Commons talk:Quality images candidates/Archive/2007-01

1. Generally good idea.
2. I dont completely understand the "Technical" section. It sugggest an image with good exposure, but bad compostion, perspective, etc. could be selected?

   that wasnt what I was thinking, I thought that if an image had Good DOF then it could be added to that section, then not only does commons have a picture resource, it has readily available section on technical properties. Gnangarra

3. I think the criteria should be similar to submission standards of typical (micro)stock agencies. Compared to FP - do not hunt for exceptionality, just evauluate quality requirements.

Proposed criteria, stated mainly in negative way

1. Copyright infrigement / unclear copyright status. QP has to be uploaded to Commons by copyright holder under suitable license.
2. Missing or misleading description. QP has to be categorized, have meaningful title and description.
3. Advertisiments and signatures in image. Copyright/authorship info of QP should be located on Commons page and/or in image metadata, and should not interfere with image contents.

1. Photographic images
   1. Low resolution / thumbnail. Photographic QP has to have at least 1.92 megapixels (=1600x1200, for example)
   2. JPEG problems. Too much compressed, too low JPEG quality settings in camera / when saving. Visible jpeg artifacts. => Use better quality settings (e.g. set JPEG "superfine", shoot RAW, save in photoshop with max. quality)
   3. Noise problems, too much noise. Be it chroma noise, luminance noise, visible grain, scrathes in scans... QP should not have distracting ammount of noise when viewd in 100%
   4. Bad exposure. Overexposure, blown out highlights, underexposure, shadows details replaced by jpeg maps... In incorrectly exposed images, signifficant details in a siginificant part are lost.
   5. Color problems. Bad white ballance. Distracting (typically purple) hazing at 100%. Color aberation. QP must have reasonable colors (which does not necesarilly mean natural colors).
   6. Improper or undefined focus, insufficient depth of field. QP should have cleraly defined focus, e.g. main subject in focus, foreground and backround out of focus. Or whole scene in focus. Counterexaple - main subject blurry, foreground even more blurry, focus is somewhere between main subject and backround. DOF could be low on purpose.
   7. Blur. Images blurred just because of shaking hand or subject moving too fast. Motion blur in QP has to have purpose.
   8. Poor lighting. Including: distracting reflections (usual problem with built-in flash), uninteded vignetting, distracting harsh shadows. Generally bad lighting making scenes with space look flat.
   9. Overfiltered. There are so many PS/Gimp filters. Rarely better image is created just by applying more and more filters...
   10. Bad or nonexistent compostion, unclear or nonexistent subject. QP should have subject and compostion of the image should support depiction of that subject, not distract from it.
   11. Bad perspective, tilt, and other distortions. An eye (or, more precisly, a brain) is a sensitive detector capable of spotting even a small tilt ... falling trees, churches, inclined water surfaces,... Images of architecture should usually be rectilinear and without too much perspective distortion.
2. Stitched images, panoramas.
   1. Panoramatic QP has to have height of 800px min.
   2. Stitching problems. Stitched images should be without artifacts, colors and and lightness should be the same across the image.

--Wikimol 22:08, 28 June 2006 (UTC)

This is great criteria, I'll work it into this, maybe this(or similar) should be added into the FP page as well. Gnangarra 12:29, 29 June 2006 (UTC)

Process to change the status

At the moment, when someone wants to change the status of an image, he has to edit the section where the image is, cut the entry, edit the section where the image will be, and paste the entry. I find it a bit difficult, and long: as it is expected to be more quality images than featured ones, the page will be extremely heavy at the end. My proposition is as follow: create four templates (e.g {{quality_nomination}}, {{quality_promote}},{{quality_decline}},{{quality_discuss}}), with different background colours. This way, changing the status of a picture simply corresponds to changing the name of the template (one single operation). It is also easier to keep a consistent timeline, as all entries would be in chronological order, regardless of their status. CyrilB 22:40, 30 June 2006 (UTC)

These work well on the image page, but how does this place the images in the sections on Commons:Quality Images.. Gnangarra 11:23, 1 July 2006 (UTC)

still thinking would the templates be something like {{QI|Exposure}} to sort in the different groups? Gnangarra 11:28, 1 July 2006 (UTC)

I thought the propsal was meant this way: templates would not be used on image description pages, but here. Either the templates would be intelligent - something like {{quality_nomination|Image_name.jpg|description --~~~~}}, so there would be just plain list of the form

 {{quality_nomination|IOmage1.jpg| ....
 {{quality_nomination|Imagssdfgfg.jpg| ....
 {{quality_promote|IOmage1.jpg| ....
 

or the templates would be used inside <gallery> descriptions, like

 <gallery>
 Image0.jpg|{{quality_nomination| description --~~~~}}
 Image1.jpg|{{quality_promote| description --~~~~}}
 </gallery>

On the side of image organization, I think that would be better done by templates than on a subpage of (Commons:Quality images). The templates would be like {{QI Animal}} , which would add the image to Category:Quality images - Animals, generating the same as Commons:Quality images/Animals. I dont know if it could be parametrized, allowing something like {{QI|Animal}} of even {{QI|Animal|Focus|Color}} --Wikimol 12:10, 1 July 2006 (UTC)

OK I understand now thanks, ideal would be the last one {{QI|Animal|Focus|Color}}. I haven't got into creating templates before if we're able to make them lets do that. I'll see what i can find out about creating the ideal format. Gnangarra 13:55, 1 July 2006 (UTC)

looking like single cat templates aren't much of a problem. I should be able to create a sample shortly Gnangarra 14:36, 1 July 2006 (UTC)

My main idea is to avoid having 2 actions to do for one image, as it increases both the modem load (important for these galleries of images) and the risk of mistake. A template {{QI|Topic|State|reason}} were "topic" is one of those you mentioned before (animal, focus or whatever), "state" is the state of the process (nomination, promotion...) and "reason" is a space to explain the decision CyrilB 21:37, 1 July 2006 (UTC)

{{QIsample}} a starting place we will need svg graphics unique to QI, currently only links to one category, I'm working on building to the ideal format Gnangarra 14:53, 1 July 2006 (UTC)

Open nominations vs. commons nominations vs. self-nominations

Kelly Martin at Commons:Village pump#Quality images suggested the nominations should be limited to self-nominations. I think it's an intesting question, and better place to disscuss is here, so I'm replying here.

