Commons talk:Quality images candidates

Reminder about withdrawn nominations

As I noticed like here, sometimes users new to QIC remove their nominations after they were declined. Just a gentle reminder that the correct way to do this is to use the Withdrawn template instead. Otherwise, it becomes difficult to keep track of how many nominations a user has made on a given day -- Jakubhal 18:15, 12 September 2025 (UTC)

• @Igor123121: -- Jakubhal 18:17, 12 September 2025 (UTC)

Closing CR less than 48h after uploading a new image version AND also less than 48h after the last comment?

Ist das jetzt hier Standard geworden? Wenn ein Kandidat einen Tag nach der letzten Stimme noch durch eine Neubearbeitung augetauscht wird, aber bevor die CR beendet ist, wurde über eine veraltete Version abgestimmt. Für welche der Versionen gilt dann das Voting? Smial (talk) 17:10, 15 September 2025 (UTC)

I assume that this is about File:2025-09-07 D500-0608 Achim-Lammerts Wörth-Christuskirche.jpg. @Sebring12Hrs: I don't really understand why you insisted on keeping this CR closed. While generally hardly anyone might care whether a CR is closed 48 h after the last vote or after the last comment, even though the latter is correct, closing CR for a re-edited image with a recent comment about this action within less than 48 h is clearly not acceptable IMO. --Robert Flogaus-Faust (talk) 08:45, 17 September 2025 (UTC)

Yes, that's the case. Thanks for your clarification. My English isn't the best, and I may not have formulated my arguments well in the discussion at CR. --Smial (talk) 11:01, 17 September 2025 (UTC)

Because to me this is the rule, and bypass the rule to turn the vote in his advantage is clearly not acceptable to me. Ok, the CR was closed early, but only 48 hours after the last vote, is allowed. On the other hand, I fully admit that I did not want any further modification to this image because I felt that it did not need it. The image is QI to me, and this doesn't need any editing to me. Don't you agree? However, I think I haven't done anything illegal, even if you don't like my behavior. And in addition, the CR was closed by another user than me, but perhaps I am completely off base and in which case other people should give their opinion. --Sebring12Hrs (talk) 13:52, 17 September 2025 (UTC)

Indeed, I have just reread the rules, and it is very poorly explained, in addition we can read that it is from the first notice, that one must count 48 hours and not the last! ("If there are no objections within a period of 2 days (exactly 48 hours) from the first review"). But in reality, in CR, it is from the last vote that we count 48 hours! Incomprehensible. --Sebring12Hrs (talk) 17:17, 17 September 2025 (UTC)

Then even the German translation is faulty:
"Die Entscheidung wird mit der Mehrheit der Meinungen getroffen, einschließlich derjenigen des ersten Kritikers und ohne Berücksichtigung des Nominators. Nach einer Mindestzeit von 48 Stunden seit dem letzten Eintrag wird die Entscheidung am Ende des Textes durch die Vorlage {{QICresult}} markiert und dann gemäß den Richtlinien ausgeführt."
"Eintrag" means any contribution, not just voting. "entry" includes e.g. comments. If only votes are to be valid, that would be nonsensical in my opinion, but then the “rules” should also refer to votes and not “entries.” Smial (talk) 17:49, 17 September 2025 (UTC)

I understand what you mean, and I find the german version cleaner than the french version, but the problem in this case is that it would be enough to comment, without voting, while waiting for a vote in favor, so that the debate can continue. Sebring12Hrs (talk) 19:18, 17 September 2025 (UTC)

@Sebring12Hrs, @Smial, the "2 days (exactly 48 hours) from the first review" statement is about regular promotions. If there are no objections against a regular promotion or decline. But the CR part is identical to the German translation: "a minimum period of 48 hours since the last entry", in German "48 Stunden seit dem letzten Eintrag". Plozessor (talk) 04:59, 18 September 2025 (UTC)

Perhaps we should include in the CR rules that the 48-hour period generally restarts if new image versions are uploaded while voting is still ongoing. I have also noticed that new, revised versions are occasionally worse than older ones. I think this would also be fair to those voters who voted for an older version and are surprised to suddenly see a significantly altered image that they have not seen before. Of course, this increases the workload for those who do the evaluation, but the only alternative I can think of is to not allow any new versions at all, and that cannot be what CR intended. Smial (talk) 08:34, 18 September 2025 (UTC)

