Commons talk:Featured picture candidates

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

Alteration of candidates through voting period[edit]

At 15:37, 17. Jul. 2016, User:The Photographer reverted his nomination of File:Consolata Cemetery 01.jpg to a previous version after I pointed out the badly done photoshopping on the actual candidate. He did so without any notice on the nomination page.

I don’t know how you feel about this, I think it’s very bad practice (I wrote so in the nomination) and playing games with FPC, for several reasons:

  • Previous voters might have voted differently on the actual candidate. So, it smells like collecting support for one image and then putting another one in its place, transferring the feature!
  • Any judgement previously expressed in the voting is compromised (in the actual case, readers might think, "What the hell did he mean?").

Do our guidelines cover this? If not, we need a rule against this IMHO, saying

  1. that any candidate is judged by users as nominated originally, we don’t need any version mixup through voting until no-one knows which vote refers to which version
  2. that nearly any alteration on it (expect very minor fixes, e.g. dust spot removal) makes it a new candidate that has to be nominated separately,
  3. and that even minor edits have to be clearly announced on the nomination page (maybe by "Pictogram voting comment.svg Comment Edit:"), so any previous voter might reconsider their voting.

Any edit on actual FPC violating this rule should lead to immediate refusal of the candidate.

Furthermore, I think the FP note on the file description page needs to say which version of the file given the featuring applies to (this information is given on featured Wikipedia articles as well, though articles undergo much more editing than files of course). Example: This is a featured picture on Wikimedia Commons (Featured pictures) in the version as of hh:mm, dd-mm-yyyy and is considered one of the finest images

--Kreuzschnabel 16:40, 17 July 2016 (UTC)

