this is commonplace nowadays. However, the data the graphs express is presented nowhere else in the article—no tables or lists (this is typical where there's a huge amount of data, such as the study produced). Without those numerical data, I couldn't reproduce the graphs properly—I mean, I could spend hours faking them, but they'd be different at very close inspection; who would bother? Now if the graphs were something special, such as a 3D representations of something that is quite difficult to express clearly in a graph, into which some original thought went, I'd be more inclined to call them original in themselves.

I note that research articles reproduce graphs from other publications when they want to, say, compare their data with those of the other study; and I'm practically certain journals don't send letters to each other asking for permission to do so, although the authors and, in the bib. at the end, the publisher, is always cited.

So where does this leave me? Remember, fair use is not allowed in WP: space. Is this free material? Tony (talk) 16:04, 14 June 2010 (UTC)

The safest way to do it would be to email the lead author of the study and ask for a copy of the data they used to create the plot. I'm afraid I'm unsure as to the status of graphs, since they're simple but the data could be displayed other ways. <shrug> VernoWhitney (talk) 20:32, 14 June 2010 (UTC)

Too much bother. What strikes me is that the graphs provide findings that are expressed nowhere in the text—that is, they provide actual data-points; however, it would be next to impossible to confabulate the data set in reverse from the graphs. I don't see why the data-points in their basic Excel charts are copyright if quoting text and explaining findings isn't. Tony (talk) 03:01, 15 June 2010 (UTC)

Check out Enguage Digitizer. It is an open-source free (GPL2) application that turns scans or screengrabs of graphs into numeric data for use in spreadsheets such as Excel or Gnumeric. One could use that in turn to create fresh charts free of any prior "design elements" but showing the same data.LeadSongDog come howl! 14:55, 16 June 2010 (UTC)

←That is a useful comment; but it exposes how ludicrous the rigid application of "copyright" is when data have to be degraded to be presented: any researcher will be upset to find their data expressed in a graph that is numerically even a little different from theirs (actually, it's falsification, since a graph implies that it's based on real data, not reconstructed). The choice of red and green as the two colours on the original graph? Wow, I could use blue and purple by clicking a button—the clicking of colour buttons on Excel can't possibly be considered inventive or artistic. Heck, I could sue the neighbour for painting their house the same colour as mine ... What if the product, in terms of data-points, is indistinguishable from the original because the reverse-engineering software is so good, or the data-points few and simple? This is getting silly. I won't bother, although thanks, I'll bookmark the link.

This is further evidence that WP needs to move on from the simplistic application of rigid copyright rules, and concede that there are instances where nothing is threatened and practicality is served by more flexibility—this is an extension of the debate about book cover illustrations in Signpost book reviews, in late April. Tony (talk) 03:09, 17 June 2010 (UTC)

You're going to have to take that stance up with the Foundation, since they only allow us to use non-free content to forward our education mission; the back-end of how WP works is not part of that. --MASEM (t) 03:47, 17 June 2010 (UTC)

The actual text of the Foundation is expressed in general terms and open to interpretation. If it weren't, why do the 280 WPs have significantly different attitudes and practices? Tony (talk) 03:55, 17 June 2010 (UTC)

I was advised to post my question here again: is it possible to upload this picture under a "fair use"-rationale in order to use it in Edwin Sutherland? -- Wo st 01 (talk | rate) 17:25, 16 June 2010 (UTC)

The answer is "very probably". First however, you should make a reasonable effort to find a free alternative (though I did a quick google and I don't think you'll have much luck). Also, you should endeavour to find out who actually owns the copyright for the image - that's not mandatory but it is extremely helpful to content reusers. CIreland (talk) 15:42, 22 June 2010 (UTC)

As of just now , the project is as close as possible with current detection methods to being in compliance with WP:NFCC #9. Prior to BJBot_3 coming online in January of 2008, there were more than 2000 violations in userspace alone. When DASHBot_5 came online in January of this year, there were more than 1000 violations (I routinely saw in excess of 1400). The daily report of NFCC 9 violations showed 76 violations as of 20:00 UTC yesterday. All of those violations have been removed. There are certainly a few remaining violations, but they are all new violations since yesterday. So far as I'm aware, this is a first. --Hammersoft (talk) 16:48, 25 June 2010 (UTC)

That's outstanding. VernoWhitney (talk) 17:01, 25 June 2010 (UTC)

I realize this is only tangentially related to this policy, but I thought editors with an interest in images might want to take a look at this. Commons has been developing a proposed policy regarding sexual content at commons:Commons:Sexual content. It is now stable and ready for review by third parties - please look it over and provide any feedback on the talk page. We want to move forward on adoption soon. Thank you! Dcoetzee (talk) 22:58, 25 June 2010 (UTC)

This actor was very famous and had many photos taken of him, even fans taking photos of him. Surely we don't have to stoop to theft from the TV programme? http://en.wikipedia.org/wiki/File:Jack_Lord_Hawaii_Five-O_title_220px.png and Jack Lord Suomi Finland 2009 (talk) 19:59, 26 June 2010 (UTC)

• Since he's dead, and no new free license imagery of him can be created, the general rule of thumb around here is we do accept non-free imagery for depiction purposes, as in this case. There might be free license imagery of him available, or somebody might be willing to release imagery of him that was previously copyrighted. If that happens, we'll replace this non-free image as soon as we are aware of the free imagery. If he were alive, we'd delete this image even if we didn't have a free license image of him at this time. --Hammersoft (talk) 14:41, 27 June 2010 (UTC)

Which non-English Wikipedias accept fair use images of copyrighted building images?