I do not generally check whether there is a new version unless the uploader of this version writes a comment. I would prefer clarification of "entry" as the latest signature with a time stamp, not just a vote, if that is really necessary. I am rather confident that there are very few supporters of Sebring12Hrs's opinion. By the way, the theoretical possibility that a nominator adds comments in an endless manner is just theoretical. If someone actually did that, they would hardly attract more voters supporting their images and they would certainly not make any friends with this kind of disruptive behavior. They might run into trouble instead. But this is also purely theoretical. --Robert Flogaus-Faust (talk) 15:03, 18 September 2025 (UTC)

The last comment on the photo in question was from the proposer, and by nature, the proposer wants their photo promoted. In this specific case, the votes were in favor of promotion. So dragging out the discussion by telling me "someone commented again" makes no sense, since it wasn't a new opinion. Coming here to lecture me is one thing, bringing evidence to prove my mistake is more complicated. Sebring12Hrs (talk) 17:04, 18 September 2025 (UTC)

And this comment by the uploader was not some irrelevant stuff that did not provide anything new, It included a notification that a new version of the image was uploaded, which had been done ca. 10 minutes before the author's entry was posted. Cancelling Smial's vote and supporting the premature closure is the reason that this entirely unreviewed version of the photo is a QI. It is really hard to believe that anyone could think that this might be appropriate, even after misunderstanding the word "entry". Anyway, we are back to square one with my statement and I am not at all in the mood to continue this debate on COM:AN/U. So this is my last entry in this thread, because it just isn't going anywhere. --Robert Flogaus-Faust (talk) 21:41, 18 September 2025 (UTC)

Why talk about COM:AN/U ???? Seriously ? Because you disagree with me, so you call the administrators for your help. As for your mood... I'll stop too. Sebring12Hrs (talk) 22:55, 18 September 2025 (UTC)

This was not a new opinion, but rather news about a modified image version that was uploaded after the last vote. This circumstance should in any case trigger a new 48-hour period to give all voters of an outdated version a reasonable amount of time to review the latest image version. My intention here is not to lecture anyone or assign blame, but to find a solution to a rule that is obviously unclear or ambiguously worded. Smial (talk) 23:39, 18 September 2025 (UTC)

I am not sure what is everyone's problem here, IMO the rules are clear - CRs can be closed only after "a minimum period of 48 hours since the last entry", and the upload of a new version together with a comment "I uploaded a new version" is surely an "entry". Thus the CR in question should not have been closed. Like Robert, I consider the potential for abuse of this rule very theoretical, and haven't seen any case of such abuse in the recent years. Also, as said above, the initial reason for this discussion was a apparently a misunderstanding, confusing the instructions for regular reviews with those for CRs. Plozessor (talk) 03:34, 19 September 2025 (UTC)

Mandelbrot Set Images by Aokoroko

A wonderful example of illustrating how to create a complex geometric fractal.
The program code is included in the file description.
I checked the program's source code and ran it on my PC. You can change the initial constants to get a different pattern, and so on ad infinitum. You can create a pattern of any size, since the program prompts you for that size at the beginning. However, it's not very fast, and you need to be patient.
Overall, this is a good QIs and meets all the QI criteria.
This is not a photo of a flower, but the result of mathematical calculations, so crop and composition are low applicable here.
The infinite number of parameter options allows for an infinite number of patterns, but I'm not worried about being flooded with such content. Again, it's not fast; creation can take several hours. In the same amount of time, you could take a lot of photos while walking the streets or in a studio, then process them and upload them to the Commons.

I prompt @Lmbuga: and @Aokoroko: here.
-- George Chernilevsky ^talk 16:23, 18 September 2025 (UTC)

I don't have an issue with these Mandelbrot images being promoted, but IMO QI is about photographs, not digital art. And some of the nominations are not even images but animated GIFs (which do not meet the QI size requirement btw). Noise, exposure, color, focus, DoF, sharpness, lighting are significant criteria for QI, but they all do not apply to this kind of computer-generated graphics. This does in no way mean that the pictures would be bad or useless - I think they are valuable and should be in Commons, just they might not be a good fit for the QI system. Plozessor (talk) 03:42, 19 September 2025 (UTC)

Non-photographic media are permitted, and to my knowledge, the size requirements do not apply to them. However, I am still not happy with mass-produced computer graphics. With these fractal generators in particular, you just play around with the parameters until the computer has produced a ‘pretty’ image. Decades ago, I spent days torturing my 286 PC with fractint, and before that my C64, and I don't really see the benefit for commons, Wikipedia and the rest of the world. Smial (talk) 10:45, 19 September 2025 (UTC)

I think you all are right, I agree with all of you. I have promoted two photos of fractals. I have promoted them to repair the damage that my previous comments may have caused, but, to be honest, I am not interested in them, no matter how beautiful they are, because I do not know any more about them and I do not see the human touch.