This is a misunderstanding of my comment. My recommendation was that this process be restricted to self-contributed work. I in no way intend to require self-nomination. Kelly Martin 23:58, 2 July 2006 (UTC)

Possibilites seems to be

1. completetely open nominations, anyone can nominate anything
2. open nominations of work created by commons users (i.e. no nominations of work from Flickr users or NASA)
3. self-nominations of anything uploaded by user (including works found elsewhere)
4. self-nominations of work created by user himself

Currently proposed is 2. In reply to Kelly Martin's commnet - maybe in the future, but I'm not sure wether to limit it to self-nominations from the beginning. It seems to me allowing nomination of others work would be natural way to promote the system - the original author would get some note on his talkpage (hey, your picture XY was revieweved and selected as Quality Image, thanks for contributing & you may want to nominate more of your images at ...)
Othervise, I'm a bit affraid it's hard to reach commons contributors. They're active on their "home" Wikimedia projects, but how many of them are active members of "Commons comunity"? How many will notice QI existence? --Wikimol 21:31, 1 July 2006 (UTC)

I agree: anyone should be allowed to nominate pictures from other users. In a way, this is also a good way to show people that their work is appreciated. And as you said, many people have little idea of what happens on commons, so they wont know there is a "quality image" category (another way could be to add a checkbox in the upload page to say "I think the image I'm uploading is of quality", but I'm affraid it would result in tons of junk to sort out afterwards...) CyrilB 22:31, 1 July 2006 (UTC)

Option 2 would be ideal, why should quality images by commons user go unnoticed and appreciated. Gnangarra 00:49, 2 July 2006 (UTC)

Back on the templates

It is easier to show what I mean. I created {{QI}} (please feel free to modify it as it is a dirty piece of code at the moment):

• Nomination of an image: {{QI|Image:U_core.png|topic=Animal|state=Nomination|reason=This is a test}}

Category Animal State Nomination Reason:This is a test

• Promotion of an image: {{QI|Image:U_core.png|topic=Animal|state=Promotion|reason=Wow,this depicts beautifully the wildlife}}

Category Animal State Promotion Reason:Wow,this depicts beautifully the wildlife

• Decline QI status for an image: {{QI|Image:U_core.png|topic=Animal|state=Decline|reason=This is not an animal, this is a magnetic core}}

Category Animal State Decline Reason:This is not an animal, this is a magnetic core

• In case it requires more discussion: {{QI|Image:U_core.png|topic=Animal|state=Discussion|reason=Are none of the magnetic cores animals?}}

Category Animal State Discussion Reason:Are none of the magnetic cores animals?

Of course, these template should be improved (color, size, ease of use), but the main idea is to have a single operation to change the status of an image. CyrilB 22:23, 1 July 2006 (UTC)

Great work CyrilB can multiple categories be used ie. Animals, exposure, focus Gnangarra 00:52, 2 July 2006 (UTC)

Great, but where the template is supposed to placed? On the image description page? I would disagree with moving the whole process there. One central place where one can overlook the whole queue and decisions is necessary.

That probably means, you will have to do two operations in one of the steps (nomination, decision, promotion), one operation in a central place, one in image description.

That's 1.3 places per status change in average, I think that's still good :)

--Wikimol 08:14, 2 July 2006 (UTC)

My idea was to put all these template on a single page (this one actually), with maybe a word on the corresponding image page (editing an image page is a light process compared to a page like Commons:Featured pictures candidates). CyrilB 09:27, 2 July 2006 (UTC)

Ok, than I agree. Than, adding categories and like seems unnecessary, and simpler template would be enough - I'll create working example.

Problems with editing can be improved if the queue is broken into sections per days. --Wikimol 10:22, 2 July 2006 (UTC)

Why not just put the template under the license box on the image page, then when someone reviews the image they change the template, I suppose for other categories it could be just a case of and additional category or 2 added to page as well. Gnangarra 09:39, 2 July 2006 (UTC)

see this Sample for what I'm suggesting Gnangarra

Yes, this would be done at the end. --Wikimol 10:22, 2 July 2006 (UTC)

Example

• 
  • Nomination: nice magnetic core --Wikimol 10:39, 2 July 2006 (UTC)
  • Review needed
• 
  • Nomination sample locomotive --Wikimol 10:39, 2 July 2006 (UTC)
  • Promotion ok, nice --Wikimol 10:39, 2 July 2006 (UTC)
• 
  • Nomination: nice magnetic core --Wikimol 10:39, 2 July 2006 (UTC)
  • Review needed
• 
  • Nomination green grasshopper core --Wikimol 10:39, 2 July 2006 (UTC)
  • Decline bad focus --Wikimol 10:39, 2 July 2006 (UTC)

--Wikimol 10:39, 2 July 2006 (UTC)

This seems fine. Now we have to sort out the discussion process (when people don't agree on an image). Should we do that on an other page (as I understand, the "quality" nomination is intended to be a quick process, so discussion should seldom occur) CyrilB 14:06, 2 July 2006 (UTC)

Second opinion page

Just to keep the page more neutral I would suggest we call it Common:Quality images second opinion. Second opinion would become a catch all page where if

• a reviewer wanted another opinion then they could put the image there
• uploader/nominator wanted to dispute a decline then they could put it there
• if a reviewer was unsure about nominating an image for FP they could put it there
• if a QI image gets edited then it could also be listed here.

I wouldn't call the page dispute because QI should be there to encourange not disputing. that was my reasoning in using decline instead of failed. Gnangarra 10:47, 3 July 2006 (UTC)

I would prefer to leave it mainly for case of disagreements, eitherwith rejection or promotion.

