Commons talk:Photographs of identifiable people/Proposal

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

Since this policy was introduced in March 2007 there has been more certainty on which types of photographic portraits we will and will not host. As a result, most deletion requests in this area are being closed more quickly and are less subject to the whims and personal views of the closing admin. There are still a few contentious areas, though, particularly:

  • images requiring consent, and
  • images where there is incomplete information about the circumstances in which the photograph was taken, so that the policy has to be applied on the basis of guesses or assumptions.

Examples include:

  • Posed photographs, such as straightforward head-and-shoulder shots and studio portraits (consent unclear), and
  • Images with indistinct or nondescript backgrounds that give no clue as to the type of location in which the photograph was taken (ie public/private is unclear).

Also of concern are certain potentially sensitive images, such as those of a child or of an unclothed adult. Some users consider that these should require a higher standard of proof that consent has actually been obtained.

In a nutshell, the proposal is that the more potentially damaging an unauthorised release of the image would be to the depicted individual, the greater the effort we ought to make to ensure that the individual has truly consented. A potentially very harmful image should require a higher level of evidence than a completely uncontroversial one.

My aim has been so far as possible to suggest rules that will not result in too much change to existing consensus. Of course, not all deletion requests will be decided in the same way – if only because of the significant variability and uncertainty that exists at the moment. The idea is to create more certainty while not moving the goalposts more than necessary.

Before anyone accuses me of instruction creep, let me say that this is an area where we have little in the way of detailed guidelines, and that as a result users of Commons’ media have little certainty about whether a particular image will be kept or deleted. That uncertainty harms Commons’ reputation among its Wiki “customers” and is bad news for the WMF project as a whole.

It would be too much to expect unanimity on these points because there are widely differing national approaches to personality rights, and because we are all conditioned by the expectations of our own societies. However, in spite of wide variations of outlook, the existing high-level rules have received a high level of support and I hope that a similar level of agreement can be achieved for these suggestions.

A comparison with the existing text can be found here.

For ease of discussion, please comment below under the appropriate heading. I will edit the Proposal page with consensus changes as the discussion develops. --MichaelMaggs (talk) 22:04, 2 February 2009 (UTC)



What does “identifiable” mean?[edit]

Photographs taken in a public place[edit]

Photographs taken in a private place[edit]

What are 'public' and 'private' places?[edit]

As a potential subject I do not "expect privacy" outside my house, but other people might have different expectation. Most of us, are not move stars, are not used to dividing world into private (no paparazzi) vs. public (beware of paparazzi). I am afraid of future endless discussions on deletion request pages about what places are public and what are private. Some more examples could help but they will not cover all the possibilities. For example some places I expect to be public, but want to be sure it is other peoples opinion too:

--Jarekt (talk) 14:19, 6 February 2009 (UTC)

I agree: work, church, weddings are public, but not patients in a hospital. (But the photo op with Julia Roberts is ok of course.) /Pieter Kuiper (talk) 14:59, 6 February 2009 (UTC)

Well, the "public" / "private" distinction has been agreed policy for a very long time, and this proposal does not change that. (There is a small tweak to the wording, for clarity, but without change in meaning}. All of the above images would remain open to discussion on this point, just as they are now. Personally, I would be reluctant to say "a church is always public (or private)" since so much depends on the circumstances. That sort of thing is still best dealt with on an individual discussion basis. --MichaelMaggs (talk) 18:24, 6 February 2009 (UTC)

I guess my problem is I do not know anything about "public" / "private" distinction has been agreed policy for a very long time, if it is written down somewhere than we should point to it. What I am concerned with is that we have a lot of images of people in public indoor spaces (like most of examples above) and it would be good to have some guidelines so we do not have to discuss them all. I do not expect "a church is always public (or private)" statement but something along the lines "most pictures taken inside buildings open to public (like ......) are public, except for .....". I assume that there could be exception to any of those rules that could and should be open for discussion, but we should not have discussion about every church/office/hospital/etc picture. --Jarekt (talk) 21:48, 6 February 2009 (UTC)
The current policy is still there - at Commons:Photographs of identifiable people. --MichaelMaggs (talk) 22:46, 6 February 2009 (UTC)
Often it is unclear, if a place is private, or not. In case of a flickr-upload it will be difficult to discuss the fact with the photographer. In my opinion we should always demand a OTSR-Ticket, if a critical photo is not uploaded by the photographer himself. -- smial (talk) 18:45, 11 February 2009 (UTC)
The range of opinions is wide, and it is interesting to compare this view with another one which was posted at about the same time. As I indicated above, we cannot expect unanimity, and the aim here is to seek a middle gound which has the greatest chance of general acceptance. --MichaelMaggs (talk) 22:46, 11 February 2009 (UTC)

Public indoor swimming pools are not public places in this sense. There might be situations where we would accept images nonetheless. For example, "home movie"-style video or images of children playing in a public indoor swimming pool might be permitted where the photographer is in the subject's family. However, there is a general presumption that such pools are private places in the sense of this proposal.  — Mike.lifeguard 05:57, 12 February 2009 (UTC)

Outdoor pools without a privacy fence should be considered public places. Those with a privacy fence should be considered similarly to public indoor swimming pools.  — Mike.lifeguard 05:58, 12 February 2009 (UTC)
We shouldn't write fixed rules for specific places, as they would be far too inflexible, but my personal view (for what it's worth) is that you are right. --MichaelMaggs (talk) 17:44, 12 February 2009 (UTC)
Of course not & that was not the intent.  — Mike.lifeguard 20:56, 13 February 2009 (UTC)

Does the photographer's presence make a difference? What about a photographer's studio?[edit]

What if it is not clear where the photograph was taken?[edit]

What is meant by “consent”?[edit]

Is it sufficient for the subject to have consented to the taking and the publication of the image, or are we requiring that they explicitly consent to it being uploaded specifically to Commons? The way this section is written now ("...and has no objections to the photograph being uploaded to Commons.") makes it seem like it might be taken to mean the latter, but surely that isn't the intended meaning? In particular, it would mean that we couldn't use any pictures of identifiable people obtained indirectly via Flickr or other such sources, at least unless we managed to contact the subject and request separate permission after the fact. —Ilmari Karonen (talk) 23:06, 4 February 2009 (UTC)

I think it should be a consent specific to Commons, because publication on Commons means that the picture can be reused by anybody, for any purpose, and possibly for commercial activity. We make the notion of publication more "sensitive". --Eusebius (talk) 07:11, 5 February 2009 (UTC)
It's deliberately phrased in a negative way ("...and has no objections...") so that we can at level 1 presume consent without the subject having to say so explicitly; that would mean losing virtually all portraits now on Commons. If it is clear that the subject does not consent, because he/she has told us so, the presumption of consent is overruled. To make this quite clear, I have changed the wording to "...and has stated no objections...".--MichaelMaggs (talk) 18:09, 5 February 2009 (UTC)

Normally, in the United States, "Consent" for the use of photographs of identifiable persons is confirmed by the use of a written release that is signed by the individual photographed and by the photographer, with positive language transferring the right to publish from the person photographed to the photographer and his assigns. That release could be scanned into a file, and this file uploaded to Wikicommons. - SSG Cornelius Seon (Retired) (talk) 23:08, 10 February 2009 (UTC)

If we have such a clear consent, that would be excellent. Unfortunately, that is extremely rare hence the need for some rules to decide what lesser types of consent Commons will and will not accept as a practical matter. --MichaelMaggs (talk) 13:06, 11 February 2009 (UTC)

Does the subject have to waive all Personality rights?[edit]

Does the subject explicitly have to agree to commercial use of the image?[edit]

Who has to give consent, when required?[edit]

"... except where the subject is a minor when consent has to be given by the parent or guardian." I would just like to point out that in some (many?) countries, consent of the minor is also required, in addition to her parents', when the subject is above a certain age at which it is considered that she understands the issue. I guess it is so in order to avoid some kinds of parental abuse and enhance the rights of children. I'm not sure it should influence our position though: I'm pretty ok with this statement, but in the case we have a parent's consent and the minor expresses an opposition, we should consider the rights of the subject and delete. I think this case is reasonably covered by the "Removal at the request of the subject" section. --Eusebius (talk) 22:05, 3 February 2009 (UTC)