I am a humanities person, I never understood the passion for kaleidoscopes () and, besides, I do not like symmetrical perfection.

No problem. Thank you, @George Chernilevsky: , for preventing problems on QI due to my lack of knowledge.--Lmbuga (talk) 03:10, 20 September 2025 (UTC)

I forgot to say that I appreciate your company.--Lmbuga (talk) 03:29, 20 September 2025 (UTC)

A joke: This GIF I made, used on 22 Wikipedias and many pages of commons, cannot be QI. IMO! (). If someone nominated this GIF, I would decline. However, would the reasons to decline be based on rules? Perhaps common sense is enough. --Lmbuga (talk) 03:51, 20 September 2025 (UTC)

How long will it take us to have images produced using artificial intelligence? Could they be nominated? Will there be rules?--Lmbuga (talk) 05:10, 20 September 2025 (UTC)

AI images cannot be nominated for QI according to the rules: QI is an award for Wikipedians, but AI is not a Wikipedian .

The speed of AI image creation is 1-3 minutes, depending on the complexity of the input description. I use DALL-E 3 for such experiments and get four variants for each query. The result can vary significantly when running the same description again and is never repeated. Most of these images contain too many stupid errors. But you can see some of the acceptable results on my Instagram page (same username as on Commons).

Unlike AI, algorithmically constructed fractals can be reproduced in repeated experiments and have mathematical meaning.

Regarding the Mandelbrot set, it still has some unresolved mathematical questions. The exact area of ​​the Mandelbrot set is unknown. The exact coordinate of the center of mass is also unknown. -- George Chernilevsky ^talk 18:45, 20 September 2025 (UTC)

Thank you. Now I understand Lmbuga (talk) 23:24, 21 September 2025 (UTC)

I did it! Pascal, ffmpeg and FASM. Image, Video, gif. Great job! Each file is written {Author: https://commons.wikimedia.org/wiki/User:Aokoroko So, ffmpeg takes a command line of 255 files and produces one "Mandelbrot Set Color Cycling Animation 600px.gif". ffmpeg reduced the images by 8 times (from 4800 x 4800 to 600 x 600), but at the same time increased the color depth from 8 bits (255 colors) to 24 bits (TrueColor, 16.7 million colors).

And now FASM. FASM.EXE - one file! No more files needed. https://flatassembler.net/download.php For Windows. This action is extremely simple - drag the SplMandelbrot.asm file icon onto the FASM.EXE icon. This will trigger the "Open with" action. This will immediately create the SplMandelbrot.exe application in the folder. PROGRAM DESCRIPTION The SplMandelbrot application draws a full-screen section of the Mandelbrot set and displays an animation. The animation direction can be changed using the arrow keys on your computer keyboard. Two different coloring palettes can be selected using the F6 and F7 keys. Five different pre-selected points of interest on the Mandelbrot set can be selected using the F1, F2, F3, F4, and F5 keys. Left-clicking zooms in (magnifies) the image by 2x, right-clicking zooms out (zooms out) by 2x. You can save information about a favorite image to a file for later use by pressing the End key and then load it from the file by pressing the Ins key. To exit, press the Esc key. The animation is 60 frames per second. This is DwmFlush. And BitBlt for 1920 x 1080. Aokoroko (talk) 22:50, 20 September 2025 (UTC)

I didn't see this discussion before, but wrote today that I see 100 QI in .webm and believe that it should not be like that. It's great, but not an image. Анастасия Львова^ru/en 21:47, 5 October 2025 (UTC)

https://commons.wikimedia.org/wiki/Commons:Quality_images_candidates/Archives_October_11_2025

https://commons.wikimedia.org/wiki/File:Mandelbrot_Set_Color_Cycling_Video_1080p_7.webm and 8, 9, 10 - Review needed (((((((((((((

Unassessed images (nomination outlined in blue)Nominated images which have not generated assessments either to promote nor to decline, or a consensus (equal opposition as support in consensual review) after 8 days on this page should be removed from this page without promotion, archived in Commons:Quality images candidates/Archives October 11 2025 and Category:Unassessed QI candidates added to the image.