• if a reviewer wanted another opinion, he can leave the image to someone else
• if a reviewer was unsure about nominating he should give it a try; deciding wether to nominate a picture is exactly as difficult as actually reviewing it
• if a QI image gets edited, it shall be put again to standard process

My reasoning is this - standard QI process should be faster and without much disscussion, so I't better to try to decide thing standrd way, and only if it fails, move to Secon opinion / Consensual review /b whatever. --Wikimol 07:09, 7 July 2006 (UTC)

Stalling

I think this is start to stall, are there any reason why we cant activate/start the process and deal with and further issues as the arise. Gnangarra 03:17, 6 July 2006 (UTC)

So, lets' start :) --Wikimol 09:08, 8 July 2006 (UTC)

Go ahead... I follow you! CyrilB 10:44, 8 July 2006 (UTC)

Promotion process

• See {{QICpromoted}}, taken from en: wp.
• The current /Promotion /Nomination ... templates unfortunately dont work on archive page without modifications -> move to main template namespace?
• some logo of Quality Images would be great, for use in the template sticked to image pages.
• it would be nice if image organization could be done "automagically" by some code in the template and categories

--Wikimol 13:19, 12 July 2006 (UTC)

agree could the template not be such that the topic choice sets the cat. Gnangarra 14:47, 12 July 2006 (UTC)

Image edits

An image was recently edited based on the review it recieved here the result was great the image has been promoted. I think we need an effective format for addressing this when it happens again. Suggest that either the image goes into the Consensual review section with a response from the editor, or alternatively the image just gets renominated with a description of the edit. Any other suggestion? Gnangarra 11:14, 15 July 2006 (UTC)

comment

I've moved three images to consensus review because I feel defects were overlooked. I'd like to remind commentators that meaningful feedback to contributors is a large purpose of this process; reviews that only consist of a few words, especially in the case of a decline, are unlikely to provide useful constructive criticism. Kelly Martin 20:45, 15 July 2006 (UTC)

Images not commented on not promoted

From discussion there seems agreement that images not commented on after an amount of time should not be promoted. Any objection to adding this to the Guidelines? How about something like "Images with no review for 15 days after being listed are removed without promotion." ? -- Infrogmation 14:28, 9 August 2006 (UTC)

Thats a fair time period, someone should be able to assess the image within that. I suggest though that images get put into a single gallery on Commons:Unassessed QI candidates, from there they can be reviewed later for promotion/decline the same as other images. Gnangarra 11:58, 10 August 2006 (UTC)

That would be okay, though I think just removing them from here and adding a "Category:Unassessed QI candidates" would be as useful and less work. -- Infrogmation 02:44, 12 August 2006 (UTC)

Sure, I think the category is a great idea. SInce "unpromoted" usually carries the connotation that image was not quality enough, whereas tagging cat ad delisting it makes the distinction precise. Drini 01:18, 13 August 2006 (UTC)

OK, I have created Category:Unassessed QI candidates and moved the first unassessed image there per discussion here. -- Infrogmation 20:28, 13 August 2006 (UTC)

Modification of guidelines

I think a modification of guidlines in section Image page requirements must be considered. Now it says:

"Quality images must be categorized, have meaningful title and description. This should include the Taxa naming for plants and animals."

I propose to change it in reference to animals and plants, because in bio-sector, a system of categorized galleries has caught on. It may be changed more or less in that way:

"Quality images must be categorized, have meaningful title and description. In case of plants and animals, conditions include the Taxa naming and placing in an appropriate, categorized gallery page."

--Pko 19:17, 30 September 2006 (UTC)

This discussion is proposing changes that was created by project "Tree of Life" who have yet to adopt this as policy, there are still ongoing discussions at Village Pump concerning this, that it would be inappropriate for QI to adopt/adapt its criteria to suit something under these circumastances. Its also inappropriate to spread the discussion outside of its current locations. Gnangarra 11:45, 1 October 2006 (UTC)

OK, let's end discussion here. But... qive me only quick explanation. It's inappropriate to adapt criteria, which still are discussed, but it is appropriate to exclude valuable uncategorized images (correctly described and placed in galleries) even if categorization of them would possibly lead to edit wars. Am I right? Pko 14:01, 1 October 2006 (UTC)

Image Size for FP and QI

I have started a discussion article on what should be the size requirements for images being promoted to COM:QI and COM:FP. The discussion is Commons talk:Discussion on Image size please participate, from this discusion a proposal for specific requirements can be developed. Gnangarra 07:22, 23 October 2006 (UTC)

Modifications

I was wondering about modifications to Quality Images, or Featured Pictures for that matter.

You see, when I went through my watchlist, apparently one QI that I provided, had been processed quite a lot and in my opinion, for the worse. I was wondering if anyone might suggest an appropriate way to react to such image modifications?

I fully recoqnice that images provided by "post processing limited" users, such as myself, may well benefit for some further PP-work (and I certainly appreciate, if someone has time to do it) - but however, if the community has identified certain picture as QI or FP, shouldn't such images be left alone at least in connection with Wikis (and I certainly understand that licenses used entitle others to modify images as they see fit) or submitted to new voting or something?

After all, how I understand the purpose of QI and FP, is that like grammatical corrections in articles, they are designed to improve the quality of the project as a whole. At the moment I am simply considering of revoking the modification and perhaps writing a message for modifier's talk page to the effect that "...as the image has been voted...would you please either leave them as such, or submit modifications for new vote...". I fully recoqnise that the modifier may well be rather new user and meant well. I certainly do not intend to "hammer" him/her with a message like "You f*cking moron... Don't you understand...", but to approach with a constructive way. However, if we take worst case scenario, perhaps modifying QIs or FPs images could be a new way for "vandalism", which would certainly not be beneficial for the project.

The images I am talking about are below. The original QI, which shows quite sufficient feather detail on the back of bird and the modified one which is clearly overprocessed to the point where all detail is lost.

So what do do? Please, advice me.

