Eatcha

Hi, There are FP uploaded by Eatcha which seem fake and not Eatcha's work. Please see COM:ANU#Eatcha. Yann (talk) 09:37, 13 July 2025 (UTC)

I would strongly reccommend that people read what UnpetitproleX has uncovered at Commons:Deletion requests/File:Zaniskari Horse in Ladakh.jpg. It makes for chilling reading. It's also a sharp reminder why we should check licenses better at FPC and double-check the origin of photos at the slightest hint that something is wrong. Eatcha went all in when creating his fake photos, changing titles, locations, authors, even Exif and cameras down to camera ID. As a programmer he knew exactly what to do to dupe us all on plenty of QIs, VIs and FPs. --Cart (talk) 01:24, 14 July 2025 (UTC)

oh god... and I do have the feeling other impostors are lurking around... - Benh (talk) 09:48, 14 July 2025 (UTC)

Don't be surprised: the bot may ask you for help

Most of us have experienced it: FPCBot, the bot which counts votes, closes nominations, and sorts new Featured pictures into the galleries, seems to take vacations every now and then. Well, in reality, this usually means that the bot has stumbled while processing a nomination and gets stuck at the same point every time it runs, so it appears to be frozen. People have to search for the cause of the problem, to fix it, and often to process nominations manually so that the bot runs smoothly again. That’s annoying and a waste of time.

But very often the reason for all the hassle is just a little typo in a gallery link, a tiny formatting error in a nomination, a renamed file or nomination subpage, etc. – i.e., small issues which a human editor can fix easily, but which still stop the bot doing its duties or halt it completely. Of course the log files contain all information which is required to find and fix these problems, but the logs are buried somewhere on the labs server; and even if we could make them publicly available it would often be difficult to find exactly that piece of information which is needed to tackle a certain problem.

Therefore we will try something new. Hopefully starting with this evening or tomorrow, FPCBot will post a short message on this talk page whenever it needs our help. In most cases the messages should be self-explanatory and and hint clearly at the problem which needs to be fixed; in other cases they will (at least) give an idea where to search. (And if a message is ever unclear or seems completely wrong, please drop me a note, then I can improve it ;–). The feature is new and will be improved over time.)

I hope that this new approach will help to resolve problems with nominations more quickly in order to avoid FPCBot coming to a standstill. Some messages will also point out problems with sorting new Featured pictures, etc.; this should help to find and sort ‘unsorted’ pictures more easily. Finally, I also hope that this will encourage more people to participate in maintaining the gallery pages, categories, etc. Best, – Aristeas (talk) 16:58, 25 July 2025 (UTC)

• Thank you for your work in trying to improve the way the bot functions for us, Aristeas. It is much appreciated that you volunteered your time and knowledge in this way. Cmao20 (talk) 22:01, 26 July 2025 (UTC)

• Thank you very much for your reply, Cmao20 – much appreciated! We will see how it is perceived when the bot posts a request for help for the first time. Hopefully people will not misunderstand this as criticism of their nominations, but rather as what it is: a request to the FPC community to fix a problem so that the bot can do its job without interruption and the gallery pages, etc., can also function properly. Best, – Aristeas (talk) 15:16, 27 July 2025 (UTC)

  I don't have a lot of time and energy for Commons at the moment, but wanted to drop the quick note that I really appreciate your work, Aristeas! If I ever see the bot asking for help, I will do my very best to assist it! Kritzolina (talk) 16:17, 28 July 2025 (UTC)

Delist and replace for Apollo images

Hi, Thanks to the website mentioned by JayCubby, all images of space exploration by NASA have been scanned in very high resolution. So we now have alternative images of FPs of Apollo. I wonder if we should delist and replace them. What do you think? I already uploaded some of them. I welcome suggestions about the color balance and light. Yann (talk) 17:59, 25 July 2025 (UTC)

I think that seems sensible. The exposure and color will sometimes need correction, but it need not be a perfect match (they were subjectively edited to begin with)

I will see if I can upload any more of them later. They are too large for me to easily edit, sadly. JayCubby (talk) 18:07, 25 July 2025 (UTC)

• File:A Man on the Moon, AS11-40-5903.jpg or File:A Man on the Moon, AS11-40-5903 (cropped).jpg instead of File:Buzz Aldrin by Neil Armstrong.jpg (or File:Aldrin Apollo 11 original.jpg, FP on English WP).
• File:Buzz Aldrin's bootprint on the Moon, AS11-40-5877.jpg instead of File:Apollo 11 bootprint.jpg
• File:Buzz Aldrin and Apollo 11 Lunar Lander, AS11-40-5927.jpg instead of File:Apollo 11 Lunar Lander - 5927 NASA.jpg

At a request from Yann I've had a look at these photos, and here are my comments.