Yes, I agree. If a minor does object, even though his/her parents have consented, I am sure we would do the right thing and delete. --MichaelMaggs (talk) 18:09, 5 February 2009 (UTC)

Where consent is required, what evidence is needed?[edit]

Level 0: Public place - Consent is not normally required[edit]

(see general feedback also) - I would like to see some level of subject permission required in some cases - perhaps where privacy could be reasonably seen to be violated. I'd like this to apply to both identifiable, and unidentifiable media - for example this image. Privatemusings (talk) 03:30, 3 February 2009 (UTC)

I don't much like that image either, but I think we would be moving too far away from the major focus here to extend it to unidentifiable people as well. Currently we can delete images out of concern for the subject's reputation, but absent that the rationale would just be "I don't like it" or "it offends me", which we generally try to steer clear of. I would delete that image under COM:PS as "not realistically useful for an educational purpose" - a basis we perhaps should use more often. --MichaelMaggs (talk) 07:48, 3 February 2009 (UTC)
Added as such to deletion requests. -Kotra (talk) 15:59, 6 February 2009 (UTC)

File:Fremont naked cyclists 2007 - 43.jpg might be a good example of something that should be OK without any clarity about consent, but only because the space is clearly public. It is a photo I took of a person—as it happens, one I know, but that is immaterial—naked in a public place. - Jmabel ! talk 02:30, 17 February 2009 (UTC)

Level 1: Private place - Consent can normally be assumed[edit]

Level 2: Private place - Photographer must explicitly state that consent has been obtained[edit]

"...uploader is trusted and known to the Commons community" seems to me to be a bit tricky. Would this apply to anyone other than admins? I've uploaded over 10,000 images, but I have no idea whether I'd be considered "trusted and known". For example, I've certainly had people question whether I was the author of an image because it had been published in (for example) a newspaper, even though the newspaper clearly credited me for the photo. Would this be an indication that I am not "trusted"? Or am I being a bit paranoid? - Jmabel ! talk 18:31, 3 February 2009 (UTC)

Good question. Perhaps it could be phrased better; certainly you would qualify! The intention was really to exclude new users, users with just a couple of uploads, or those with a history of copyvios. We don't want an unknown user who created an account yesterday to be allowed to get away with asserting whatever they want (or every copyvio specialist will soon see the loophole). I suppose that's the same as saying that we do not accept blindly statements by unknown users alleging that their upload is OK. We ouight to avoid "trusted user" as that already has a specific meaning on Commons. Can you think of a better way of phrasing it? --MichaelMaggs (talk) 18:48, 3 February 2009 (UTC)
I have suggested using "only where the information appears trustworthy", which may be better as it relates to the quality of the information rather than attempting any judgement on the extent to which the uploader is "trusted". Does that work? --MichaelMaggs (talk) 21:42, 3 February 2009 (UTC)
I'm pretty open on how it's worded, but worried about how it will be applied in practice. I wish there was some formal way on Commons for me to be noted as having a record of significant and positive contributions other than applying for an administratorship that I don't particularly want. - Jmabel ! talk 04:10, 6 February 2009 (UTC)
Anyone else like to chip in here? I would say that even a brief review of your contributions would be enough to show that you clearly qualify. --MichaelMaggs (talk) 07:39, 6 February 2009 (UTC)

The fact that "identifiable minor in a private place" is enough to unconditionally raise an image to this level seems a bit strange, at least if "private place" is interpreted broadly enough to include e.g. school classrooms. If this interpretation is followed, it seems that a lot of pictures like File:DRC classroom.jpg, File:Classroom, Myanmar.jpg, File:Djibouti classroom.jpg, File:Mogadishu UNOSOM classroom.jpg and indeed many of the images in Category:School children would have to go, since few if any of them come with an explicit statement of consent (and in most cases tracking down the subjects and their parents, or sometimes even the photographer, seems well nigh impossible). Of course, to some extent this may be intentional: while the children in those images may have consented to being photographed, I doubt most of the photographers explicitly contacted all their parents too, and so, if we really subscribe to the rule that pictures of identifiable minors always require parental consent, they'd indeed have to be deleted. But it does seem a bit like throwing the baby out with the bathwater. —Ilmari Karonen (talk) 23:26, 4 February 2009 (UTC)

It's true that nobody has (yet) brought those images up. My expectation is that if they did, then parental consent probably would be required. Unfortunately, however we draw up the rules, we are going to find some images that have to be deleted, but that doesn't matter provided there is consensus that the rules are reasonable overall. I would be reluctant to start introducing special cases (eg a classroom is always taken to be a public place), as the debates would then never end. But did you have some suggestions as to how the proposals might be improved? --MichaelMaggs (talk) 18:20, 5 February 2009 (UTC)
I must say it seems to me more natural and sane this way (because more compatible with my own national laws, I guess), but I understand there is a debate. --Eusebius (talk) 07:09, 5 February 2009 (UTC)
  • This might be a language issue on my side. What does "reasonable embarrassment" mean? Should it not be "unreasonable embarrassment", as in level 1? Pruneautalk 21:52, 6 February 2009 (UTC)
    • ✓  Done with slight copyedit. --MichaelMaggs (talk) 17:59, 9 February 2009 (UTC)
I agree with Karonen. How could class photos be images that "might realistically be expected to cause the subject harm, distress or embarrassment"? Why would unposed photographs or even posed photographs in the case of minors be "typical examples" of images in this category? /Pieter Kuiper (talk) 21:43, 11 February 2009 (UTC)
Because many contributors here (not you, I think) consider that minors should be dealt with differently from adults. The range of opinions is wide, and it is interesting to compare this view with another one which was posted at about the same time. As I indicated above, we cannot expect unanimity, and the aim here is to seek a middle gound which has the greatest chance of general acceptance. --MichaelMaggs (talk) 22:49, 11 February 2009 (UTC)

Level 3: Private place - The subject or parent/guardian must provide explicit consent[edit]

"it is expected that OTRS permission should be available on upload". Does that mean that "OTRS pending" status are not acceptable, and that such images must be kept as deleted until the ticket is confirmed? No problem with that, it's just not perfectly clear to me. Plus, will we use the generic OTRS template? I'm afraid it would be easy for a photographer to provide or use a license-related OTRS ticket in order to reference a fictional subject consent. --Eusebius (talk) 22:10, 3 February 2009 (UTC)

Requiring OTRS for "Photographs of an identifiable minor in a private place showing any disease or medical condition" may not sound unreasonable at first sight, but then somebody will request deletion of File:Windpocken.jpg (uploaded 2005), File:Mary mac dougal 1.jpg (published 1924), etcetera. Somebody may think that the permissions of File:Neck Turner.JPG are not good enough. I would oppose such deletions. /Pieter Kuiper (talk) 08:09, 5 February 2009 (UTC)