PS. For purposes of conversation, I have posted this issue on both QI and FP talk pages. --Thermos 20:47, 28 November 2006 (UTC)

Original

Modified

Why not handle it like Wikipedia? Images can be changed like Featured articles, if it's a bad edit everyone can revert it. If it needs futher discussion go to FP removal candidates. It would be bad to a priori decline edits. I don't like second uploads like ImageOfATree_edit.jpg because a) the currently linked pages will never benefit of the version change and b) the version history gets lost. --Ikiwaner 17:40, 29 November 2006 (UTC)

"Grace period and promotion"

Commons:Quality_images_candidates#Grace_period_and_promotion says "If there are no objections in period of 2 days (exactly: 48 hours) from review, the image becomes promoted, or fails. If you have objection, move the image to Consensual review state."

But there are several images which have been reviewed long back (say, much more than 48 hours ago), which have not been moved to the Consensual review state, and have not been promoted/rejected. Am I missing something obvious? Kprateek88 12:20, 2 January 2007 (UTC)

It means that the closing process is less than perfect... However, such an image has indeed failed or been promoted independently of the fact of not being archived. For a promoted image it is legitimate to add the template "Quality Image" immediately after the grace period has expired even if it has not been removed from the page. In general, I would suggest a little more attention and effectiveness from the closers... - Alvesgaspar 15:41, 11 January 2007 (UTC)

As the closers work in their spare time we can't expect them to be more attentive. But I expect from those people who tag their own images just after grace period also to help archiving. For me it's OK when a pic gets promoteed or declined after grace period since feedback to the contributors is one of the goals of FPQI. --Ikiwaner 17:30, 11 January 2007 (UTC)

you mean QI-LadyofHats 18:16, 11 January 2007 (UTC)

perfectly right. --Ikiwaner 22:12, 11 January 2007 (UTC)

Rewrite of the guidelines

As we had a discussion about having too much bad pictures and not enough good ones here I feel encouraged to ask to rewrite the guidelines for Quality images.

Today it's often not clear what exactly the difference between a Featured Picture and a Quality image is. Therefore many times images don't pass at Quality image because the reviewer wishes a perfect image. Perfect images are FP. We must accept non perfect images to illustrate the vaste amount of articles existing in Wikipedias. Let's try to quantify the amount of imperfecntess allowed for a QI.

We should also consider that most images are viewed as thumbnails. So there is no real use of being able to print large posters from a QI. You can print posters out of FP.

Proposal

Purpose

The purpose of quality images is to encourage the people that are the foundation of Commons, the individual users who provide the unique images that expand this collection. While featured pictures identifies the absolute best of all the images loaded into Commons, Quality images sets out to identify and encourage users efforts in providing quality images to Commons. Additionally quality images should be a place to refer other users to when explaining methods for improving an image.

new Additionally Quality images should strike out the top 5% of the images uploaded at Commons so usually a Wikimedia project valuable picture from a skilled or lucky amateur photographer should pass as a QI.

Resolution requirements

Graphics located on Commons may be used not only for viewing them on a screen. They may be also used but also for printing. or for viewing on very high resolution monitors. We can't predict what devices would be used in the future neither, so it is important that pictures being nominated have a large enough resolution, largest dimension exceeding 1600 px. new: largest dimension at least 1280 px. 1280x1024 is kind of standard for bigger displays so an image can be viewed full screen on such a monitor independent of aspect ratio.

new In case of very sharp images (i.e. downsampled ones) or images with fewer detail resolution may go down to 800 px largest dimension (SVGA resolution). Reason: High resolution with few detail is a waste of disk space. Those images can easily get upsampled where needed.

Image quality

Digital images can suffer various problems originating in image capture and processing, such as noise, problems with JPEG compression, lack of information in shadow or higlight areas, or problems with capture of colors. All these issues should be handled correctly new: when shown full screen on a cumputer monitor. This means that high resolution images can have significant unsharpness, colour aberrations or noise at 100% as long as they appear correctly on a standard 1280x1024 px screen.

What do you think? --Ikiwaner 23:34, 18 January 2007 (UTC)

I think you are on the right track. Images that are both amazing in content and technically should be "Featured" candidates. "Quality" should be tolerent of minor flaws so long as the image in some ways stands out significantly from the majority of photos. -- Infrogmation 22:47, 19 January 2007 (UTC)

Definately heading in the right direction, there should be a higher tolerance of technically flaws in the backgrounds. ie jpeg compression, or over exposed areas of sky/ground. Gnangarra 04:54, 20 January 2007 (UTC)

Full ack. Users who aren't equipped with state-of-the-art cameras should have a chance to get their image promoted as QI if it is outstanding with respect to composition or subject, too. --MRB 11:02, 20 January 2007 (UTC)

Thats why this was created in the first place to recognise/encourage the efforts of those who dont have state of the art equipment yet upload many very good images that are the back bone of commons, no point having only 6-10 contributors recognised with 2-300 images and the rest of the recognised quality images available being from US government agencies. QI should be recognising 15-20% of the images uploaded where the largest dimension exceeds 1600px well within the capability of most 4mpx point and shot cameras. Well that was my reasoning behind developing QI in the first place, that said QI should develope into what the community wants. Gnangarra 12:58, 20 January 2007 (UTC)

Please just don't dumb it down too much. Overexposure? JPEG artifacts? 600px size? Sorry, but QI should be what the name states. --Dschwen 15:07, 20 January 2007 (UTC)

I agree with Ikiwaner idea, but I think that minimal resolution should be higher that 1280px. 1600px will be good. Lestat 17:23, 20 January 2007 (UTC)

There are just few monitors that could show a 1600 px shortest size image full screen. Samsung SyncMaster 305T is one. It costs about 1500 €. --Ikiwaner 18:35, 20 January 2007 (UTC)

Excuse me there was a caveat in the new guidelines. To achieve what I stated image dimension should be 1280 px largest size which is not that far from 1600 px. --Ikiwaner 15:30, 21 January 2007 (UTC)