• As usual, we don't get that much more information from such super-high scans of film photos just because we can now scan them down to the level of film grain. Most of the images don't get much sharper, only some details are now clearer. We can also see minute damages to the film (scratches, drops, etc.) better, but they are of no consequence.
• I've checked the white balance, and it's actually close to perfect in these new scans, which is refreshing. I don't think we should mess with that in any way. Usually you check WB against white objects, but we have no way of knowing how white are of the objects really are. It's not like they brought a ColorChecker to the moon. Grey is out too, since we've seen from Moon samples that the ground has a sickly yellow-green-tint to the grey. This also reflects the sunlight, and casts a very slight greenish tint on everything on the Moon. The only ref we can use is black. Assuming the laws of physics are the same on the moon as on Earth, the deepest shadows and the void of space should be pure black, and they are (give or take a film grain or two). This correct color is the biggest win with these new scans.
• I'd leave the exposure as it is in these. Photo conditions on the Moon are extreme due to the sharp contrast between highlights and shadows. I've tested to tone down highlights and whites, but the results are never satisfying. Let's keep the white glowing and the shadows sharp and stark.
• However, like with most hi-res film scans, there is an awful lot of color noise introduced when you make these scans. This is down to a lot of factors, including the photoelasticity of the polymer film. (This problem is not as big when you scan old photos made on rigid surfaces like glass plates.) Looking at the Moon surface at 100%, it looks more like confetti at a pride parade than lunar dust.

I have made versions of all the four photos available in this folder, where I've reduced the color noise by 67%. Please note that this is not in any way reduce the normal noise in the photos, it just unifies the colors better. But turning the pink, magenta, blue and green pixels into the the same color as their nearest overall color (we are talking pixel level here) does reduce to overall size of the files significantly since it doesn't have as many deviant color pixels in it to keep track of, even though no compression is made. I'm just mentioning this since folks may wonder why the new files are so much smaller; that just how digital files work.

You may use the versions I've made as you please. The color noise reduction is the only edit made on them. --Cart (talk) 13:51, 28 July 2025 (UTC)

@Cart: Thanks for your message. EXIF data says: "Horizontal resolution 2,019.64 dpc, Vertical resolution 2,019.72 dpc". I have tried to match the color balance and light of images provided by NASA. I think this is the best model we have. Yann (talk) 14:01, 28 July 2025 (UTC)

Like I wrote, the color balance and light in your edits are fine, it was just the underlying color noise I was addressing. --Cart (talk) 14:08, 28 July 2025 (UTC)

FPCBot asking for help

The gallery link Animals/Fish#Order : Aulopiformes (Grinners) in the nomination Commons:Featured picture candidates/File:Chile lagarto (Synodus lacertinus), La Paz, Baja California, México, 2024-12-20, DD 64.jpg does not point to a valid section on Commons:Featured pictures/Animals/Fish. (The section after the # in a gallery link is valid if and only if it corresponds letter for letter to a subheading which is immediately followed by a <gallery> element.) Therefore one or more new featured pictures are added to the Unsorted section of Commons:Featured pictures/Animals/Fish. Please sort these images into the correct section. Thank you! / FPCBot (talk) 21:01, 28 July 2025 (UTC)

• I believe the issue here was that the gallery for the correct family - lizardfishes - did not exist. I should have fixed the problem, if someone would like to check my work. Cmao20 (talk) 21:33, 28 July 2025 (UTC)

• Hello Cmao20, yes, that was was the reason, and you have completely resolved the issue. Many thanks! Best, – Aristeas (talk) 21:37, 28 July 2025 (UTC)

I confess that the bot’s message is a bit technical (well, quite appropriate for a bot, isn’t it? ;–) and not easy to understand. The reason is that at that point in the code we do not know what exactly is wrong with the gallery link; to provide a more meaningful message I would have to add more tests (we will see if that is feasible). But since you, Cmao20, immediately understood what it was about, the message seems to work after all. Let’s go a little bit more into detail in order to understand what is going on.

• From the perspective of the nominator: When Poco a poco nominated the lizardfish photo, he knew that the gallery page for fish photos sorts the FPs by families, but that there was not yet a subsection for the Synodontidae (lizardfishes) family. Therefore he used

  (i) the next level, the order of Aulopiformes (grinners), for the gallery link: Commons:Featured pictures/Animals/Fish#Order : Aulopiformes (Grinners). That’s completely fine, of course (I absolutely do not want to criticize you, Poco a Poco! I am just describing how things work).

  Other common ways to deal with this situation are:

  (ii) to use just the name of the gallery page for the gallery link and to add a hint like “(no section for Synodontidae (lizardfishes) yet)” after it;

  (iii) to use the final link with the correct section although it does not exist (yet) and the link does not work, in this case: Commons:Featured pictures/Animals/Fish#Family : Synodontidae (Lizardfishes). All three solutions have advantages and disadvantages.