Peter, once a consensus set of rules has put put into place as policy, there would be little point in your simply opposing deletions, as admins close deletion requests on the basis of policy not votes. You are an experienced guy here, and if you take the view that the images you mention ought to be OK, maybe you can suggest how the proposal could be re-worded to achieve that? If you are of the view that no consent is needed, or that it should be implied, do please say. I doubt that either option would achieve a consensus, but maybe I am wrong. The old image will be OK, anyway, as it is unlikely that the subject is still alive. --MichaelMaggs (talk) 18:28, 5 February 2009 (UTC)
That is why I am bringing this up now. I tend to oppose deletion requests not related to copyright or scope (and attack images are out of scope). I am not in favour of a new text, because any new rules will cause problems. I had a look at this "Level 3", to see what kind of images in the repository might be affected. And I concluded that the reasonable wording is likely to have unreasonable consequences.
In my opinion, requiring OTRS for all "Level 3" photos is not a good idea. Uploaders obviously have internet access, but people in photos often won't. It would also be much more difficult to ensure that an email of a parent or some subject was authentic. Are the OTRS people prepared to do this? /Pieter Kuiper (talk) 19:28, 5 February 2009 (UTC)
File:Mary mac dougal 1.jpg seems to bring up a few issues not so far covered by this proposal. The first one is that we can't obtain consent of the subject, since according to the source the subject is dead (and apparently was so already at the time the picture was published). Do we, or should we, demand consent of the dead, and if so, who would give it?
The second issue is that much of the current wording seems to be written mainly with user-created images in mind. In this case, the image was published 85 years ago in a respectable medical journal; it seems reasonable to assume that the journal at the time felt they were legally and ethically justified in publishing the image (although obviously their situation is not exactly the same as ours), and it seems a bit absurd to apply the exact same standard to it as we would for someone's personal photo that had not been previously published. I'm not trying to suggest that we should arbitrarily apply a double standard to user-created photos on one hand and the rest on the other — but it does seem to me that e.g. {{PD-old}} or {{PD-USGov}} images sometimes present issues which this proposal, as currently written, may not adequately address. —Ilmari Karonen (talk) 22:21, 5 February 2009 (UTC)
See the first line of the policy: it "deals with photographs of identifiable living people". Only the living have a reputation that could be damaged. No problem at all with old images. --MichaelMaggs (talk) 22:54, 5 February 2009 (UTC)
As for cases like File:Windpocken.jpg, it seems to me that the main problem is a disagreement between the definition and the examples. I can't see any way in which that picture "might be realistically – without consent - be expected to cause the subject significant harm, distress or embarrassment"; but if the examples are to be taken as definitive (rather than merely illustrative), then it nonetheless automatically falls under this level due to depicting a "minor in a private place showing any disease or medical condition" (although we might perhaps be able to weasel out of this one, for this particular photo, by arguing that the person — who has surely grown up by now and looks nothing like the baby in the photo any more — isn't actually "identifiable" under the definition given). —Ilmari Karonen (talk) 22:35, 5 February 2009 (UTC)
How would you deal with that example? --MichaelMaggs (talk) 22:55, 5 February 2009 (UTC)
How would you deal with it? Or with File:Jacobsen syndrome patient.JPG, not quite 18, but from a reliable source (published), and with a good free copyright license. Useful photo. Still, I would support deletion if the girl requested Commons to do that. But I am against this proposal's absolute language requiring OTRS confirmation for images of minors "showing any disease or medical condition". Any? Kids on crutches, or in wheelchairs, or with acne? /Pieter Kuiper (talk) 13:51, 6 February 2009 (UTC)
In several ways. First of all, I'd like to see the policy explicitly state that the examples given for each level are not normative, but merely typical examples, and should not be blindly applied where they clearly contradict the definition given for the level.
That makes sense. I'll do it. --MichaelMaggs (talk) 18:38, 6 February 2009 (UTC)
✓  Done --MichaelMaggs (talk) 18:04, 9 February 2009 (UTC)
Second, I'd like to see more discussion about the examples themselves. Exactly why does "showing any disease or medical condition" automatically bump everything up one level — does it reflect some local cultural taboo against showing signs of ill health that I'm unaware of? Actually, the more I look at it, the more I start to suspect that "showing any disease or medical condition" is actually meant to be an euphemism for something else: probably "looking funny". Taken literally, it seems quite silly to imply that having a bandage on my arm (a wound is a medical condition, right?) or suffering visibly from a cold would automatically require a higher level of proof that I'd consented to being photographed.
Would you support getting rid of the medical examples entirely? See below under #Medical conditions. --MichaelMaggs (talk) 18:38, 6 February 2009 (UTC)
✓  Done Medical condition text removed, as this does not seem to be gaining a consensus. --MichaelMaggs (talk) 18:04, 9 February 2009 (UTC)
I should have reacted to that before. To me, all the issue about photographs of identifiable people boils down to: "does it reveals about the person something that should be kept private?". From this point of view, my state of health (read the first person as impersonal and hypothetic) and the diseases I've been subject to are something pretty private, one of the most private and sensitive subjects, one which I'd like to protect in a particular way, just like the aspect of my naked body, for instance. I think it is normal that health-related disclosed information should at least be considered with more attention by reviewers, and it does not shock me that it could automatically make the picture go up one level. I could do without though, provided it was written somewhere that pictures showing medical conditions need at least extra care, and often extra protection (I mean privacy protection, not file/page protection). --Eusebius (talk) 18:47, 9 February 2009 (UTC)
Third, for that particular photo, I'd try to contact the uploader. They signed up with their real name and stated they were the photographer, so it's quite possible that we might be able to contact them and that they either a) are the child's parent, b) did have explicit parental consent, or c) can contact the parents to obtain it. After all, it does no harm to have better proof of consent than would strictly be necessary. But I still don't think this level of proof should be necessary for this image, given that it shows nothing harmful, distressing or embarrassing, and especially given that the subject is not even likely to be particularly easily identifiable. —Ilmari Karonen (talk) 18:03, 6 February 2009 (UTC)

Legal issues[edit]


According to this section, a work is unacceptable on Commons if it is "arguably illegal" in the source country. It is also said that existence of personality rights in the source country is not enough to make an image unacceptable. I fail to understand how these sentences can be compatible, to me personality rights may make an image illegal in the source country, no? In the case of France, I think a huge batch of pictures could be "arguably illegal" (a photographer was once condemned for having taken/published a picture of a guy sitting with his girlfriend on a bench in a public park, as far as I can remember). I think I haven't fully understood this part. --Eusebius (talk) 22:17, 3 February 2009 (UTC)

Every Image of non covered people or at least women is arguably illegal in most islamic countries; should we realy delete a photo of Condolezza Rice visiting a US base in Saudi Arabia or Afganistan? sугсго 10:46, 4 February 2009 (UTC)

I'm not saying that. I'm saying the phrasing is not clear enough and might be seen as encompassing this kind of case. --Eusebius (talk) 11:38, 4 February 2009 (UTC)
The "arguably illegal" wording has been there for some time, but it now seems highly dubious that there really is consensus for it to stay. It has certainly not been enforced. The community has not supported restrictions based on local law regarding photographs of antiquities in Egypt, and it is highly unlikely that the community would support deletion of photographs of women taken in Islamic countries (should that indeed be illegal, as suggested above). Given that the consensus seems to be that Personality rights in France should be ignored as a basis for deletion, there seems no point now in retaining the wording. I have deleted it, and we will see if there is any consensus for it to be put back in again.
On the actual question you raised, there is a technical distinction between "illegal" (ie a criminal act) and "unlawful" (an act that someone else has the right to restrain - eg on the basis of a Personality right). So there is technically no conflict in the wording. --MichaelMaggs (talk) 18:43, 5 February 2009 (UTC)
Thanks for improving my English then! --Eusebius (talk) 21:41, 5 February 2009 (UTC)

Moral issues[edit]

It is said that images "unfairly obtained" are normally unacceptable. What is covered by this sentence? How does it relates to the "Unposed photographs where an identifiable unclothed or partly unclothed adult in a private place does not appear to be aware of the presence of the photographer, including underwear and swimwear shots", which are put in the level 3 category (so acceptable under conditions), or to the "Unposed photographs where an identifiable adult in a private place does not appear to be aware of the presence of the photographer", which are level 2, thus acceptable with a declaration of the photographer)? --Eusebius (talk) 22:21, 3 February 2009 (UTC)

The wording "unfairly obtained" was if I remember rightly originally based on the code of conduct which applies to newspaper publications in the UK. I don't think it has given us any problems so far. It clearly must mean something different from the wordings you have quoted, and I think it could come into play where, for example, the photographer has obtained the image by tricking the subject in some way, or by committing fraud. The wording is quite broad, which is good as it gives the community the opportunity to apply it to a variety of situations should the need arise. --MichaelMaggs (talk) 18:57, 5 February 2009 (UTC)

Can an image be made allowable by adding the {{Personality rights}} template?[edit]

Re-use of the image[edit]


Probably OK[edit]

Probably not OK[edit]

Removal at the request of the subject, photographer or uploader[edit]

Would like to see some instructions here on how to appraoch the different level images when an issue arises.