I want to know which other Wikis I can post an image of the Air France head office in (so far the image is on the English and French Wikipedias, and it is not eligible to be posted on the Spanish Wikipedia) WhisperToMe (talk) 16:48, 28 June 2010 (UTC)

meta:Non-free content has a table with pointers to the individual project policies, so that should be a good starting point. VernoWhitney (talk) 16:58, 28 June 2010 (UTC)

Thank you very much :) WhisperToMe (talk) 17:03, 28 June 2010 (UTC)

Resolved

I am seeking an exemption from the Non-free content policy for the ongoing Wikipedia:Contributor copyright investigations/Arab League in order to convert the list of images to a gallery to aid in sorting out copyright violations from allowable subject matter. Anybody for or against my proposal? VernoWhitney (talk) 15:19, 29 June 2010 (UTC)

I'll note that if such an exemption is granted, it will be temporary. On completion, CCI pages are blanked. And while CCI pages can hang around a long time, I think this one could be cleaned up relatively quickly with the attention of an image contributor. --Moonriddengirl ^(talk) 15:30, 29 June 2010 (UTC)

Why do the images have to be displayed on the page rather than just linking them? (i.e. use [[:File:Bagpuss.jpg]] to yield File:Bagpuss.jpg). Seems unnecessary to faff about getting exemptions. CIreland (talk) 15:47, 29 June 2010 (UTC)

Mostly because there are 312 (if I counted right), and I don't want to have to click through to each one to see whether it's a map or a flag or a photo in order to sort them. VernoWhitney (talk) 15:58, 29 June 2010 (UTC)

• I don't want to throw bureaucracy into the face of common sense. Please don't take this in that way. However, I don't see the justification of convenience applying here. Each image still needs to be reviewed, not just glanced at in thumbnail mode or in a gallery (same effect). The image description pages need to be seen in order to evaluate each image on a case by case basis. Yes, it's a lot of images. That's unfortunate, but we sometimes run across situations like that. If you want to group them together by flag, map, photo, logo, etc. then group them that way in separate linked lists. While technically speaking, exemptions to WP:NFCC #9 can exist, in practice they don't. Looking at Category:Wikipedia non-free content criteria exemptions, the only page in that category is a page for a bot that hasn't been run in two years . The other categories in that category all link non-free files, rather than displaying them. Even Wikipedia:Graphic Lab/Illustration workshop does not permit the display of non-free images in its workspace, and they have a much stronger justification for doing so than is the case here in my opinion. --Hammersoft (talk) 16:10, 29 June 2010 (UTC)
• There are a couple ways of doing this without requiring the exception.
  • If you have the Navigation Popups gadget enabled in your preferences, hovering over an image will bring up a thumb of that image and other useful details. It is sufficient to rapidly determine the diff between flags, maps, and photos.
  • You could create a file off size (local computer) that would be the page with all the images within in. Then, opening a spare browser tab, you can copy that file into the edit window and PREVIEW the page, which will show you all the images without actually storing it.
  • I would expect that there's some functionality of AWB that could be used here to help, how, though, I don't know. --MASEM (t) 16:18, 29 June 2010 (UTC)

Don't ask - just do it if that is the easiest way for you to do it. Prodego ^talk 16:21, 29 June 2010 (UTC)

Prodego, VW's problem is that if he "just does it", a bot will revert him. CIreland (talk) 16:22, 29 June 2010 (UTC)

Yes, something will have to be done about that. Prodego ^talk 17:06, 29 June 2010 (UTC)

I did just do it and then fought of DashBOT since it apparently doesn't respect the category and I thought I got the bots|deny tag set up correctly, but then Hammersoft removed the images and the category and told me to come here, so here I am. And yes, I am aware there are other ways to do it, but popups doesn't work on the computer I'm stuck with most often due to some wonky configuration issues, and the whole point of the category is to make maintenance such as this easier. If other people were willing to slog through it as it is currently it would already be done and I wouldn't be here, so I'm trying to make my job easier so it actually will get done. This seems to be the same logic used at Wikipedia talk:Non-free content criteria exemptions/Archive 1#Proposed addition for the one page that still is in the category. And re:Hammersoft, I'm not intending to be snarky, but part of your reason for opposing granting exemption is because exemption isn't granted? VernoWhitney (talk) 16:30, 29 June 2010 (UTC)

Ignore that and just get the bots to ignore the page. You want the images to be visible for a short time to improve Wikipedia - you shouldn't have to come here to do it. (In other words - Hammersoft should have let you do what you were doing, you did the right thing.) Prodego ^talk 16:56, 29 June 2010 (UTC)

You can also review the old revision of the page for reference. –xeno^talk 17:08, 29 June 2010 (UTC)

That works for the 5 images that already were in the gallery - unless you're suggesting that putting everything into a gallery like I want to, watching it get reverted, and then working from the old revision is somehow more kosher than getting a temporary exemption. VernoWhitney (talk) 17:21, 29 June 2010 (UTC)

As several people have said, a temporary use on a page explicitly dedicated to sorting out copyright violations is fine. I restored the category; it should be removed when the images have been sorted. The whole point of the "exceptions" system is to allow for this sort of thing, and people would not be suggesting so many alternatives if there were really no benefit to seeing the images. — Carl (CBM · talk) 17:37, 29 June 2010 (UTC)

Thank you. VernoWhitney (talk) 17:51, 29 June 2010 (UTC)