  (iv) One could also create the missing section right away and use it for the gallery link; this is a bit risky as long as we do not know whether the candidate will succeed or not.

• What the bot used to do: If the nominator uses solution (i), i.e. a higher-level heading for the gallery link, the bot used to sort the new FP into the first <gallery>...</gallery> element it found after that heading. At the first glance everything seemed to be fine; at the second glance this was a mess, because very often the new FP did not belong to that <gallery> element. You can easily see this in our example: the first <gallery> element belongs to the Chlorophthalmidae (greeneyes) family, so putting the new FP into that section is plain wrong. The bot has used this faulty practice for years, producing many hard-to-find sorting errors.
• What the bot does now: To get rid of these sorting errors, I have implemented a clear guideline for the bot: It only considers gallery sections in which the <gallery> element follows immediately after the subheading. This rules out any ‘abstract’, superordinate headings, like in our example the heading ‘Order : Aulopiformes (Grinners)’; if a gallery page is sorted by families, like the fish gallery page, the bot now considers only the subheadings for the families. Therefore, if the nominator uses solution (i) from above, the bot no longer puts the new FP into some <gallery> element after the superordinate heading, but into the ‘Unsorted’ section. I used to search for images in ‘Unsorted’ sections with a separate program and to sort them one by one into the correct section. That the bot now asks for help allows everybody to find these unsorted FPs and to assist it by putting each FP into the appropriate place.
• Alternatives: If the nominator uses solution (ii) or (iii) from above, the result is the same – the bot must put the new FP into the ‘Unsorted’ section. This is a progress, of course. Solution (iii), i.e. linking to the not-yet-existing section, was discouraged in the past because the bot sometimes put FPs into almost arbitrary sections if the gallery link pointed to a missing section; that was another reason for hard-to-find sorting errors. I have fixed this, too, so the bot will just put the new FP into the ‘Unsorted’ section where we can pick it up.

TL;DR; – Best practice: When nominating an image for which the correct gallery section does not exist yet, nominators are free to select the solution of their choice from (i) to (iii). Personally I would suggest to use (ii), i.e. to use only the name of the gallery page in the gallery link and to add after the link a hint like “(section for Synodontidae (lizardfishes) does not exist yet)” because this is the most transparent solution. In any case, if it is obvious after a few days that the nomination is very likely to be successful, it makes sense to create the missing section right away and to change the gallery link to that section. This allows the bot to sort the new FP right into the correct section; and that means the least work for all of us. Best, – Aristeas (talk) 09:44, 29 July 2025 (UTC)

Thank you very much, Aristeas, for your explanation and code improvements. I've been updating some of the links (with a newly created family secction) just before nom closure in the past when it was about to get promoted but I forgot to do it in this case. One of my current noms has the same issue, this would be how (ii) looks like, right? Poco a poco (talk) 18:06, 29 July 2025 (UTC)

Hello Poco a poco, you are welcome. Yes, this is a good idea – it provides the final gallery link and is transparent by making clear that the section does not exist yet. Thank you very much, – Aristeas (talk) 18:44, 29 July 2025 (UTC)

New feature: notify the original creator

Everybody knows: When an image gets promoted to Featured picture status, FPCBot notifies the nominator and the uploader of the image by posting a template on their user talk pages. (If nominator and uploader are one and the same user, only one template is added.)

But sometimes a third user deserves a notification: the original creator. For example, consider a nomination like this one. The photo was originally created by Mithun Kunwar, retouched and uploaded by Radomianin, and nominated by Kritzolina. So FPCBot will notify both Kritzolina and Radomianin – but we should also notify Mithun Kunwar as the original creator, of course.

Until now nominators have (hopefully) handled this by copying the notification template to the talk page of the original creator. FPCBot was not able to do this because it is not easy to identify the original creator of an image: the Mediawiki software tells us the name of the user who has created the nomination subpage and the name of the user who has uploaded the image, but it is much more difficult to find out who has originally created an image. And we will never find a perfect solution for this, there are just too many different ways in which the original creator is declared in image descriptions.

But we can define a simple way to declare the original creator. When you nominate an image which was created by an user who is different from uploader and nominator, just make sure that the nomination contains this phrase: {{Info}} ... created by [[User:...]]. Actually this phrase is already present if you create the new nomination as usual by clicking the blue button “Create new nomination” on COM:FPC – you must just fill in the creator name. I have taught FPCBot to look out for this phrase. If the bot finds this phrase in a nomination and sees that the user mentioned here is different from the uploader and from the nominator, it will notify the creator with the help of a new template, {{FPpromotedCreator}}, and also mention that user in the monthly summary of new FPs.

