Commons talk:Photographs of identifiable people

Archives 1

Image of an overweight child

Would the child in this picture be deemed identifiable? And should this image be used on childhood obesity? Jmh649 (talk) 18:54, 19 September 2010 (UTC)

Yes, in either case identifiable. -- smial (talk) 19:04, 19 September 2010 (UTC)

Yes it should be used or no its should not be used as identifiable? --Jmh649 (talk) 20:22, 19 September 2010 (UTC)

In Germany it could not be used without permission, in case of children permission by the parents. Other countries I don't know. I do not like the way personality rights are dealt with at commons, but i must live with that. On the other hand this is a question of fairness... -- smial (talk) 22:31, 19 September 2010 (UTC)

If you were going to use a contemporary photo for a topic like that I would strongly recommend getting subject's parent's permission. Probably easier to get some adult who'd let you use their own childhood photo for the purpose. Better yet, I'd consider uploading an image where the face is not at all identifiable (possibly by way of a Gaussian blur). Blurring this photo wouldn't do, because the original is already published here and hence identifiable. - Jmabel ! talk 03:24, 4 November 2010 (UTC)

I pixelated the face. cheers, Amada44  ^{talk to me} 16:44, 4 November 2010 (UTC)

Hey, it's not that he's overweight - his pants are just too tight. Give the kid some slack. Wnt (talk) 05:53, 11 January 2011 (UTC)

Re-use of the image, and reassuring image donors

I think the 'Re-use of the image' section should further emphasize that putting something under the public domain or a CC license does not open the door for abuses of the right of publicity or defamation off-wiki. It is important for image donors to know that these acts are just as illegal as they would be otherwise, no matter the copyright status of the image, and that punishment can be enforced even on, for example, abusive re-uses of public domain US government images.--Pharos (talk) 20:43, 3 November 2010 (UTC)

This is exactly what the {{Personality rights}} tag is for. Dcoetzee (talk) 04:49, 4 November 2010 (UTC)

Well, yes. The idea is that this should be more explicit and further explained on the Commons guideline page as well.--Pharos (talk) 06:45, 4 November 2010 (UTC)

Oh you meant the section on this page. I guess that couldn't hurt, although personality rights are also extremely complex, varying from nation to nation and even within nations widely, so we should avoid making bold sweeping statements that aren't true everywhere. Dcoetzee (talk) 07:42, 4 November 2010 (UTC)

"people openly taking photographs" transform a private space into a public space???

Apropos this. So it's OK for us to host the pix if a bunch of paparazzi bust in on some poor sod doing his business in a a locked privy? ...on a couple getting busy in a locked bedroom? Since (once the door is open) they'd be 'openly taking photographs'? Sems a novel theory that "people openly taking photographs" transform a private space into a public space. Thoughts? Considering re-doing my undone edit. --Elvey (talk) 10:34, 6 December 2010 (UTC)

I didn't understand your edit either and nearly undid it myself, but was going to wait for a day. House rules on photography have basically no effect on privacy, the best I can tell, which your edit seems to imply. I believe it is "expectation of privacy"... in a bathroom or bedroom, absolutely that exists, but in most other places in the company of multiple people you don't know (i.e. not close friends and family) then there is usually no expectation of privacy. There is certainly no expectation of privacy in many places where photography is restricted for other reasons. The specific example your edit was made to is a "large private party or concert", not one of the types of places you mention above, and house rules on photography in those specific examples do not create an expectation of privacy, so your edit would seem to be at least misleading, and if you intended it to apply to a concert or similar setting, then incorrect. Carl Lindberg (talk) 16:03, 6 December 2010 (UTC)

I've never seen a bathroom with a "No photography" sign before; there is generally nothing restricting photography in a bedroom or bathroom, except for that expectation of privacy. (Which I might acknowledge sans people; poking around someone else's home bathroom can be pretty darn intrusive.) But the way your change was inserted, it read to me to be about restrictions printed on tickets, and I think people completely understand that no matter what that says, even if they bother to read it, that's not going to stop pictures of them showing up on Facebook, and people are going to think they're silly or weird if they object to it. You have no social expectations of privacy there.--Prosfilaes (talk) 19:00, 6 December 2010 (UTC)

I think that the first draft better caught the point that photography is permitted. The "large numbers" part also strikes me as odd. Plenty of times there is an event ranging from a Christmas party to a meeting of political volunteers in which only one person has a camera, but the general tone generally encourages that person to take photos. Wnt (talk) 05:57, 11 January 2011 (UTC)

It's about an expectation of privacy. More photographers tends to lower the expectation of privacy. Anyone think it's OK for us to host the pix if a bunch of paparazzi bust in on some poor sod doing his business in a a locked privy? ...or a couple getting busy in a locked bedroom? Since (once the door is open) they'd be 'openly taking photographs'? No, then our policy should reflect that.--Elvey (talk) 21:50, 4 February 2011 (UTC)

No bathroom or bedroom I've ever seen has a sign saying no photography. Please, please stop conflating the signage issue with bathrooms and locked bedrooms.--Prosfilaes (talk) 22:13, 4 February 2011 (UTC)

Prosafiles, Please stop beating your wife. In other words, I have no idea what you're talking about in your accusation. What if the photographers are all friends and family? or all employees? This page proves lots of folks see 'the signs; that's a helluva lot of images, so your edit summary claim is clearly false. --Elvey (talk) 17:23, 5 February 2011 (UTC)

You conflate signs and bathrooms. The two don't go together. I don't know what you're claiming about signs in your link; this is the first time you mentioned "there will be recording" signs.--Prosfilaes (talk) 18:26, 5 February 2011 (UTC)

And stop beating the strawman. The question is not about bathrooms or bedrooms. No one has advocated permitting photograph of people so engaged, and no matter what this page says, they won't pass DR. Those aren't the real questions here.--Prosfilaes (talk) 18:29, 5 February 2011 (UTC)

AGAIN,what the **** are you talking about? The fact that "many people are openly taking photographs" does not ensure that there's no expectation of privacy. Anyone who thinks so is not thinking clearly. Because that's true, as I'm sure you agree, we need to find some acceptable wording that's not false and misleading. FFS! P.S. your recent edit summaries suck.

It generally does; the fact that someone or ones can break into a sealed room and take pictures of acts that would never have commenced if there were to have been photography in the first place is irrelevant.--Prosfilaes (talk) 21:37, 5 February 2011 (UTC)

It almost always does. And the fact that people are not taking photographs also does not mean you have an "expectation of privacy" either -- it is just one factor. For example, in a museum where there are "no photography" rules, that does not create an expectation of privacy. Bathrooms, bedrooms, sure -- but outside of that, if you are in the company of people you don't know, there is usually no expectation, regardless if they are or are not taking photographs. But, people taking photographs is a pretty strong indication that you *will* be photographed, so act accordingly -- the photographs would hardly be a surprise in that case. If it's only family, then it may exist -- hard to fathom a place of work being a place where you would expect privacy, outside of bathrooms, dressing rooms, and the like. Carl Lindberg (talk) 22:43, 5 February 2011 (UTC)

Voice recordings

{{Personality rights}} links to Personality rights, which mentions commercial exploitation of voice. Should that be mentioned in sound files' descriptions, and how? It is not clear to me if the text in {{Personality rights}} is applicable. --AVRS (talk) 16:16, 18 April 2011 (UTC)

A bit unclear on Consent of the subject?

In it's current form, the has the following text:

Consent of the subject (who is a non-public figure) is required even for photographs taken in public places in the following countries:

• China
• Czech Republic
• Germany
• Libya

(incomplete list)

The list was originally started by User:Docu (diff) with edit summary see talk. Later User:Isderion added Germany to the list (diff). China was then added (diff) by User:Jameslwoodward. Since Libya has also followed. For me - as a non-native English speaker - it is a bit unclear what exactly the consent is for. For China it's easy - User:Jameslwoodward even links to Commons:Licensing#People.27s_Republic_of_China which states it is for publishing the photo. For Germany you can photograph people in public without their permission if they are in a group of people, but if you pick someone out, you have to ask the person, before it's legal to take the photo (as I have been explained privacy laws in Germany by a German back in April - this may be wrong?). For the Czech Republic I have no idea and let's leave Libya out of the question for now. My question thus is: What consent is this list for? The taking of the photo, the publishing of the photo or the commercial reuse of the photo? Because I think that should be more clear. --Henrik (heb: Talk · Contributions · E-mail) 06:56, 26 May 2011 (UTC)

On a note (this should have been included above), I'm impression that User:Docu's reference to the talk-page is to Indentifying children by name, which may give the intentions of the original list, but from my perspective doesn't give a clear picture. --Henrik (heb: Talk · Contributions · E-mail) 07:03, 26 May 2011 (UTC)

As per de:Recht am eigenen Bild Austria and Switzerland have similar laws and should be added to the list. As of my knowledge a permission is required for any (public) use of such an image. Commercial use in the sense of use for advertisement/promotion does always require written permission, likely in all countries. --Túrelio (talk) 07:14, 26 May 2011 (UTC)

Commons is not an encyclopedia -- WP might be interested in the question, but as a practical matter, Commons doesn't care which rule applies -- any of the three rules stated above would prevent us from using an image.

If I may clarify Túrelio's comment -- if Coca-Cola wanted to publish an ad with a photo of me with a caption saying, "Jim Woodward likes Diet-Coke", it would require my consent no matter where the image had been taken. On the other hand, an image of me looking at a statue could be used in a calendar or on a commercial website without my consent unless it had been taken in one of the named countries.      Jim . . . . Jameslwoodward (talk to me) 10:32, 26 May 2011 (UTC)

On the same point, the summary advice included in the tutorial (see Commons:Usability_issues_and_ideas/Archive_3#m:Licensing_tutorial.2Fen) is somewhat misleading as it appears to apply to the US primarily and is rather bad advice if you are in one of the countries mentioned. --  Docu  at 11:25, 26 May 2011 (UTC)

There is a U.S. slant to all this, as if the Foundation could be liable in the United States for simply hosting particular images, that is a major concern and can be a reason for deletion. Differing privacy rights in other countries generally do not apply outside those countries, and it becomes more a matter of Commons policy and the opinions of users here if we should remove images on those grounds. Often it is possible to blur parts without actually deleting them, and for "notable" people they are often public figures anyways with lesser privacy protection for this type of thing. Anyways, the page is describing possible issues which may surround re-use of the images here. Usage in the listed countries may well require additional permissions which Commons does not have, whereas usage in other countries may not. The list being referred to is about privacy rights, so it is more about the publishing and not the taking, I'm pretty sure. Carl Lindberg (talk) 17:08, 26 May 2011 (UTC)

Are you suggesting that the privacy rules in the countries listed concern the reuse of images already on the Commons and not to the uploading of the images on to the Commons in the first place? If so, I am not sure what you say is entirely accurate. The Libyan position, for example, appears to be that "[a] photographer may not show, publish or distribute a photograph unless the people depicted in the photograph have consented, unless the photograph is of a public event or of officials or persons enjoying public renown, or the public authorities have given permission for its publication for the general welfare. Notwithstanding the preceding, no photograph may be shown or circulated if doing so would result in detriment to the honour, reputation or social standing of the person depicted in the photograph." (My paraphrase; emphasis added.) It seems to me that uploading a photograph taken at a private event without the consent of non-famous persons shown would amount to circulating, distributing, publishing or showing the photograph in question.

I think this guideline is due for a rewrite as it does not really provide sufficient guidance to users as to the extent privacy laws of countries outside the US are to be respected. I note that such laws are really non-copyright restrictions, so the fact that we have this guideline is an exception to the general policy against applying the non-copyright restrictions of countries outside the US. Do we apply such laws with their full rigour, or only to the extent of privacy law in the US? Here is an example. Let's say the law of Ruritania simply states that no photograph may be distributed or published without the express consent of every person depicted. On the other hand, under US law (drastically simplified) such photographs can be distributed or published if taken in a public setting. Should the guideline be that the laws of both Ruritania and US must be fully complied with (in the same way as we require photographs to be in the public domain in both their source countries and in the US), which means that the more restrictive law will govern, or is it sufficient if the privacy law standards of the US are satisfied? — Cheers, JackLee ^–talk– 02:51, 28 May 2011 (UTC)

We don't need to refer to Ruritania, take Germany, which is quantitatively more relevant. Law of Germany clearly forbids the publication of images of non-public figures who haven't consented expressedly or at least indirectly to publication; for details see de:Recht am eigenen Bild (in German). This right is indeed enforced in Germany and usually the unwillingly depicted person does win the legal suit. In case of a violation likely we (the hoster) as well as a re-user is liable. To respect this restriction is rather easy with original uploads to Commons. However, it's less clear with images of already published on Flickr or like. --Túrelio (talk) 08:08, 28 May 2011 (UTC)

Ruritania is a fictitious country that I was just using as an example. — Cheers, JackLee ^–talk– 14:51, 28 May 2011 (UTC)

Do we apply such laws with their full rigour, or only to the extent of privacy law in the US? We are only absolutely bound to follow US laws, since that is where the Foundation can be sued. Others are more of a community decision... there are lots of national laws which we ignore due to their potential adverse affect on Wikipedia or other projects (Germany law on Nazi symbols, UK "perpetual copyright" on the King James bible, and others I'm sure). In general, we do try to follow the privacy laws I think, but if certain countries have certain aspects which we think go a bit too far, we may decide to ignore them. For example, if Gadaffi tried to claim that every single photograph of him is a detriment to his honor, would we automatically take them down? Doubt it. I can't find it at the moment, but I remember the only photo we had of someone from Europe, taken in public when he visited Canada, was requested to be taken down under their interpretation of the law in his country, and it was denied. Most of the time though, we should be able to respect the privacy laws, I'd think. We can blur faces of people incidental to the point of the photo, and that sort of thing -- there are usually a lot of ways short of deletion we can try to conform to privacy issues, and I think we usually try. We could split out a country-by-country summary of privacy laws to another page; this page should try to document the policies we actually follow, and starting to add every single aspect of all privacy laws may have the impression that we are changing policy to match. Carl Lindberg (talk) 14:13, 28 May 2011 (UTC)

I accept that the Foundation is only required to comply with US laws, but I think we really need to provide more guidance than simply "we do try to follow the privacy laws [of other countries]..., but if certain countries have certain aspects which we think go a bit too far, we may decide to ignore them". Can we not try and achieve consensus on precisely what situations we will regard as violating privacy and which ones we will not? If we are not rigorously following the privacy laws of countries other than the US, then we should not have a page listing those privacy laws as that will just cause confusion to volunteers. — Cheers, JackLee ^–talk– 14:51, 28 May 2011 (UTC)

Without seeing the laws in particular, it is hard to make a general case. It is entirely possible that a country may try to break new ground and add aspects that we feel go too far; perhaps we will not follow those. This stuff can be quite tangled; for example see this -- a Frenchman won a lawsuit based on the image right in France against a publishing house; eventually it went to European court where they ruled France had violated European law with that judgement (it unfairly interfered with the publishing house's right of free expression). In general, these rights do not automatically override rights of free expression; they are usually balanced against each other (even in the countries which have these laws) -- it is all part of what is "illegal". Treatment here may be somewhat similar; we are probably going to balance our educational mission and the photographer's rights against them. Photos taken in public are going to be fairly contentions (see the battles over File:HomelessParis 7032101.jpg). That said, the policy has said we do follow local law (regarding identifiable people anyways). I don't like the list in the first section, to be honest. It ignores a lot of subtleties and exceptions in those laws, which may be very relevant. The rest of the text seems to more illustrate the policy as I have understood it. But yes, those local privacy laws may very much affect whether we keep an image, if it is felt it is illegal in the country it was taken in. So I think we should try and document them somewhere, be that on this page, or a page dedicated more to privacy rights (this policy covers more than just that; we should be careful about defamation etc. in the titles or descriptions of images too). Carl Lindberg (talk) 16:19, 28 May 2011 (UTC)

From my perspective (and somewhat in connection with my original question) I think that the best would be, to try to create a sort of "guide to privacy rights for countries". That said - since it's only a very few of Commons contributors who are actually lawyers - there is a danger, that someone might think it a sort of definite presentation of regulations. The background for my initial question is that recently I was made aware of the Danish legal stance (which I think very few danes are aware of):

The predominant point of reference, is that any publication of a portrait photograph requires consent [of the person depicted]. The reasoning for this, is that such a publication might provide the depicted person with discomfort, possibly with other information such as name, of the pulication for all with access to the internet, and the considerations of this discomfort is judged as more important than a possible interrest in publication.

A portrait photograph is defined as a photograph, with the pourpose of depicting one or more specific person(s).