• I strongly  Oppose the new text. Remember that very often pictures are seen in full resolution, even when they are bigger than the monitor, just to observe the details (like we do with paintings). Also, we should be able to get perfect printings of QI images in full resolution. I agree with Dschwen in that we should not degrade the standards of QI. Maybe a new award could be created to apply to beatiful pictures that do not technically qualify for QI. - Alvesgaspar 17:39, 20 January 2007 (UTC)

All pictures get pixelated if you zoom in enough-> Don't judge large images at 100% on screen. I agree that we shall get good prints like for Wiki Readers out of QI. But this is given for the new proposal since a print has about 200-300 dpi compared to a screen with 70-100. A new award would not do anything since you'll have to clarify then what the difference between 3 different quality steps are (good images, QI, FP). That will be more complicated than explaining the difference between FP and QI. If Alvesgaspar just wants to accept QI that print perfectly then what would be the difference to a FP? --Ikiwaner 18:29, 20 January 2007 (UTC)

•  Info I support big resolutions for both Quality Images and Featured Pictures. While FP should be perfect in composition / technically, when judging a QI the fact that the person who shot the photo is usually an amateur (like most of us) should be taken into account. But we should keep requirements, what I've seen before on JPEG artifacts, noise, blown highlights and image size is not acceptable. The quality of an image can mainly be judged on: composition, grain (noise), image size, dynamic range (blown highlights, too much shadow), colours. The first ones are important for the visual comfort:

• • JPEG artifacts are a major quality issue, QI should not have major artifacts
  • Blown highlights are not acceptable. This is a quality factor, not respecting it makes a non-QI.
  • Image size: the majority point & shoot are > 4MP. I don't see why people couldn't upload images in a reasonable resolution (at least 1280x1024, preferable > 1600x1200). I agree that 6MP, 10MP or more MP are maybe too much, I don't imagine many people printing out posters from images on Commons every day. Images in lower resolutions should be rejected, I guess users intentionally upload them in small resolutions to keep people from using them (as wallpapers, printing...). This opposes to the whole idea of Commons. Wanting to have a QI image on Commons but not wanting people to use it is ridiculous.
  • Noise: too much noise damages image quality. But we should keep in mind that P&S cameras have as much noise at ISO 100 as other DSLRs at ISO 800.

REFOCUSING

before you start discusing about what can be tolerated or not, i would like to remind you some small details that you may need to have in mind.

Commons

on [Commons:Welcome|the welcome page] of commons you can read what the project is i site " ...Wikimedia Commons is a media repository that is created and maintained not by paid-for artists, but by volunteers. ..." and even when today the main propouse for commons is to cover the wikimedia proyects. is also true that they aspire to create a free image repository . wich means many pictures are acepted even if their use is not yer clear. actually as such one can hardly know wheather an image will be usefull. becouse the proyect is considering that with time new uses will come in, als the wiki proyects grow. in a normal "paid for" image repository, images are considered to be printed, and or as digital media. AND that includes poster size. and apart from disk space, lets face it is easier to get a small version of a big file, but IMPOSIBLE to get a big one from a screen size one. one can not get more detail from there is none.

even with this there has been tons and tons of images uploaded to commons that are close to be called TRASH. too small, few details, bad quality. imposible to be used, even in screen size. one of the main problems of commons together with organizing the incredible amount of images that come in, is between others, to be able to sort out the trash.

Feature Pictures

from the begining feature pictures has been focused in trying to find the best of the best. Sometimes having higher standards than many editors i know. i trully believe some of the best photographs on the world would have troubles becoming featured in commons. but any way.. in resume what featured picture looks for . simpled placed is "professionalism" not only that but they also pay atention to the actual posible use for such images. together with the complicity of it. i still remember how complicated is to get a picture featured just becouse it was taken in a zoo. in resume, what featured picture looks for is. professional looking, usefull, and tecnically perfect images. the so often called WOW factor.

Quality images

oposite to the wow factor .Quality is something that can be more easily recognised. there can be a really boring picture of a completly grey, flat stone, and still have outstanding quality. starting by size, color, composition, etc. Quality images tryed to achieve a more public aproach on images in commons. it tryes to sort out images that are good to be printed, good to be used, good to work with. surely some quality images will have a bit of wow factor, but that is not what QI is about. if well quality plays a good part in achieving that "wow".

My opinion

problem with what quality is.. everyone has a diferent concept on what this means. so even when one can measure light, contrast and size of an image. everyone will have a diferent idea on what is tolerable, what is needed and what is not. as a grafic designer i know that one can by propouse deform an image luminosity, one can even overexpose an image, in order to achieve quality. something that for many may sound rather crazy. but the same goes for color, contrast, focus, etc. so that clear to read sentences like "overexposed=bad" are not really usefull.

in order to understand what is quality and what can be acepted and what not. we need to know what are we looking for, and have a list of priorities.

• is it more important for an image to be seen clearly on thumbnail, or screen size, or both?
• is it more important that an image colors can be printed, or can they be screen colors?
• are there especific formats for especific sort of images (SVG, PNG)?
• is there a limit to jpg compresion, or is this format not acepted at all?
• is there a minimum size for a file? and why not a maximum?
• do the size allowed depends on other factors, like historical content, or format (svg)?
• are black and white images acepted same as color? and if so what about sepia?
• are altered (computer or camera) photografs allowed? and if so then artistic images too?
• do the main subject MUST be completly in focus? or should all image be in focus?
• what is more important, the complicity to take a photo, or that the main subject isnt cut?
• does composition goes above subject dificulty or the other way arround?

you start to get what i mean? and all this questions have not even touched the actuall content of the image. or weather this content is showed in a "good" way.