Aokoroko (talk) 17:12, 10 October 2025 (UTC)

Commons:Quality images candidates#Consensual review rules

See Commons:Quality images candidates#Rules

The link doesn't work. What should be there?

Probably Commons:Quality images candidates#Timetable (day 8 after nomination) is somehow related, but after so many time I am still not sure how. Анастасия Львова^ru/en 09:55, 10 October 2025 (UTC)

Both links are working, at least for me. Commons:Quality images candidates#Rules leads to the "Rules" section between the "Consensual review" heading and the topmost image that is in consensual review on the page Commons:Quality images candidates. The time table is on the very end of the page. --Robert Flogaus-Faust (talk) 11:54, 10 October 2025 (UTC)

No need to check both (of 3) links, but I understood the problem - it is broken on all pages but English ones. @Красный: , глянь как разбирающийся в переводах, что там сделали не так? Анастасия Львова^ru/en 12:19, 10 October 2025 (UTC)

There's no such section as #Rules in any version besides the English and Simple English ones. These types of links don't work with translations, I think we need to make some separated Rules page instead, or, at least, try some anchors. Красный ^{wanna talk?} 12:49, 10 October 2025 (UTC)

Можно ещё попробовать заставить переводчиков проверять как на их языке этот раздел называется и делать прямые ссылки через язык и отсылку к разделу, но это, как мне кажется, чересчур. Красный ^{wanna talk?} 13:03, 10 October 2025 (UTC)

Unfortunately, the translations seem to be an inconsistent mess. If you change your language consistently to Russian, Commons:Quality images candidates/ru#Правила seems to work. However, if your language is English and you just change it to Russian with the link within the Quality Images Candidates page, the link would be Commons:Quality images candidates/ru#Rules. Anyway, the rules seem to be always in English on the Russian page, as far as I can see. The link provided never works and I always get an awful mishmash of Russian and English text. In the German version with German as the preferred language, Commons:Quality_images_candidates/de#Regeln would work, but even though most of the rules section was translated to German, there is still a jumble of German and English and the links provided on the page do not work. --Robert Flogaus-Faust (talk) 09:01, 11 October 2025 (UTC)

So we have 2 topics here about the need to change something in the rules, but personally I still feel myself not brave enough to suggest how (I'm in the QI project just for about 2 years, not so long) :) Анастасия Львова^ru/en 09:07, 11 October 2025 (UTC)

I was bold and did. Sorry, now I see that it was a request about Withdrawn also, it is not done (I am not sure if it is important, because if someone is oppose you can just wait, the same result as with withdrawning, just more slowly). Анастасия Львова^ru/en 14:49, 27 October 2025 (UTC)

Link to nowhere

In: Consensual review rules: Commons:Quality images candidates#Rules – it links nowhere Gower (talk) 20:49, 11 October 2025 (UTC)

@Gower: Merged with an open topic on the same page... Анастасия Львова^ru/en 21:02, 11 October 2025 (UTC)

Guidelines

How about adding to guidelines:

• that nominating the same picture twice or multiple times doesn't help
• explaining difference between comments and discussion.

As a newbie it wasn't obvious for me, so maybe those informations would be helpful? Gower (talk) 20:52, 10 October 2025 (UTC)

Nominating the same picture twice actually helps, if the picture haven't got any comments first time. You either get comments or someone promotes it. I did that a lot this spring cleaning unassessed images category greatly and, as I know, some other contributors do it too. Красный ^{wanna talk?} 22:28, 10 October 2025 (UTC)

@Красный, @Lvova said to me that nominating twice makes promotion postponed and I should stop that. Gower (talk) 06:49, 11 October 2025 (UTC)

Ah, you did not wait until the first nomination had ended? That is indeed not useful. But if the image ended up as unassessed image after the time to review expired, there is no reason to not try again. Kritzolina (talk) 06:54, 11 October 2025 (UTC)

It's just a language problem now. I told you not to support images twice, and nominating means nominating :) So they are about the fact, that if an image was not promoted, you can do another try; but you really should not /promote twice. Анастасия Львова^ru/en 07:06, 11 October 2025 (UTC)

Ну ведь можно же проверить сначала, что я ему сказала, прежде чем отвечать, Красный :-/ Анастасия Львова^ru/en 07:09, 11 October 2025 (UTC)

Thanks for explanation, if so, could we add those suggestions/informations to official guidelines for newcomers? Gower (talk) 07:49, 11 October 2025 (UTC)

