Commons:Featured picture candidates/File:State Emergency Service of Ukraine (MChS) Mil Mi-8MTV picking up water near Nezhin.jpg/2

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

File:State Emergency Service of Ukraine (MChS) Mil Mi-8MTV picking up water near Nezhin.jpg, not featured[edit]

Voting period is over. Please don't add any new votes.Voting period ends on 24 Mar 2015 at 23:26:16 (UTC)
Visit the nomination page to add or modify image notes.

State Emergency Service of Ukraine (MChS) Mil Mi-8MTV picking up water near Nezhin
Julian, have a look at the version at File:Ukrainian MChS Mil Mi-8 Belyakov.jpg for something closer to reality. Sadly, that realistic photo is even lower resolution, but this one has had the full 500px treatment applied. The chroma noise I'm referring to is also in the dark still water rather than just the choppy water that might conceivably generate a rainbow effect. When you compare the two, it's clear this version has had the saturation cranked-up, thus emphasising the inherent chroma noise. We wouldn't accept such a poorly post-processed JPG from a Commons photographer. -- Colin (talk) 12:01, 16 March 2015 (UTC)[reply]
You might be right, I would expect the colours to come from foggy air from the blown away water (which would spread) but it's probably greatly emphasized. The problem with the "from a Commons photographer" argument is that I'm okay with that - we can make "our" photographers learn how to take and edit photos that are perfect for Commons - I think I learned this to some extent myself (not perfect but good), from FP discussions. I'm thankful for that, it made me a better photographer/editor. But we can't do that with other photographers (at least not as easily) and there are subjects that just aren't captured by Commons photographers. I'm torn here, I see your argument but I also think the photo has something special that deserves recognition.  Neutral for now. — Julian H.✈ 12:26, 16 March 2015 (UTC)[reply]
I think sometimes we extend our "point out minor fixable flaws trying-to-be-helpful advice" to become reasons to oppose, which is wrong. We tend to get away with it for Commons photographers who then fix the utterly insignificant CA in their 36MP image :-). It fails, spectacularly, when we get a high-quality high-resolution portrait from a third-party and then pixel-peep it to death. So I agree that we have to learn to just live with what we've got sometimes, and weigh up the merits and flaws, even if we know the photographer/post-processor could have done better. If this was 20MP with that degree of chroma noise at 100% I'd let it pass since it would be invisible at 5MP never mind 2MP. But at just-HD resolution, it isn't close to meeting the definition of "our finest" on the technical front. I agree that "wow" from the perspective/subject is impressive, but the small size seriously diminishes its utility -- you need about 5MP to print in a A4 magazine to high quality. -- Colin (talk) 13:08, 16 March 2015 (UTC)[reply]
  •  Support I hear ridiculous comments that do not take into consideration where this photo was taken. surely someone will recommend using a tripod. This section is contaminating some kind of obsession for some users. --The_Photographer (talk) 13:14, 16 March 2015 (UTC)[reply]
  •  Oppose Just not thrilled by it and main subject ridiculously small. Taking a picture in the air does not necessarily make it great, as proven here. - Benh (talk) 14:43, 16 March 2015 (UTC)[reply]
If you think that we need change our ridiculous size requirements for images. You can propose this in this section talk page. This image meets the minimum requirements. It's funny, you call the ridiculous requirements with a ridiculous comment. --The_Photographer (talk) 17:06, 16 March 2015 (UTC)[reply]
Can we stop requiring legislation for/against absolutely everything for just a moment, and allow people to use their brains -- we are not computer programs. Our minimum size guidance is a lower bound beyond which images "are typically rejected unless there are 'strong mitigating reasons'". We are free to oppose for images above this, or support for below. No justifications have been given for why the image is only 2.67MP. -- Colin (talk) 18:32, 16 March 2015 (UTC)[reply]
I think the concern is not the size of the image itself, but the subject (i.e. helicopter) within the image. In response to that particular concern though, I would say the subject is not just the helicopter, but also the mrotor-wash underneath. -- KTC (talk) 17:48, 16 March 2015 (UTC)[reply]
I agree with you: subject is more than just the helicopter. -- Colin (talk) 18:32, 16 March 2015 (UTC)[reply]
  •  Oppose I'm mostly concerned about the chromatic noise or aberration which is quite visible on the water surface even in this size. The picture seem to have somewhat magenta cast in it. Perhaps the problems could be partly fixed though. --Ximonic (talk) 16:08, 16 March 2015 (UTC)[reply]