If QI is suposed to be a help for commons users in how to improve their images. then the nomination process should be much more than just a "short reason" but an actual advice. i doubht a normal human being can do much with a simple "bad quality" reason. or would understand much of a simple "bad jpg compresion". only few of us really understand what this means. so why not to try giving a bit more of advices than just reaons? peronally i find that there is no image that could not be improved. wich is probably a professional deviation. but in order to find a middle point between all this questions and the opinion from everyone. we need to get a better nomination system and more clear idea of what we want. When i think about it maybe it would be usefull to make a similar system like we used to have to qualify images during my time on the university. you divide the image in 3 aspects. first the image quality, second the composition, and third the "content" wich is not other than how faithfull is the image to what it wants to show. a relation between a mesage and how it was this achieved. each student gives a note to each of these 3 aspects on each image. this allows the author of the image to recognise its weak points as well as the strong ones, without loosing by that a chance to "pass" and since all students would give notes , one also aboids the "personal opinion" factor. finally you add the average sum of every aspect. and you end with a number that allows you to know weather the image has in overall quality or not.

i would even go a bit further and make not one but 3 quality seals. one for "sufficient" quality, another for "good" quality, and another one for "outstanding" quality.-LadyofHats 21:28, 20 January 2007 (UTC)

... and mine

I don’t think the guidelines criteria are really that difficult to achieve without professional knowledge or expensive equipment. But please don’t understand me wrongly: to create a high quality picture is not necessarily an easy task: we need to know some basic principles about photography and also to be persistent. Very often we have to repeat the same shot over and over again until we get the ideal composition, lighting, exposure, sharpness, etc. Very seldom a casual snapshot qualifies as a quality image. Much more often the perfect image is chosen among a series of shots made with different settings and under different conditions. In other words, the only secret to get great pictures is hard work, patience and some talent to recognize beauty whenever it crosses our eyes. This is a great place to learn from other’s experiences but we have to be humble and attentive in order to be able to profit from it. Also, we must know the pros and cons of our own equipment and use it accordingly, not trying to do things which can’t be done. For instance, I know that my single-lens non-DSLR camera (I have a Konica Minolta Dimage A200) will never be able to shoot a bird like Mdf (who uses powerful telephotos) or a building panorama like Dilif (who uses a high-quality DSLR camera and sophisticated panorama/stitching techniques). Are these limitations really impeditive to obtaining high quality pictures? I don’t think so.

To modify the rules in order to get more Quality Images is like cheating without a real profit because the overall quality of incoming pictures will remain exactly the same. Do we really need to do that to draw the attention of more talented and hard- working users (or to make the present users to work harder...) ? I also don't think so. -- Alvesgaspar 01:34, 21 January 2007 (UTC)

I agree with Alvesgaspar. Quality Images is good the way it is. There does not need 3 different standards of quality just to satisfy the people with little experience, patience, or luck with taking pictures. Taking pictures is trial and error. --Digon3 18:41, 21 January 2007 (UTC)

I also agree, and I strongly oppose the proposed changes. Commons images may be used for many purposes, and it is simply not good enough, in my view, for an image to look ok only as a thumbnail. At the very least, users may want to enlarge the image on screen and scroll across it, and they should be able to do so without the image degenerating too much. Obvious technical flaws such as extremely low resolutions, major JPEG artefacts and so on are easy for anyone to avoid, even with a cheap camera or scanner, and there ought to be no question of any image being called a Quality Image with really obvious flaws of that type. Other issues such as colour balance, composition, depth of field and so on shouldn't be judged as rigorously as with Featured Pictures, but in my view the standard ought to remain reasonably high or else the award becomes meaningless. That may mean that some less-talented photographers can never aspire to gain QI status, but that's an acceptable price to pay. If we don't have enough QIs, the solution is to encourage more users to apply and not to reduce standards. --MichaelMaggs 21:05, 21 January 2007 (UTC)

example image with too low resolution according current requirements

It makes no sense to me to set resolution requirements for QI higher than for FP. There are not few of such FP that wouldn't pass todays QI requirements (this, this, this, this, ...). I would wish to have more superb looking 12 MP images too. But let's face reality: There are almost no such images here. One reason is that it takes 6 minutes to upload a 8 MP 2.5 MB image with an analog modem connection.

I'd like to remember that the new proposal is no revolution that will "dumb down" our level that it becomes meaningless. It only implies that an image should look good when shown full screen. Everything bigger is poster printing and pixel peeping. I do not understand people like Alvesgaspar whining around that there are not enough FP (same applies to QI) when the level of rejected images is so high. Even images of users like MDF with excessively expensive equipement get rejected at QI. Others have won prizes in image contests and get rejected as FP (this). This is not encouraging an amount of people that could satisfy Commons needs for good pictures. Let us do this gentle update of the requirements. --Ikiwaner 23:07, 21 January 2007 (UTC)

• • why dont we make a list of statements so people can post weather they agree or disagree on each one. so to sort out a conclution of this conversation. becouse in my eyes there are points in wich we agree, and some were we dont. so to aboid discusing up to infinitum, lets just try to land some final points -LadyofHats 23:04, 21 January 2007 (UTC)

Yes. This could be:

1. new purpose
2. Resolution requirements together with image quality
3. Resolution requirements for very sharp images (paragraph 2)

--Ikiwaner 23:14, 21 January 2007 (UTC)