• Carl, I don't agree with this. I'm not going to edit war over this of course, but the addition of the exception has not achieved consensus here. Verno, yes I'm saying an exception is a very rare thing. In practice, we don't have them. That's reality. There is a possibility of one, but we'd need a strong rationale for doing so. The only rationale that's been provided is that it would be more convenient for you, a person encumbered with a computer that does not satisfy your needs, to do something. That's not adequate justification for an exemption. Yes, I'd like to see the work done. No, I don't want to see policy violated out of a need for convenience for a person lacking the computer they need. That's not a strong reason. Rather the opposite. Is there some other reason that lends credibility to this request other than your need of convenience? --Hammersoft (talk) 17:57, 29 June 2010 (UTC)

• Prior to Verno's recent work there, nobody has touched that CCI since you last did in January. CCIs are only opened where there is strong reason to believe copyvios exist. I would suspect that if an exemption means we rectify the possible widespread copyright problems with any kind of timeliness, it's probably well worth it. Whatever copyvios remain from this guy have gone ignored for five months. And if the work isn't done in timely manner, it seems like it would be easy enough to restore it to its non-exempt status. --Moonriddengirl ^(talk) 18:02, 29 June 2010 (UTC)

• I don't see that as an argument to grant an exception. That's an argument that the area needs more volunteers. As I noted, the graphic lab has a considerably stronger reason to have an exemption than the here proposed convenience of a single editor. They don't have an exemption. I fail to see a need to grant an exemption in this case for an editor with a lacking a computer who finds it inconvenient to process these images in the policy compliant ways proposed. --Hammersoft (talk) 18:06, 29 June 2010 (UTC)

Let me explain my thoughts. It seemed to me that people were suggesting all sorts of "behind the back" techniques that were intended to do exactly what the gallery tag would do: to show the images so that the maps could be sorted from the photographs, etc. I was about to just write a quick toolserver program that would display all the images for VernoWhitney. But then I realized it would be farcical to write a program to show the images, just to avoid adding a single line to the page itself to allow the images to be shown. Other people have suggested similarly convoluted strategies, such as putting the galleries in an older revision to hide them from the bot.

My impression is that this "exception" is temporary and related only to sorting out one batch of images. I presume that will happen pretty quickly, and then the exception can be removed. The "exemptions" section of the NFC policy seems to directly relate to this sort of management. — Carl (CBM · talk) 18:07, 29 June 2010 (UTC)

• (edit conflict) x 2. Let me just point out that there was no policy violation, as the page was in the category and WP:NFCC#9 specifically allows exemptions. Nowhere do I see it stated that there must be consensus for the category to be added, so I was violating no policy by adding it. As to the justification, the page is being used to ensure compliance with the policies of WP:COPY and WP:COPYVIO, which means it is being "used to manage questionable non-free content" as defined at Wikipedia:Non-free content#Exemptions. Cheers. VernoWhitney (talk) 18:07, 29 June 2010 (UTC)

• First, there is and was a policy violation. Adding the page to the category was done by the editor asking for the exemption. I objected to it, and asked the editor to seek the exemption here. No consensus has been achieved that this exemption is proper. Yet, the category has been added back. I find that improper. If we don't need consensus to add a page to the exemptions category, then anyone can add their userpages with userboxes containing copyrighted content. That's obviously not how it works.
• Second, the "convoluted strategies" are to assist the editor with a computer that is apparently lacking. Yes, we could provide an easy solution without the convoluted strategies. Yet, we're doing so for a single editor. If this "solution" were for the good of the many, rather than for the good of the one, I'd be happy to grant an exemption for the entire subset of CCI pages carte blanche. But that's not the case. We're asking for an exemption here for a single editor. If you want me to sort the images into maps, flags, and photos, hell I can do that right now without having to resort to violating WP:NFCC policy to do so. Just give me the word that that is what is needed, and I'll do it right now. --Hammersoft (talk) 18:11, 29 June 2010 (UTC)
  • That would be wonderful. I have only a vague idea of what is being sorted out. VW and/or Moonriddengirl: can you clarify what needs to be done, exactly? — Carl (CBM · talk) 18:13, 29 June 2010 (UTC)

• Just for reference, I just went through the 45 images in this section, sorting them into four groups (flags, photos, maps, graphics/logos). It took me 12 minutes. My work was edit conflicted, and what Verno did made my edit impossible (not his fault of course) so I didn't apply it. But, 45 images in 12 minutes to alleviate the need to sort. If I had applied my save (can't now), we wouldn't need the exemption anymore. --Hammersoft (talk) 18:30, 29 June 2010 (UTC)

• (edit conflict) First, Hammersoft, I'd love more volunteers. :) Sadly, my perpetual talk page ad hasn't proved that successful, any more than my last sad little attempt to drum up interest at AN. :( (Though I must say, we had a temporary influx that was very nice!) If you know where they're lurking, please please please send them on. :) I'm not the one working on this CCI; I'm currently working on a batch of infringing art articles, but what's needed here is for this content to be sorted, reviewed, repaired and/or tagged as necessary so this issue finally put to bed. That seems to be what he's doing at the moment: , , . Verno's been putting in overtime at CCI lately. Since images are not my major area, I'm afraid I've been bypassing this particular CCI for some time. :/ If you want more specifics and are willing to do the legwork, you might want to stop by Verno's talk page or leave him a TB, since he seems to be in "work mode." As of a few days ago, we had 28 CCIs, and I do believe he's trying to get that number down before the backlog gets much larger. --Moonriddengirl ^(talk) 18:35, 29 June 2010 (UTC)