  • lvl 0-1, discuss with uploader
  • lvl 2, discuss with uploader, tag as OTRS required or contact OTRS agent to confirm
  • lvl 3 contact COM:AN

May need to also consider a specific COM:CSD#Speedy_deletion identifiable person option so that admins are able to act quickly and clearly where its necessary... Gnangarra 01:47, 3 February 2009 (UTC)

That makes sense. Given the potential for huge reputational damage at level 3, uploaders ought to make sure proper permission is in place before uploading, and speedy deletion of those which lack permission sounds the right way to go. --MichaelMaggs (talk) 07:53, 3 February 2009 (UTC)
✓  Done See new section "What should I do if I find an image on Commons which breaches these rules?", below. --MichaelMaggs (talk) 17:50, 3 February 2009 (UTC)

Avoiding problems[edit]

What should I do if I find an image on Commons which breaches these rules?[edit]

New section, added at the suggestion of Gnan, above. --MichaelMaggs (talk) 17:50, 3 February 2009 (UTC)

(Other general feedback)[edit]

More to say later, haven't checked for loose ends and minor issues but my general impression is that this is a stunningly thorough piece of work, and the guidelines, dividers, criteria and so forth that it lays out will do a great deal to reduce some of the conflicts we have had between good faith contributors who might not necessarily agree. Strongly urge the careful review of this for any errors or loose ends (without changing the general sense, the multiple levels are extremely useful and should be kept) and then the adoption of this as official formal policy. We will still have edge cases and conditions where we may not agree on where things go, but we should have far less than before. With the gray areas narrower we can err on the side of safety without throwing out vast swaths of things that might well be OK to keep. Bravo. Thank you thank you. ++Lar: t/c 23:14, 2 February 2009 (UTC)

echo the sentiment that this is really strong, necessary work - thanks! - I've posted previously that it's desirable for some aspects of the policy ('moral issues' for example) to apply to all media, not just 'identifiable people' - 'upskirts' and 'downblouses' for example don't clearly identify the subject in my view, but I believe we should expect the subject's permission. That's clearly 'off topic' for a page discussing 'identifiable people' though - so this is really a big thank you from me - cheers, Privatemusings (talk) 23:40, 2 February 2009 (UTC)
I also agree this looks really good, and gives a clear direction on identifiable person type photographs. Suggest later expansion to country specific guidelines/laws so that editors are able to understand it application in places they dont know. Gnangarra 01:39, 3 February 2009 (UTC)

This is a great work and I would like to thank MichaelMaggs for that. I don't fully agree with every single detail, but I'd gladly enforce such a policy. As a pretty psycho-rigid Frenchman working in the field of privacy, I would have liked a higher level of explicit consent expectation at some points, but I understand the rationale behind every section, my doubts have been explicitly addressed in the text and I promise I would behave :-) It seems to me that this text is detailed enough to provide a clean basis for our work. I can't see any obvious "loose end", but I'm not infallible. I will make a few remarks about consistency and details in the relevant sections. Again, good work. I hope it will become an official policy soon. --Eusebius (talk) 21:58, 3 February 2009 (UTC)

Thanks for trying. It is a neat simple scheme. I like the concept of reasonable expect. "Can I investigate two possible. This demonstrates both of them.File:Bundesarchiv Bild 183-1987-0702-020, Berlin, SEZ, Duschen.jpg. All the Bundesarchive pictures I believe should be kept.

  • Firstly- if the image is old enough- it should be accepted even if today it would be differently categorised.
  • The Lass who appears topless is in Germany so would accept that being photographed in public like this would be acceptable- I imagine that if this were in Italy- then she would see this as an intrusion.
  • The lad in trunks in this context would accept the photograph. Under current UK law, his parents social worker could see this as a serious act of negligence on behalf of his parents. It is better if no consent is asked.
  • All these images have been approved by the national authority for Commons publication- that should be reason enough for publication.

Theres a few ideas. --ClemRutter (talk) 22:59, 3 February 2009 (UTC)

To me, that image is OK as it appears to have been taken in a public place. --MichaelMaggs (talk) 19:07, 5 February 2009 (UTC)
  • First of all, many thanks to you, Michael, for these proposals. A really good and detailed work! There are some smaller points where I do not fully agree with but it is, of course, much more important to find a broad agreement. I join Eusebius in his questions under single headings above and have only one general proposal touching local laws of personality rights that - in my understanding - are imperative and should clearly be accepted as such. For example, in some European countries identifiable is explicitly not limited to the face and also means identifiable by a near person (not only by strangers on the street). Laws also diverge in allowing unconsented publication of photos taken in public and private places, and so on. It would be helpful to write an overview over local laws comparable to COM:FOP. If wanted I am gladly ready to help researching and writing them. --Martina Nolte (talk) 10:28, 4 February 2009 (UTC)
Thanks for the comments Martina. National laws on Personality rights do indeed diverge widely between countries, and I think you are arguing that we should never keep an image if the subject could, under local law, object to it on the basis of Personality rights. That's a perfectly respectable position to take, but I am afraid it does not seem to be supported by consensus here over several years. It seems fairly clear from many deletion requests that unless the subject actually asks us to remove the image, the fact that he or she could potentially do so is not enough for the image to be deleted. My proposals reflect that consensus as I understand it. --MichaelMaggs (talk) 19:07, 5 February 2009 (UTC)
Why does Commons respect and adopt local laws regarding COM:FOP, but should not do so in regard of personality rights? I don't understand where the difference could be seen. In most cases of pictures that I am thinking of it would be sufficient if the uploader states that the person agreed to publication under a free licence. --Martina Nolte (talk) 13:40, 7 February 2009 (UTC)
The distinction, I think, is that FOP is a freedom to photograph free of copyright restrictions, whereas Personality rights relete to non-copyright restrictions. Commons generally ignores Non-copyright restrictions. --MichaelMaggs (talk) 18:13, 9 February 2009 (UTC)
  • Some years ago my German, now naturalized American sister, dared to allow her daughters (3 years and 5 years old) to play on a public beach in en:Colorado Springs. The problem was: The girls have weared their panties, but not their bikini top. This caused so much trouble, that she never allowed that again, neither in other American states. This is a clash of civilization between the US and the rest of world, including Arabia. So all attempts to find a solution for all states seem to be rubbish. Let wp.en choose their pics and let the interwikis choose their pics to illustrate the articles. 11:12, 4 February 2009 (UTC)
So all attempts to find a solution for all states seem to be rubbish. So it is! {{Personality rights}} clearly states "Dear user, you are responsible for whatever you want to do with that image. Check the laws!" Our German Wikipedia friends are indeed invited to make a german translation, just in case someone has a language problem with that english only template. -- 14:52, 4 February 2009 (UTC)
Yes, a clash of cultures indeed. The best we can do is to warn re-users that they may in some countries infringe local rights. --MichaelMaggs (talk) 19:09, 5 February 2009 (UTC)

I have been browsing around a bit, and one of the more problematic images is of a person fully dressed, shot in a place open to the public, and in use. I am talking about File:Friki.jpg. It is in use to illustrate en:Nerd Pride Day and "Geek" in Spanish, and that is also the meaning of the filename ("freak"). There is no evidence that subject is a self-declared nerd, there is no indication of posing or consent. Not quite clear if the material with skin tones that he is reading is sexual in nature. Does it matter much? Does it make a difference if the guy is 17 or 19 or 23? I think it is an invasion of privacy. It is probably too hard to formulate general rules. /Pieter Kuiper (talk) 20:16, 9 February 2009 (UTC)