My intention was to add Denmark to the list, but I couldn't quite figure it out. --Henrik (heb: Talk · Contributions · E-mail) 15:51, 29 May 2011 (UTC)

┌─────────────────────────────────┘
I think Clindberg's point is that we should not be creating a summary of all the privacy laws of different countries (like we do for copyright laws at "Commons:Licensing"), because it is not our intention to comply strictly with all of such laws. Instead we should just develop a Commons policy on the matter which may be more liberal than the privacy laws of some countries. — Cheers, JackLee ^–talk– 16:23, 29 May 2011 (UTC)

No, I think it would be useful to have such a page. I would think we can follow them most/all of the time, but there may be situations where there are other considerations. It would help to inform any such discussions, for sure. I don't like the bare list in the first section here, as it seems a bit simplistic to say "photos of people in public are not allowed", which is what that was -- I think it's quite a bit more nuanced than that. There are indeed "portrait rights" in some countries, which are a little different than privacy even, and we may take those into account as well (but we may decide that attempts to deny us *all* photographs of a person may be too much, for example). Those have been case-by-case thus far, and having a better grasp of local laws (and thus what rights people in those countries are accustomed to) would be very helpful. Carl Lindberg (talk) 18:04, 29 May 2011 (UTC)

Sorry, must have misunderstood. My view, expressed above, is that such a page is not going to be very useful unless we indicate exactly which laws we are not going to require editors to comply with. Otherwise, the page will just serve to confuse people. ("According to the page, Ruritanian law requires this, that and the other thing! Why are we not complying with it?") — Cheers, JackLee ^–talk– 15:28, 31 May 2011 (UTC)

That's a fair point... but I guess mine is, it is entirely possible that particular countries have a version which goes into territory that other countries don't, possibly in conflict with the goal of supporting the other projects, and if so it may become a community decision on how best to balance them. It's hard to have an absolute policy when we don't know what the laws actually are. The policy was indeed written from a common-law (US, UK, Canada etc.) take on privacy rights, which was reinforced by the Foundation board resolution (emphasis on pictures taken in private places; no guidance on pictures taken in public). Civil law countries often do have different boundaries (again, not specifically mentioned in that resolution). The guideline does state that we do try to follow these variants as well. But, Europe seems to have a tradition of various forms of rights such as these, and when they come into conflict, the rights must be balanced equally against each other -- we probably have that same situation here. If it is at all possible to conform to the local laws by blurring portions, changing file names or descriptions etc. (i.e. anything short of deletion) we should absolutely do it. And often, deletion may well be the most appropriate way to handle it, even if it does harm the projects. Also, not responding to a specific request from a person just because their country's laws allow it isn't really a good idea -- we should weigh the merits requests from people pictured in public in all countries, I think. But, the right of free expression (and the educational purpose) should be balanced, in my opinion -- there are possibly situations where following the letter of the law to the very end may seem to be too much. Knowing the laws, and the particulars and nuances, can very much help to inform those cases. The Germany case, as far as I can tell, is fairly strict -- but there are several exceptions. One is obviously that "notable" people have less expected privacy, so the rules don't apply as much to them, as was mentioned. Another is people actually attending an event; the privacy rights are much less for photos taken of the event. I think there are few others as well. I think, really, that the more stringent privacy rights like that usually come with a number of exceptions and other factors to think about, and is not always clear-cut. It is hard to get that across in a simple list. Carl Lindberg (talk) 14:54, 3 June 2011 (UTC)

@Henrik: if the situation you mention is correct for Denmark, you should add Denmark to the list. If you look at the contributors of prolific uploaders from other countries listed (e.g. Germany mentioned by Turelio), you will notice that there are hardly any portraits taken in public places of random non-public figures. --  Docu  at 02:39, 2 June 2011 (UTC)

I've just talked this through with another Dane - who is a bit more skilled in the legal aspects than me - and Denmark should be on the list for the above reasons. I will add it and clarify the introduction to the list a bit. --Henrik (heb: Talk · Contributions · E-mail) 06:06, 3 June 2011 (UTC)

FYI: just in case anybody here wasn't already aware of it, as of May 29 the WMF Board unanimously approved a Resolution:Images of identifiable people. --Túrelio (talk) 08:12, 2 June 2011 (UTC)

I have now expanded the list a bit and given it a bit more descriptive introduction (diff). The location and/or form is certainly open for changes, but it should serve as a starting point. --Henrik (heb: Talk · Contributions · E-mail) 06:26, 3 June 2011 (UTC)

Cool, that is a definite improvement. A few things though. First, "commercial use" is {{personality rights}}, and is not a reason for deletion -- that needs to be made clear (or maybe we just remove those parts to eliminate confusion, as there is no policy to delete images where that is the only issue). In this context, "commercial use" is completely unrelated to the term "commercial use" as used in copyright, and so long as Wikimedia is not using it commercially there is no issue. We are focused on images which are not legal to publish. The Germany section seems under-specified; that does not mention the fact that public figures have considerably less protection here, and also does not mention that photos taken at events the subjects are attending also do not really get this protection. The Danish section is indeed interesting. Do we feel that means that a) we need to proactively delete all portraits of Danish people (taken in Denmark anyways) unless we have OTRS permission, or does it mean that we will automatically respect any request from that person to delete it, or c) give strong weight to that law, but examine each case individually? The European Court of Justice decision I mentioned earlier states that rights of free expression must be balanced against these type of rights. I guess I'm of the opinion that we should not automatically delete photos taken in public places in most cases, unless there seems to be risk of unfairly disparaging the subject, but we should also strongly weigh these laws when deletion requests from the pictured person come in for photos taken in such countries. Carl Lindberg (talk) 15:10, 3 June 2011 (UTC)

It's a good thing to detail the list, but we still need to introduce it at the beginning of the guideline. --  Docu  at 18:27, 5 June 2011 (UTC)

┌─────────────────────────────────┘
Could we add a summary of the Dutch copyright law as it applies to publication of a person's image? So far I have found COMMERCIAL APPROPRIATION OF A PERSON'S IMAGE page 8 of which has "The Auteurswet protects the individual against unauthorised publication of his or her portrait" and "Section 21 of the Auteurswet provides that publication of the portrait is not authorised in so far as the subject, or after demise of the subject, one of his or her surviving dependants, has a reasonable interest in opposing publication." Auteurswet is copyright. The work compares the law in several jurisdictions, for example page 12 has "Alabama, as in various other jurisdictions in the United States, the right to the use of a person's image is protected under the tort of invasion of privacy." -84user (talk) 19:18, 9 June 2011 (UTC)

I kind of like the idea of a page that lists country-by-country rules, as it may be helpful to both uploaders and re-users. However, I'm not convinced that it needs to be part of this page, as part of an "official guideline". Perhaps we should split it off, with a ==See also== to it. WhatamIdoing (talk) 17:41, 10 June 2011 (UTC)

I'm not sure if the usefulness of the guideline should be limited to uploaders in the US. For user publishing images from elsewhere, e.g. (frequent sample in previous comments) from Germany, it's a fairly important part of the guideline and needs to be mentioned specifically. That said, the layout of the list could probably be done differently. --  Docu  at 07:21, 11 June 2011 (UTC)

If the bulk of problematic uploads would come from just a few countries, I'd agree it would be worth briefly outlining the relevant portions of those countries' laws within the guideline. Listing full details for several countries would be overkill IMO, with the detail overwhelming the main message. I think we should already be cutting back on the detail, especially for countries with a smaller presence on Commons (e.g. Libya). A table might be a good way to summarise the information here, with a link to a page listing the full details. --Avenue (talk) 11:27, 11 June 2011 (UTC)

Commons guidelines tend to be fairly specific for each country. Have a look at Commons:Licensing with its #Country-specific_laws part. Freedom of panorama would be pointless without specifics for countries. --  Docu  at 06:42, 19 June 2011 (UTC)

What does "identifiable" mean. Perhaps we need two definitions.

This guideline does not define what "identifiable" means. I found a proposal page here that does contain a definition. I haven't investigated the history to see if that is an old version of this guideline or something that never was. However, it defines "identifiable" as "might be recognised in public by a stranger on the basis of the photograph", typically meaning the face or some very distinctive feature is included, and it "is not enough that the subject is able to recognise him or herself, nor that an intimate partner might be able to do so". I think this guideline must have a definition of "identifiable" and that the above might be reasonable for the legal aspects of the guideline but not necessarily for the moral aspects.

When the WP Medical Project last looked into the issue of medical photographs here and here (which are typically of patients), we found that leading medical journals take the view that all patient photographs were "identifiable". Their definition of "identifiable" would appear to include identification by the person, family members and friends. The use of a black band over they eyes is nowadays regarded as totally ineffective at making the image unidentifiable. I would argue that the pixelation of the obese girl in this guideline is similarly infeffective.

I propose that we consider these two definitions of "identifiable". How do these definitions fit with

• The type of image
• The name and classification of the image on Commons
• The use of the image in Wikipedia and other users of Commons images.

My current view is that the File:Childhood Obesity.JPG image should fail this guideline when suitably amended. It, by its name and classification is a medical photograph, and as such should come under the stricter definition of "identifiable" that includes identification by the person and people who know the person well. We should also consider disallowing photographs that make use of inadequate privacy measures like black bands on the eyes and pixelation. Even if the image was renamed and reclassified to be merely "Girl eating ice cream.jpg", its use in an article on Wikinews or Wikipedia on obesity should also fail this guideline.

The issues of consent wrt patient images is also worthy of discussion and note here I propose we discuss that later and stick to considering "identifiable".

Colin (talk) 08:10, 9 June 2011 (UTC)

For the courts in Germany, as far as I know (IANAL), identifiability means that someone who knows the depicted person could identify her/him in an image, even if the face of the person is not shown. For details see here (in German). --Túrelio (talk) 08:17, 9 June 2011 (UTC)

IMO the first definition is much more reasonable, because I do not believe that a picture of two square inches of skin is "identifiable" (and the medical journals apparently do).

There are two ways of looking at the act of 'identification': We could worry about a stranger seeing the image online, bumping into the person at the bus stop, and saying "Oh, you're the guy whose picture I just saw on Commons!" Alternatively, we could worry about a user seeing the picture online, and saying "Hey, that's Joe!"

IMO the first case is the important one. Joe isn't normally hurt by being identified by people who already know him (unless, say, the person who knows him is his parole officer, and Joe's robbing a bank in the image). But the first case might be awkward for the subject. Thus I support the "might be recognised in public by a stranger on the basis of the photograph" version, not the "my best friend might realize that's me" definition.

Also, I think that we should address the issue of revoking consent. A person might give consent, and then change his mind later (e.g., if the image becomes such a sensation that he's stopped every time he leaves his home, or if parents gave consent for publication of an embarrassing image of their child, who has since become an adult and wants it removed). There are practical limits to what we can do, but I think we ought to offer at least limited revocation: We'll take it off Commons, but we can't help the fact that other people have copied it in the meantime. WhatamIdoing (talk) 17:54, 10 June 2011 (UTC)

Permission is a big issue, which includes revoking consent. I'd like to deal with that but in another section. I want to concentrate on what determines "identifiable" for now. I'm comfortable with a "harmless" photo being judged on the stranger-identifies-you threshold. I don't think this is adequate where the picture is potentially hurtful or embarrassing. It is also inadequate IMO where specific permission has been given but breached on the grounds that the image is not identifiable to strangers. Here's two examples:

• Someone takes a picture of your overweight daughter at McDonalds without your permission or even knowledge. Some short time later she goes to her school and the class go to Wikipedia to look stuff up. Meanwhile "Obesity" has become a featured article and is on the main page with your daughter as the lead image. The entire class clearly see your daughter held up as an example of obesity to all of Wikipedia's 8 million readers. They recognise her hair, her skin colour, her favourite t-shirt, her jeans and even the McDonalds they all go to. When the squint their eyes, it is even more obviously her. You can work out what sort of teasing will follow.
• Your father goes to his GP with an irregular mole on his cheek that he's had a while but finally decides to get investigated. He's referred to a specialist who diagnoses cancer and wants to take a picture of it. He produces a form asking for permission to use the picture for educational purposes, which your father agrees to. The specialist uploads the photo to Commons. Commons does not restrict use to "educational purposes" so his permission isn't valid but the specialist has cropped the photo to show only the face below the eyes and thinks it is no longer "identifiable" so this guideline does not apply. You see your father's face on Wikipedia when reading about his type of cancer. His irregular mole, his nose and mouth are all obvious to you. You help him take the specialist to a tribunal for professional misconduct.

The medical journals can play safe because they expect patient photographs to be taken these days with permission forms filled in. This guideline is trying to set the grounds for where permission is not required. Colin (talk) 19:46, 10 June 2011 (UTC)

In the second example, I see that the photographer has breached his agreement with the subject, but that seems to raise broader issues than just identifiability. What if someone took photos in a museum against house rules, and posted them here? I think that's much the same issue. --Avenue (talk) 22:35, 10 June 2011 (UTC)

Permission and consent are related and you are right that they apply to inanimate objects too as private property comes with the owner's restrictions on whether photographs can be taken and for what purpose. Could someone point out where Commons has a guideline on permission? I can only find stuff on copyright. I do want to discuss permission, whether here or elsewhere, but I want to clear up the "identifiable" issue first. It seems this guideline is trying to cover the "whether you need permission/consent" aspect. But it states up front that it does "not apply to photographs where the subject is unidentifiable" yet offers no definition of unidentifiable.

I agree the photographer has breached his agreement, but he could claim he didn't need to ask permission anyway because the person is not identifiable to strangers. I don't think that is acceptable. Colin (talk) 13:42, 11 June 2011 (UTC)

I think that the first example is covered by #Moral issues, "Those that unfairly demean or ridicule the subject" and "Those that unreasonably intrude into the subject's private or family life". Declaring a girl to be obese is demeans the girl and invades her privacy. Such a photograph (or, more precisely, such a use of that photograph) is unacceptable unless the person has specifically given permission for it.

I don't think that the second case is covered here (although it is likely a breach of the contract). Also, it's something that could likely be improved by cropping the picture even further, to focus more on the mole. WhatamIdoing (talk) 22:18, 14 June 2011 (UTC)

I don't want to derail this discussion, but I just stumbled across the essay Commons:Patient images, which seems somewhat relevant. It deals more with consent than identifiability, but does appear to advocate the "black band across the eyes" treatment (among others). --Avenue (talk) 02:43, 15 June 2011 (UTC)

I'm aware of that essay. Perhaps I should have linked to it at the start. The talk page of that essay is more informative than the page itself. For example, the quote from the BMJ that "Black bands across the eyes are wholly ineffective in disguising the patient". Colin (talk) 07:11, 15 June 2011 (UTC)

I'm disappointed there hasn't been more response to this issue, given that a recent WP Signpost highlighted the need to improve and enforce this guideline. I think, correct me if I'm wrong, that we agree the obese girl picture is demeaning to the girl and that the technique used to disguise her is not guaranteed to make the image unidentifiable to herself or people who know her. I'll try to draft some suggested changes and I think we may need to try to delete the image in order to establish consensus (as well as for the obvious reason). Colin (talk) 07:11, 15 June 2011 (UTC)

Sorry I haven't had/got time to contribute properly, but I agree this is a definition we need to be clear about. I'm sympathetic to your broad definition of identifiability. One thing I want discussed is post-identifiability. We realised in the COM:SEX proposal that an uploaded image (say with the head chopped off) was still in some sense "identifiable" because at any later time the photographer could release just the other half of the image. Similarly file metadata could be used to make connections if it is detailed. 99of9 (talk) 12:58, 15 June 2011 (UTC)

So, basically you think it should not be possible to illustrate obesity with a photograph of a person, as all of them will be "identifiable" under new guidelines? I'm not sure I agree that pixelation is not effective. I would agree that medical images should be treated very carefully, as that type of information is very much guarded by privacy and other special laws, but I don't see a good reason to apply the same stringency to every other type of photo. Better I think to come up with suggestions to better anonymize such photos so that we can illustrate, not simply resorting to deletion. Carl Lindberg (talk) 13:03, 15 June 2011 (UTC)

Err... are you assuming that nobody would ever give their consent to release a (suitably anonymized?) picture of them? We can illustrate anything if we have clear consent. --99of9 (talk) 13:12, 15 June 2011 (UTC)

Right, so now this guideline is morphing so we need permission of people taken in public, which is not what the board resolution was about. Carl Lindberg (talk) 13:13, 15 June 2011 (UTC)

No, I never said that. I'm just saying that no matter what the definition of identifiability is, phrases like "not possible to illustrate" will never be the case. --99of9 (talk) 13:17, 15 June 2011 (UTC)

Sorry, I guess I meant "not possible without explicit permission from the pictured person". Carl Lindberg (talk) 13:29, 15 June 2011 (UTC)

Deletion-requests of pictures of one or more specific person(s) taken in a public space in Denmark

For info: I have started this topic on the Village pump. --Henrik (heb: Talk · Contributions · E-mail) 09:13, 17 June 2011 (UTC)

POW

The "Convention relative to the Treatment of Prisoners of War" makes illegal to publish some photographs of prisoners of war. Specifically its article 13 protects POW from "public curiosity". I think we should add a section on this on this page. Images that are likely to fall in this category shouldn't be hosted on Commons. --  Docu  at 06:42, 19 June 2011 (UTC)

I agree. But this rises another issue, which I'm not quite sure how to handle. Currently Commons already hosts a number of POW-pictures taken after 21 October 1950 (when this convention came to force), i.e. from Category:Prisoners of war in the Falklands War and Category:Prisoners of war in the Vietnam War (and there are further examples in Category:Prisoners_of_war and Prisoner of war). Should we have a principal discussion regarding those already hosted photos that are in violation of article 13 and if so where? Commons:Village pump or Commons:Village pump/Proposals? --Henrik (heb: Talk · Contributions · E-mail) 07:18, 31 August 2011 (UTC)

I have started a discussion about this subject on Commons:Village pump#Third_Geneva_Convention_article_13. Everybody is encourage to contribute. --Henrik (heb: Talk · Contributions · E-mail) 13:37, 5 September 2011 (UTC)

China

• GreatWallofChina3600ppx27.jpg
• 
• 
• 
• 
• 
• 

The policy is not really clear about photo with chinese people's face in it. It says :

• Taking a picture of a person in a public space: Does not require consent
• Publishing pictures of a person in a public space: Does not require consent (as per above)
• Commercial use of a published picture of a person in a public space: Requires consent

According to the Chinese Civil Law Article 100[1] photos of regular people may not be used for profit (commercially) without consent.