• And you did just before I posted. So back to Vermeer with me. --Moonriddengirl ^(talk) 18:36, 29 June 2010 (UTC)

• My offer was rejected. C'est la vie. But, once his sorting is done there's no reason to maintain the galleries, is there? They can be converted back to linked lists very easily (a couple of minutes work at most, and I'll do it happily). --Hammersoft (talk) 18:38, 29 June 2010 (UTC)

• I should think so. I'll ask him to explain here if he has further need and how it might be implemented, and I personally appreciate your willingness to pitch in. As I said, our need is dire! --Moonriddengirl ^(talk) 18:51, 29 June 2010 (UTC)

• Okay, my sorting is done and I don't know of any further work that would need the galleries that can't be done just as easily via looking in history, so if you could turn it into a linked list again while keeping the comments next to the image links, that would be much appreciated. VernoWhitney (talk) 19:06, 29 June 2010 (UTC)

• Ok, I'll get to it right now. --Hammersoft (talk) 19:11, 29 June 2010 (UTC)
• Done. I've removed the exemption category and the DASHBot deny tag. --Hammersoft (talk) 19:18, 29 June 2010 (UTC)

• Thanks! VernoWhitney (talk) 19:23, 29 June 2010 (UTC)

• FWIW, I support the temporary exemption if it makes it easier on the user completing the oft-thankless task of cleaning up copyright infringement. –xeno^talk 18:43, 29 June 2010 (UTC)

We generally can't have non-free photos of living people, but is there such a rule that forbids the use of a non-free photo of a wikinotable animal, or could one be used with the appropriate rationales? Mjroots (talk) 05:50, 2 July 2010 (UTC)

It would depend greatly on whether a free photo could be created. For example, a non-free photo of a common zoo animal would probably be replaceable, but a non-free photo of a very rare animal might not be. I think this would have to be analyzed on a case-by-case basis. The rule for people is just that, in almost every case, non-free photos are replaceable, and we have included this in the policy because it is such a common question. — Carl (CBM · talk) 10:31, 2 July 2010 (UTC)

The animal in question is someone's pet, but it is notable enough for an article on Wikipedia. Mjroots (talk) 19:07, 2 July 2010 (UTC)

If it's a pet, it's tame, and we can reasonably expect a free image can be obtained. I'd consider the same restrictions. --MASEM (t) 20:39, 2 July 2010 (UTC)

• I think the same abstract situation applies to animals as it does to humans. Is the animal alive? If alive, is creating free license imagery impossible for (insert fantastic reason, like life imprisonment for example). There's not a question of whether it's easy or not to create free imagery. The question is can it be created. --Hammersoft (talk) 20:12, 2 July 2010 (UTC)

• • If we are talked a specific animal (in this case a living pet) then yes, the same idea as to why we refuse non-free of living people apply. I would also argue that what we could call "common" animals - ones that can be found in a zoo or are wildlife that easily co-exist with humans (deer, squirrels) etc. - should nearly always have a free image over non-free barring common-sense exceptions. All other animals - those hostile to humans, living far far from populated areas - we can allow non-free since there is no reasonable expectation someone can get a free image of them. --MASEM (t) 20:39, 2 July 2010 (UTC)

• Agreed. --Hammersoft (talk) 20:48, 2 July 2010 (UTC)

• I'd disagree on the "hostile to humans" part. I've got photographs of wolves and grizzly bears, and I'm trying for other local "hostiles" such as mountain lions. The hard part isn't staying safe, it's finding them in the first place. --Carnildo (talk) 22:40, 2 July 2010 (UTC)

All those can be found in zoos or nature preserves.so yes, free photos are reasonable. --MASEM (t) 22:43, 2 July 2010 (UTC)

• I was wondering about this when Varanus bitatawa was on ITN a couple of months ago. This is a critcally endangered lizard which has only just been described by Western science, and which only lives in a remote part of the northern Philippines. I would say that it is a candidate for non-free image usage: I for one wouldn't be shocked by a non-free image appearing in the article. Physchim62 (talk) 23:48, 2 July 2010 (UTC)

• I agree with that assessment. --MASEM (t) 00:46, 3 July 2010 (UTC)

Okay, so kicked off by an inadvertant use of the non-free File:Jessica Biel on the cover of the magazine Gear.jpg which made it to ANI, and my subsequent removal of the image from Gear (magazine) to be replaced by their logo, I'm broaching the topic of under what conditions is it acceptable to use a Non-free magazine cover. And so I pose two related questions:

1. Is it acceptable only if the particular issue is being discussed or are they allowed for the depiction of the magazine in general?
2. Does it make a difference if there is a free logo (i.e., {{PD-textlogo}}) available to be used for the magazine in general?

As it stands {{Non-free magazine cover}} explicitly states it is to be used "to illustrate the publication of the issue of the magazine in question". This was originally placed into the template back in August '05. This particular phrasing/narrow interpretation was challenged at least in June '06 and February '08 with no changes being made to the template. The arguments for changing the template and the general use of magazine covers generally revolve around the more flexible allowances in other similar templates, such as {{Non-free newspaper image}} which is allowed "to illustrate either the publication of the article or issue in question" and {{Non-free comic}} which allows it to illustrate "the periodical comic book series of which this issue is a part".