Yes, that one does merit discussion. The title "freak" to me sounds potentially defamatory, and I would think this section of the guidelines would apply: "You should bear in mind that defamation may arise not only from the content of the image itself but also from its description and title when uploaded" (the same wording appears both in the existing policy and in this proposal). I would rename. --MichaelMaggs (talk) 21:07, 9 February 2009 (UTC)
Renaming might help a bit if it only were for commons, but the problem that this image is used to illustrate gl:Friki etcetera would still remain. Should the context in which images are used matter to commons? /Pieter Kuiper (talk) 21:23, 11 February 2009 (UTC)
No, I don't think it should. We are here just to host the images, and what the wikis and indeed other users do with them is outside our remit. In the same manner, we do not delete an image which is alleged to be factually incorrect or "wrong" in some way if the image is in use: see COM:PS#"Neutral point of view". --MichaelMaggs (talk) 22:54, 11 February 2009 (UTC)


  • I can't wait for the future discussions about some photo of "identifiable unclothed or partly unclothed adult" where level of professionalism of the model and/or photographer will be judged based on the image itself.
  • Some images are non-controversial and some have special rules because of: nudity, age, disease or medical condition of the subject. It might simplify the policy if we lump all the special rule cases together and treat them the same. I did not study the proposal in enough detail to see how much it would change. Is such lumping can be done than the policy would be easier to remember and enforce, and future "decision trees" (like File:Derivative Works Decision Tree.svg) would be much less "bushy". --Jarekt (talk) 14:15, 4 February 2009 (UTC)
The decision tree could be made simpler, but only at the expense of less consistency with current consensus. Make some suggestions by all means. --MichaelMaggs (talk) 19:19, 5 February 2009 (UTC)
  • Ambitious work, but new detailed rules are likely to cause loads of deletion requests for previously uploaded images. It will often not be practical to obtain consent. /Pieter Kuiper (talk) 07:49, 5 February 2009 (UTC)
When we are formulating policy we ought to be thinking about what is right, and going from there. Once the policy has been agreed on that basis, if we have images that don't satisy it they will have to go. Otherwise we have no rules but just a straight vote on every image. Do please mention specific examples of images that you think should be kept but which will fail these rules. --MichaelMaggs (talk) 19:19, 5 February 2009 (UTC)
Some good examples have been brought up: classroom photos, Bundesarchiv photos of clothing-optional beaches, photos of kids with chicken pox. I am afraid that Martina Noltke will find many, many more. /Pieter Kuiper (talk) 19:34, 5 February 2009 (UTC)
All of those I have addressed above, but I accept that you may disagree. --MichaelMaggs (talk) 19:49, 5 February 2009 (UTC)

Great but[edit]

Jimmy image

The last couple of months i noticed some people who wanted an image deleted (for whatever reason). They just wanted to get rid of it and started looking for reasons to get it deleted. I'm afraid this proposal is going to be used as an excuse to get rid of images some people don't like. Take as an example the Jimmy image. I don't like Jimmy, i hate him and i want to get rid of the image. You say you have approval, but i don't believe you. Send it to OTRS. How are we going to prevent this kind of misuse? Multichill (talk) 11:07, 6 February 2009 (UTC)

To me, this is level 2 at most (unposed photographs where an identifiable adult in a private place does not appear to be aware of the presence of the photographer), and the subject may even be aware. Either way, this photos does not need OTRS, and anyone requesting deletion on the basis you suggest would find the DR closed very fast as "keep". --MichaelMaggs (talk) 18:13, 6 February 2009 (UTC)
✓  Done To clarify that OTRS should not be demanded, I have added "An OTRS consent would also be fine, but OTRS should not be required at this level". --MichaelMaggs (talk) 18:16, 6 February 2009 (UTC)
And see File:Raël giving a Cellular Transmission.jpgGeni (talk) 12:23, 6 February 2009 (UTC)
I have no idea what is being shown there. Is it some ritual where the presence of spectators with cameras/cameraphones could be expected? If so, it counts as a public place. --MichaelMaggs (talk) 18:32, 6 February 2009 (UTC)
spectators yes cameras less so. Takeing pics of Raël isn't allowed.Geni (talk) 19:43, 9 February 2009 (UTC)
Allowed by whom, though? That sounds like a non-copyright restriction which is not a reason for Commons to reject an image: see Commons:Non-copyright restrictions. --MichaelMaggs (talk) 19:58, 9 February 2009 (UTC)

Another issues is with camera resolutions getting higher what used to be unidentifiable people in the background are becoming identifiable.Geni (talk) 12:23, 6 February 2009 (UTC)

I'm still not clear to me how we're going to prevent misuse like i described above. Multichill (talk) 08:31, 9 March 2009 (UTC)

It seems to me that the level-based classification provides us with enough tools to avoid them (it is at least very clear in the Jimbo example, Raël example would probably need circumstancial discussion). --Eusebius (talk) 12:00, 9 March 2009 (UTC)
Misuse would be apparent in a deletion discussion, as suggested below we place a cut off date for images prior to implimentation where we presume consent after that, then consent is required anyway so its not going to be open to anymore abuse then current discussion process is. Gnangarra 14:09, 9 March 2009 (UTC)


Some of the comments above have objected to minors being treated more "strictly" than adults, particularly at level 3. Would there be consensus to scrap the distinction between minors and adults? --MichaelMaggs (talk) 18:29, 6 February 2009 (UTC)

No, because minors cannot themselves give informed consent. Also the laws of Florida, where the servers are housed, make a distinction. ++Lar: t/c 06:11, 7 February 2009 (UTC)
Do you know what the distinction in Florida law is? Pruneautalk 15:11, 7 February 2009 (UTC)
I don't think anyone is proposing to get rid of the requirement for parental consent for pictures of minors — what I questioned above is whether, in addition to that, it really makes sense for pictures of minors to automatically be bumped one level up in the "level 1–3" system. Specifically, I'd like to see some more discussion on whether "an identifiable minor in a private place showing any disease or medical condition" and "an identifiable unclothed or partly clothed minor in a private place" are really examples of cases that should automatically require level 3 proof of consent (i.e. explicit statement of consent via OTRS). (I think we can all agree that pictures showing "a minor in any type of sexual pose", where legal at all, should indeed require this level of proof.) Note that the former category could be interpreted to include images like File:Windpocken.jpg, while the latter could include e.g. File:Baby diaper.jpg or File:Bundesarchiv Bild 183-1984-0322-016, Klein Bünzow, Krippenkinder im Dampfbad.jpg. —Ilmari Karonen (talk) 19:09, 8 February 2009 (UTC)
There is something to be said for the argument that unclothed minors should not automatically be bumped up one level, especially images of babies where unclothed or clothed in just a nappy/diaper is a pretty normal day-to-day state. "Unclothed" does not have the connotations with a baby that it might have with an adult. What would you say to changing the level 3 example from:
  • "Photographs of an identifiable unclothed or partly clothed minor in a private place, or of a minor in any type of sexual pose. (Clearly, this applies only where the image is legal.)"
  • "Photographs of an identifiable minor in any type of sexual pose in a private place. (Clearly, this applies only where the image is legal.)" ?
--MichaelMaggs (talk) 18:26, 9 February 2009 (UTC)
The latter formulation is more clear, but more permissive. I'd like to keep a reference to nudity, because (as medical condition) they reveal something about the person that people usually do not show. --Eusebius (talk) 18:57, 9 February 2009 (UTC)
See proposed wording I have quoted below. It is more general than nudity. --MichaelMaggs (talk) 20:07, 9 February 2009 (UTC)

Medical conditions[edit]

Some of the comments above have suggested that what would be covered is unclear. Would there be consensus to scrap reference to medical conditions? --MichaelMaggs (talk) 18:29, 6 February 2009 (UTC)

  • Symbol support vote.svg  Support removing the section. There are many cases where a medical condition cannot be realistically expected to cause harm. When such a condition can be expected to cause harm, we can always use the basic definition of level 2 or 3 to delete. ("The hosting on Commons of images at this level might be realistically – without consent - be expected to cause the subject significant harm, distress or embarrassment.") Too much detail won't help in this case. Pruneautalk 21:54, 6 February 2009 (UTC)
    • ✓  Done Medical condition text removed, as this does not seem to be gaining a consensus. --MichaelMaggs (talk) 18:15, 9 February 2009 (UTC)
  • Symbol oppose vote.svg  Oppose (too late?) I think a word about medical conditions should be present, maybe only to say that they can cause more harm and should be considered with care. --Eusebius (talk) 18:54, 9 February 2009 (UTC)
Eusebuis, you don't need to oppose, just to ask :) I think the only way to include both points of view would be to have some general wording such as "Depending on the risk of harm to the subject, some images showing medical conditions may warrant a higher level than would otherwise be the case, as may the manner in which a minor is depicted". I have added a suggestion along those lines. Does that help? --MichaelMaggs (talk) 20:05, 9 February 2009 (UTC)
Looks ok to me. --Eusebius (talk) 20:13, 9 February 2009 (UTC)


The proposal uses the term legal, but do not define according to which jurisdiction is has to be legal. The usual Commons way - legal in Florida, in the country of the image's orgin and in the country of the upload - has been deleted recently (was #Acceptability [1]). I support this deletion, but we need a new one!?!