  • Pictogram voting comment.svg Comment I agree while editing a picture can change dramatically all the result, the same RAW file can give a bad or a good result according to the editing work. And of course when I vote, I vote also for one editing work and not for another. Christian Ferrer (talk) 18:15, 17 July 2016 (UTC)
For info to meet at the end of your comment, it is possible to protect an image from uploading, maybe once promoted an image should be protected... Christian Ferrer (talk) 18:22, 17 July 2016 (UTC)
The protection idea has some issues. It wouldn't affect admin FP photographers, and we would probably start getting protected edit requests from photographers if they needed to make a minor change that we would normally have allowed during the FPC (removal of a spot that they just noticed, etc). INeverCry 18:34, 17 July 2016 (UTC)
  • Symbol support vote.svg Support Once a file has been nominated, only minor changes should be allowed. I don't like the idea that I may vote support for an image, and then that image is changed to something very different, making my support vote go for something I didn't even see when I voted. Changes like the above should be plainly prohibited, and, if done, should disqualify the image. We already have the alternative procedure if a nominator or other editor wants to present an image where significant changes have been made as a clearly indicated alternate. INeverCry 18:26, 17 July 2016 (UTC)
  • Symbol support vote.svg Support per INeverCry + Kreuzschnabel. --Alchemist-hp (talk) 18:41, 17 July 2016 (UTC)
  • Symbol oppose vote.svg Oppose For any of those 3 proposals. Reading them I've the impression that we would be overreacting. The source of the problem here is that The Photographer applied a relevant edit to a running nomination for which votes have been made, and didn't mention it in the nomination page. That was not okay. The solution for that would be to introduce a guideline to ensure that it shall be ALWAYS documented in the nomination page that there is a new version and what changes have been applied in it. If you like, all those who had voted should be pinged, but I wouldn't go further than that and increase burocracy and introduce restrictions in a well working system. Poco2 18:42, 17 July 2016 (UTC)
  • So I vote, and then the photographer/nominator changes the image significantly, and I have to have the bother of being pinged to see if I want to change my vote? What if I'm not around for a few days to see the changes? My vote gets counted dishonestly in that case. This kind of practice subtracts from the integrity of FPC. INeverCry 18:53, 17 July 2016 (UTC)
  • Honestly, I do not think the system is working too well now. Increasingly, images are being edited through the voting process, following advice such as cropping or brightening, until the image being discussed has become a quite different one from the one I gave support a week ago while I wouldn’t support the edited one, maybe for overprocessing. More and more, voters change their vote therefore. As for pinging, I just feel bothered by being asked to re-review a nomination again and again. It feels like assisting image optimization. That’s why I really do think that any considerable edit makes a different image which demands a separate nomination. – Generally, IMHO a candidate on FPC has to be the final result of an image creation process, and it’s that result which is to be discussed here, not some raw material that still needs major editing. We do not create, we do judge. --Kreuzschnabel 19:08, 17 July 2016 (UTC)
  • You are using the worst case (the picture is edited in a way I wouldn't support anymore) to try to establish a heavy process that would jeopardize one of the main points of FPC: improving the images. You are not willing to re-assess a picture if there is an update but you are expecting that the nominator nominates again and again until all meaningful feedbacks to improve the image run out. Well, that would be the case for those wo want to improve the pictures they nominated, others may ignore improvement proposals to avoid having to go through the process again. So, at the end of the day you are either punishing the nominator (they will get fed up) or hindering the improvement of the pictures. Poco2 20:12, 17 July 2016 (UTC)
  • What happens if someone votes for an FPC, and then they have to take time off to travel, or they're sick, or busy with work, and they come back to see that the FPC has passed but it isn't the same as what they voted for because changes were made after their vote? Why should voters be less important than nominators? INeverCry 20:38, 17 July 2016 (UTC)
  • INeverCry, how often have you got this problem?. On the other hand, I can tell you that I had several times the same problem, but in the other direction with people opposing for a picture, that was improved, but they couldn't change the vote on time. But, my god, that's life and this kind of off-WM things should not play a big role about how we define our guidelines. Furthermore it isn't clear to me the consecuences of documenting which version was awarded FP, is that an argument no to improve it never ever again?. Please don't immobilize FPs and FPCs based of hypothetical cases or fears about what could happen, we need lean processes. Poco2 20:57, 17 July 2016 (UTC)
  • Another interesting question would be how often does an image that needs significant changes during FPC actually end up passing? With your own works, Diego, when I see your best here at FPC, I know without the slightest doubt I'm supporting. You've had many images here that pass with a landslide of support because the images are just beautiful. That's when FPC is really enjoyable for me, not when I have to keep checking back to see if some tedious changes have been done and whether they "worked" or not... INeverCry 21:11, 17 July 2016 (UTC)
  • Well, that's up to the reviewer. If a reviewer believes that the image needs important improvements he shouldn't support or he should oppose until they are implemented. Following the proposal here that would always be a oppose, as an essential improvement (which would be an essential change) cannot be done for that nomination. Please, understand that you'd punish everybody here not only those noms where some tediones changes are done. I do very often improve my noms, specially when others nominate them as there is no way that I keep my potential 15,000 FPCs (I've been often surprised by successful FPs nominated by somebody else that I couldn't think of) up for a flawless FP nom. Poco2 21:21, 17 July 2016 (UTC)
  • Thanks for letting me know, INeverCry, I reviewed it and looks good enough :) Poco2 19:01, 18 July 2016 (UTC)
  • Symbol support vote.svg Support Strongly per INC and KS.--Jebulon (talk) 22:13, 17 July 2016 (UTC)