So, is it ok or not to host on commons photos with chinese people in it without consent ? Because it seems allowed to take the photo and publish it, it's only when the image is used that the reuser has to take it into consideration... and probably only in china. --Lilyu (talk) 11:42, 31 July 2011 (UTC)

Using it "commercially" is another term for what is known as "personality rights" or "publicity rights". You can't use one of those photos to, say, endorse a product, or something like that, i.e. using a person's likeness or reputation to try to sell stuff (in this area, the meaning of "commercial use" is completely different than "commercial use" in a copyright context). This is pretty common actually, and not really anything peculiar to China (though as always the details will differ between countries, and differs between US states actually). Many photos have the {{personality rights}} tag added to remind re-users about it, though really anyone involved in doing such work should know about those laws, and obtain a model release or other relevant permission from the person photographed. These commercial rights are non-copyright restrictions, and are distinct from privacy rights (where it becomes illegal to simply publish a work, and where Commons itself could be in violation -- those are where we consider deletion). So yes, we can host them. But if you are making an advertisement, I would steer clear from using them unless you obtain additional permission from the pictured person. Carl Lindberg (talk) 13:25, 1 August 2011 (UTC)

Moving 'Country specific consent requirements'-section to a seperate page

Looking at the Country specific consent requirements-section it has somewhat quickly become quite big, so I'm thinking of moving it to a page of it's own (suggesting Commons:Country specific consent requirements). Any thoughts/comment on that? --Henrik (heb: Talk · Contributions · E-mail) 10:20, 26 August 2011 (UTC)

I support your suggestion. Perhaps we can let the summary-table stay here and move the details to a sub-page. BTW, the section has to grow even more due to the fact that most non Anglo-American law systems including many big and important countries do not allow in principle to publish or distribute photographs of identifiable people without their consent, even if the picture was taken in a public space.--JordiCubero (talk) 02:58, 27 August 2011 (UTC)

Leaving the summary-table on Commons:Photographs of identifiable people makes good sense to me. --Henrik (heb: Talk · Contributions · E-mail) 10:12, 29 August 2011 (UTC)

I think moving the details to a separate page is a good idea. It might be useful to retain a copy of the summary table in the new page too. --Avenue (talk) 00:34, 30 August 2011 (UTC)

I have now moved it. The table is extracted to an independent page for template-inclusion as it seemed like the most simple way to avoid double-updates. --Henrik (heb: Talk · Contributions · E-mail) 08:48, 1 September 2011 (UTC)

Thanks, that looks good. I've added a navbar and expanded the note about reaching more details. --Avenue (talk) 13:53, 1 September 2011 (UTC)

There is an argument for hosting the country-specific consent requirements on Wikipedia, as this is encyclopaedic information that (a) isn't a matter for Commons folk to interpret or judge and (b) would then come under the sort of policy scrutiny that WP encourages such as better sourcing. Colin (talk) 10:18, 2 September 2011 (UTC)

Moving this stuff to WP:EN could be an option to consider, of course it may be of encyclopaedic interest. On the other hand, the focus is very strongly on specific image use policy problems of Commons here. On WP, others will make changes, perhaps without being aware of those needs, and the article could get less useful for Commons. I'm not sure. At least, let me some time, I have got some countries in the pipeline and would like to complete this here (I'm not a registerd user on English WP and usually don't contribute to English-language WP articles).

BTW, I wonder if common law countries like India should be added to this list. I don't know any details about India, but I think they all have this "no-expectation-of-privacy-in-public-space-rule" and therefore do not considerably differ from the standard Common politics (influenced by the Anglo-American system). At least, the exceptions mentioned up to now for India are perfectly viable without adding anything to the usual Commons criteria (since the use of embarrassing or derogatory pictures is not acceptable here).--JordiCubero (talk) 19:38, 2 September 2011 (UTC)

US/UK

Commons:Country specific consent requirements doesn't cover the US or the UK. This seems a strange omission - surely someone knows these cases and can add them to the table and the subpage? Rd232 (talk) 23:32, 15 September 2011 (UTC)

I was also wondering what the laws in the US and UK are. I wouldn't be surprised if there were no restrictions in the US (as we typically rate freedom of speech over privacy), but it would be nice to know for sure. Kaldari (talk) 23:39, 15 September 2011 (UTC)

Above the table, it says "The following is a list of countries where consent is needed for one or more of the mentioned situations." So if consent is not required in the US and UK, their omission doesn't seem strange. But perhaps it would be better for the table to include all countries, so that it isn't left unclear whether a country has no restrictions or just hasn't been added to the table yet. --Avenue (talk) 00:42, 16 September 2011 (UTC)

Above the table it also says This list is incomplete: Just because a country isn't listed here, it does not reflect a fact that everyone is free to take/publish/commercially use pictures of people in public spaces in that country. I think the table should include all countries. Rd232 (talk) 06:20, 16 September 2011 (UTC)

Generally, you can take photos unless there is an "expectation of privacy" (among family and friends, or in bathrooms, locker rooms, etc.) Commercial use of course is always protected; that is {{personality rights}}. Carl Lindberg (talk) 03:48, 16 September 2011 (UTC)

Are you speaking of US law, UK law, or both? Kaldari (talk) 21:11, 16 September 2011 (UTC)

Normally not OK

I was looking into the Normally not OK examples, and is trying to source them a bit. Initially I'm going to use them as examples in a talk about privacy discussions on Commons. Secondly it might be beneficial to link the examples on the Commons-page. What I have found so far is:

• A man and woman talking, entitled "A prostitute speaks to her pimp" (possible defamation)

Examples in Commons:Deletion requests/Image:Prostitute-from-lviv-ukraine-speaks-with-her-pimp-highres.jpg and Commons:Deletion requests/File:Prostitutes in the street of Reeperbahn.jpg

• An identifiable child, entitled "An obese girl" (potentially derogatory or demeaning)
• Partygoers at a private party where photography is not permitted or is not expected (unreasonable intrusion without consent)

I see the Burning Man cases such as Commons:Deletion requests/File:Burning Man 228 (241613953) crop.jpg and Commons:Deletion requests/File:Burning man 2.jpg

• Nudes, underwear or swimsuit shots, unless obviously taken in a public place (unreasonable intrusion without consent)

Basically the same as above

• Long-lens images, taken from afar, of an individual in a private place (unreasonable intrusion)

Commons:Deletion requests/File:Beach in Italy (302214719).jpg and Commons:Deletion requests/File:Pregnant woman nude on the beach.jpg.

This leave me with two questions:

1. I can't find the discussion about the deletion of the "obese girl" (even mentioned in a image caption on the Commons page). Any pointers?
2. Any argument for not linking to above discussions to provide a bit more weight into the "not okay" listings?

In kind regards heb [T C E] 05:37, 19 April 2012 (UTC)

The obesity deletion discussion is here: Commons:Deletion requests/Image:Childhood Obesity.JPG. --Avenue (talk) 22:46, 22 May 2012 (UTC)

Personality Privacy rights, et al

(title modified for clarification, boldly: as I think this is what the OP is referring to. Feel free to revert if I am wrong.)--Canoe1967 (talk) 18:51, 24 May 2012 (UTC)

Hi everyone, While reviewing past board resolutions (don't ask, I'm a geek, I know), I came across the resolution on Images of Identifiable People. After I re-read it, I admit to being a little surprised because I don't recall there being a ton of discussion about it here. It's possible that I've missed something, but I wonder if Commons has made any progress toward the parts of this that are directly pointed at this project, namely:

• Strengthen and enforce the current Commons guideline on photographs of identifiable people with the goal of requiring evidence of consent from the subject of media, including photographs and videos, when so required under the guideline. The evidence of consent would usually consist of an affirmation from the uploader of the media, and such consent would usually be required from identifiable subjects in a photograph or video taken in a private place. This guideline has been longstanding, though it has not been applied consistently.
• Ensure that all projects that host media have policies in place regarding the treatment of images of identifiable living people in private situations.
• Treat any person who has a complaint about images of themselves hosted on our projects with patience, kindness, and respect, and encourage others to do the same.

An (admittedly superficial) review didn't seem to show a lot of movement on the policy question in the first bullet point, so I'm wonder if there's some way that my team or the Foundation can support the commons community in this. Would it be helpful if we were to get some suggested draft language from Legal/Community Advocacy? I'm not saying we should, it's just one idea for how the WMF could support the commons community in fulfilling this Board resolution that's nearly a year old. If there are other ways that we could be helpful, I'm happy to listen here and try to find the ones that make sense for this community.

I'm not here with any predefined outcome in mind: really, I'd like to open a discussion about how the WMF (and specifically the Legal and Community Advocacy Department) can be helpful in working toward the position laid out in the Board resolution. Thanks! Philippe (WMF) (talk) 01:10, 20 May 2012 (UTC)

I may be confused. I think the consent would have a clash with legal issues on privacy. If I take a picture of a friend in my house, he may allow it placed in commons but not wish to have his name known to WMF. WMF would have to go on my word that I have his permission. Anyone that sees an image of themselves could add a speedy delete tag and/or email the adresses that cover that?--Canoe1967 (talk) 01:25, 20 May 2012 (UTC)

I know OTRS is backlogged, but this is what I imagined the service to be when it comes to issues like this. I know this will also run into some of the issues we have now over the PORN that we got, but as far as I am aware, I am not sure a lot of things were done. We just tag an image and call it a day. User:Zscout370 ^{(Return fire)} 03:04, 20 May 2012 (UTC)

Could we just create a policy that simply states that WMF is CYA and any uploader can have his butt sued for images? I removed a statement in an en:WP BLP article after the the BLP asked in help desk. I did this without question, speedily, and in good faith. We had a discussion on the talk page as to wording and reliable sources and the statement was replaced with a more accurate one that I think the BLP had to accept legally.--Canoe1967 (talk) 03:42, 20 May 2012 (UTC)

It seems to me that would not comply with the Board resolution, which was not to CYA, but was directed to the community. So it's not a matter of the WMF covering, but rather of the commons community coming into agreement with the resolution passed by the Board, as I read it. Philippe (WMF) (talk) 13:16, 20 May 2012 (UTC)

• I don't have time to answer properly just now, but two things that I'm aware of were done: COM:PEOPLE now includes the clause expecting evidence (at least uploader assertion) when consent is required, and {{consent}} was created and discussed. The latter has not gone into widespread use yet - I would suggest that anyone who uploads photographs of people tries it out on their own uploads to see if it needs any other parameter options. --99of9 (talk) 04:47, 20 May 2012 (UTC)
  • I note that while you're obviously correct about the statement above, it's also important to say that COM:PEOPLE includes this statement:

"If you would like to proactively assert compliance with these guidelines for a particular photograph or video, you can add the {{consent}} template to the file's description page. Please refer to the template documentation for further instructions. Use of this template is not required for compliance with these guidelines or other Commons policies."

A template without policy enforcement behind it does not satisfy my reading of the Board resolution. Philippe (WMF) (talk) 18:12, 21 May 2012 (UTC)

• I think the intent of that sentence is to say that there are other ways to assert consent, not just that one particular template. But I agree a bit of clarification could help. --99of9 (talk) 22:31, 21 May 2012 (UTC)

• I put all versions of {{consent}} on a file of mine that isn't used much yet: File:Watermark sample.jpg. Feel free to use it as a sandbox file. Should we put a link to the template on the upload page, or possible have a bot add it to all files in category:people to get uploaders to notice it? It seems good enough to be policy to me.--Canoe1967 (talk) 05:18, 20 May 2012 (UTC)

• The upload wizard will need a dropdown menu with each of the possible consent|parameters. --99of9 (talk) 11:38, 23 May 2012 (UTC)

• Note: this was discussed at VP (archive) last month. The issue hasn't gone away. For instance, we could more explicitly address privacy as a reason for deletion (see Commons_talk:Deletion_policy#Privacy). Rd232 (talk) 13:46, 20 May 2012 (UTC)

• The issue is serious and should be discussed. I am a bit worried though, that such resolutions are approved without as much scrutiny as changes in our guidelines. It is unclear how the resolution is supposed to affect existing images taken with undocumented consent. I suppose those should be kept, unless deemed very problematic. There is also the problem about declared consent being contested - by the subject, by somebody claiming to be the subject or a third party. The second is a serious threat to some classes of images (I remember a case about dead soldiers). The consent template would satisfy the resolution, but not these other problems. It was also introduced as a suggestion and should get more eyes before made official or put in wide use. --LPfi (talk) 09:33, 21 May 2012 (UTC)
• At this point should we decide where to actually discuss it. This thread may be buried in time soon, as it seems others have been. On the matter of the consent templates, they could be added to all the images in category people to draw attention to the issue. Changes to all the upload pages including ones like Commonist, Android, etc. may have to be made as well. A statement such as: 'If this image has living people then one of the consent templates MUST be used?'--Canoe1967 (talk) 16:28, 21 May 2012 (UTC)

I like that. Does it have teeth? Is the policy sufficiently strong to back it up? Philippe (WMF) (talk) 17:42, 21 May 2012 (UTC)

I think if a bot master had a bot add:

A query has been raised regarding consent of identifiable persons: It is not clear whether all identifiable subjects have consented to the publication of this image or video. If you created this media, and it was taken in public or you have the consent of the subjects (or parental consent for minors), please replace the template parameter 'query' with a more appropriate one from Template:Consent. If this is not rectified, the media may be discussed for deletion as a violation of Commons:Photos of identifiable persons.

...to every image in category people it would bring the issue forward to scrutiny. Just a thought though. Does present policy actually require some type of consent tag now with all images containing people?--Canoe1967 (talk) 17:51, 21 May 2012 (UTC)

My read is that the Board left some wiggle room. I think any attempt at proactively updating policy and strengthening it would be a welcome sign. I tend to think that mass-using a bot to drop a massive number of templates is likely to have a flashback effect that would be unwelcome. In that case, I think what we would see is a hardening of resolve by those opposed (who would also, quite rightly, point out that the time requirements needed to satisfy those would be extreme). I would suggest, instead, some sort of triage system that begins with higher-risk images (if such a thing can be identified) and works toward some threshold. Philippe (WMF) (talk) 18:17, 21 May 2012 (UTC)

• Could we create a category 'images of people lacking consent templates' type thing? A bot could add all the people images to this new category. I don't think there is a way to add all images of people robotically unless they are in the people category now. Then editors could go through all the files in the new category manually and add approriate tags to the obviuos ones?--Canoe1967 (talk) 18:31, 21 May 2012 (UTC)

  I think Canoe1967 is absolutely right that this discussion should be taken to a good place. One cannot be supposed to watch this page as it is busy and more or less English only. And any decision will affect a huge amount of images.

  On the subject matter: The change should be about new uploads. There the key things is having the templates and making the upload forms handle them smoothly.

  The policy has required consent, not documentation of consent. That could be read as uploading an image means implicitly assuring consent. Deleting perfectly good images because of changes in policy should be the last alternative (I have supposed Commons is a stable image repository in that sense). I think we should delete old files only if there is a clear violation of privacy (handling of complaints should probably continue as now).

  --LPfi (talk) 18:57, 21 May 2012 (UTC)

• I agree that we shouldn't just bomb consent|query on every pre-existing image of a face, but I would point out that pre-existing policy reads "...the subject's consent may be required as described at Commons:Photographs of identifiable people. In all cases, the burden of proof lies on the uploader or other person arguing for the file to be retained to demonstrate that as far as can reasonably be determined... that any required consent has been obtained." I'd say that past uploads should be given good faith, but if there are red flags flying, they should go to DR. --99of9 (talk) 11:38, 23 May 2012 (UTC)

The policy is just rubbish made during the Jimbo pron witch hunt and will get abused by the haters to get rid of images they don't like. Trick question: Find the board member who was an active Commons community member around the time of that resolution.

We don't need a board that sends commandments from their ivory tower, we need a board that actually helps us to improve things here. Now they're only making things worse: Rubbish policy, we don't abide to it and yet more material for the haters to dump on Jimbo's talk page. Good job board! Multichill (talk) 22:13, 21 May 2012 (UTC)

I can see your points. Until this thread was started I assumed that WMF was CYA and uploaders are to be responsible for permission under threat of legal action. I also mention that I read somewhere that WMF can dictate policy unilaterly because they control the servers type thing.--Canoe1967 (talk) 22:22, 21 May 2012 (UTC)

Move discussion to:

Commons talk:Privacy policy? We could still have a bot tag all images in Category:Unidentified people? There are only around 10,000 images there, so that may not irk too many people, justify the tags, and create more notice of the issue?--Canoe1967 (talk) 20:04, 21 May 2012 (UTC)

If that is the destination, I would suggest moving the whole discussion there to provide context and prevent forking. Dankarl (talk) 21:25, 21 May 2012 (UTC)

Oops. Should it be Commons talk:Photographs of identifiable people instead? That other section is for how commons handles our user data, I think.--Canoe1967 (talk) 21:51, 21 May 2012 (UTC)

That sounds like a good idea. If there are no objections, I'll move the discussion there so that it can be more easily carried forward. :) --Maggie Dennis (WMF) (talk) 11:09, 23 May 2012 (UTC)

Sure, go ahead. There is also some related discussion happening at Template talk:Consent. --99of9 (talk) 11:28, 23 May 2012 (UTC)

Moved. :) --Maggie Dennis (WMF) (talk) 13:58, 23 May 2012 (UTC)

Private place

What is a private place, speaking of what I know, sport photographies can be done in privately owned stadium. I'm really not against enforcing a stricter policy on personality rights, but we have to consider all the pros and cons, and tune the policy. --PierreSelim (talk) 22:10, 21 May 2012 (UTC)

See Commons:Photographs_of_identifiable_people#Normally_OK. It seems that each country has different laws though.--Canoe1967 (talk) 22:25, 21 May 2012 (UTC)

Are people talking about privacy rights or personality rights? They are not remotely the same thing. Personality rights have never been cause for deletion -- they are analogous to trademark, using someone's name or likeness for advertising and things like that, and are sometimes known as publicity rights. As far as the U.S., privacy rights happen when there is an "expectation of privacy". That is definitely not the case for sports photography. Carl Lindberg (talk) 22:27, 21 May 2012 (UTC)

I think the OP meant privacy rights. I think most courts would find that if an image from commons is used for financial gain of personality rights it wouldn't matter if the image was even considered public domain and is covered by other laws?--Canoe1967 (talk) 22:33, 21 May 2012 (UTC)

Thank you Carl Lindberg for your enlightning comment. I'm refering to the wording of the WMF statement, as Commons is an international project the wording must be better than that. I cannot be expected to understand that private place means expectation of privacy (and I won't blame anyone for misunderstanding such wording). --PierreSelim (talk) 08:57, 24 May 2012 (UTC)

Private places and public places are discussed in the first 3 sections of the project page. Is this discussion adequate, both for scope and clarity, considering also the discussion of moral issues farther down the page? Dankarl (talk) 21:50, 24 May 2012 (UTC)

Organize the thread?