In the interest of full disclosure I will answer my own questions with the opinion that the use for depection of the magazine in general fails WP:NFCC#8, see in particular the corresponding guideline at WP:NFCI #1 which states that cover art is not suitable for identification without critical commentary. (I will also note that this portion of the guideline appears to contradict the established practice regarding at least book and album covers, and so that can be an extra-credit question: How to resolve this dichotomy?). For my second question I feel that it also (when lacking critical commentary) fails WP:NFCC#1 as the encyclopedic purpose of general magazine identification is served by the logo itself.

Your input is appreciated. VernoWhitney (talk) 19:17, 10 June 2010 (UTC)

The only time - outside of use for identifying the magazine - to use a magazine cover is if the cover itself is a subject of critical commentary. The only immediate case of this that comes to mind is the darkening used by some magazines in O. J. Simpson murder case. --MASEM (t) 19:41, 10 June 2010 (UTC)

I guess I wasn't quite clear enough. From what I can tell it's fairly well established that you can't use, for example, a magazine cover with a person on it just to have a picture of that person, barring situations such as the one you just mentioned. The question is more along the lines of can a particular magazine cover be used to identify the magazine at all? VernoWhitney (talk) 19:55, 10 June 2010 (UTC)

I want to thank VernoWhitney for raising this issue in such a considered and constructive way, especially since he was proposing an image deletion his attitude and good faith is particularly impressive. I find the image policies inconsistent and very difficult to understand.

VernoWhitney mentioned the question of critical commentary, which I feel exists for the cover of Gear magazine (it and the first cover image are discussed in the article) whereas the cover image of GQ featuring Ryan Gosling seems to have been an arbitrary choice, and the article makes no mention of the cover being notable. In contrast film posters and DVD covers do not seem to require any critical commentary at all and seem to be used purely for identification.

I do hope a more consistent and fair policy can be formulated and then applied widely and not selectively. I appreciate all efforts to clean up this mess even if it does not result in retention and it means many images need to be deleted, at least it will be fair and understandable going forward. -- Horkana (talk) 19:48, 10 June 2010 (UTC)

Oh, I misunderstood, but I will say this: I believe a logo for the magazine should be used before an arb. cover. This avoids several issues - we're using minimal use (it's usually the primary logo that is of interest , not the magazine layout), there's no issue with which issue to use, and sometimes it may be possible replace the cover with a free logo that fails the threshold of originality. --MASEM (t) 20:02, 10 June 2010 (UTC)

Existing practice seems to be all over the map. In some articles about a magazine a cover is used in the infobox (e.g. U.S. News, The New Yorker, and GQ), in others just the magazine's logo (e.g. TIME), some use an old PD cover (e.g. National Geographic), and in at least one (e.g. Vogue) no image is in the infobox but a cover is used in a section with commentary about the cover. —RP88 ^(talk) 20:13, 10 June 2010 (UTC)

Yeah, I'm hoping to get consensus to either a) go through and start picking out the ones magazines that use non-free covers only for identification or b) change the NFC guidelines and the template and make it clear that covers are allowed for identification regardless of subject matter (album/book/newspaper/magazine/etc.). VernoWhitney (talk) 20:17, 10 June 2010 (UTC)

Cover art is being used to identify the subject of articles (and no strict requirements that there be critical commentary about the art itself is needed). Whether this is correct or even advisable, I presume is your question. I know that if we suddenly say that cover art isn't OK, it is going to upset a lot of people (and then we are probably also going to have a lot of weird one line sentences added to articles like Journalist Mr. X said the cover art is "phenomenal". to try to add "critical commentary" in order to keep the precious images. As is, it's sort of like we have a working compromise, in that we allow one non-free image of cover art in the article about the work (whether it be a book, record, or magazine). Maybe a special case can be made for magazines, since they always have logos? But don't books have text-only title pages? And can't we just crop the artist and title text out of album covers or use the info on the spine? Slippery slope indeed.-Andrew c ^[talk] 21:29, 10 June 2010 (UTC)

I would say that only in the case of newspapers, magazines, and other periodicals which have a logo should we use the logo over a random cover, if one is able to isolate the isolate from the rest of the cover; if this isn't possible, a cover should be used instead. For normal books, the book cover is best. --MASEM (t) 23:23, 10 June 2010 (UTC)

Why not a title page instead of a book cover? I also think this creates an "unfair" situation where one magazine gets to show it's cover, but another does not. I think magazine cover layout is an interesting topic, or at least almost as interesting as book cover design and illustration. There is also cultural significance with "cover stories" and "covergirls" and so on. I mean, perhaps I am just playing devil's advocate, but if someone is asking us to work on drawing distinct lines in the sand, your suggestion seems a tad arbitrary. I'd be curious if there were an examples of a newspaper or magazine cover that would be required to be shown due to a lack of a logo. Do you have any examples in mind? Perhaps we could just do away with freebie magazine covers used only for identification of the subject of magazine articles. -Andrew c ^[talk] 00:05, 11 June 2010 (UTC)

If there's consensus for it, then I think that (no freebie magazine covers for identification) is the line in the sand I'd like to go for, since it's already supported by the template. (As far as I know every other template supports the use of a cover for identification.) VernoWhitney (talk) 01:10, 12 June 2010 (UTC)

The purpose is not just to show the logo, it is to show the whole way the magazine or newspaper presents itself on the news-stand -- something significantly more informative than just the logo. That is why it makes sense to show the whole cover. I think you will find that is almost universally the standard applied for newspapers. If there is an oversight in the "magazine" template then it should be corrected. It also serves to make the treatment of newspapers and magazines all of a piece with how we treat other media properties. Jheald (talk) 17:53, 13 June 2010 (UTC)