  1. Only the law in Florida counts?

sугсго 10:16, 10 February 2009 (UTC)

There is definitely no consensus for that, but neither is there consensus that we should take into account all local laws such as personality rights, rights to images of Public domain antiquities that are claimed in Eygpt etc, hence the deletion of that section. All we can reasonably agree on is that the image must be fully legal in Florida, and that a variety of Non-copyright restrictions, including some foreign law restrictions, are not taken into account. Users are free to argue the point, though, on a case by case basis. That is not ideal, as it is intrinsically uncertain, but it's as far as we can go at the moment. Policty does develop over time, and maybe in another couple of years the issue might be clearer. --MichaelMaggs (talk) 17:41, 10 February 2009 (UTC)
Michael, I don't understand your first sentences clearly. We are dealing here with images of living persons, thus "local laws", of course, means: personality rights. Your Egypt example concerns property rights or did I get it totally wrong? --Martina Nolte (talk) 19:17, 16 February 2009 (UTC)


Support the proposal[edit]

Please say why.

  1. Comprehensive, reworked and discussed proposal. Lycaon (talk) 11:05, 6 February 2009 (UTC)
  2. I have minor quibbles with some bits that are too permissive, and some that are too restrictive (per my personal preference) But this proposal is comprehensive, respectful of the rights of others beyond what the law requires, reasonable in what needs to be done to satisfy it (not onerous to fit within) and will be a substantial improvement over what we have now. I can live with my quibbles. So... support ++Lar: t/c 04:39, 10 February 2009 (UTC)
  3. I think this is really good work, and very necessary and useful - I too reckon there might be tweaks that could help in due course (but wouldn't mind a medium sized period to see how it actually pans out - perhaps 3 months or so?) - this is really good policy work in my boook... Privatemusings (talk) 06:17, 10 February 2009 (UTC)
  4. This policy is clearer and more precise than the current one. I believe it will be easier to reason on this basis. It gives general guidelines that look reasonable to me, and leaves room for the application of local Personality Rights when the subject shows up. Issues on which users/admins strongly disagree have not been eliminated by definitive rules (so there is still room for arguments in the DRs) but rather pointed out and isolated from the issues on which some kind of consensus can be found. --Eusebius (talk) 06:44, 10 February 2009 (UTC)
  5. Because we need to start somewhere, like all policies//guidelines etc they are going to get changed(improved) over time. This is comprehensive in what is trying to be established, IMHO until its policy we arent going to able to tweak appropriately until its truely tested via deletion com:an etc.. Gnangarra 12:14, 11 February 2009 (UTC)

Support the proposal with reservations[edit]

Please indicate specifically how you would like the proposal to be reworded to deal with your reservation.

  1. --Martina Nolte (talk) 01:34, 10 February 2009 (UTC) More detailed than the older one, so that I do not vote against it. I could only support the proposals if local laws of personality rights would be respected as obligatory (comparable to COM:FOP). User's and contributer's wish to get and keep much images, in my eyes, cannot overrule legal issues, or otherwise: law cannot be overruled by project consensus. The guideline would invite to offend laws and thus harm Wikimedia's reputation.
  2. I like the fact that this proposal clarifies the current policy, but I personally (and possibly many others) did not realize that we have a policy that can be used to delete images with correct licenses. Current and proposed policies affect many indoor photographs, especially older ones or transferred from flicker or Wikipedia. I am afraid we will loose a lot of them. On the other hand I agree that if Commons is breaking some laws by hosting some kind of image than the image should be removed.--Jarekt (talk) 03:56, 12 February 2009 (UTC)
    Um, if I understand correctly, your reservation is that Commons has been working to a longstanding policy, namely COM:PEOPLE, that you were unaware of? --MichaelMaggs (talk) 17:35, 12 February 2009 (UTC)
    Let me try to clarify. I have always seen COM:PEOPLE as a general guidelines, but not as requirements which can be used in deletion request. I also have never seen deletion request based on COM:PEOPLE (possibly because I mostly work with outdoor and historical photographs). I am afraid that we do have a lot of images if people indoors which we will loose. If those images are breaking some laws (of Florida and country of origin?) than I support it, otherwise I am afraid of loss of images and time and effort people put into organizing them. --Jarekt (talk) 13:44, 17 February 2009 (UTC)
  3. -- smial (talk) 00:02, 14 February 2009 (UTC) Ein Schritt in die richtige Richtung, der hoffentlich für mehr Klarheit sorgt. Allerdings ist die Definition, was eine identifizierbare Person ist, deutlich zu eng gefaßt, es sind außer dem Gesicht zahlreiche weitere Merkmale für eine eindeutige Identifizierung denkbar, Verpixelung, Crop oder eine Hinterkopfansicht reichen sehr oft nicht zur Anonymisierung. Weiterhin sollte bei Uploads aus fremden Quellen wie flickr, suicidegirls usw. grundsätzlich eine Einverständniserklärung o.ä. der abgebildeten Personen verlangt werden, sobald auch nur geringer zweifel an der Öffentlichkeit des Aufnahmeorts besteht. Bei Uploads durch bereits bekannte Wikipedia-Fotografen kann man gerne großzügig AGF walten lassen, aber nicht bei fremden Quellen. -- smial (talk) 00:02, 14 February 2009 (UTC)
    Would you (or somebody else) please provide an English version of your comment? --Eusebius (talk) 08:12, 14 February 2009 (UTC)
    using a rather helpful google translator I get "A step in the right direction, hopefully to provide greater clarity. However, the definition of what an identifiable person is clearly too restrictive, there are in addition to the numerous other facial features for a unique identification possible Verpixelung, Crop or a view of the back very often not the anonymization. Furthermore, should uploads from external sources such as flickr, suicide girls, etc. basically a declaration of consent, etc. of the depicted persons are required, even when little doubt in the public of the image exists. For uploads using already well-known Wikipedia-like photographers can generously AGF lenient, but not with foreign sources.", thinking along these lines I can agree with the points raised about the definition. Though I can see the value of assumption in the positive for Flickr type sourced images I think we should be taking the copyright approach where theres doubt then delete. I can agree in some latitude for regular contributors but again these people should already know the requirements and address them before the uploading, we are human and do make mistakes. translation of my comment Denken in diese Richtung kann ich mit den Punkten, die über die Definition. Obwohl ich sehe den Wert der Übernahme in die Positivliste für Flickr-Bilder stammen Art Ich glaube, wir sollten dabei die Copyright-Konzept, bei dem Zweifel Theres löschen. Ich kann in einigen Spielraum für regelmäßigen Beiträgen, aber auch diese Leute sollten schon wissen, die Anforderungen und sie vor dem Hochladen, wir sind Menschen und machen Fehler Gnangarra 13:17, 14 February 2009 (UTC)
    Sorry for my german comment, but I'm EN-1, can read most english text, but have strong difficulties to discuss complex topics. -- smial (talk) 13:36, 14 February 2009 (UTC)
    Might be it sometimes takes a little time but I'm ready to translate your and other's German contribs in English. I think discussing this important issue should not be handicapped by language. I know an handful of good photographers on de-WP that do not upload on Commons for different reasons (and one of them we are discussion here) and I think it might be very useful to get them "in the boat". Wenn jemand eine Übersetzung von Englisch ins Deutsche möchte, bitte Bescheid sagen. --Martina Nolte (talk) 15:15, 14 February 2009 (UTC)

Oppose the proposal[edit]

Please say why.