Should we organize this discussion somehow?

• It should be made clear that privacy rights are far different than personality rights.
• Should the consent templates be made policy and mentioned in upload instructions. They may need and edit as to consent to upload does not mean allowing use as personality rights.
• How do we deal with existing images.
• Can we had a speedy delete email address for people who didn't give consent to the templates.

No need for this one. We already have COM:OTRS. --99of9 (talk) 22:43, 23 May 2012 (UTC)

I was thinking of people that come across a picture of themselves that they did not give consent to upload. They may not be familiar with OTRS and other methods of asking for removal. Either an email address or 'click here to have image removed' right on the image page and easy to see. It would be nice if there was a way to have it deleted within minutes, but I doubt that is possible. There may be abuse of this feature, but isn't it better to have the images removed and then restored if it was a false positive? A block/warning list could be made of IP, email, and user names that do abuse the system. This may satisfy the 3rd point made by the OP. Treat them with respect, type thing.--Canoe1967 (talk) 18:34, 24 May 2012 (UTC)

• Are there points from the Talk:Consent templates that should be mentioned here. Just some thoughts that may move the discussion forward.--Canoe1967 (talk) 14:55, 23 May 2012 (UTC)
• A standalone RFC might be a more flexible way to organise this wideranging discussion. Rd232 (talk) 06:45, 25 May 2012 (UTC)

Identifiable people

There is a useful discussion from last year on this page, including two possible definitions of identifiable and the suggestion that different images might require different definitions. Please read and continue the discussion here.

If we allow the less inclusive definition of identifiable to apply to uncontroversial images, it will simplify the process and make more images available from 3rd-party sources. For examples where this could make a difference, please see File:Moose on the playground.jpg, File:Working an oyster net.jpg, File:ATVs in the woods, Alaska 2010.jpg; none of these locations is unambiguously public, none of the photos would let a stranger spot the subject on the street, but I think all would be recognizable to friends, family and co-workers. Dankarl (talk) 22:30, 24 May 2012 (UTC)

I'd agree that the weaker definition of identifiability should apply in most cases, i.e. where there are no overriding moral or legal concerns. The man on the ATV looks identifiable by a stranger at full resolution. For images like that, a {{3rd party consent}} template could help us record when consent can be established for images from 3rd parties (e.g. through comments by their uploaders on Flickr). --Avenue (talk) 05:20, 25 May 2012 (UTC) Never mind, I now see the "hosted" argument for {{consent}} does the trick. --Avenue (talk) 09:20, 25 May 2012 (UTC)

Does it in this case? If so only because there is enough information in the photostream to surmise a relationship with the photographer/poster. That is a different kind of identifiability. Do we consider it? Dankarl (talk) 13:06, 25 May 2012 (UTC)

{{consent}} - further cases

Please feel free to respond to the individual bullet points directly.

• To prevent a lot of unnecessary DRs, I suggest adding a deceased parameter to the set. In that regard, how many years would we hold off posting images of deceased persons? Dankarl (talk) 00:18, 25 May 2012 (UTC)
  • In general, I'm pretty sure the right to privacy expires upon death. Although, photos of a death scene may be different -- California ruled that the family had some privacy rights for those.[1] Pretty sure I remember a similar French case on death scene photos; they noted that European privacy rights also expire on death but there was some sort of human dignity protection granted to the family on such photos. Carl Lindberg (talk) 14:41, 25 May 2012 (UTC)

• I would also suggest an unidentifiable parameter reflecting whatever consensus we reach about what constitutes identifiability. Dankarl (talk) 00:18, 25 May 2012 (UTC)
• In response to User:Avenue's comment under Identifiable People (directly above) for the case Consent:Hosted what do we mean by evidence? Do we mean an explicit statement, or can we also infer permission because (for instance) the Flickr photostream provides evidence of a family relationship relationship between the photographer/poster, and the fact that the photograph has been posted for some reasonable period? (I am thinking here only of uncontroversial cases, i.e. cases that do not appear to raise moral issues of intrusiveness and where we have received no objection from the subject.) Dankarl (talk) 13:07, 25 May 2012 (UTC)

• "Evidence would normally be an assertion" ... but I guess if there's good evidence of other types, it would be accepted in a DR. I don't regard longevity as good evidence - subjects don't always scour Flickr, and it can take years for them to notice themselves, if ever. --99of9 (talk) 13:30, 25 May 2012 (UTC)

• For the case of Consent:Published, what do we mean by "Professional editorial standards"? do we mean modern standards? And how do we evaluate? Does this mean an automatic pass to "mainstream" sources? to Federal websites? Dankarl (talk) 13:07, 25 May 2012 (UTC)

• If the standards are disputed, it can be discussed in a deletion review. I don't think it would be possible to write a watertight definition that dealt with all edge cases well. How do you define mainstream? (I suggest that asking about examples might be easier) --99of9 (talk) 13:30, 25 May 2012 (UTC)

The question of "which sources?" was part of what I was asking. It is not at all clear to me what we mean by this case or how we would evaluate, and with this degree of ambiguity waiting for the deletion review process to reach some overall consensus would not be efficient. Dankarl (talk) 13:58, 25 May 2012 (UTC)

So, for example, if you find a picture in a book published by a publishing company which uses professional editors, then for our purposes it's ok to assume that the subject of that picture consented to the publication of their picture. Same with most newspapers with professional editors. Blogs - no. Self-published - no. There are some in-between cases, but unless you can name a big category of publications that are ambiguous, I don't think it's worth trying to guess in advance. --99of9 (talk) 14:25, 25 May 2012 (UTC)

• Does anyone no the local laws in Israel? There is concern about this image right now: Commons:Featured picture candidates/File:Woman outside Sepulchre-2.jpg. I may just email my embassy there. I could also ask my foreign affairs dept. To check the laws in all countries that we have embassies in, gov.ca.--Canoe1967 (talk) 16:49, 25 May 2012 (UTC)
• I've shooted a lot of photo during the general assembly (in France it's a private event where only members can join) of the french chapter Wikimedia France and posted it on our local wiki. The president asked me if I could put the pictures of him on Commons therefore I used {{consent}} with full agrument (he knows it can be used commercially), I hope it's ok with what is being discussed here, I'm not sure to understand everything. An example here: File:Rémi Mathis - AG WMFR 2012-05-26.jpg. --PierreSelim (talk) 14:56, 30 May 2012 (UTC)

Add new countries here

Commons:Country specific consent requirements/table

I have emailed my embassy in Tel Aviv because of concerns over this image: Commons:Featured picture candidates/File:Woman outside Sepulchre-2.jpg. I also asked them if our foreign affairs dept. has a list that may be more complete. This section could be edited by them or us to add more countries; unless we want them to attempt editing the table. I know that would scare me.--Canoe1967 (talk) 18:09, 25 May 2012 (UTC)

The Canadian embassy emailed me back and recommended that I email the Israeli embassy in Canada. That is the next step. I may just email our foriegn affairs dept. and see if they have a list of countries as well.--Canoe1967 (talk) 17:07, 30 May 2012 (UTC)

Israel

I was referred to the Protection of Privacy Law, 5741 – 1981. "Chapter One, 2. Infringement of privacy is any of the following:(3) photographing a person while he is in the private domain;(6) using a person’s name, appellation, picture or voice for profit;" I don't know how to add this to the table. Column 1 would be a yes, and 2 and 3 would be with permission? Should I upload the .pdf if gov.il documents are public domain? There is a Hebrew version online as well. Just google Protection of Privacy Law 5741 1981. Both versions are on gov.il sites.--Canoe1967 (talk) 19:33, 30 May 2012 (UTC)

• (note) I have since uploaded the English and Hebrew .pdfs.

Hebrew and English--Canoe1967 (talk) 23:19, 27 July 2012 (UTC)

 Done

Lawbrain

I found a wiki that has links to laws in many countries. I think they are official links.

• http://lawbrain.com/wiki/International_Privacy_Law --Canoe1967 (talk) 20:07, 30 May 2012 (UTC)

Mexico

It appears that Mexico requires consent for publishing of photographs of identifiable people. This is the link I found. The other entries appear to be similar to that presented here, except for Mexico which doesn't find a mention here. cheers. —SpacemanSpiff 07:56, 1 August 2012 (UTC)

Do you know how to edit the table? Add Israel above if you wish.--Canoe1967 (talk) 12:50, 1 August 2012 (UTC)

Commons:Country specific consent requirements/table Finally found it.--Canoe1967 (talk) 12:58, 1 August 2012 (UTC)

 Done--Canoe1967 (talk) 13:15, 1 August 2012 (UTC)

It isn't the technical bit of adding it to the table that I was uncomfortable with but whether the source itself was right (and my interpretation of it). I was hoping someone familiar with these issues would take a look instead of me, so I'm glad you've done it. cheers. —SpacemanSpiff 13:41, 1 August 2012 (UTC)

If you look at the table page linked above it has text below for each country. I made new sections and just put our links there.--Canoe1967 (talk) 14:23, 1 August 2012 (UTC)

proposing extending beyond photos to likenesses, nonphoto images, video, and inaccuracies

I propose to edit this guideline to extend its applicability to visual images of identifiable persons and to likenesses of identifiable persons. Most text would not change. However, an accurate drawing of a person can be just as illegal as a photograph of that person, depending on jurisdiction; a drawing may be made directly from a subject or from a photograph of a subject; and likenesses which can present legal problems include voice recordings and autographs. Fanciful and imaginary drawings and likenesses, apart from whether they would be original research, may also present legal problems akin to those of photographs. Inaccurate photographs may also present legal problems if viewers believe they are accurate; and, inherently, every photograph is inaccurate to a degree, some more so, and it is not always possible to tell whether an image is a photo or not.

I'm not a lawyer.

The proposed addition is approximate and would be the first section after the lead. It follows:

==Nonphotographs, video, and inaccuracies==

This guideline also applies to images other than photographs and to other likenesses, such as voice recordings and autographs, to moving images, and to images of doubtful accuracy. Exceptions, if any, are noted.

I'll wait a week for any comment. Thank you. Nick Levinson (talk) 19:40, 15 September 2012 (UTC) (Corrected formatting by adding a line break: 19:49, 15 September 2012 (UTC))

Other issues aside, how do you establish whether a drawing or caricature involved a violation of provacy? Dankarl (talk) 20:52, 15 September 2012 (UTC)

If a photo violates privacy, an otherwise-identical derivative drawing of the photo violates privacy. As it is possible to make a drawing look like a photograph, the test could be what a jury believes it is viewing. Where I'm less clear is with nonderivative drawings and realistic-but-inaccurate drawings, which, as I noted, would be subject to law that likely varies by jurisdiction. Unrealistically inaccurate drawings probably would not violate privacy regardless of content because people would not be misled about their realism, and caricatures would likely be in that vein. It could be a question for a jury. I'm being legally cautious and we should be, too. Nick Levinson (talk) 21:02, 16 September 2012 (UTC)

I don't see at all that a drawing would regularly violate privacy. Can you provide a law or court case on the matter? Perhaps an identical drawing of a photo would violate privacy. But in general, I don't think there's any legal prohibition on drawing Bill Gates at his dinner table, even if he would have legal protection for a photograph taken there.--Prosfilaes (talk) 21:40, 16 September 2012 (UTC)

If it's a near-identical drawing of a privacy-infringing photo, you may have a point, although even in that case the drawing is not nearly as problematic as a photo (it's not nearly as good enough as proof). But that would not extend to any other kind drawing and I don't see why we need to modify policy in that regard. If the drawing removes the elements of the photo which make it a privacy violation, no problem. 99.99% of the time, it's going to be photos/videos, and possibly audio recordings, that are privacy issues -- could mention videos and sound recordings, sure. It really helps to point to court case decisions on relevant matters to back up your changes. Fanciful drawings do not pose any issues at all, unless they are defamatory -- and that would be nearly impossible, as those would usually be taken as one person's opinion (and free speech), and not a statement of fact. Are there any court cases which back up your contention that fanciful drawings are an issue? Also, original research is not a Commons policy, at least in that way. Quite the opposite. See COM:NOR. We don't want to be a platform for artists trying to promote themselves, but that is a separate area of scope. The main question is whether the drawing can reasonably be used in an educational context, which many drawings can. Carl Lindberg (talk) 22:08, 16 September 2012 (UTC)

Photographs are not central to the New York right of publicity. What matters is a "picture" or a "portrait", thus a drawing is included, as is a mannequin. A picture or portrait even without a face can cause a violation; so can a picture of poor quality with small faces; so can a picture that was somewhat edited; so can a picture of a person who looks like a famous person if used confusingly; and so can a picture in a Wikipedia article if the article is written as an advertisement by a representative of a company the article is about, although that's against two policies/guidelines. A research article qualified because of publication by someone else to promote a product without permission, and it qualified as a right of publicity issue, quite apart from copyright. A fictional biography of a famous person can violate. Sources from judicial opinions, per annotations in a secondary source, for this are on this talk page (until archived), at the topic/section #propose clarifying identifiability (to be posted shortly, probably within the hour).

This makes it likely that a caricature can fall afoul of the right of publicity. A caricature of a famous movie actor infamous for his religion, if used to advertise, say, competing movies or competing religious products without his consent could be violative even though it's a caricature. Indeed, the principle is that a caricature of a little-known person who gave consent but which looks like the famous person who did not consent and where the intended audience would have been fooled into thinking the latter was advertsing products could be violative.

It seems even possible that a house can be included, if the house is recognized as a certain person's home, in which case pictures of distinct houses hosted at Commons even without people being visible may give rise to violations just as people pictures can. The references I saw to houses as implicating the right were vague. If you need that investigated, feel free to let me know.

Wikimedia's prohibition on original research need not apply if, for example, a fanciful (obviously inaccurate) drawing is of significant importance on its own worthy of being reported and is itself sourced, and then the image of the drawing is misused for advertising or trade use. No original research need have happened in the creation of such Wikimedia content.