The newspaper template already allows for a cover image to identify the newspaper in general. I am aware that the magazine template is out of line with the other templates (and general use), but it appears to be the only one in line with the guideline at WP:NFCI #1. Am I correct in understanding that your stance is that both this template and the NFC guideline need be changed? VernoWhitney (talk) 18:51, 13 June 2010 (UTC)

I can understand that some newspaper/magazine/other periodicals will have a very identifiable look by the arrangement of elements on the page (WSJ and USA Today come to mind), but we always seek out free over non-free. Logos of many magazines and newspapers have a good likelihood of being free due to failing the threshold of originality (being just a font), and thus for identification of the work for WP's purpose, we should aim there. If the layout of the newspaper is commented on in the body, then I can see the use of a sample cover to show that. But for most of these, layouts of covers change issue to issue, and thus not significant. --MASEM (t) 19:02, 13 June 2010 (UTC)

NFC #1 says that a non-free item should not be used if it can be replaced by something that serves the same function. For the reasons I set out in my comment, the logo does not serve the same function as the cover. The informational value (and fair-use non-controversy) of showing such a cover is accepted for all media, and explicitly okayed in the guidelines. Magazine covers are no different.

As has been gone over endlessly, the issue is not whether something is "commented on in the body", it is that it adds to reader understanding. Yes, obviously each issue has a different cover; but the cover style of a particular title is something characteristic, which is why we inform our readers by showing it. Jheald (talk) 15:09, 14 June 2010 (UTC)

I don't disagree that we have an accepted standard of allowing some NFC as identification for a notable work (where there is critical commentary on the work, not necessarily the image), though have argued this is a present bane in dealing with discographies, etc. But that's not the issue. The consideration here is trying to reduce the amount of non-free work without harming the encyclopedia. Free >>> non-free as long as it serves the same purpose. The point you're arguing - that the visual aspects of the cover or front page layout of a magazine or newspaper is a critical factor - is what I take issue with, because for many of these newspapers or magazines, there's nothing special about the layout on the day-to-day/week-to-week periodical. The key identifying element nearly all the time is the logo - the thing that sticks up on the top of the magazine or newspaper to make it easy to find when searching a newsstand and what people are going to associate the work with. As many of these logos likely fail the threshold of originality, they can be used freely over a non-free cover without any loss to the reader. I am sure there are cases of the standard periodical cover design being iconic itself - and thus discussed in other sources - which would allow for a cover image to be used in the body of the article to demonstrate this, still leaving the magazine logo in the infobox and allowing for easy identification.

I will also point out that we don't have direct allowance (or disallownace) for magazine and newspaper covers even though they are presently used a lot though not consistently. We have two allowances that we can consider, #1 (Cover Art) and #2 (Logos). By "cover art" that implies creative arrangement of elements to make a work, as would be the case for many books, albums, and video game covers. I don't think this can easily be applied to a newspaper front page layout - its there for visual impact, and incorporates copyrighted images in a manner that attracts the eye but in a functional manner. I'm not saying with 100% it isn't art, but it is definitely less "art" than most other cover media. On the other hand, the Logo provision easily allows for such usage.

And as I write the above, I realize a third problem comes to mind. Here or on another page, there was a discussion about images from the flotilla raid that were press corp photos - images we specifically forbid using even under claims of fair use (because that claim is very weak w.r.t. commercial opportunities). If we use a newspaper or magazine cover that includes news agency photos somewhere within them, we run into the same problem - yes, the newspaper/magazine has the permission to republish them (we hope), but we do not. That makes including covers with these difficult.

We also add in the problem of people wanting to have a personal preference for which cover of a periodical to use. Newspapers aren't as a problem since rarely a single story takes up the whole page, but if you take a magazine that usually has a singular image for the front, you can imply bias one way or another if it is not carefully chosen.

I strongly recommend that we normalize all periodical works like newspapers and magazines to using logos to better meet the NFC. This does not disallow the use of a general cover image if the cover layout is significant and discussed, or of a specific cover image if that cover is discussed in detail relative to the newspaper/magazine. But we reduce non-free use without harming the work or reducing the understanding of an article by using logos over covers for all such periodicals. --MASEM (t) 15:42, 14 June 2010 (UTC)

If someone were to take a selection of magazines that you followed, blacked out their logos, and then showed them to you, I think you might be surprised at how readily you would be able to identify them, one from another. The point is not that the designs are "iconic", it is that they are characteristic. And that is why showing them adds something useful to reader informativeness.

This relates to "identification" too. If you look up all the various discussions there have been about WP:NFCI #1 over the years, you will find an underlying double rationale presented: (a) that this helps readers know that they have come to the right article; and (b) that it shows what the publishers have chosen for themselves to make their item identifiable. As the thought-experiment with the blacked-out covers shows, showing a characteristic cover layout does add to (a), but even more it squarely it informs the reader per (b) -- which is the key reason that we accept that, as a blanket class with no further questions requires, cover images satisfy NFCC #8.

The other side of why we accept cover images is that they are universally accepted as "fair use" -- even eg French Wikipedia allows them, operating under EU rules. In the real world this kind of use of this kind of outer matter is simply not controversial. Nobody will chase Wikipedia for using them; nobody will chase any of our downstream reusers; nobody will think anything bad of us for having used it. Use of a cover by other media to stand for and identify a magazine is simply part of the landscape. There's no NFC #2 issue, because the agency knows this is how covers are used -- it is already priced in by the agency knowingly licensing it for use as a cover image.