I am still testing this new feature to rule out any unwanted side effects, but it will be up and running next week. Happy nominating! – Aristeas (talk) 19:20, 30 July 2025 (UTC)

Technical details: You can put arbitrary text between the {{Info}} and the created by [[User:...]], but they must be in the same paragraph. You need to use the exact wording created by [[User:...]]; the bot also tolerates the uppercase spelling Created by and two or three common variations like created and uploaded by, but you are on the safe side when you stay with the default created by [[User:...]]. The user name can be specified either in the short form, like [[User:Kritzolina]], or in the usual long one [[User:Kritzolina|Kritzolina]]. And it doesn’t matter if you use the phrase in a nomination for which you don’t need creator notification – if the creator is the same person as the uploader or the nominator, it is just ignored. Of course, this feature does not work with Flickr user names etc., only with Commons user names. Aristeas (talk) 19:20, 30 July 2025 (UTC)

• Thank you very much, Aristeas, for this excellent enhancement! Notifying the original creator as well is an important and long-overdue step - especially when authorship isn't obvious. Your solution is pragmatic, elegant, and at the same time respectful of the effort and creative contribution of others. It's truly impressive how thoughtfully and consistently you continue to develop FPCBot - that's by no means something to be taken for granted. With sincere appreciation and warm regards, -- Radomianin (talk) 19:40, 30 July 2025 (UTC)
• Yes, great work! Thanks a lot! Yann (talk) 21:37, 30 July 2025 (UTC)
• Aristeas, a very nice and thoughtful idea and well implemented. Giving you access to the FPC Bot coding, is one of the best ideas lately here at FPC. I've added this new template as a step in the Commons:Featured picture candidates/What to do after voting is finished#Manual procedure. All the steps should be there if the Bot for some reason (Heaven forbids!) should be out of commission for a while. This is not very likely, now that the Bot is in Aristeas' very capable hands, but just in case... (power outage at the Bot server/earthquake/whatever) --Cart (talk) 12:13, 31 July 2025 (UTC)

• Many thanks to all of you, much appreciated. The new feature will be available starting with tomorrow, August 1. – @Cart: Great idea to mention {{FPpromotedCreator}} in the description of the manual procedure! This is indeed very useful, especially because that short manual is also the only kind of documentation we have for FPCBot’s duties (there was a separate description, but it has been deleted by Eatcha … maybe we should restore and update it). I have added {{FPpromotedUploader}} to the description, it was missing, too. – And indeed, it is very possible that we will need the manual from time to time, namely when I make a stupid mistake myself and the bot gets stuck as a result ;–). – Aristeas (talk) 18:40, 31 July 2025 (UTC)

FPCBot asking for help

The gallery link Animals/Mammals/Primates in the nomination Commons:Featured picture candidates/File:Davidraju img15(Macaca mulatta) Rhesus macaque.jpg does not specify the section on Commons:Featured pictures/Animals/Mammals/Primates to which the image(s) should be added. Therefore one or more new featured pictures are added to the Unsorted section of Commons:Featured pictures/Animals/Mammals/Primates. Please sort these images into the correct section. Thank you! / FPCBot (talk) 21:03, 31 July 2025 (UTC)

 Done by Cart, thank you very much! – Aristeas (talk) 15:51, 1 August 2025 (UTC)

FPCBot asking for help

The gallery link Animals/Arthropods#Order : Hemiptera (True Bugs) in the nomination Commons:Featured picture candidates/File:Northern flatid planthopper nymph (81276).jpg does not point to a valid section on Commons:Featured pictures/Animals/Arthropods. (The section after the # in a gallery link is valid if and only if it corresponds letter for letter to a subheading which is immediately followed by a <gallery> element.) Therefore one or more new featured pictures are added to the Unsorted section of Commons:Featured pictures/Animals/Arthropods. Please sort these images into the correct section. Thank you! / FPCBot (talk) 05:01, 1 August 2025 (UTC)

 Info To put it more clearly: The gallery section Commons:Featured pictures/Animals/Arthropods#Order : Hemiptera (True Bugs) is missing a subsection for the family of Flatidae to which this photo belongs. We need to create that subsection and to move the new FP from the ‘Unsorted’ section to that new subsection. – Aristeas (talk) 08:58, 1 August 2025 (UTC)

 Done by Cart, thank you very much! Footnote: I will not search for ‘unsorted’ FPs as frequently as I did in the past; now that the bot reports them here, everybody can help with sorting them. – Aristeas (talk) 15:50, 1 August 2025 (UTC)

•  Comment I must say something similar. I won't be around forever to fix the galleries of nomination so that they don't result in the FPC Bot asking for help. The way that Aristeas has set this up, I hope that the good people who make nominations at FPC will also do their part in maintaining it too. The QIC is a collaborate effort of all who participate in it, the FPC should be the same. --Cart (talk) 16:15, 1 August 2025 (UTC)