Nick Levinson (talk) 14:18, 18 September 2012 (UTC) (Corrected a misspelling (mine): 14:25, 18 September 2012 (UTC))

Which is only tangentially related to this guideline. None of what you mentioned is particularly important to the Wikimedia Foundation, because it doesn't commercially use anything here. Moreover, it goes beyond what labeling will handle. A trivia question was ruled to imply the sponsorship of an athlete and thus violate his publicity rights. Ultimately, I don't see that we can or should worry about this. People should know the laws that affect them and take responsibility for following them themselves. The laws that affect advertising are complex and not directly related to Wikimedia's providing of images or materials.--Prosfilaes (talk) 08:38, 23 September 2012 (UTC)

We do not delete things on the basis of publicity rights. We may tack a {{personality rights}} tag on a work to inform re-users of that potential issue, but usage on Wikipedia will hardly ever violate those. We are primarily concerned about privacy rights, which are completely different, as if a work violates privacy, then it becomes illegal to publish and host here. The two doctrines should not be confused, and are treated in very different manners here. While both are potential issues with photos of identifiable people, only privacy would cause something to get deleted. Definitely agreed that the right of publicity can cover an awful lot more types of works than just photos, but (at most) that means we add a warning tag. Sound recordings though are one area where the right of privacy could easily crop up; that should probably be mentioned, although the topic of the page is about photographs. Also, the "original research" policy is specific to Wikipedia, not all Wikimedia projects in general. Carl Lindberg (talk) 15:27, 23 September 2012 (UTC)

It is tangential, but that's why I propose extending the guideline to all likenesses, with emphasis on photos because they'd be the most frequent cause of violation, but at least mentioning applicability to other likenesses. If we never delete or rewrite anything because of publicity rights, statistically we probably are failing in our legal responsibilities. The risk is more significant than we see, but we're probably not being caught often, and we're probably not catching editors editing on behalf of their companies for pay and who have undisclosed conflicts of interest; many editors are editing articles likely subject to conflicts of interest quasi-anonymously from IP addresses, according to a study a while back. Thus, many are likely to be writing disguised ads in Wikipedia, and that leaves the Foundation legally exposed. Text is better controlled with sourcing, verifiability, and so on, but nontext is more problematic. I provided a small slew of case precedents at the topic/section #propose clarifying identifiability; they're just a handful but significant because most cases are decided by settlement or summary judgment or, because they are not appealed, judges may not have to write opinions, and those cases are often based on existing case law. Thus, judicial precedents have much more weight than their number might suggest. I don't know what the "trivia question" sentence refers to, so I can't judge that. I agree that privacy and publicity rights are different and have been trying to straighten out other people's confusion on that; it is one reason for addressing it where editors will see it. Nick Levinson (talk) 19:28, 23 September 2012 (UTC)

I think there is a misunderstanding here... when it comes to the "free" status, we generally only consider copyright law (and copyright-based restrictions). Anything else (such as trademark) is a non-copyright restriction (see Commons:Non-copyright restrictions) does not cause something to be deemed non-free, and is not a reason for deletion. Of course, Wikimedia must follow the law itself, so if Wikimedia's own use violates the law, then we need to do something about that. For privacy violations, simply publishing the photo is a problem, so we will delete those. You did come up with a number of situations in that other discussion which can speak to the issue of identifiability in *privacy* situations, and yes, the policy should give those as examples -- thank you for bringing them up. When it comes to *publicity* rights though -- a completely different concept -- those are generally only an issue in very commercial contexts (i.e. using someone's likeness to promote a product or something along those lines), sort of a personal version of a trademark, and use in Wikimedia projects virtually never is an issue, so no we are not failing our legal responsibilities. If there ever is an issue, we simply remove that one particular use -- but as long as there is a viable non-infringing use for the photo, we keep it on Commons overall. This policy is more geared towards privacy rights since that is the only thing which truly affects Commons; if by some weird twist an article on Wikipedia is deemed an advertisement, that gets fixed on Wikipedia. Privacy rights are a concern, yes, but publicity/personality rights, hardly ever, at least here. 65.26.25.5 21:23, 23 September 2012 (UTC)

No, I don't think that WMF has to worry about legal liability for others using Wikipedia for ads. A car company ran an ad saying something like "what baseball player made his 5000 home run in 1980? Joe Jenkins. Like Joe Jenkins, our sedan can go the distance..." They lost when Joe Jenkins sued. And yet we shouldn't warn on every page that use of facts from this page could violate publicity rights. Nor should the oats article have a warning pointing that claims that oats lower your cholesterol may not be used on your oat-containing food advertisements in the US. If you're trying to straighten out other's confusion, why are you trying to introduce publicity rights to this essay on personality rights?--Prosfilaes (talk) 02:39, 24 September 2012 (UTC)

Publicity rights is the same thing as personality rights. Did you mean privacy rights in that last references (since that is what this essay is primarily about, excepting the "commercial use" column in the table)? Carl Lindberg (talk) 07:09, 25 September 2012 (UTC)

I created an essay, Commons:Personality rights.

About the above:

• The cases I cited on this talk page were not privacy but publicity situations.
• Copyright issues are irrelevant. Being non-free or free is irrelevant. Being noninfringing or not is irrelevant.
• I never proposed all of those warnings (oatmeal, etc.) and no one else did, either, so we don't disagree on that.
• Privacy rights are already covered and I didn't propose for that.
• We agree on publicity being the issue but violations likely occur more often than we realize. Stating a polcy or guideline would educate more editors.
• This isn't only about ads run elsewhere using Wikimedia content. This is also about content in Wikimedia being edited into a functional ad, not called such but serving as one, and in that case a violation of a publicity right exposes the Foundation. Having a policy/guideline reduces that risk, at least probably by mitigating the damage.

Nick Levinson (talk) 17:10, 25 September 2012 (UTC) (Corrected a link & spacing: 17:17, 25 September 2012 (UTC))

A new essay, Commons:Nonphotographs, video, and inaccuracies, by me, has been added. The shortcuts to it are COM:NONPHOTO and COM:NONPHOTOS. I plan to link the guideline to the essay in about a week, but I'll wait a week for any comment. Thanks. Nick Levinson (talk) 15:36, 23 October 2012 (UTC)

I'll believe it exposes the Foundation to risk when someone from the Foundation says so. Nor do I see why having a proposal saying "if you do vandalize Wikimedia for your financial gain, bad things could come to the WMF" is going to help anything.--Prosfilaes (talk) 11:34, 24 October 2012 (UTC)

At this time, it's just an essay. I didn't write the law; legislatures and others did and sources agree. The statement is not anywhere as vague as suggested ("'if you do vandalize Wikimedia for your financial gain, bad things could come to the WMF'"). Because it's more specific, it educates editors and should. It may be that existing procedures, such as against editing with a conflict of interest, reduce (not eliminate, but reduce) the Foundation's liability for damages, but those procedures don't all prevent creating an ad within Wikimedia and disguising it as noncommercial content and thereby leading to a right-of-publicity issue. Editors who seek to be cautious will likely find protective advice useful, when it's specific enough to inform. Nick Levinson (talk) 15:31, 25 October 2012 (UTC) (Corrected my template and syntax errors and misspelling and clarified: 15:40, 25 October 2012 (UTC))

The only shortcut to the new essay is COM:NONPHOTO; the other was deleted. Nick Levinson (talk) 21:14, 27 October 2012 (UTC)

Linked. Nick Levinson (talk) 16:17, 6 November 2012 (UTC)

propose clarifying identifiability

I propose to edit this guideline to explain that identifiability means merely that one third party being able to identify the person is enough for identifiability; it is not necessary that many people be able to identify her or him, as with a celebrity. In one case in the U.S., a person claimed identifiability even though the picture was only from behind her and the judge permitted the case to proceed to trial on that issue, meaning that the judge considered it not impossible that her claim was true, but I don't know if the legal issue was tested at an appellate level.

I am not a lawyer.

The proposed addition is approximate and would be the first subsection in the Legal Issues section. It follows:

===Identifiability defined===

Identifiability exists even if only one person other than the subject and the creator of the image or likeness can identify the subject.

I'll wait a week for any comment. Thank you. Nick Levinson (talk) 20:02, 15 September 2012 (UTC)

Nick

• Please provide a verifiable citation to the case you mention.
• Please see the discussion above regarding the need for interaction between the level of identifiability and the expected degree of privacy accorded to the subject matter.
• I would suggest "identifiability can mean" Dankarl (talk) 20:46, 15 September 2012 (UTC)

It would take too long to find a source for that case as it was some years ago, but the more general point that as a matter of law a full face is not necessary to identify someone is probably sourceable, if we need a source for that. As I'm not writing an article but proposing a policy and most policy points are not sourced but are decided, which legal points do we need sources for in order to proceed with this? If it's not a long list, I can probably get sources for some or maybe all of them.

Privacy is not the only concern and therefore not the only test. Publicity is a separate and nonantonymous concern, albeit not one arising from most use within Wikimedia projects but only afterwards in reuse.

I didn't find the particular discussion to which you referred, although several discussions on identifiability generally seemed to wallow around a bit.

The language you propose raises a question: By implication, it says that a single third person identifying the subject may not be enough for identifiability. Someone raised a question of whether an intimate should also be excluded along with subject and creator. Is that who you meant by introducing ambiguity or was there some other category of recognizers?

I favor caution in favor of protecting people's rights. We can usually figure out other ways of educating readers about subjects, so that we need not be conflicted about privacy and publicity rights.

Nick Levinson (talk) 21:14, 16 September 2012 (UTC)

Please find the case before changing policy. There's always enough time -- just don't change the policy until we can back it up with cites. That text is much too restrictive, in my opinion. In some cases, one other person being able to identify the person may be enough, but in other situations, it may not be. Your proposed addition has none of that context; it should hedge a lot more than that. Without court case cites, I oppose this change. Also, being allowed to proceed to trial is not at all the same as a final ruling -- that is just saying the question is close enough to examine in more depth. I think it is true that the face does not necessarily need to be shown, and it may also matter what kind of private facts are being revealed -- if they are the type of facts that are damaging if people close to the person find out, such an interpretation may be accurate, but if they are facts which family already know but are only damaging if recognized by the greater public, it's possible a different standard would be used. All aspects of the situation should be considered. We should be basing the policy on the law, so either court cases or balanced legal opinions should be used. Carl Lindberg (talk) 22:14, 16 September 2012 (UTC)

I may have accidentally found the case, although this one is older than I would have thought, so it may not be the same, but it'll do, especially since this one is at a newer stage and from the State's highest appellate court, resolving doubts we both had. And several judicial holdings in New York State summarized in annotated statutes relied upon by attorneys are on point. All of these concern N.Y. Civil Rights Law, sections 50 and 51, which create a right of publicity in the State. I have not compared California's law, which governs Wikimedia's servers, but I'm not expert enough to be sure the laws of the 49 States besides California don't also govern in some factual situations.

• "[A] photograph of the nude plaintiffs, mother and child, which showed their bodies full length as viewed from a position behind and to the right of them, and which did not show their faces, nonetheless revealed sufficiently identifiable likenesses to withstand defendant's motions for summary judgment." In State court: Cohen v. Herbal Concepts, Inc., 63 N.Y.2d 379, 482 N.Y.S.2d 457, 472 N.E.2d 307 (N.Y.Ct.App. (State's highest court) 1984), as annotated in New York Consolidated Laws Service: Annotated Statutes with Forms (hereinafter CLS) (LexisNexis, 2001 (ISBN 0-327-10666-2)), Civ. Rts. L., sec. 51, main vol., [anno. group] 20, 2d paragraph.
• "[E]ven though reproduction of ... picture was of poor quality and ... faces were small", the people in the picture could assert claims, because the picture was of a World Series baseball team and thus the people were identifiable. In State court: Shamsky v. Garan, Inc., 167 Misc.2d 149, 632 N.Y.S.2d 930 (Sup. 1995), as annotated in CLS, Civ. Rts. L., sec. 50, main vol., [anno. group] 12, 2d paragraph, & sec. 51, main vol., [anno. group] 13, 5th paragraph.
• "[T]he question is whether the figure is recognizable, not the number of people who actually recognized it." The "picture" must be "a clear representation of the plaintiff, recognizable from the advertisement itself." The "fact that she no longer looked like she did in the picture" did not matter. She had been in a 1922 silent movie and the case was decided in 1971. The advertised product's first 4 letters matched her name. In Federal district court, ruling on State law: Negri v. Schering Corp., 333 F.Supp. 101 (S.D.N.Y. 1971), as annotated in CLS, Civ. Rts. L., sec. 50, main vol., [anno. group] 12, 3d paragraph, & sec. 51, main vol., [anno. group] 13, last paragraph.
• The "portrait or picture" must "at least approach ... likeness" and must be "a clear representation of a person whether by photograph, statue, imitation or word painting", thus "more than mere picture of scene suggested by a play". In Federal district court, ruling on State law: Levey v. Warner Bros. Pictures, 57 F.Supp. 40, 62 U.S.P.Q. 376 (D.C.N.Y. 1944), as annotated in CLS, Civ. Rts. L., sec. 51, main vol., [anno. group] 7, 5th paragraph.
• Even though "the portrait ... is not an exact reproduction, slight changes having been made in the pose ... so as to adapt it to the purpose of the advertisement", the portrayed person still has a cause of action. In State court: Loftus v. Greenwich Lithographing Co., 192 App.Div. 251, 182 N.Y.S. 428 (App. Div. 1920), as annotated in CLS, Civ. Rts. L., sec. 50, main vol., [anno. group] 13, 1st paragraph. But, the "[f]act that magazine had touched up photograph because, in the original, the subject's hair blended into the background did not give rise to cause of action for ... invasion of privacy where the magazine merely made an arbitrary delineation of the subject's coiffure which provided an inoffensive and natural appearing result." In State court: Bass v. Straight Arrow Publishers, Inc., 59 App.Div.2d 684, 398 N.Y.S.2d 669 (1st Dept. 1977), as annotated in CLS, Civ. Rts. L., sec. 50, main vol., [anno. group] 13, 2d paragraph.
• When a "look-alike's face" that "seems indistinguishable" from a real celebrity's face and advertisement's "context ... clearly implies that he or she is real celebrity", the "question ... is whether audience at whom advertisement was aimed" would have been fooled or not. In Federal district court, ruling on State law: Allen v. National Video, Inc., 610 F.Supp. 612, 226 U.S.P.Q. 483 (S.D.N.Y. 1985), as annotated in CLS, Civ. Rts. L., sec. 51, main vol., [anno. group] 20, penultimate paragraph.
• The "portrait or picture" need not be a photo as long as it is "recognizable as likeness". In Federal district court, ruling on State law: Ali v. Playgirl, Inc., 447 F.Supp. 723, 3 Media L.R. 2540, 206 U.S.P.Q. 1021 (S.D.N.Y. 1978) (criticized in WR Acorn Eng'g Ass'n v. Ehrlich-Rominger Architecture 2000 U.S.Dist. Lexis 723 (D.C. Ore. 2000)), both as annotated in CLS, Civ. Rts. L., sec. 51, main vol., [anno. group] 7, 6th paragraph.
• "Publication of a fictitious biography of a public figure constitutes an unauthorized exploitation of his personality for purposes of trade". Starting and ending in State court and in Federal Supreme Court in between: Spahn v. Julian Messner, Inc., 59 App.Div.2d 684, 18 N.Y.2d 324, 274 N.Y.S.2d 877, 221 N.E.2d 543 (N.Y.Ct.App. 1966), vacated, 387 U.S. 239, 18 L.Ed.2d 744, 87 S.Ct. 1706 (U.S.Sup.Ct. 1967), on remand, 20 N.Y.2d 752, 283 N.Y.S.2d 119, 229 N.E.2d 712 (N.Y.Ct.App. 1967), on rehearing, 21 N.Y.2d 124, 286 N.Y.S.2d 832, 233 N.E.2d 840, 30 A.L.R.3d 196 (1967), appeal dismissed, 393 U.S. 1046, 21 L.Ed.2d 600, 89 S.Ct. 676 (U.S.Sup.Ct. 1969), "and remittitur amd", 20 N.Y.2d 759 (N.Y.Ct.App. 1967), as annotated in CLS, Civ. Rts. L., sec. 51, main vol., [anno. group] 16, 1st paragraph.
• A magazine "article relating to clothing styles" and having a photo of an actor was not protected by the First Amendment against the actor if the article really served the purpose of an advertisement because of "covert agreement between magazine publisher and clothing manufacturer." In Federal district court, ruling on State law: Grant v. Esquire, Inc., 367 F.Supp. 876 (S.D.N.Y. 1973), as annotated in CLS, Civ. Rts. L., sec. 51, main vol., [anno. group] 20, 13th paragraph.
• A "manikin" is a "portrait or picture". In State court: Young v. Greneker Studios, Inc., 175 Misc. 1027, 26 N.Y.S.2d 357 (1941), as annotated in CLS, Civ. Rts. L., sec. 50, main vol., [anno. group] 9, 4th paragraph, & sec. 51, main vol., [anno. group] 7, 3d paragraph.
• If an author's "research article" was published without consent and if that publishing promoted someone else's product sales, that violated the author's right of publicity. In State court: Ravich v. Kling, 17 Misc.2d 683, 187 N.Y.S.2d 272 (1959), as annotated in CLS, Civ. Rts. L., sec. 51, main vol., [anno. group] 21, only paragraph.

I went a little beyond what you requested, but all of these have interesting implications for Wikimedia's projects, if New York law applies to them or insofar as California law is essentially similar.

That First Amendment protection can be inapplicable to a magazine article because it was secretly an advertisement is a principle that could be applied to content in Wikipedia articles and not just to reuse. I understand that a study found that many articles about companies are edited from IP addresses that belong to the respective companies, meaning that thousands of editors are almost certainly not being identified as having conflicts of interest and are probably editing articles consistently with promotional or advertising principles. Imagine that company X adds to the company X article a picture of a smiling person using the company's product. That's likely a violation, albeit more by the company than by Wikimedia. Wikimedia then would be dependent on editors' diligence in keeping conflicted editors from editing and on not accepting a payment from company X, which it might be remiss on regarding IP-based editors if it has no procedure for identifying IP-related conflicts of interest.

Nick Levinson (talk) 14:45, 18 September 2012 (UTC) (Corrected for what I meant by my erroneous "newer" & clarified points: 15:31, 18 September 2012 (UTC)) (Corrected by striking out legal citation that belonged to a different case (my error) & without affecting discussion of case, for which correct citations still remain: 23:03, 22 September 2012 (UTC))

Sounds like there may be enough here for a help page Commons:Identifiability. A summary needs to go in the guideline here, but it's often easier to do a summary when there's more detail available elsewhere (and the guideline can point to that page for more details). Rd232 (talk) 16:14, 18 September 2012 (UTC)

I'm considering it and have started a draft, maybe as an essay, so that others with legal skill can tighten it up. The guideline itself, though, should not just summarize but should state what editors should do, and that should be cautious and steer clear of a person's rights, not try to go right up to the dividing line between lawful and unlawful. Nick Levinson (talk) 16:50, 19 September 2012 (UTC)

I created Commons:Personality rights as an essay and will probably create a redirect from Commons:Identifiability to the essay on another day. Nick Levinson (talk) 17:01, 25 September 2012 (UTC)

Redirected. Nick Levinson (talk) 21:21, 27 October 2012 (UTC)

Personality rights and applicable jurisdiction

{{Personality rights}} might give the impression that only the local laws of the reuser are to be considered. This is important for the reuser, but does not protect the subjects of the photo against reuse in countries where (some) personality rights do not apply. I would like it stated very clearly that also local laws where the image was taken should be respected.