  1. This policy will lead to the needless deletion of many useful images because of the requirement of consent for "Level 2". I do not understand how the "typical examples" listed there "might realistically be expected to cause the subject harm, distress or embarrassment". Many of these cases will be images of people at work, at play or at school, just doing very ordinary things. /Pieter Kuiper (talk) 18:41, 12 February 2009 (UTC)
    I'm also worried about this possibility. I think that the policy calls for simply a believable assertion by the uploader barely makes this acceptable to me. And I do mean barely. I think this part of the policy needs to be more permissive, but I don't know that I would oppose the changes without other reasons for doing so.  — Mike.lifeguard 20:59, 13 February 2009 (UTC)
  2. See Pieter and #Great but. Multichill (talk) 08:31, 9 March 2009 (UTC)


Has anyone asked Cary whether expanding the role of OTRS in this manner is an acceptable use of resources? We are already struggling to keep up with permissions (see probably 2 or 3 locations just at this moment lamenting the backlog in permissions queues) - do we really need to be adding to the burden? Is there another way of achieving the same thing or something similar without using OTRS? OTOH, perhaps it's fine & there's no problem except in my brain.  — Mike.lifeguard 06:05, 12 February 2009 (UTC)

I honestly think this is a non-problem. OTRS comes in as a requirement only at level 3 and there are very few uploads that fall into that level, and even fewer where anyone will actually submit a proper consent that needs to go to OTRS. This will make a trivial difference to OTRS workloads, but will allow admins a clearer mandate to delete quickly images that are potentially very harmful withiout having to go through a lengthy DR procedure with its built-in 7 day waiting period .--MichaelMaggs (talk) 17:41, 12 February 2009 (UTC)
I think a formal acknowledgement and acceptance of the change in scope for OTRS would be required for me to accept this. Furthermore, I think the presumption that we should be deleting images straight away may be problematic. I will probably expand on that in a few minutes after reviewing some recent changes I've not yet read. Depending on what they say, that may or may not be valid at this point.  — Mike.lifeguard 20:54, 13 February 2009 (UTC)
I think that objection is valid only insofar as we end up deleting images which don't have a high potential for harm. This policy must not be taken as a carte blanche for people to nuke images they don't happen to like. Given recent events, I think that's a reasonable reservation.  — Mike.lifeguard 21:20, 13 February 2009 (UTC)
Again, Mike, I am quite tired of this reducing the "recent events" to imputation of "personal taste" (subtextually meaning pov guided censorship). Alleging this to other people again and again (altough they delivered much more detailed and objective arguments) is propaganda and spin. Of course, the proposed guideline or an even more restricted one could and probably will lead to a bundle of deletions. On the other hand we will get - in the long run - a much higher quality and legal standard. We would get - insted of the deleted ones - more good photos on Commons, because quiet a lot of photographers consciously do only contribute on their local WP (some even aks expressivly to NOT remove their pics to Commons!) and they will only be willing to contribute here if they are sure that the platform on which they publish is highly reputable. I talked to nearly 50 people from de-WP personally in the last weeks on this question so that I do not state only a personal fantasy here. Organisational questions like OTRS capacity should not lead to unright guidelines. IMO we better should think about how to organize RC control and quality management. I am planning to prepair some suggestions and then bringing them into the forum. Also in this point I am convinced that - once the community shows a readiness to implement a higher legal and quality standard - we will get enough helping hands. --Martina Nolte (talk) 15:46, 14 February 2009 (UTC)
I think it's entirely fair to characterize some people's recent actions as inappropriate (you'll note, I'm sure, that the community at large agrees with that assessment). Whether that has or should have any bearing on this proposal is an argument that should be made and I'm aware that I haven't made a convincing argument, only an assertion. I think it's a tenable one though.
I'm also well aware of the objections of some enwiki and dewiki contributors (in particular) to Commons, and I find them to be mostly baseless. This is one area where they're not, but I don't think it's a good reason to not use Commons. Nevertheless, we should work on improving the current policy - not because of instrumental benefits like dewiki photographers uploading good images, but instead for principled reasons: We shouldn't violate people's privacy. That's why I think the proposal should probably go further in doing so than it does currently - a fact which is apparently invisible. As a side-note, I'm not concerned about deletions per se but rather bad deletions. Losing images is fine (good, actually) - as long as they're the right images.
As for my comments about OTRS, I was referring specifically to using OTRS to archive model releases and my only objection was that Cary should probably consent to expanding its scope. It may not have been clear to you, but I was not saying the whole proposal should be scrapped based on that concern.
 — Mike.lifeguard 17:11, 14 February 2009 (UTC)

Unidentifiable people[edit]

While this policy applies to identifiable people only, I am concerned that this means that any and all violations of privacy are acceptable if you just pixellate out someone's face, or crop it out or whatever. As has come up time and time again in conversations regarding CheckUser, once violated, privacy remains violated (it cannot be "taken back") - thus I think there can be an argument about how much identifiability should be a factor in these decisions.

For example, here I argued that identifiability is just one factor to consider among the others which this proposal discusses. So, this is more conservative in the sense that non-identifiable images can be considered. I'd be interested to hear what others have to say about that view. That my arguments were apparently convincing in that case is heartening. I think there was another DR where I made a similar argument, but I forget what/where it was.  — Mike.lifeguard 21:11, 13 February 2009 (UTC)

I'm not as good as you when it comes to arguments... I think we should stick to identifiable people, but I also think the notion should be a bit wider (kind of similar point of view, then). I think that relying on a visible face or the ability of a stranger to recognize the person is too narrow: when privacy is violated, consequences can sometimes be more serious (for the concerned person) when the sensitive information is disclosed to close relatives/friends. I don't really care if some guy in another country sees a picture of me (let's say a naked and drunk picture, and let's precise that it's totally hypothetic) with my face cropped. He probably can't know it's me, even if he has another pic of me. But if my brother/sister/mother/best friend see that, they would certainly recognize me and it would cause me harm. --Eusebius (talk) 21:46, 13 February 2009 (UTC)
I'm going to be away for the next week. Mike and Eusebius, why don't you try knocking some ideas around and see where you get to before I get back? All the best, --MichaelMaggs (talk) 22:40, 13 February 2009 (UTC)
Who, me? Oh, ok, then if I'm supposed to talk seriously, I guess I should moderate what I've just said (and have a nice break). I pointed out the fact that a picture that would be considered by us as of an unidentifiable person could cause a privacy-related harm, but I'm not even backed by French law here, at least not in a systematic way (I think we can take the French law as a fairly protective example). French courts usually require that the face or some distinctive body feature be visible, or that strangers (judges) be able to recognize the person by basing themselves on other pictures. I'd be even more protective if I were a judge, but thanks God, I'm not. I think we should consider the potential damage of pictures of unidentifiable persons (unidentifiable by "careful reviewers", not by close relatives) as possible, but exceptional. I don't think we should mention it in the current policy. --Eusebius (talk) 23:15, 13 February 2009 (UTC)
Yes, well I actually overstated things as well... some kinds of intrusion can be remedied by pixellating out the face or something (File:Childhood Obesity.JPG for example), but other's can't (such as the example I mentioned above. This sort of equivocation is why I think we should prefer to make arguments at a DR rather than allowing (and even encouraging) unilateral speedy deletion. It's possible (in fact, true) that people will take varying stances when considering these issues - in unclear cases I would not want one administrator to make the decision in isolation.  — Mike.lifeguard 20:54, 14 February 2009 (UTC)
Do you think something be added to the current proposal to precise things, in the "What does “identifiable” mean?" section maybe? --Eusebius (talk) 21:00, 14 February 2009 (UTC)
Yeah but be careful in defining, I think the less precise the definition the more room editors will have to use reasonable judgement. I think this image could be swung many ways depending on the definition, while the face is partly obscured place of employment enhances the ID, is it in the public view in a zoo or because the people were in an area that isnt accessable to the public are they in a private place. IMHO maybe we should look for a few examples and ensure we get a reasonable balance initially then be able to tweak as the policy gets actual rather than theoritical use. Gnangarra 01:37, 16 February 2009 (UTC)

Current statement is:

“The subject of a photograph is “identifiable” when he or she might be recognised in public by a stranger on the basis of the photograph. Normally, that will require that the face be visible, but occasionally there may be other identifying characteristics (unusual tattoos, one arm missing etc). 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 (eg on the basis of some characteristic such as a mole which would normally be hidden).”