Subject

I would add “subject” to the list. I think it’s important for reviewers to consider the luck factor as well when reviewing pictures. You can demand higher quality images when the subject is cooperative and luck is reduced, but where luck is a major factor, such as wild life photography, a good deal of the reviewer’s consideration should be the luck factor. Remember this could also apply to being in the right place at the right time with a camera, (for example, the amateur low quality video images on every news channel of the 2005 Boxing Day tsunami comes to mind). Images of some historic event would be rare indeed, and if someone is willing to donate such an image to commons, I’d certainly consider such an image QI or FP even if there was some substantial flaw, because such images would be incredibly rare and valuable to Wiki, whether they meet the technical criteria or not.
I have been reviewing other images nominated to QI and FP, and I must say I see some fantastic images at times, very artful and professionally done, but I’m often left wondering how useful they are for what commons is for which is a repository of material for the various Wiki projects. At times, I am left wondering if the purpose of QI and FP is art photography and we are loosing focus as to what commons is meant to be. Don’t get me wrong, I fully support the need for some criteria to be met in order for the images to be QI or FP, but I am somewhat amazed at how difficult it truly is to get a good image promoted. Ikiwaner has already provided some very good examples of this so I won’t add further to that point.
I guess everyone knows by now that my interests are in underwater photography. To be honest, I don’t get disappointed if an image is not promoted, with a few exceptions where I truly thought an image would or should have been promoted. I still find it difficult to accept wildlife in its natural habitat being rejected because of the “noisy” surroundings, but those are exceptions rather than the rule. I have found that just by nominating some images, they get exposure and end up in an article somewhere, whether they are promoted or not.
To the point at hand, someone mentioned that it takes a lot of effort and retaking shots before you get that quality image and it takes a lot of work to get those great shots. I don’t deny that, and if your subject is very cooperative, such as a structure, a house, an apple, a diagram, a tree etc., I agree that you should be able to get an excellent shot as the luck factor is reduced. Indeed, for those subjects, I’d even encourage some stronger critique. In contrast, where the subject is uncooperative, it’s difficult to get a second chance to shoot it, like a burning building, a moving car, a boxing match, a golf ball in flight, wild life. The luck factor is extremely high in those instances. For fast moving fish, I can tell you from experience that luck is a good 90% of the image. First you have to have the luck of finding the subject, and then you have to try to shoot it before it swims away. For each dive trip I go on, I can easily discard 80% of the pictures. I get so many with tails cut off or from the rear etc. It really is not easy. And if you run into something rare, chances are you may never run into that species again.

Rare unicornfish, tail cut

Unicornfish from the rear

Adding pictures here of the rare whitemargin unicornfish as examples. The first image has a tail cut off; the second one is slightly from behind. These fish swam by in only a few seconds and I only got two shots off. Unfortunately no lucky shot and that is a shame because these fish are somewhat rare and very shy and it is very tough to get near them to shoot a picture. There is no second chance to set up a shot like that. I haven’t nominated these because of some obvious flaws, but the very rarity of the subject should add a lot to a reviewer’s consideration. I have several reference books on reef fish and marine life which I use to help identify creatures after dives and you would be surprised how many poor quality pictures there are in these published books. This means the editor/author didn’t have many choices to begin with, and if a book editor/author is crunched for choices where they usually pay royalties for using said images, you know a perfect underwater shot would be rare. Sure, you can find some fantastic images in dive magazines, but the magazines pay a lot for those images, and not surprisingly those fantastic shots are most commonly of slow moving subjects, don't see many of fast swimmers. I’d need a lot more expensive gear and practice to be able to shoot images like those in these magazines, but to be honest, if I were able to shoot images like that, I’d be selling them, not giving them away. I’d need to in order to recover any costs from the highly expensive gear. Anyway, I guess I’ve made my point and as you can guess, I’d put subject very high on the list. Jnpet 04:01, 22 January 2007 (UTC)

QI and FP discussions

I agree with Alvesgaspar, Digon3 and MichaelMags, and I also strongly oppose the proposed changes. Wikimedia Commons is a free database for media files, and, as far I can understand, for all kind of media. So the pictures should also be useable in print media, dia shows, or similar things and not only for the medium internet. So its our duty to ensure that our pictures have a high quality standard.
So QI ist there to show the user of this database that the picture, he wants to download, is reviewed, that it has a high quality and that there is no licence problem.
FP is for me a totally different thing. Is not only a perfect QI as Ikiwaner mentioned above. For me FP is a picture with a high technical quality like a QI and additional with a high artistical or aesthetical quality - a so called "eyecatcher". -- Simonizer 12:31, 22 January 2007 (UTC)

That exactly sums up my thoughts on the topic. I cannot agree more. --Dschwen 13:14, 22 January 2007 (UTC)

May I ask what is difference from what Simonizer says to this? --Jnpet 06:19, 23 January 2007 (UTC)

It doesnt differ so much the purpose part, but it differs from conclusions below the purpose, which permit serious flaws and thus contradict the high quality standard. --Dschwen 07:51, 23 January 2007 (UTC)

Ok, but then there seems to be a contradiction. Some of the examples and indeed if I understand the criteria of the rules as explained by Ikiwaner , it would appear QI has higher requirement standards than FP. He showed some examples of FP that would not pass for QI. Isn't that a contradiction? Someone mentioned not dumbing down QI, but isn't this just proposing to dumb it down to at least match FP level? The alternative is to increase the standards of FP to surpass QI. The point is, if everyone agrees that FP is a superb QI, shouldn't either the QI or FP criteria be adjusted to ensure that FP indeed is a superb QI? To be honest, it seems to me the only way of ensuring that FP is a superb QI is to ensure that FP nominations have first been promoted to QI. In other words, once QI has been awarded, then the image can be nominated to FP. You would certainly avoid these contradictions as has clearly been pointed out by Ikiwaner. Jnpet 09:52, 23 January 2007 (UTC)

It is not true that QI has higher requirements than FP either in theory (i.e. according to the guidelines) or in practise. Of the four examples given three are almost 2 years old, from a time when the promotion criteria were different (and applied differently) and one is a painting. One has to understand that during this period not only better equipment became available and less expensive but also our selection process had a clear evolution, as a result of more and much better pictures being uploaded everyday to Commons.

If we go through recent FP candidates reviews we will verify that the lack of technical quality is, by far, the most common reason for not promoting a picture. In general, poor quality pictures are discarded very early in the process, like here, here and here. In other cases, technical quality is not enough, like in this picture which IMO would easily meet all QI criteria. Much less frequently (and I couldn’t find a suitable exmaple) a picture is promoted despite the fact of not meeting technical quality criteria. These cases are obviously rare exceptions to the general rule.