I, as a photographer, cannot use my copyright to add any restrictions on the use of a photo, because that would make the photo unfree. On the other hand, under Finnish law, I might be prosecuted for offering photos of identifiable people for commercial use without their consent (if photos are used where there are no such restrictions in local law).

The local law where the photos are taken should be applicable, as it effects reasonable expectations of subjects and possible defamation will have effects in the country of origin. The fact that a reuser might not risk being prosecuted is not the same as the the act not being an offence.

The warning is of course not binding, but I think making the reuser aware of the broader meaning of "applicable laws" might be important to the subjects of the photo, the photographer and the uploader and do no harm.

--LPfi (talk) 10:56, 17 September 2012 (UTC)

• I think there's a much more general problem: there's a table of country-specific consent requirements, but no explanation as to how they affect the general policy stated above. Suppose a country requires consent to publish images of people in public places. Does any picture featuring people taken in such a country require the {{consent}}-tag? Or does every picture uploaded from such a country require that tag? As stated, the policy could even be read to suggest that any picture that would be pulished to such a country would require tagging (although that would be dumb.)

A more consistent policy statement might be, for example: "Wikimedia Commons policy requires consent documentation for images taken in, or uploaded to Commons from, countries which require such consent for photography, or for (commercial/non-commercial) publication." 24.177.121.137 04:54, 23 September 2012 (UTC)

• It is very much source-country dependent I would think. I have had similar "reservations" regarding, the photos I have taken of identifiable people - even in public - in Denmark. In July 2011 I created a "personal" template, to cover this predicament. Through a number of updates and changes it was yesterday "formalized" a bit more, by being put it into {{Denmark no consent}}. Something similar might be useful for the case of Finland, since Finland and Denmark, have somewhat similar legislations regarding photography of individuals in public space. In kind regards, heb [T C E] 10:42, 11 October 2012 (UTC)

Dead people?

Currently, this guideline doesn't distinguish at all between living and dead people. This differs from the WMF board resolution, and from the guideline's checklist summary, both of which restrict themselves to living people. These inconsistencies have already led to some confusion.

The guideline does start off saying that it is about "taking pictures" of identifiable people, which would suggest it's mainly about living people, although it later goes on to discuss uploading photographs in general. This includes photos from external sources, which often hold many historical images of people who are now dead. Including dead people in the guideline's scope seems to exceed what the WMF has called for, and will probably overcomplicate the parts dealing with subject consent.

I suggest we revise the guideline to say that it applies to pictures of identifiable living people almost throughout. A brief mention of the recently dead in the moral issues section could be useful, though. The parts dealing with the international legal situation should also cover pictures of people who are no longer living, where this reflects local law. --Avenue (talk) 14:43, 3 December 2012 (UTC)

Agreed. I'm considering actually whether a radically shorter version of COM:PEOPLE might not be better (much clearer in policy terms), with as much as possible moved to a subsidiary "how to" page. Rd232 (talk) 12:48, 4 December 2012 (UTC)

Certainly the guideline needs something "radical" done with it. It doesn't even define what "identifiable" means (see discussion above from over a year ago) -- it isn't just a black and white issue. I don't think moral/ethical issues are handled at all well by the Commons community, which essentially defaults to an "unless it is illegal, don't restrict me" approach. My feeling is there are some areas where WMF should pay an expert or consultancy to draft the rules here and just impose them on the community. After all, it is their site, not ours. Colin (talk) 13:52, 4 December 2012 (UTC)

I didn't mean to imply that the guideline is fine apart from the living/dead people issue. I agree the guideline needs to be a lot clearer. Identifiability, consent, which countries' laws should apply - these are all thorny issues. I picked on the living/dead issue to start with because it seemed relatively simple (although where it interacts with the thornier issues it quickly starts to get murky). --Avenue (talk) 14:05, 4 December 2012 (UTC)

So how long after someone is dead are they "fair game"? And does that apply to all shots? Surely there are photographs for which the subject expected privacy (so did not want the photograph taken, never mind released) or where the photographer was expected to not release the photograph widely, and where there is no justification for ever posting on Commons. Sort of a "public interest" test. A voyeuristic/paparazzi shot doesn't become ethical when the subject dies. A boyfriend/girlfriend intimate photograph shouldn't be exposed to all when one of them dies. The only thing that happens when someone dies is their legal rights change and their ability to give consent disappears. Surely the ethical issues remain? Colin (talk) 15:14, 4 December 2012 (UTC)

If you ask different people those questions, I'm sure you'll get different specific answers, and likewise various legal jurisdictions will have different rules. For instance, the law in Florida requiring subject consent (section 540.08) says that this is no longer necessary 40 years after the subject died (see clause 5). (Note that under their regime, the subject's spouse or any of their surviving children can give consent after the subject dies.) There is nothing sacred about that figure of 40 years; it is just the answer Florida's legislators settled on as providing the best practical solution to these issues in their jurisdiction.

The WMF board resolution gives us a clear mandate to strengthen and enforce this guideline as it applies to media portraying living people. While pictures of dead people aren't unimportant, I think we should put the issues surrounding them to one side until we have successfully addressed the more pressing problems regarding living people. In the meantime, perhaps we should include a general proviso stating that despite the guideline not explicitly covering pictures of people who are no longer living, similar issues may apply to them. --Avenue (talk) 00:21, 5 December 2012 (UTC)

I think the logic of the policy is to use what the local legal requirement is. Elsewhere (I forget where) I've seen 15 years after death. Obviously this is an area we need to work on getting more info, and maybe establish a backstop "if in doubt, assume X years after death". Rd232 (talk) 10:07, 5 December 2012 (UTC)

The guideline actually states (in the Legality section) that images are unacceptable if they are arguably illegal where they were taken, uploaded from, or in the USA (because that is where the Commons servers are). I think that goes some way beyond our practice for copyright issues, in which we don't usually worry about the law where the uploader sits.

The USA bit also seems problematic because I gather there is no federal law on such issues, and the rules in each state are therefore different. The US servers are in Florida, so "40 years after death" might be a logical default. But we also have servers in the Netherlands (including I think one media backup server),[2] and we don't usually worry about the law there. I've also seen people say that the location of the WMF's headquarters (now California) is what really matters, and the recent WikiTravel litigation was filed in California. At least for celebrities, some personality rights persist until 70 years after death in California (see California Celebrities Rights Act). --Avenue (talk) 12:11, 5 December 2012 (UTC)

Redraft

I've started a redraft at Commons:Photographs of identifiable people/2013. So far I've left out the "guidance" section, and it's possible it might be better left out (put in a subsidiary guidance page instead, rather than in the main policy). Comments, suggestions? Rd232 (talk) 18:35, 5 December 2012 (UTC)

Per the discussion at Commons:Requests for comment/images of identifiable people, could we add a sentence recommending (but not requiring) the use of the {{consent}} template to indicate the consent status of photos of identifiable people? Kaldari (talk) 03:59, 10 December 2012 (UTC)

It's mentioned under point 3 "evidence" of the "subject consent" section. Do you want to make it more prominent, or change the wording? Feel free to suggest or tweak. Rd232 (talk) 15:07, 10 December 2012 (UTC)

Partygoer examples are confusing and misleading

We current state that it's OK to post photographs of "partygoers at a large private party where photography is expected", but it is not OK to post photographs of "partygoers at a private party where photography is not permitted or is not expected". How can someone possibly determine whether or not photography is "expected" at a private party? Regardless of whether it is expected or not, a private party is an event in a private place. We should be requesting consent for such images, not assuming that everyone who goes to a party is OK with giving up their privacy.

I would like to suggest either removing both of these examples or replacing them with a single example stating that it is not OK to post photographs of partygoers at private parties without consent, otherwise these examples undermine the entire guideline. Kaldari (talk) 05:41, 28 December 2012 (UTC)

I think it would be better to replace the first with another example that makes the same essential point more clearly (that public situations can occur in privately owned places), e.g. a press conference or a red carpet event like the Category:Time 100 Gala. It should be worded more carefully too. "Privately owned" is clearer than "private", for instance.

The second example is also unclear. I think the main idea there is that it's not okay to post photos taken where this would violate the subject's reasonable expectation of privacy, even if the situation itself is not especially private. I think an example illustrating this would still be worth including, although I can't think of a better one at the moment. --Avenue (talk) 14:20, 28 December 2012 (UTC)

OK, I've tried to adjust both of the examples per your suggestions. Hope that sounds better. Kaldari (talk) 22:57, 28 December 2012 (UTC)

That's much better IMO. I've reworded one of them slightly. --Avenue (talk) 09:26, 29 December 2012 (UTC)

Anonymizing the image

I've removed the pixelated obese girl and the advice on using such techniques to "anonymise" an image. The original unpixelated image is on the internet and by dragging & dropping the pixelated version onto Google Image Search it is trivial to locate it. This highlights the need to ensure privacy concerns are dealt with at upload time rather than applying a sticking plaster later. The internet never forgets. Another technique for anonymising images by the use of blank bands over the eyes, once used in dusty old medical textbooks and journals (never intended to be seen by the general public), is no longer considered acceptable by medical publishers. -- Colin (talk) 15:40, 28 December 2012 (UTC)

I agree with that anonymising is highly problematic and has far too often been used without due consideration of the risk that identification remains possible. Another technique for anonymising images by the use of blank bands over the eyes, once used in dusty old medical textbooks and journals (never intended to be seen by the general public), is no longer considered acceptable by medical publishers. - a source for that would be very helpful. Rd232 (talk) 00:46, 29 December 2012 (UTC)

We had the black band technique used in television. Our president thought it was a bad idea and got himself showed "anonymized" in that way. I think the episode put an end to the bands over here. --LPfi (talk) 01:29, 30 December 2012 (UTC)

• "You must seek the patient’s written consent to publication in the BMJ if there is any chance that he or she may be identified from a picture, from its legend or other accompanying text. Patients are almost always willing to give such consent. We no longer publish pictures with black bands across the eyes because bands fail to mask someone’s identity effectively." BMJ
• "Informed consent should be obtained if there is any doubt that anonymity can be maintained. For example, masking the eye region in photographs of patients is inadequate protection of anonymity." Uniform Requirements for Manuscripts Submitted to Biomedical Journals - Ethical Considerations in the Conduct and Reporting of Research - Privacy and Confidentiality

The "Uniform Requirements" guidelines are pretty much the standard for all medical journals, and the BMJ is a leading UK medical journal. There are plenty other examples, but they just repeat this. In my opinion, patient images with black bands across the eyes are a signal that the image is not suitable for Commons. The patient may have not given consent at all (with the doctor assuming the bands were effective) or wished some degree of privacy however feeble, and probably only consented to appearing in a medical journal. Colin (talk) 18:17, 30 December 2012 (UTC)

I've added a subsection on the problems with anonymising an image, with some examples. -- Colin (talk) 22:04, 30 December 2012 (UTC)

The examples seem misleading to me. Many viewers will be very familiar with these artworks, and some could probably identify them from a small portion of the image, maybe even of just the background. The poses are iconic, the style of a painting can be a giveaway, and so on. This is very different from the situations we would typically deal with here, of photographs of real people who are typically not portrayed in a characteristic pose or situation. Better examples might be produced by applying these treatments to the photo of Larry Sanger from the lead section, for instance. --Avenue (talk) 22:29, 30 December 2012 (UTC)

For any example to work and sink in, it needs to be someone who you would recognise if only the attempts at anonymising hadn't been done. Nobody other than the nerdiest Wikipedia geeks would recognise Sanger even without pixelation or cropping. The examples are illustrative rather than experimental proof. It is difficult to find a free image of someone that everyone would normally recognise. The issue with identification isn't that some stranger might recognise you. Who gives a damn about them. The issue is that your family, your boss, your potential employer, your school mates, or your girlfriend might recognise you. Commons:Deletion requests/File:Barceloneta Girl.jpg (NSFW) is an example where the Flickr link and Google Image Search both identify the "anonymised" subject. Anyone on holiday with that girl at Barcelona (especially if the image was more recent) would identify her quite easily despite the pixelation: the headband, sandals, location and date all make identification possible. -- Colin (talk) 23:12, 30 December 2012 (UTC)

My concern about anonymised paintings is that paintings are too different from photographs, so they don't illustrate the issue well. My thought with the Sanger pic wasn't that he was widely recognisable - it was that anyone reading through the page should remember seeing it at the top of the page a few minutes ago. But we also have plenty of photos of celebrities that most people would recognise, and that would make useful examples, e.g.

--Avenue (talk) 13:50, 1 January 2013 (UTC)

I can see the advantage to a photograph compared to a painting. But it is quite hard to find one that illustrates many of the issues of just hiding (part of) the face of the subject. We need readers (from all countries) to be familiar with other parts of the image (the body, clothes, surrounds) in order to be able to put themselves into the position of some who may then recognise a non-notable person they are familiar with (or are). To me, it doesn't matter if the Mona Lisa example is a bit over-obvious. Indeed the humour of such a pathetic attempt to hide the Mona Lisa helps the lesson sink in. I'm not opposed to a photograph but the problem with that Brad picture is that it is all too easy for someone to make it difficult for nearly anyone to recognise him, but then all we'd be left of is a picture of a suit. We don't tend to have problems with people taking pictures of folk wearing suits. Now if you can get a picture of Brad with his clothes off, there might be more interest :-). Colin (talk) 16:05, 1 January 2013 (UTC)

I wouldn't mind if just one of the examples was over-obvious, but when all three of them are, I feel it diminishes the credibility of the whole thing. It makes it seem like we can't find any real life examples. Sure, we can't display truly problematic images here, but I think we should try to find some realistic examples if possible. --Avenue (talk) 23:43, 2 January 2013 (UTC)

I like the examples using artwork. I also don't recognize the person in Avenue's example, and wouldn't even if there were no black bands. Avenue's example (assuming the file name gives the person's correct name) assumes that everyone recognizes American actors, which I think is less plausible than assuming that people will recognize the most famous couple of paintings or statues in the history of the world. WhatamIdoing (talk) 19:48, 18 January 2013 (UTC)

Another redraft

New draft: User:Colin/People

I think we all agree the current guideline is inadequate. The guideline has gone largely unchanged since its creation five years ago, despite a WMF resolution that it should be strengthened and better enforced. There have been several attempts at drafting a new version (one here and another more recently here). Some problems with the existing version are:

• The layout and structure are poor, with information scattered about and repeated.
• The country-specific table dominates and isn't particularly useful on its own -- the sub-page with the details is necessary to make any informed decision about country-specific laws.
• The guideline doesn't discuss in detail consent or identification, both of which have degrees and nuances and should be considered in the context of the image.

Other problems I'm seeing with recent deletion discussions are

• A confusion between copyright licence restrictions (permission required, commercial use) and the subject's rights (permission required, commercial use), which are completely separate. I've seen admins get the two confused.
• Amateurish attitudes towards dealing with privacy and responding to images with legal or ethical concerns. Pixelating the face of a person who is the clear subject of a photograph and who's photograph should not have been taken or published without consent is, in some ways, similar to filing off the serial number of a bike one has just stolen.

To resolve these problems, I've created yet another draft: User:Colin/People. I would appreciate your comments - put them here rather than on some other talk page. I think the problem with drafts is that they can get bogged down with discussions until deemed 100% perfect. But this is a Wiki so I'd much prefer if we got the material onto the guideline page and worked on it there. If this draft is considered an improvement, even if it has some issues, I'd like to propose we move it over to the guideline and then clearly tag any contentious sections as such. For these contentious sections, it will then be apparent to readers that the advice there is not yet stable. A lot of the existing text is still present, just better organised and presented IMO, as well as some new sections of consent and identification.