What I suggest we might add (although the current statement is ok for me) (please correct my english):

“The subject can often be made unidentifiable by technical means aiming at hiding the face (like pixellating or cropping). However, this might not be sufficient in some particular cases, and if subjects manifest themselves complaining that the photograph is still harmful to their image, their demands should be carefully considered even though their face might not be directly visible.”

Please comment. --Eusebius (talk) 08:11, 17 February 2009 (UTC)

I support what you're getting at there. Unlike celebrity images and historic images, images that are illustrative of something (attire, activity, condition, whatever) almost always can be replaced, so we should err on the side of respecting the depicted person's wishes. ++Lar: t/c 15:55, 17 February 2009 (UTC)
support with Lar --Martina Nolte (talk) 18:27, 17 February 2009 (UTC)

Sure, but I was talking about images where the subject is definitely not identifiable yet a privacy violation exists in taking the picture (as opposed to in the picture itself... see the DR I linked above for example).  — Mike.lifeguard 19:14, 17 February 2009 (UTC)

I'm not talking about this DR in particular, but if the person cannot be identified, the picture cannot be considered as a privacy violation, since no private information can be linked to any individual. --Eusebius (talk) 19:24, 17 February 2009 (UTC)
There is no definitive way to know for certain that a person cannot be identified. Therefore erring on the side of caution is appropriate. ++Lar: t/c 22:18, 17 February 2009 (UTC)

Further options?[edit]

I really don't see a clear way forward here. It seems to me that there are some fairly strongly polarised views with insufficient incentive to compromise. Maybe somebody else could do better than me in seeking an acceptable middle ground. Or perhaps there may be just one or two suggestions here that could be split out and which might find consensus? I am happy to continue contributing but don't feel I can take the lead role going forward. --MichaelMaggs (talk) 17:11, 7 March 2009 (UTC)

Really? from what I'm reading there hasnt been any discussion for 3 weeks, the last of that was saying the we can differ on the side of caution(exercise commons sense) as to unidentified people. IMHO I think its ready to be put to the community for adoption as a policy Gnangarra 01:04, 8 March 2009 (UTC)
It would be an error to err on the side of caution. I can only repeat myself: This policy will lead to the needless deletion of many useful images because of the requirement of consent for "Level 2". I do not understand how the "typical examples" listed there "might realistically be expected to cause the subject harm, distress or embarrassment". Many of these cases will be images of people at work, at play or at school, just doing very ordinary things. /Pieter Kuiper (talk) 08:34, 8 March 2009 (UTC)
There are lots of useful images we cannot use. Wrong licenses, personal rights issues, copyright problems. Why do personal rights not apply to pix of people at work, at school, or at play? Why should personal rights stand behind copyright? -- smial (talk) 15:43, 8 March 2009 (UTC)
Pieter: I understand there is a cultural clash here. Simply being publicly visible on the internet while at work or at school would be considered as a harm (in itself) for almost everybody I know. I find it very difficult to explain and share this with somebody with very different feelings about that, this is why I'm not sure my remarks are very useful to the debate. --Eusebius (talk) 15:53, 8 March 2009 (UTC)
Pieter, what if we were to take the approach of assuming good faith with all images uploaded before 1 March 2009 we say some thing like, for images uploaded before the 1 March 2009 there is a presumption of consent from the subject of the photograph as such these photographs are excluded from the requirement for written consent. Then if we recieve a request for an image to be deleted its up to the person making the request to establish they are the subject. The date choice is based on my opinion that its better to select a recently passed date to avoid mass uploading to circumvent the policies implimentation. Gnangarra 00:06, 9 March 2009 (UTC)
That would solve some problems. But suppose we would get another donation like Bundesarchiv - how would you deal with that? The proposed policy would exclude many useful images from being uploaded. /Pieter Kuiper (talk) 08:42, 9 March 2009 (UTC)
Yes it would. But where is the problem? Some or many images may have to be deleted, some or many cannot be upoaded. But with the remaining we could be sure, not to violate any personality rights. Very many of the possibly affected images are useless at all or can be substituted by uncritical images. I believe human rights and/or personality rights should not be valued minor to e.g. copyright. And I believe commons must not be a conglomeration of rubbish or any photo ever taken or showboater. Quality counts. -- smial (talk) 10:51, 9 March 2009 (UTC)
Smial seems to desire a clean and sanitized Commons, where naked breasts are tucked away in remote categories. I fear that non-copyright reasons for deletion will just lead to more censorship. /Pieter Kuiper (talk) 17:52, 10 March 2009 (UTC)
I made some photos of piercings/body modifications that are possibly a bit better than most of those pix at commons. But as I have no permission to show them, they are not uploaded. That's all. This has nothing to do with censorship. Too much garbage here. -- smial (talk) 09:39, 11 March 2009 (UTC)
I'd like to prepair with interested people a proposal for a "task force" of users who would help to contact or e-mail uploaders of Level 2 images to give statements that consent has been provided, or to help Level 3 uploaders how to get releases, to formulate and to translate specimens for letters and releases and so on. Such a team could help saving a lot of pictures from deletion. I understand that there is a fear of loosing (to) much pictures (and perhaps discouraging new uploads, too) and perhaps people like Pieter Kuiper are more willing to accept the actual proposal if such a helping team is provided. I would, of course, offer to take part in such a team. What about that idea, Pieter?--Martina Nolte (talk) 17:13, 10 March 2009 (UTC)
That seems very encouraging, and shows the extent to which Martina is prepared to compromise to find some middle ground here. Pieter, are you prepared to compromise too? --MichaelMaggs (talk) 18:07, 10 March 2009 (UTC)
Was that a compromise?? Gnangarra's suggestion was more reasonable, and I would consider it, but Nolte's task force will lead to the removal of loads of images, for example from the Bundesarchiv donation. For how would the task force contact these people? /Pieter Kuiper (talk) 18:31, 10 March 2009 (UTC)
Yes indeed, it was a very substantial compromise as she and many other users here come from societies in which personality rights are strongly protected, which could mean no portaits of living people here at all without subject consent. Gnangarra's suggestion is also helpful. So, would you support Marina's suggestion subject to Gnangarra's cutoff date? --MichaelMaggs (talk) 19:13, 10 March 2009 (UTC)
With a cutoff date being the date of introduction of a new policy, a task force will not be necessary. Yes, societies are different, but why impose restrictions on societies that do not feel a need for this? Why should Germans worry about personality rights of Americans? I do not believe the participants in this discussion are representative for commons. I am not in favour of these new restrictions, but if you want to run this, I will not object against a test period. That should make others aware of the consequences, which will either lead to a wider consensus in favour of the new regime or to a reversion to the current situation. /Pieter Kuiper (talk) 19:46, 10 March 2009 (UTC)
Why should Germans worry about personality rights of Americans? Nobody said that. Actually Commons lets rule American law over local laws - not only of Germany, but (as far as I know) most European countries and others more. For this reason I already proposed a guideline for personality rights comparable to COM:FOP to meet all ends. Btw, Pieter: my name is not Nolte, but Martina or - if you want to stay highly polite - Martina Nolte. Thanks. --Martina Nolte (talk) 21:40, 10 March 2009 (UTC)
At german WP the DÜP (Dateiüberprüfung) is a tool (or a team) to 'keep' as many images as possible, not to delete as much as possible. Many uploaders simply do not know about licenses, permissions, personal rights etc. and we try to help them. -- smial (talk) 10:02, 11 March 2009 (UTC)