WP policies are designed to address practical realities. The bottom line here is that there is no practical benefit in our not using a cover in this situation -- not to us, not to anybody else. It would merely - completely unnecessarily - impoverish our readers. Jheald (talk) 16:50, 14 June 2010 (UTC)

I disagree with your thought experiment. You'd not only have to block the logos, but all headline text, unique photographs, and the like, and leave being only the graphic elements and general layout. Again, some are easily id'd by that (National Geographic comes to mind), but those are expectional. The logo is an indicator that just as clearly, if not moreso, assures the reader they are at the right page, that a cover image, particularly if the logo is buries due to the size of the image. (This, btw, is a very weak need. In paper, yes, when you are flipping through pages, that is a good argument for using a visual indicator of being on the right page, but not for electronic versions).

Also, this is not about fair use; covers as used as they are now certainly fall within fair use. NFC exceeds fair use to assure minimal non-free material is used to address the educational nature of the material. Unless the cover's layout has a clearly stated iconic aspect that is discussed, one is simply presuming the cover has iconic nature to use it over a simpler and likely free (due to lack of originality) logo. --MASEM (t) 17:29, 14 June 2010 (UTC)

The point is not whether the cover is iconic, it is whether it is characteristic. A cover of Time magazine, for instance, can immediately be told apart from a cover of The Economist, even with the logos blacked out; and both would readily be distinguished from the New Statesman or Atlantic Monthly etc. A cover of Architectural Digest, say, or BBC Gardens Illustrated, with their restrained rather formal style sets these titles apart from a more mid-market title such as BBC Gardeners World magazine. The cover of Radio Times conveys a different personality than TV Quick. And so on. Showing the readers a characteristic cover therefore adds distinctively to the informativeness of the article for them. As with other forms of cover art, we hold that this is sufficient to justify what is, given that the entire purpose of the cover is to make the product recognisable and identifiable, a very slight fair-use taking. Jheald (talk) 11:17, 16 June 2010 (UTC)

There's the catch. As the template and the underlying NFC are written we don't hold that this is sufficient to justify the use of magazine covers without critical commentary. VernoWhitney (talk) 12:27, 16 June 2010 (UTC)

Exactly - we cannot presume something's iconic or a key characteristic without a source. If there was no other way to provide an identifying image for a periodical, I see no problem with covers and thus allowing the implicitly iconic/characteristics without statement be used, but we have a way to use possibly free content; if there's no source to say anything about the iconic nature of the cover, then the free content (or reduction of non-free by using just the logo as opposed to the entire cover) replacing the cover image is not harming the reader's understanding of the article. If the cover is documented as iconic as I'm sure it is, this warrants an example of said cover in the body of the article, but for most periodicals this just isn't the game. --MASEM (t) 12:34, 16 June 2010 (UTC)

A number of points to clear up in the above. (1) At issue is not reader understanding of the article, it is reader understanding of the subject of the article. (2) The question of whether something adds or does not add to reader understanding is an editorial judgment to be made by editors, and if necessary the wider wiki community. It is not dependent on sourcing. (3) We are not just seeking an image for the sake of having an image; at stake is informing the reader. (4) It is not required that the cover is a "key characteristic" of the title; what is being claimed is that to show the characteristic cover-style shows something generally informative about the title, which more than justifies the (universally accepted in the real world) fair use of a cover image. Jheald (talk) 17:26, 16 June 2010 (UTC)

So if I skip past my objection that covers are allowed for so many others things, and then consider the statement "We cannot presume something is iconic" the problem is then what kind of critical commentary is required to justify including one cover and not another? We are left with inconsistent difficult to understand guidelines like we have now, where only a few articles include magazine covers, either by oversight or by expert Wiki editors who have fought to keep a notable cover image.

To give an example. The cover of Gear magazine that brought me here in the first place, I would argue is the most notable cover that magazine ever had, but we got here because although there was some commentary on it another editor didn't think it was notable enough. There is no clear line.

This all seems like justification for what we have now, and unfortunately doesn't make the situation any clearer or easier to understand for editors. -- Horkana (talk) 12:58, 16 June 2010 (UTC)

I'm saying that for periodicals - newspapers and magazines - the logo is the right thing to have in the infobox, and avoid the use of covers there. If the cover is iconic (as sourced by references) or other similar reason, it can be added in the body. For other works, which are not periodically published, cover art makes sense. --MASEM (t) 13:05, 16 June 2010 (UTC)

Okay I'll admit that is actually a little clearer: The default position is no covers for periodicals, you say. Still it will be difficult difficult to know what is the vague criteria of notable enough to include a cover. -- Horkana (talk) 14:27, 16 June 2010 (UTC)

Actually to be clear: right now, the default position is that a periodical cover is appropriate for identification in an infobox (Which is probably why there's a lack of consistency of this in actual articles). However, as a result of this conversation, I believe we should re-assess that position, and consider that nearly always, a logo is going to be able to serve the same purpose for most periodicals, and these types of logos are ones that would have a good chance of being free content due to lack of originality; thus by replacing the cover with logos, we better serve the free content mission. There are cases like Time where we would then have the logo and then likely a separate cover image in the body of the article to highly its iconic nature, but that cover image is rarely going to be necessary as most periodicals don't have such iconic imagery. --MASEM (t) 15:07, 16 June 2010 (UTC)

This should be a fun project. By my count there are 3,124 images tagged with the template which should be gone through, checked for usage, possibly modified to create a logo, and then possibly moved elsewhere in an article or removed entirely. VernoWhitney (talk) 15:29, 16 June 2010 (UTC)