-- Colin (talk) 13:32, 2 January 2013 (UTC)

• I made a couple of edits that I hope will be taken as friendly. I think there is also a need to clarify in the "Consent" section that in countries such as the U.S., there is no need for explicit consent for photos taken in a public place. Yes, we say that elsewhere, but it should be restated or people will think these sections contradict one another. - Jmabel ! talk 17:02, 15 January 2013 (UTC)

I'm reluctant to repeat and summarise the complex "whether required" issue that is only covered an inch or so further up. But I agree we need to make sure it doesn't contradict or give the impression consent is always required. Does this edit help? If not, have you got a suggestion? -- Colin (talk) 22:38, 15 January 2013 (UTC)

• I made a few minor wording changes, but otherwise it looks good. The organization of the material is a huge improvement over the current guidelines. Thanks for working on this. Kaldari (talk) 22:25, 15 January 2013 (UTC)

Thanks for the edits, both of you. Colin (talk) 22:38, 15 January 2013 (UTC)

• I think it is very well done, and would support moving over to the guideline on the understanding that we would continue to work on contentious sections (if any). --Skeezix1000 (talk) 22:40, 15 January 2013 (UTC)

 Comment I think in some respects it is quite good, but it is quite wordy and hard to scan. It reads more like essay than policy; it's important that policy can be grasped as quickly and easily as possible. Compare the formatting (headings, bullets) of my Commons:Photographs of identifiable people/2013 draft. Also Commons:Photographs_of_identifiable_people/2013#Scope is quite important. I seriously think my draft is a better starting point for structure and clarity, and that your version should be merged into mine, keeping any additions/clarifications as brief as possible and possibly tucked away in explanatory footnotes. That would mean importing especially the Identification section and the ones below it, but also some points from the sections above it. Rd232 (talk) 00:31, 16 January 2013 (UTC)

• FWIW, I prefer Colin's. Yes, it reads like an essay, but it's a highly readable essay. I'd be more inclined to solve the issue of clarity by adding a "nutshell" summary at the front of Colin's version. - Jmabel ! talk
  • I strongly disagree that a "highly readable essay" is suitable for a policy document. For one thing, don't forget that non-native speakers will be reading the English version too - especially (but not exclusively) if there's no translation in their language(s). Generally, if things need to be said in essayish form in relation to policy (and yes, this is often helpful clarification for those willing and able to read it), they're put into a separate essayish document which is appropriately linked. Rd232 (talk) 07:45, 16 January 2013 (UTC)
    • I think the essay form is better for the purpose of this document. There are so many shades of grey that a short easily grasped policy cannot cover even the most important issues (which is clear from your draft, where policy is intermingled with all kinds of considerations). That said, we might want a short policy page, which just states what laws we follow and what moral issues we find important, and leave all the discussion to an essay-like guideline. --LPfi (talk) 08:42, 16 January 2013 (UTC)
    • Rd232, as you know, I was aware of your draft and incorporated some of your text/changes into the one I'm proposing. And some of the issues and problems with people-images you raised on at the RFC influenced this draft. I wanted to take a different approach. I didn't want a terse bullet-point guideline which might be suitable only for experienced admins such as yourself. That's fine for when everyone is up-to-speed and just needs some rough notes to remind them of the rules. But this is a complex subject and new editors need more help to get their heads round the issues. The other draft isn't complete and doesn't address many of the problems I wanted to cover or clarify in the draft I made. I don't think there is anything in your draft that isn't in mine, or addressed in other guidelines. The issue of scope is covered in a different way. IMO it is vital the guideline is readable and not a stilted set of points. The text needs to introduce ideas and develop them. Of course it is a draft and there may be room to trim some fat if folk agree some text is more wordy than needed. And I agree with Jmabel a nutshell (both at the top and perhaps for some sections) would be good. I may be able to enlist one of Wikipedia's fine writers to help copyedit and polish and tighten the prose, but I dont' think it is at all bad as it is. I suspect I won't convince you, Rd232, because you're naturally invested in your own approach. Let's see what other people think. -- Colin (talk) 08:52, 16 January 2013 (UTC)
      • Well obviously we are both invested in our own approaches :). Nonetheless, I strongly disagree that your essay approach is right. You think my bullety/highlighty approach assumes admin-level knowledge? Not at all, it just tries to give the key points clearly, without in any way giving false certainty. I think your essay approach assumes a willingness to engage and learn that the vast majority of readers will not have. Fine, provide it for those willing, but don't force it on everyone as part of the key policy document. And frankly, why should users have to learn legal terms like "right of publicity" when all they need to know is "if it was taken in a private place, it needs permission, if it was taken in a public place, it depends on local laws - list here - (oh and for some commercial uses there are additional considerations)"? You really cannot underestimate readers' attention span on the internet. Rd232 (talk) 13:09, 16 January 2013 (UTC)
      • The guideline is needed for both eduation of users planning to upload and also as somewhere that experienced folk point to when involved in, for example, a deletion discussion. It needs to be readable by those who will be affected by certain laws they've never heard of before, but also usable by folk trying to apply the rules. Your draft also covers the right of publicity, yet intermingles it with the quite separate privacy issues that are controlled by private/public place concerns -- which don't affect the right of publicity at all. I'm afraid I find the text of your draft to be extremely confusing, as evidenced by the comments at the RFC. Please, let's see what other folk think because I don't think we will persuade each other of the merits of the others's draft. -- Colin (talk) 13:38, 16 January 2013 (UTC)
        • General education about laws should be but on Wikipedia, not Commons - the page can and does link there for more info. My draft is clear on the public/private place concerns because that's the basic distinction - and one users can fairly easily grasp in principle (even if there can be unexpected legal complexities of definition). "Some commercial uses" are clearly distinguished as potentially needing specific consent; they are not "intermingled" in the page. Yet the English Wikipedia article en:Personality rights actually does mingle them - maybe you should address that before trying to write a single Commons policy page that tries to do the job of splitting that article into publicity rights and privacy rights. Rd232 (talk) 16:02, 16 January 2013 (UTC)

I found Colin's version to more clearly enunciate the principles and the rationale. I didn't actually find it too wordy or essay-like. That's not to disparage Rd232's version, but of the two, I preferred Colin's. I appreciate that it may just be a personal preference, but that's my two cents. --Skeezix1000 (talk) 16:42, 16 January 2013 (UTC)

It appears that the discussion has wound down, and there is consensus to move the draft onto the guideline page. The more discrete issues appear to have been addressed (and in any case were not fundamental to the draft). Rd232 raised broader concerns, but no one else seems to share them and he acknowledged in his January 16 comment that the guideline could be adopted during his wikibreak. --Skeezix1000 (talk) 21:36, 23 January 2013 (UTC)

I agree. I wanted to allow plenty time for folk to comment but this doesn't appear to be a guideline that gets much traffic on Commons hence the low participation here. Colin (talk) 22:38, 23 January 2013 (UTC)

Agreed. You did leave lots of time, and you promoted this discussion at the Village Pump, the RfC and on people's talk pages. --Skeezix1000 (talk) 16:39, 24 January 2013 (UTC)

I unfortunately haven't been able to put as much time into this as I would have liked (mainly because it requires a longer stretch of connected effort, than I have had the room for lately :(). Overall I much prefer the Colin's draft (in it's current version anyway) to the current guidline. I do however miss the [[Commons:Photographs of identifiable people/2013#Scope|scope part from Rd232's draft - specifically something that clearly explains i.e. living as being more than just being alive in some jurisdictions (in Denmark for instance it extends at least three weeks following the death and up til ten years pending on the circumstances, according to a technical reply (not tried at court) from the Danish Data Protection Agency) and photographs in this context having a broader meaning also covering paintings, drawings and other forms of imagery (though this is partially shown under identification). I think adding the scope-section under legal issues would be of great benefit to the draft.

Also I miss a bit more clear distinguishing between taking (or producing per the above comment on other media forms) and publishing. Though I don't readily have a clear-cut idea how to do this right now, I think something along the lines of bolding publishing the photo from the intro text and adding a remark like this guideline simply deals with the publishing of media with potential privacy issues, specifically publishing them to Commons would do it, but I think that part needs to mature slightly :)

That stated I think that both Colin and Rd232 has done a great job with their respective drafts and I applaude the time and effort they have put into it. In kind regards, Henrik/heb [T C E] 12:17, 26 January 2013 (UTC)

It was indeed a sad thing that no people joined up on the discussion, however they have had plenty of opportunity I think. I have completed the move. If nothing else it might get people out of their buckets :) Please also see new section in the bottom of this page. --heb [T C E] 13:22, 26 January 2013 (UTC)

USA and others

I think the draft (like Rd232's draft an perhaps this page) is too USA centric. It states "normally" about things where the USA standpoint is unusual (at least looking at the country specific requirements) and "some countries" for the normal legal requirements. We have to abide to Florida/California/whatever law, but there is no reason to have it as norm otherwise.

For people accustomed to local law it is of course difficult to write from a general perspective, but I hope there is consensus about trying to.

The worst issue is the one about straight forward photos in public places. I think consent is generally required for publishing such photos, where the individual is the primary subject of the photo. I hope we could find some other example for "usually not needed". For me the obvious one is where the subject is an event or a busy public place, where it is hard to avoid identifiable individuals (with a decent resolution) - or a politician giving a speech. I am not sure what example should be picked.

--LPfi (talk) 09:07, 16 January 2013 (UTC)

I agree, but I disagree that Commons:Photographs of identifiable people/2013 is USA-centric. Read it again, maybe - I was specifically trying to be less USA-centric than the current policy. Rd232 (talk) 13:11, 16 January 2013 (UTC)

• I've tried to address the USA bias in this edit. I hope that helps. I agree that this guideline mustn't be written with the assumption the reader is based in the USA (or UK). Colin (talk) 19:39, 16 January 2013 (UTC)

Policy section

I've undone some of this edit by LPfi that introduced at "Policy" section. I think the intention was to summarise the key points, though that is normally what the lead should be doing. My main problem with it is that this is a (proposed draft) guideline and we can't just elevate it, or bits of it, to "policy" which kind of a protected term here. Are there parts of that section that should be in the lead? Or do we think a "Key points" section is needed? Or something else? -- Colin (talk) 09:43, 16 January 2013 (UTC)

Well one fairly obvious solution is to have both a policy and a guideline. The Policy would be based on my version (Commons:Photographs of identifiable people/2013) and the guideline on yours (User:Colin/People). Rd232 (talk) 13:13, 16 January 2013 (UTC)

That's not an obvious solution at all. We don't have policy and guideline covering exactly the same things. If there was a parent policy page, it would cover general legal and ethical principles that form the basis from which guidelines are developed. It would concern the rights and responsibilities of the subject, the photographer, the owner of the image, the uploader, the reusers and Commons as a host. And it should be written by the WMF, not a bunch of amateurs. Colin (talk) 13:38, 16 January 2013 (UTC)

• It depends what you mean by "exactly" the same things. English Wikipedia has some examples of clarificatory essays supplementing policy; that's really what you're trying to do here. It's helpful done that way, and unhelpful, even dangerous, done the way you want to do it.
• If there was a parent policy page, it would cover general legal and ethical principles... - no, that's fundamentally misunderstanding what a policy page is. A policy page basically tells you what to do, not why the policy is what the policy is, nor provide the general education needed to understand why it exists or contains the content it does.
• it should be written by the WMF - an even more alarming misunderstanding of what policy is. WMF does not handle content issues except where legally required; if it did, it would risk becoming legally responsible for all the content, which is not tenable. Rd232 (talk) 15:50, 16 January 2013 (UTC)

• Well this is a distraction. The point is that both drafts have a huge amount of the same words in them. Except, bluntly, one of them is readable and easy to understand and the other not. That's not a starting point from which to make one policy and one guideline. Wikipedia has a clearer distinction between policy and guideline, and the guidelines are meant to follow from policy, to show how to put it into practice in certain areas, for example. On Commons, it appears the distinction is less clear, but generally the really important stuff is policy. Generally speaking, the policy and guideline pages on Commons are vastly inferior to Wikipedia, which reflects on the lower participation levels and perhaps lower ability with English too. As for the WMF bit. Well the owners of the site determine the scope (educational content) and I believe this should extend to legal and ethical standards about what they are prepared to host. They should get lawyers and ethics experts in to draft such standards. The community can then write guidelines drawn from this. The WMF have already issued a request that we improve this guideline and we've done nothing so far. Don't kid yourself that the community have ultimate say on these things. See en:Wikipedia:You don't own Wikipedia. -- Colin (talk) 16:12, 16 January 2013 (UTC)
  • I'll reiterate, since everyone else is reiterating: adding a good "in a nutshell" executive summary at the top of Colin's version should address Rd232's issue. - Jmabel ! talk 16:51, 16 January 2013 (UTC)
  • Well, fine, if you can get your version adopted, with or without revisions in light of my comments, good luck to you. I'm resuming my wikibreak in earnest so I won't participate any further in this discussion for a while. Rd232 (talk) 18:52, 16 January 2013 (UTC)

That's very disappointing. I hope there are revisions and changes because they would reflect a community working together in a wiki towards improving the guideline. All the many previous attempts at fixing this guideline have failed, largely because of endless talkpage discussions by one or two folk. We need lots of eyes and lots of helpers. I think what I've proposed is such a huge improvement on what we have that it would be a big step forward to build upon. But I'm keen to see what other folk think. Colin (talk) 19:10, 16 January 2013 (UTC)

I've added a nutshell. Obviously it has to be brief. I'd expect the lead to be a better summary of the key points, which in turn is expanded upon by each section in the body of the guideline. Colin (talk) 19:53, 16 January 2013 (UTC)

I think the nutshell should give a frank summary of the gist of the page. The point is to give general guidance, to expand on the page topic, and to help readers decide whether they need to learn more about this topic. I think the following may more closely satisfy these criteria. "Never upload a photograph that may make your subject uncomfortable without his or her consent. Many countries require the consent of the subject prior to publication. Some countries do not permit any photography of people without consent." --Walter Siegmund (talk) 21:56, 16 January 2013 (UTC)

I prefer Colin's stab at the nutshell. The reference to photos that "may make your subject uncomfortable" is too vague and isn't really the test. Moreover, the second sentence seems to imply that it's simply an issue of consent, depending on country, whereas the analysis is actually more complicated than that. --Skeezix1000 (talk) 13:53, 17 January 2013 (UTC)

The more one looks at it, particularly country rules, the more complex it can be. It might seem simple if one's only concern is USA photographs and publishing. While many people will agree with your "may make your subject uncomfortable" test and use that as their own guideline for photography or publishing, others take a different approach. The actual test is a different one, which doesn't always align with "uncomfortable". Actually the main aspect of the nutshell I think is the first sentence. Everything flows from there. There are other legal, moral and ethical concerns that Commons does not respect, and indeed fights strongly against (rightly or wrongly). Colin (talk) 14:17, 17 January 2013 (UTC)

Employees

I think that the section about employees is not about employees, but where there are special expectations at the time the image is taken. The "employee" might as well be unpaid and without a formal relationship (other than what is established by turning up with a camera). As I understand the purpose the section is about expectations about how the images are going to be used, partly irrelevant of for whom the photographer is working. I might of course have misunderstood. --LPfi (talk) 13:49, 17 January 2013 (UTC)

What brought that section into being is the considerations at the Patient Images essay I orked on. Ultimately, employment rules on photography are influenced by "expectations at the time the image is taken" but they might be stricter or laxer than one's personal expectations. For medical staff in the UK, my understanding is that some NHS trusts have employment rules that state written consent is needed for all photography of patients. Furthermore, the GMC, which is the professional body for doctors, has additional guidelines on photography and other forms of imaging, stating when consent is required and how to ensure patients aren't identified. I would imagine there are employment guidelines for other professions, particlarly for vulnerable groups such as teachers - pupils, social workers - clients. While such concerns aren't necessarily grounds for deleting an image from Commons, they are something uploaders and photographers should be aware of, hence the warning. Colin (talk) 14:06, 17 January 2013 (UTC)

True. That part might be OK. But there is a "and any contract (assumed or formal) with the subject", which is as relevant for non-employees. A non-employed photographer should of course be aware of any formal contracts, but the "assumed" part is harder. It might be that "assumed contracts" are mostly relevant where the area should be thought of as being private, but I think there may be important cases yet not covered. --LPfi (talk) 09:15, 18 January 2013 (UTC)

I've removed that bit. You are right that the contract (implied or otherwise) with the subject concerns all photographers. Colin (talk) 11:13, 18 January 2013 (UTC)

Uploader request

File:Iraqi army t55 tank.jpg It seems one of the subjects didn't give consent. Anomizing is probably not an option. Would someone delete the version that he is referring to? I think he wants to keep some versions so I don't know how to tag it.--Canoe1967 (talk) 21:43, 6 January 2013 (UTC)

The version with the person has much better resolution than the others (which are not different versions, but other images). I think a soldier by his tank has little expectations or rights to privacy, so deleting the image is a bit hefty. On the other hand I am not sure the person is important to the image, so editing the image may indeed be an option (is he identifiable in the original version?). --LPfi (talk) 09:20, 16 January 2013 (UTC)

The uploader added a Jan/2013 request to a 2011 section of the help desk talk page. I came across it from my watchlist. Iraq is not on our list and we don't know if it was a private military compound or not. He may be getting a hard time about it so we may wish to honor his request.--Canoe1967 (talk) 18:48, 16 January 2013 (UTC)

In general I think we should speedy such request, I don't know the particulars of this image, but with service personnel in conflict zones, an ill advised picture may lead to a lot worse than bruised egos, compromised security, endangerment of familes etc.--01:51, 1 May 2013 (UTC)

New version!