 Support The fixed version is much better to say the least. Thanks! Could be a little bigger picture but for me it's impressive enough. --Ximonic (talk) 13:54, 17 March 2015 (UTC)[reply]
See comment below. --Ximonic (talk) 16:30, 19 March 2015 (UTC)[reply]
  •  Oppose as per Colin. Yann (talk) 20:23, 16 March 2015 (UTC)[reply]
  •  Oppose per other opposes. --Alchemist-hp (talk) 22:47, 16 March 2015 (UTC)[reply]
  • ✓ Done Fixed color balance, chromatic aberration, color noise, oversharpening, perspective, color aberration in light, distortion --The_Photographer (talk) 23:20, 16 March 2015 (UTC)[reply]
  •  Support Ю. Данилевский (talk) 13:21, 17 March 2015 (UTC)[reply]
  •  Support --Palauenc05 (talk) 15:51, 17 March 2015 (UTC)[reply]
  •  Support new version -- KTC (talk) 16:34, 17 March 2015 (UTC)[reply]
  •  Comment This is actually worse. Rather than go back to the photographer and ask for a more realistic (less processed for 500px style) image, the new version is about as authentic as a colourised b&w photo. The EXIF data explains what has been done. Firstly, colour noise reduction at level 90/100. This has removed all colour detail. The patch of land in the corner is now just various tones of green and the sea is just tones of blue. But more importantly (since this is an image of a helicopter picking up water in a bucket), where have the red ropes that held the bucket gone? They have just been merged into the sea like so much digital colour noise. Next, the blue hue adjustment is -13 (modest) but the purple hue adjustment is -100 (max) which has transformed all purple in the image into blue. Next the blue and purple luminance have been increased to +36/+38. This does return the blue of the helicopter to closer to the original photograph. However, the sea is still nothing like the original which was a typical brown/blue rather than deep blue. Lastly the JPG has been saved in "ProPhoto RGB" rather than "sRGB" (or even "AdobeRGB) which are the only two standard colourspaces for JPG. ProPhoto RGB is a colourspace used internally by Lightroom/Photoshop and is completely unsuitable for 8-bit JPG. Some computer browsers, most non-professional computer image viewers and all mobile browsers will fail to display this in anything approaching the "correct" colours (likely rather desaturated). But after this must post-processing, "correct" is anyone's guess. There's no change on the over-sharpening, perspective or distortion, though I didn't think the latter two were even an issue. If I were the photographer, I'd be upset that someone had removed all the colour detail, all the purple, and had ensured mobile and computer users see completely different colours. -- Colin (talk) 21:41, 17 March 2015 (UTC)[reply]
Instead of criticizing the work of others(500px comment), do not be afraid to fix it by yourself. The ProPhoto RGB color space encompasses over 90% of possible surface colors in the CIE L*a*b* color space, and 100% of likely occurring real world surface colors documented by Pointer in 1980 --The_Photographer (talk) 11:48, 18 March 2015 (UTC)[reply]
  • I hope you do get what you write. If your monitor encompasses what ProphotoRGB has to offer, it's fine for you, but mine merely is a sRGB, and I bet most people are in the same situation. Also what Colin tries to explain you is that the colorspace covers so much that your 8bit-JPEG is not big enough of a container for it. It gives you a mere 256 levels per channel, which is a reason why it might not be suitable to represent, says, a gradient from the lowest to the highest value of "color", without visible transition. Hopefully I wasn't too wrong. - Benh (talk) 21:10, 18 March 2015 (UTC)[reply]
  • There are no monitors that can display ProPhotoRGB, it exists merely to enable lossless colour transformations during pre-processing. Rather than argue about it, you could just Google. There isn't anyone who knows anything who would recommend a ProPhoto JPG. If your job was to sort screw nails by size, you'd use a 15cm ruler with mm scale rather than a metre stick with cm scale. Bigger isn't necessarily better; it's just bigger. -- Colin (talk) 22:24, 18 March 2015 (UTC)[reply]
  • I think you made an honest good-faith attempt to improve it but this just shows the problems with fiddling with other people's JPGs, and that none of us are professional post-process artists (which some photographers outsource their work to). A modest amount of colour noise reduction would help the original candidate photo, but I wouldn't go any further than that personally. At FPC we need to review a third-party image for what it is, if we are unable to request an improvement from the photographer. Even if the candidate is too saturated/contrasty/noisy to please our FPC tastes, that's what the photographer thought was an improvement on his out-of-camera version (assuming that's the history). His choice/taste and we should respect the colours he chose for his image -- even if that means we dislike it at FPC. -- Colin (talk) 23:14, 17 March 2015 (UTC)[reply]