  • Diego, please understand my proposal is in no way about punishing anybody for anything. My point is to keep the processes of a) optimizing an image and b) judging an image being kept clearly apart. In which way does my proposal affect your nominations? I only propose that any edit should be announced within the discussion (which, to me, is a matter of course, a question of respect towards those who bothered to vote on my nomination), and that the results of major edits (i.e. any editing that is likely to change others’ opinion, such as croppings) should be nominated separately because you never can make sure that any voter for the previous version will take notice of the the change to reconsider their voting. Again: I want final results to be nominated and assessed here, and if they still need editing, it’s not too much asked IMHO to withdraw them and put a fresh nomination on, so the votes on the different versions stay clearly separate. I don’t see any punishment or hindrance in that, just a cleaner assessment process. --Kreuzschnabel 22:15, 17 July 2016 (UTC)
  • Symbol oppose vote.svg Oppose Sorry for almost a day of delay, the change is already commented in the nomination page. I have always said the changes in the images on the nominations and this is no exception. I think that Kreuzschnabel proposal is excesive --The Photographer (talk) 22:55, 17 July 2016 (UTC)
    • That’s exactly my point! Firstly, the edit has not actually been announced (i.e. simultaneously) but rather mentioned more than 7 hours after the revert (15:37 vs. 22:51). Secondly, there has no minor edit been done but the image itself has changed dramatically (from badly done cutout to busy bokeh) so it’s virtually a different rendering. I think this is not OK even when announced in time, because you cannot make sure that each previous voter has a chance of re-revieing the (different) image and reconsider their voting. Minor edits are within reason, severe alterations such as this are not. You should have withdrawn you nomination and put up another one with the altered image. Nothing against this from my side. But never do change candidates within an ongoing review process. --Kreuzschnabel 06:38, 18 July 2016 (UTC)
      • I understand your point, however, this has more to do with common sense of voters and nominator and you can see a good example of what I say here. --The Photographer (talk) 17:37, 18 July 2016 (UTC)
  • Symbol oppose vote.svg Oppose - At QIC, there are many instances of photos being edited to pass the criteria. Here, we've often had the same thing. However, I absolutely would support a rule that requires all changes after nomination to be clearly specified on the nomination page, with everyone who's voted automatically pinged. I also think that radical changes that go beyond spot removal, noise reduction, mild to moderate changes in light and contrast and suggested crops should trigger automatic opposition if the nominated picture is not withdrawn. However, I'd leave that up to the judgment of nominators and reviewers, rather than try to create a definition of "radical changes" and enforce it. -- Ikan Kekek (talk) 04:02, 18 July 2016 (UTC)
    • That’s more or less exactly what I suggested. No radical changes on active nominations. Did I make myself so unclear? --Kreuzschnabel 06:31, 18 July 2016 (UTC)
      • Yes, I'm afraid you do, because I would allow changes much more substantial than just dust removal. A crop can change a picture's composition so significantly that I change from opposing the picture to solidly supporting it. Ikan Kekek (talk) 07:07, 18 July 2016 (UTC)
        • And wouldn’t you agree that, after such a significant alteration, it would be wiser to open a fresh nomination on the altered image, instead of having votes for both version mixed up in one nomination and risk misuse of the votes of those who do not agree with the alteration but don’t notice the alteration either before voting period ends? I prefer the former. I do not want to forbid changes, I just propose to open a fresh nomination on a candidate that has been significantly changed during voting period, withdrawing the nomination on the previous version, in order to certainly know which voting has been given for which version of the image. That’s all. TL;DR: Minor edits: announce them and go on. Major edits: close the nomination by withdrawal and open a new one to avoid voting mixup between versions. Nothing is forbidden, it would just be clearer to assign votings to image versions. --Kreuzschnabel 10:17, 18 July 2016 (UTC)
          • Your opinion is noted, but as I said, I don't see the great usefulness in trying to define "major edits" in a binding rule, especially as some crops produce a negligible effect while others change a photo greatly, and the number of pixels often isn't what makes the difference. The other thing is that to a large degree, I think you're trying to find a global solution for a specific problem that was already solved by individuals deciding to object and vote against the nomination in question. Why create a rule when current consensus voting procedures are arguably working fine? -- Ikan Kekek (talk) 18:10, 18 July 2016 (UTC)
  • Symbol oppose vote.svg Oppose No, please. I do hear you - but what we need is common sense and transparency. And fewer rules. --Martin Falbisoner (talk) 05:53, 18 July 2016 (UTC)
    • I am myself surprised that such a rule seems necessary, as for me it’s a matter of course not to change a candidate while it’s being discussed. Some do not think that way, though, and I do not want my support on any candidate to be transferred to a considerably different image which replaced it. --Kreuzschnabel 06:31, 18 July 2016 (UTC)
    • As for me, I would oppose !--Jebulon (talk) 07:53, 18 July 2016 (UTC)
  • Mild oppose We should have reactive candidacies where appropriate. If I point out a small problem in a restoration, for example, I'd consider it being fixed better than the vote carrying on unstoppable, possibly to the point of featuring the image with the problem. There is a limit to how much editing should be done, and you should notify all pre-extant voters after any really significant change, but it's hard to legislate that cutoff point. Adam Cuerden (talk) 02:12, 19 July 2016 (UTC)
  • Pictogram-voting-question.svg Question What happend if a picture already FP is altered by the author? --The Photographer (talk) 02:36, 19 July 2016 (UTC)
    • I've had featured pictures be given major recrops or desaturations or the like by random people before. I tend to just revert and notify the modifier, but we don't really have any procedure for checking for alterations to FPs. While "Oh, I missed a dust speck" isn't a problem, we probably should be actively monitoring. Adam Cuerden (talk) 02:55, 19 July 2016 (UTC)
    • That’s why I suggested to mention the file version the feature has been given. --Kreuzschnabel 07:37, 19 July 2016 (UTC)
      • A good idea, but it will need some sort of check to spot when they don't match. In some cases it's going to be a simple, expedited update ("he removed some dust specks") in other cases, a clear revert ("someone cropped the hell out of it... ignoring the warning that CSS image crop was in use and cropping it would break its usage on Wikipedias"), and sometimes, it may be we should rerun it ("A problem was fixed, but this substantially changes the file") - that said, we're going to need to lift the 2-nom limit a bit for that last type; the whole system's pointless if no-one actually asks for a re-review. Adam Cuerden (talk) 08:39, 19 July 2016 (UTC)