Per the Another redraft-discussion above, I have moved the overall accepted draft into the position of the guideline. The old guideline - with all it's contributions - can now be found at Commons:Photographs of identifiable people (old). In kind regards, Henrik/heb [T C E] 13:17, 26 January 2013 (UTC)

Per my comment on your talk, I'm not sure if the move was carried out correctly. I think the guideline page should retain the history of the guideline page, not just of the draft. I wondered if it was possible to edit the original guideline page to include the new draft text with an edit summary that had a link to the draft version -- this is what Wikipedia seems to suggest but I'm confused about it. Also the above **(old) page still has category and guideline banners which it should not have if it is no longer the guideline. -- Colin (talk) 14:06, 26 January 2013 (UTC)

The old guideline - including all history of contributors and editors can still be found on Commons:Photographs of identifiable people (old). To the best of my knowledge, the MediaWiki software doesn't allow merging of two pages history. Since the current (new) guideline contains the history of all who contributed to that, and the old one is preserved with all it's history, I don't think it can get much better than that :). Other than that you are right about the categories and banners on the old page. It was an oversight on my part, which has now been remedied. In kind regards heb [T C E] 14:09, 26 January 2013 (UTC)

Well I'm no expert so just want to make sure the right thing is done. The new version incorporates a load of text from the old one, so the guys who wrote that deserve credit. What we have now just makes it look like the guideline is only a month old, which seems wrong. Colin (talk) 14:38, 26 January 2013 (UTC)

I think it is possible to merge history with: delete one version, move the other, undelete all. Yann (talk) 14:51, 26 January 2013 (UTC)

There are fairly detailed instructions in English at en:WP:HISTMERGE, if that's what we want to do. I don't know if there is an equivalent page here at Commons. WhatamIdoing (talk) 15:55, 26 January 2013 (UTC)

That approach works quite well, if and only if the old page has not been modified since the new one was created. Otherwise there will be confusing intermingling of versions of the two pages, with no hint in the history about what happened. If there is no overlapping history one problem remains, namely that it seems like the new page took the place of the old one from its first revision (the history does not show the then current name of the revision). --LPfi (talk) 21:34, 26 January 2013 (UTC)

I'll see if I can fix it up. Gimme 5 minutes... Kaldari (talk) 20:35, 27 January 2013 (UTC)

OK, the old history has been restored and I gave people credit for the new version in the edit summary. Hope that works. Kaldari (talk) 20:45, 27 January 2013 (UTC)

Works for me. Thanks everyone. Colin (talk) 08:28, 28 January 2013 (UTC)

I'm sorry for any confusion that I may have caused. I wasn't aware of the EnWiki policy (and I don't think Commons have a similar), but I am glad that a solution that everyone could be happy with was found :) In kind regards, Henrik/heb [T C E] 15:00, 29 January 2013 (UTC)

Surreptitious sketches and paparazzi paintings

This edit expanded the scope to include paintings and drawings of people (collectively "images"). Is this really the case? What are the legal issues with such media? Artworks are different from photographs in that there is a significant investment of time in performing the capture, and a degree of separation from reality, whether deliberate or for artistic or caricature purposes. There may also be some dispute that the artwork is actually of any one person. From what I can see wrt the legal country-specific cases, the law mainly deals with photography (which I think it reasonable to extend to moving pictures without the need to be explicit about it).

So I'm wondering if this is an unnecessary complication. It certainly makes the lead more complex to read. Do we really have any problem with surreptitious sketches and paparazzi paintings? We don't need to solve all problems involving people and Commons, and the guideline is named "Photographs of identifiable people". Thoughts? -- Colin (talk) 15:44, 29 January 2013 (UTC)

I don't think that "surreptitious sketches" is intended. How could you tell whether a drawing was "surreptitious"? If I see someone and sketch them from memory next week, when the person isn't around, is that a "surreptitious sketch"? It's kind of a silly idea.

On the other hand, I don't think that we would actually want to host, say, nude sketches of someone's ex-girlfriend or paintings of ex-husbands engaged in defaming activities (e.g., shooting up a school). It's not about unfairly obtaining the sketch or painting, but about distributing images that could get us tied up in a libel or defamation lawsuit. WhatamIdoing (talk) 17:15, 29 January 2013 (UTC)

Fair enough, WhatamIdoing, but I think the issue you've raised is easily addressed by revising the first section of the defamation section by adding some words in brackets so it reads: "Images (including non-photographic images) must not unfairly ridicule or demean the subject." In terms if the edit in question, I agree that it is unhelpful and confusing. Paintings and drawings give rise to different issues than photographs and we just can't tack on a reference to them in the lede. The new language should be reverted until there is some consensus on adding it. --Skeezix1000 (talk) 17:54, 29 January 2013 (UTC)

I can't talk about other countries, but in at least the Danish legislation, the scope covers all form of images (in Danish: billeder) and not just photographs, including such things as portrait paintings (taking a photo of a person in public, going home, and then paint a painting from the photo and publish it is exactly the same situation as publishing the photo), caricatures, videos, croquis and other non-photographic works. As I mentioned previously here, I think a more precise approach to scope would be beneficial (beside photo also covering common used terms here like living and publish). --heb [T C E] 18:43, 29 January 2013 (UTC)

I think it is simplest for this to be a guideline on photographs because (a) that's its name, which also determines the scope (b) photographs are overwhelmingly the issue. That the legal and ethical issues with photographs can also be applied to some extent to drawings, paintings, and to a lesser extent audio, is something that surely can be left for common sense and case-by-case examination. Part of the purpose of a guideline is to establish and record consensus on a common area of concern. I may be wrong, but I don't see this as a common area of concern. Perhaps folk can cite examples of problematic other images that would benefit from the guideline explicitly including them? So I think what's happening here is that folk are considering all the areas that might be relevant (which is good) and putting them into the pot (which isn't necessarily good). One compromise might be to include a statement at the end of the lead, which says that there are similar issues wrt other media such as artworks, video and audio. But I think it too strong to state (as Rd232's draft did) that the guideline fully extends to those areas -- because the law in particular does not.

Wrt scope, I agree we should continue to consider this. I don't think there is anything in Rd232's draft's scope section that isn't included here: We have a whole section on Identifiable; the beyond-death aspect is also included in the lead and the personality rights section; the other media issue is being discussed. -- Colin (talk) 09:02, 30 January 2013 (UTC)

I can only agree that "photographs are overwhelmingly the issue" and because of that I don't think there is any point in changing the title of the guideline. That said I find the remark of "the law in particular does not [extends to those areas]" to be very country- (US-) centric, which is okay, but in that case that should be mentioned. I checked into German, Swedish and French laws which also appear to use the more generic term image or portraits rather than photograph. I wouldn't be surprised if most of Europe have similar approaches to this.

For the alive/dead part I wonder if the statement that "[i]n most countries, these issues only affect images where the person is identifiable and still alive" is actually true. Of course if it is, it is no problem, but I do believe that it is over-simplified and perhaps even misleading. --heb [T C E] 10:48, 1 February 2013 (UTC)

My point about the guideline not "fully extend"[ing] in those areas is indeed that internationally it does not. There are some countries where the law on portrait paintings is equivalent to portrait photographs but this isn't universally true -- hence once can't just assume they are equal, which is what Rd232's draft's scope section did. The fact that the legal issues become even more varied and confusing when one considers paintings and drawings, say, is a motivation for leaving it out for now. On a case-by-case basis, should someone AfD a painting with such issues, then really it needs folk to investigate the relevant laws rather than assume paintings=photographs.

As for the alive/dead part, you may be right. I wonder if that can be rewritten to "The legal and ethical issues are greatest when the subject of an image is clearly identifiable and still alive. However, there may still be problems after the subject is dead or if they cannot be (easily) identified."

Does that help? Colin (talk) 11:44, 1 February 2013 (UTC)

Stating that "The legal and ethical issues are greatest when the subject of an image is clearly identifiable and still alive. However, there may still be problems after the subject is dead or if they cannot be (easily) identified." is from my point of view better, though I would like it to be a bit more clear-cut. I.e. "Though the legal and ethical protection often is most prominent when the subject is alive, in some jurisdictions the legal protection may extend years post death. Not being able to clearly identify the subject may also introduce another set of issues that should be considered carefully." In kind regards, Henrik/heb [T C E] 14:32, 3 February 2013 (UTC)

I think that Skeezix's idea about "Images (including non-photographic images) must not unfairly ridicule or demean the subject." addresses my main concern. WhatamIdoing (talk) 03:09, 2 February 2013 (UTC)

It would definitely make it less ambiguous, which I think must be the goal here. In kind regards, Henrik/heb [T C E] 14:32, 3 February 2013 (UTC)

Add "meta-data" to "clues may be obtained from...." sentence

Recommended change, to be applied to all language version:

Outside of the image, clues may be obtained from the image title, description, origin, source url, geolocation and date.

To

Outside of the image, clues may be obtained from the image title, description, origin, source url, and meta-data including but not limited to geolocation and date. Davidwr (talk) 03:58, 14 April 2013 (UTC)

re Identification section

This passage of the "Identification" section looks pretty off to me:

"Placing a black band over the eyes was historically used to hide patient identity in medical publications but is no longer considered effective.(Source: ICMJE) Pixelated features can sometimes be revealed by squinting one's eyes. These crude attempts to anonymise images may damage the value of an image to Commons to such a degree that it has limited or no realistic chance of being used."

OK, crude attempts are not good. So why not use uncrude techniques? It's easy enough to do. The passage should say something to this effect:

"Placing a black band over the eyes was historically used to hide patient identity in medical publications but is no longer considered effective, so don't do that. Pixelated features can sometimes be revealed by squinting one's eyes, so don't do that. An image editor's whirlpool feature may be be reverse-engineered, so don't do that. Just putting a black square over the face is effective but jarring and ugly, so that's not recommended. Using your image editor's blur to convert the recognizable features into an unrecognizable blur is simple, completely effective, and usually reasonably acceptable aesthetically. Ditto cropping identifying features. So do one of those. Of course, anonymizing the image in this manner may (or may not) damage the value of the image to some potential users."

I propose to rewrite the section along these lines if there's no objection.

Also, re the image examples under "Problems with attempts at anonymisation"... the first two are OK, but the third (David) seems pretty effective (assuming it was a random person and not David). Instead, the section should be called something lie "Examples of good and bad anonymisation", with a couple examples of each.

There are reasons not to anonymize, sometimes, and these are discussed. If those reasons are not applicable, then anonymyization should be considered, and some useful ways to do it described. Right? Herostratus (talk) 02:16, 1 May 2013 (UTC)

I disagree with the proposed text, which for a start is way too informal. Please can you give (on this talk page) some actual examples of good and justified anonymisation. I disagree with you on the third example (David). Remember we are talking about cases where the person has not given their permission to be photographed or for that photograph to be published widely online. Do you think that cropping off the head makes it ok to publish a photograph of someone with full-frontal nudity without the necessary permission? An example where cropping to anonymise (to some degree) an image may be acceptable is patient photographs -- but in those cases consent is required as standard ethical practice. Colin (talk) 06:41, 1 May 2013 (UTC)

I'm not sure that you really can give a good example, because it depends on the circumstances, not just the visual appearance of the photo.

Imagine that there is some skin disease with a distinctive appearance. I happen to see it on someone in CC-BY-licensed Flickr photo, maybe someone at a public event. I crop the photo to remove half the picture. What's left is a good picture of the skin disease. Nobody (except maybe the person himself or herself, who sees that skin disease every day) simply looking at the picture would be able to say who it is. It's just somebody's shoulder, right?

But we're not trying to protect that person's privacy merely through visual inspection of the photo. What happens if you take that photo to a reverse-image search? Does TinEye or Google Images find it? (Oops, now everyone who searches knows what the person's face looks like.) Is there a link back to the original at Flickr? (Oops, now everyone who clicks the link knows what the person's face looks like.) Is there a description at Flickr? (Oops, now everyone who clicks the link knows when and where the photo was taken, and maybe even the person's name.) Is the photographer's name included? (Oops, now people know what social circles this person might belong to.)

The medical standard these days seems to be that you can use things if and only if there is no chance of the patient recognizing the image of himself. This pretty much limits you to microscope slides, and even then they're getting pretty paranoid. Our standard is a bit looser, since we're primarily concerned with whether other people can recognize you, but the cropped photo of David fails even that. WhatamIdoing (talk) 15:03, 1 May 2013 (UTC)

I guess there are two angles here. There's the picture that clearly shouldn't be published in itself such as a nude taken without permission, that some folk on Commons think is just fine as long as we chop off their head or blur their face enough. Then there's the picture that has already been safely published but becomes an issue because you've attached a medical diagnosis (or labelled them as drunk or a prostitute or whatever else might be defaming or hurtful). Both cases are problematic without permission and I think it would be hard and possibly legally wrong for this guideline to advise anyone that cropping/blurring might avoid them being sued.

The Google image search thing is interesting because one doesn't have to stumble upon the diseased shoulder and wonder "who might that be"? Google images works the other way too, making it all the more likely that the defaming medical misdiagnosis will be found by someone with an interest.

The medical standard you mention only applies without consent. If you have consent then you can publish whatever the patient has agreed to. Which is great. Wikipedian Doc James says he has no problems getting consent for the photographs he takes and uploads.

It is also worth noting that stock image sites are even fussier. We don't require a model release, just your claim you got consent if someone asks. If I was a publisher of a book, I'd rather pay a few quid for a photograph with a model release and indemnity from the site, than take my chances with something found on Commons. Colin (talk) 16:15, 1 May 2013 (UTC)

What to do

Is there anything practical that we could add to Commons:Photographs of identifiable people#Removal requests for a person who discovers an unwanted picture of himself (or herself) on Commons? Even "Contact any current admin by e-mail" might be better than nothing. WhatamIdoing (talk) 00:24, 9 June 2013 (UTC)

• If we do write something, we need to make it clear that the subject of a photo cannot always get the photo removed. - Jmabel ! talk 03:11, 9 June 2013 (UTC)

...unless they are a public figure called Jimbo and the image is hilarious. Penyulap ☏ 04:38, 9 June 2013 (UTC)

• Penyulap: I didn't say that they can never get it removed, I said they cannot always get it removed. More often than not, we accede to requests like this. I assume I know what image you are talking about, and we would almost certainly have readily granted a similar request from anyone other than perhaps a public political figure. - Jmabel ! talk 16:03, 9 June 2013 (UTC)

So we'll need to be careful not to promise more than we can deliver. On the other hand, there are situations in which we routinely delete non-consensual photos, e.g., naked pictures uploaded by ex-boyfriends. If a victim discovers this page, what can we tell her to do? WhatamIdoing (talk) 15:39, 9 June 2013 (UTC)

• I'm not sure that is particularly different from any other deletion request. It would almost certainly be granted, with the possible exception that (for example) the petitioner was a professional model and the ex-boyfriend a professional photographer, and similar images of the petitioner were widely published. In that case, the fact that the photographer happened to be an ex-boyfriend would probably be irrelevant. Or are you concerned about finding a way to avoid the Streisand effect. - Jmabel ! talk 16:07, 9 June 2013 (UTC)

The easy cases are when it is "out of scope" such as images which are neither educational in their own right nor used in a Wikimedia project outside the commons, or if there is a likelihood of lack of permission from the copyright holder, such as "I gave my boyfriend this picture I took of myself." Those can be handled quietly through normal deletion processes. On the other extreme are images which clearly don't meet any deletion criteria. For those the only real option would be office action or a court order.

Anyone seeking to have material removed should know about the en:Streisand effect, and they should know that not only will many mirrors not remove the material quickly or at all, but that some web sites make a point of hosting material which has been deleted from Wikimedia projects. Davidwr (talk) 19:05, 9 June 2013 (UTC)

So it sounds like the options are these:

• If you are the photographer, then use CSD for copyvios (probably more useful for parents who took pictures of their kids, or do people really take that many photos of themseleves?)
• If it's out of scope, but otherwise unremarkable, then use the usual deletion process (just like anyone else)
• If it's sensitive material (so you don't want to attract public attention to it), then ??? We have to assume that the person is a newbie, which means that the person can't just go ask a friend who is an admin. WhatamIdoing (talk) 15:29, 10 June 2013 (UTC)

It seems to me that to be in compliance with the resolution on Images of Identifiable People, in any case where the subject objects, except for unambiguously public locations or unambiguously public figures, we need unambiguous evidence of permission. Such images should be taken down promptly and quietly until such permission is established via OTRS. This would probably cost us some good images but it would contribute to public trust in the process and would also avoid having to make a lot of fine distinctions and difficult judgement calls in advance. The only thing that gives me pause is how to avoid false claims by people who are not the subject but simply object to the image, while still respecting the subject's privacy. Dankarl (talk) 16:34, 10 June 2013 (UTC)

Wrt WhatamIdoing's original question, would they even find or know about this guideline? This is a guideline for User:xxx not for Joe Public. It may help any admin or advise any deletion discussion, but as far as helping viewers to know what to do about images they don't want here, surely that's the job of a Help page somewhere else. -- Colin (talk) 18:32, 10 June 2013 (UTC)

As far as I know there is no help-page on the subject. Until such as time (if it should ever arise) I think this page should "carry the burden" from the lack of a relevant help-page. Omitting good information, simply because it should be another place that may or may not be created in the future, doesn't sound right to me...

On the subject of what to do, I'm much on the same page as User:Dankarl, though I think that false claims is a relatively small problem, that can be handled through demanding that requests are sent to OTRS. --heb [T C E] 12:46, 11 June 2013 (UTC)

There was some discussion a couple months ago of having a "request removal" button as part of the image page; this would bring up a box with buttons for various cases - copyright, inappropriate image, and otherwise, which would then launch the appropriate response process. I don't remember whether lack of subject consent was specifically mentioned but it certainly could be. I don't have time today to chase this down but if needed will take a look tomorrow. Such a process would require a well defined and easily explained response path - part of the thinking behind by suggestion of 10 June, above. Dankarl (talk) 13:05, 11 June 2013 (UTC) The link is Commons:Village pump/Archive/2013/05#May 16 subsection Making it easier for problematic files to be brought to our attention sub-sub section Technical feature Anyone who knows how to link a subsection or sub-subsection directly please do so with this, thanks Dankarl (talk) 14:16, 12 June 2013 (UTC)

There is another discussion on a similar topic at Commons talk:Criteria for speedy deletion#G7:_Personality_rights. WhatamIdoing (talk) 03:03, 16 June 2013 (UTC)

• Would it not be better as these could be sensitive issues to actually suggest that they to email a bureaucrat rather than just "any admin", I dont think this is a big volume issue and crats in general are more active as a group than admins are so random chance is unlikely to hit someone who is otherwise occupied. As crats have gone through multiple reviews they should be more able to act more discreetly, have a more indepth policy knowledge and should be able to address the concerns more adequately depending on the circumstances. Gnangarra 04:17, 16 June 2013 (UTC)

Yes, that would probably work, but we need to be sure that the instructions and scope are very well-defined and standardized across multiple help pages. The current situation is really a mess.   — C M B J   07:22, 16 June 2013 (UTC)