It's hard for me to answer the question properly because I'm angry about how much time I've had to spend, and even after all that, there's still a misunderstanding. Yes, the rules probably need to be supplemented, but also with the recommendation not to do 250 reviews in an unfamiliar project in three days without the intention of tracking whether the identified issues have been addressed. Анастасия Львова^ru/en 08:28, 11 October 2025 (UTC)

А где-то была индикация того, что он этот пост сделал по результатам обсуждения с тобой? Ну серьёзно, не ожидай от людей дара медиума, это трудновато местами. Красный ^{wanna talk?} 09:58, 11 October 2025 (UTC)

Это показывает уровень того, насколько я с этой ситуацией замоталась; ты прав. Анастасия Львова^ru/en 10:01, 11 October 2025 (UTC)

I also suggest to add information about withdrawing a nomination. I suggested this here in September. As a newcomer, it was not clear that 'withdraw' was expected instead of removal edit. I feel pretty strongly that the instructions are not complete enough for newcomers, but not sure how to influence others here to consider. E bailey (talk) 16:10, 26 October 2025 (UTC)

I'm surprised.

I'm surprised. Two or four photos have just appeared on QIC, at around 10 p.m. on 14 October 2025. I promoted one of them (of Lvova) several days ago, and it appears to be promoted. I don't understand any of this. Why do these images keep coming back?

The picture I promoted a few days ago is File:Павловский проспект (Ломоносов) 01.jpg --- Lmbuga (talk) 21:55, 14 October 2025 (UTC)

Probably it's an answer? :) Анастасия Львова^ru/en 22:02, 14 October 2025 (UTC)

Ok, but the photo appeared promoted by me at a time when it had not been promoted by me. Of course, I had promoted it long before. After promoting that photo, I must have promoted at least twenty more, and is at the forefront--Lmbuga (talk) 22:13, 14 October 2025 (UTC)

No problem. I'm not mistaken. It's just a warning. Something may be happening, but I'm not mistaken.--Lmbuga (talk) 22:25, 14 October 2025 (UTC)

MediaWiki_talk:Gadget-QICvote.js#Sometimes_messages_are_posted_in_the_wrong_places. Анастасия Львова^ru/en 22:27, 14 October 2025 (UTC)

Livio

See COM:ANU#User:StPaul.jpg. --A.Savin 11:26, 24 October 2025 (UTC)

I don't understand what's wrong with this user: Edit war? I'm not aware of this whole story of course. Sebring12Hrs (talk) 15:20, 25 October 2025 (UTC)

+1. Photos are very good. Юрий Д.К. 09:04, 27 October 2025 (UTC)

User StPaul is supposed to be a sock of an indef blocked and globally locked user. --Robert Flogaus-Faust (talk) 10:35, 27 October 2025 (UTC)

Revert massive edit

@Cheima fezzani: I'm not sure what you were trying to accomplish here, but you deleted a lot of reviews added by other people. Please be more careful with your edits, and maybe you could explain what this change was about. ---- Jakubhal 16:26, 27 October 2025 (UTC)

• I have added some of your nominations. Is that all? It looks like you edited old version of the page. Please don't do it again. -- Jakubhal 16:30, 27 October 2025 (UTC)

Can a blanked out image still be QI?

I've had to blank out the Bansky artwork in File:At New York City 2024 054.jpg - which is a QI - due to the absence of FOP in the US. Can it still remain a QI despite the blanking out? Thanks. Mike Peel (talk) 18:09, 27 October 2025 (UTC)

• Hard to speak for everyone, but I definitely wouldn’t promote it in this version. It’s a very heavy edit, and the result is a completely different image from the one that was originally a QI. -- Jakubhal 15:59, 3 November 2025 (UTC)

File:20240614_great_blue_heron_wethersfield_cove_PD200579.jpg

The Discussion template for the 20240614_great_blue_heron_wethersfield_cove_PD200579.jpg image shows error. I tried to repair, but failed. LexKurochkin (talk) 14:00, 3 November 2025 (UTC)

Jakubhal fixed it just before I could do it. The link to the raw version broke the template. --Robert Flogaus-Faust (talk) 14:32, 3 November 2025 (UTC)

It is the = character, It was included in the link (but was not necessary). I've seen it before. We cannot use it without <nowiki> inside QI page templates (perhaps all mediawiki templates). -- Jakubhal 15:53, 3 November 2025 (UTC)

Thank you! LexKurochkin (talk) 16:06, 3 November 2025 (UTC)