You do not want to ask NASA or a painter to fix their colors. --The_Photographer (talk) 11:48, 18 March 2015 (UTC)[reply]
  • I don't think editing or fixing another's work or FP candidate is the problem. Well all make wrong choices from time to time. I just don't think it was a good idea to overwrite the image, instead of uploading it as an alternative. - Benh (talk) 06:00, 18 March 2015 (UTC)[reply]
Well its not a ALT, its the same image, same angle, same size, same author... --The_Photographer (talk) 11:48, 18 March 2015 (UTC)[reply]
  • Agree. The modest change I suggest would be fine for an overwrite but changing the colours is really too much of a change -- and needs to be documented (it's no longer solely the original artist's work). -- Colin (talk) 08:23, 18 March 2015 (UTC)[reply]
Why not do that by yourself? --The_Photographer (talk) 11:48, 18 March 2015 (UTC)[reply]
  • You don't prove him wrong though. And his criticism is well structured and thought. It's not like hesays "it's shit" period. I was amazed with the rope issue, and this alone should trigger a revert on ur changes. - Benh (talk) 22:02, 17 March 2015 (UTC)[reply]
I guess it's OK when that someone is "the photographer" himself ;) - Benh (talk) 21:57, 17 March 2015 (UTC)[reply]
I am not the photographer --The_Photographer (talk) 11:48, 18 March 2015 (UTC)[reply]
  •  Comment @Ahonc and The Photographer: There is a problem with this nomination in that over a dozen support/oppose votes were cast on the earlier photographer's image that Ahonc nominated, and about four votes on the significantly altered version that "The Photographer" uploaded and which has an important flaw that may justify reverting (the ropes carrying the bucket are no longer visible). It would have been better if this upload were done to an alternative file and an Alt added to the nomination (providing the nominator agrees to such disruption to his candidate). One option is to ping all those who voted prior to the image change and ask them to reconfirm or change their vote on the new image -- but this assumes the flawed current version is not reverted. Alternatively, revert the modification and ping those who voted recently.
At present, it is impossible to tell which votes apply. My view is that latest version does not represent the image the photographer donated to Commons, has a serious flaw, and should be reverted. Respecting the photographer's artistic choices is more important IMO than fiddling with the image to please the FPC crowd and gain a gold star. I would support making a new version that had very modest colour noise reduction applied, while ensuring that the ropes holding the bucket remain, and no global changes to colour. But looking at the opinions expressed, I am doubtful that would be sufficient improvement to sway enough to support. -- Colin (talk) 11:57, 18 March 2015 (UTC)[reply]
I know, you can do it :) --The_Photographer (talk) 12:01, 18 March 2015 (UTC)[reply]
  • @Halavar, Palauenc05, Yuriy75, and Ximonic: The new version you have voted on has two significant problems: the ropes holding the bucket have vanished with over-aggressive colour noise reduction, and the colours (particularly purple) have been radically altered from what the photographer uploaded. The photographer has not sanctioned these changes, which represent deterioration and alteration of his artistic choices as image-creator (whether you like those choices or not: they are his). It is not our job to significantly change his image merely to suit tastes at FP. I propose the new version is reverted. If you like the new version, then it should (per the licence) be uploaded as a new file that clearly states what the changes are and who made them (as required by the licence). -- Colin (talk) 07:57, 19 March 2015 (UTC)[reply]
I pull my support for now since I noticed the new issues. But in my opinion there is no reason to make this all such big of a problem as it currently looks. I suggest that 1) either this will be reversed, a new alternative will be made with softer methods (perhaps like how was done with the Vietnam picture). 2) or in respect to the creator we just reverse this anyway because of the trouble this has caused or might cause and we continue without the alternative. In that case there won't be very good chances for a FP, I think. --Ximonic (talk) 16:30, 19 March 2015 (UTC)[reply]
Confirmed results:
Result: 11 support, 9 oppose, 1 neutral → not featured. /Yann (talk) 09:58, 25 March 2015 (UTC)[reply]
Retrieved from "https://commons.wikimedia.org/w/index.php?title=Commons:Featured_picture_candidates/File:State_Emergency_Service_of_Ukraine_(MChS)_Mil_Mi-8MTV_picking_up_water_near_Nezhin.jpg/2&oldid=154687686"