Or alternatively, 3,124 images which fully satisfy NFC 1 to 10 and should be left well alone. Jheald (talk) 17:32, 16 June 2010 (UTC)

Yeah, which is why I want to make it clear this would need to be by-consensus change to NFC. Nor would it be a rush to replace them - it's not like we had "free" images suddenly tagged non-free and needed removing. --MASEM (t) 15:37, 16 June 2010 (UTC)

Clearly it needs consensus and should be done over a period of time as it isn't in keeping with standard practice up until now, but I'm not seeing that this requires a change to NFC, as I thought WP:NFCI #1 already covered it - it's just the part of the guideline which isn't followed by every other template or common usage. VernoWhitney (talk) 15:56, 16 June 2010 (UTC)

Note that WP:NFCI #1 establishes that it is sufficient that a cover is being used for identification for it to be acceptable; it doesn't say that it is necessary that the cover is being used for identification, nor that identification is the only justification for use. It is simply that identification is a particularly weak justification for use, but NFCI #1 underlines that even this is acceptable. Jheald (talk) 20:03, 16 June 2010 (UTC)

Except for the fact that it says "not for identification without critical commentary" (emphasis added). VernoWhitney (talk) 20:08, 16 June 2010 (UTC)

Look at the wording again. It specifically addresses this. There needs to be critical commentary on the underlying thing -- the magazine in this case -- not necessarily on the cover itself. Jheald (talk) 20:17, 16 June 2010 (UTC)

I've read it, it says "that item" (i.e., the issue of the magazine), not "the series of which that item is a part" (i.e., the periodical). VernoWhitney (talk) 20:22, 16 June 2010 (UTC)

I wrote it. The intention was never to be as narrow as you make out. "That item" could equally well be the magazine in general. As Masem points out, that is the customary understanding; and policy is supposed to follow customary understanding, not the other way around. Jheald (talk) 20:44, 16 June 2010 (UTC)

I don't see that it was ever made clear in the discussion leading up to your change, and if it was perfectly clear then the template would've been updated and I likely would've never started this conversation. Precision matters. VernoWhitney (talk) 21:55, 16 June 2010 (UTC)

Precision matters, yes. But unlike Anglo-Saxon statutory construction, WP policy interpretation is not predominantly based on microscopic lexical analysis. It's recognised that policy drafting is a work in progress, a groping towards expressing in words how we do things, rather than revealed truth engraved in stone. Hence the well-worn phrase "policy follows consensus", rather than the other way round. Clear consensus is needed to overturn customary understanding. Jheald (talk) 22:22, 16 June 2010 (UTC)

I understand that, and I would be fine if the consensus is to clarify the NFC guidelines and the template to firmly establish that covers are allowed for identification regardless of subject matter, so long as the policy/guideline is clarified. VernoWhitney (talk) 22:31, 16 June 2010 (UTC)

(edit conflict) Now that's getting into the part that's subjective no matter how you look at it, which is how much commentary is sufficient to pass WP:NFCC#8 (and this applies to every non-free image). I personally look for multiple sources which allow at least a couple of sentences describing the image (such as for Brandi Chastain for Gear where three sources are cited). VernoWhitney (talk) 15:14, 16 June 2010 (UTC)

The answer is very simple. No commentary is required to pass NFCC #8. The test is whether the image sufficiently adds to reader understanding (of the subject of the article). Jheald (talk) 17:16, 16 June 2010 (UTC)

Apparently we'll have to agree to disagree about that one, but it's getting off topic anyways. VernoWhitney (talk) 17:41, 16 June 2010 (UTC)

Check the archives. It's been discussed often enough. Jheald (talk) 20:03, 16 June 2010 (UTC)

Jheald's right on this aspect - the present is acceptance of non-free imagery in the infobox or lede about the work being discussed in the page is acceptable (with some cavaets like free replacement, BLPs, etc.). The counter-example is the use of non-free in discographies or the like where there's no discussion of individual elements of the list/table.

To be clear, I am not proposing any NFCC changes. I am proposing discouraging the use of "cover art" for periodical works in favor of the logo for identification purposes (in infobox or lede). --MASEM (t)

As I read it using non-free imagery in the infobox is acceptable because the article itself provides the critical commentary. VernoWhitney (talk) 21:19, 16 June 2010 (UTC)

But if we can replace it with something free that serves the same purpose (and in the infobox, it is likely only for identification), then clearly we need to do that. Even if the logo is possibly non-free, just an image of a logo is less a problem than a cover with logo and several other creative elements. --MASEM (t) 21:21, 16 June 2010 (UTC)

Even if I'm wrong on the critical commentary part, we agree there. VernoWhitney (talk) 21:56, 16 June 2010 (UTC)

My contention is that it does not serve only for identification (in the narrow sense of reader recognition). The whole cover is also valuably informative in conveying how the magazine presents itself (as per the examples I gave above), and that also needs to be taken into account. Jheald (talk) 22:03, 16 June 2010 (UTC)

And that would be part of the community consensus decision to determine if this is truly the case. Myself, I can see this being true for a small number of magazines but not the general majority, and thus would rather see a logo in the infobox, and in the exceptional cases, the cover image used in the body of the magazine next to text to describe the icon nature of that cover style. But that's myself. --MASEM (t) 22:26, 16 June 2010 (UTC)

Given the lack of recent replies and the failure to reach a clear consensus here, would an RfC be the next step? VernoWhitney (talk) 13:30, 22 June 2010 (UTC)