From the above it follows that the only reason for forcing pictures to be qualified as QI prior to a nomination for FP would be to free the FP candidate’s page of substandard nominees. But is it really necessary? Alvesgaspar 12:42, 23 January 2007 (UTC)

Thanks for the explanation, that cleared up a few things and it certainly makes sense.
As for your last question, it may not be necessary, but I think it would clear up a lot of confusion. Anyone new to QI and FP would not know the difference as it stands now until someone is kind enough to explain it, indeed I was confused at first and started nominating to FP before I even realized what the alternatives were. By putting a step procedure in place, with clear guidelines will make it very clear to any newcomer who wishes to contribute. I would also think that it would make it easier for new reviewers to understand the criteria for QI and FP respectively where a reviewer looks for minimum quality, licensing and subject requirements for QI and then can have real personal and highly critical look at the aesthetics of an image for FP nominations. Instead it seems rather confusing still and I get the feeling QI and FP are reviewed with the same criteria in mind by some reviewers.
That's one part of the reason, the second is that if an FP first had to be awarded QI, I think it would add a lot of prestige to FP and any such image would also be seen as highly prized. It would probably free the FP candidates page of substandard nominees as you say, but that too adds to the prestige of FP and I'd have thought it would make it easier for reviewers if they wouldn't also have to put their QI hat on to review FP nominations. Jnpet 14:15, 23 January 2007 (UTC)

We're trying to increase the number of good-quality candidates for FPs at the moment, and I'm afraid that forcing candidates to have already passed QI would reduce the number of FP images substantially. I agree that new users often don't understand the difference, which results in too many sub-standard images being put up for FP, but can't we deal with that by making the rules clearer, and having good cross-references at the top of the QI and FP pages? --MichaelMaggs 17:15, 23 January 2007 (UTC)

Is it quantity or quality that you want? I’ve often found that they exclude each other.
Regarding your last questions, I think making rules clearer is not going to help much. Keep in mind that Commons is open to all Wiki languages and projects. If clearer rules in English are the solution, you have already excluded a good number of non English speaking Wikipedians. To be honest, I find even native English speakers aren’t bothered reading instructions at times. If you are looking to increase FP and QI nominations, I would have thought the first thing to do is to bring it across all languages. Take a look at the Commons home page. Click on a language listed and you see the home page is instantly translated maintaining the exact format and layout of that page. Who did that? How can we get that done to the QI and FP pages? Just getting these pages into Japanese or Chinese and I’d estimate your exposure could easily be doubled or tripled.
Only issue I see here is there could be communication problems when nominations come in from various languages. Have already noticed a couple of comments on FP when someone complains about something they don’t understand to the point of being rude, which isn’t very encouraging. Linguistic tolerance is key. In most instances the red for oppose and green for support is very clear in any language. Where explanations are made, I’m sure we can get some volunteers to assist, in fact put that in the guidelines that if anyone bilingual could assist with comment translations, for their specific languages, please do so. You could even have standard translated response templates like (Out of focus), (Not clear), (Noise), etc. to make it even easier to bring it across the languages. As it is now, English is excluding potential contributors, so we make English the “bridge” language between all contributors. Jnpet 06:17, 24 January 2007 (UTC)

P.S. I’ve inserted some headings since this discussion has apparently moved beyond what the initial proposal was regarding. Didn’t intend to derail Ikiwaners initial proposal. Jnpet 06:17, 24 January 2007 (UTC)

Going back to QI prior to FP requirement idea. What if all images which pass QI automatically get sent to FP as a nomination? That way, the issues pointed out above regarding the confusion between the two gets addressed, plus it also addresses the issue of increasing FP nominations. A simple process, where first round evaluation asks the question is it QI? Second round evaluation asks is it also FP? A good image would sail through to FP, whereas others may only be awarded QI and fail in FP. It also means that a nominator need only nominate an image once, whether seeking only a QI or aiming for an FP, the starting point is the same regardless. It would eliminate junk in FP giving it more prestige, it would simplify overall nomination process and eliminate confusion, it would increase both QI and FP nominations and should give a QI to FP ratio that clearly shows he difference between the two, and on a plus side, someone nominating an image only expecting a QI, may be surprised to find him/herself with an FP. It could make things more fun for nominators. Thoughts? Jnpet 02:17, 25 January 2007 (UTC)

Seems a neat idea to me! Support. ++Lar: t/c 02:37, 25 January 2007 (UTC)

This is good suggestion, only question QI is for self published images only. NASA, Flikr etc images under the current criteria cant be QI, Gnangarra 05:24, 25 January 2007 (UTC)

Let's change the QI criteria. Open it up for all nominations.
Another thing that came to mind is that we could also add POTM and POTY to further streamline this. After FP is awarded an image is automatically nominated for POTM. At the end of the year, the 12 POTM are nominated for POTY.
The only outstanding issue would be how to incorporate POTD into this, pehaps a POTD can be determined by the most votes given to an FP on any given day. If there is no FP promoted on a specific day, the runner up of the previous day's FP could be promoted to POTD.
Could this work? Jnpet 06:18, 25 January 2007 (UTC)

• Nominating automatically all QI pictures to FP status would be an unnecessary burden for the FP candidate’s page because the profit of that measure would be minimal. In most cases a quick look is all we need to verify that an image is not “featurable”. Anyway I think it is also a responsibility of the QI reviewers to verify if the picture under evaluation has any chance of being promoted to FP and act accordingly. Sometimes I have suggested to the author the nomination of their pictures or I did it myself. Remember that FP and QI are quite different things. The QI collection forms a media database of high quality pictures which can be used for various Wiki and non-Wiki projects and which we turn to when we need a good image of a specific subject. So it is perfectly correct to have several QI of ducks or butterflies. The FP collection is, on the other hand, a kind of “Wiki show-case” or “gallery of distinguished pictures", that we are proud to exhibit to everyone as the best the Wikipedia can offer. In this case, it makes little sense to have several FP of ducks or butterflies of the same species. IMO, having few FP is not a big problem per se. The real problem is having a very low percentage of good quality images in the enormous flood being uploaded to Commons every day Alvesgaspar 14:00, 25 January 2007 (UTC)

• • Ok, let's recap. Here are the problems:
    • Not enough QI or FP nominations.
    • Huge Some confusion over the difference between the two, no clarity.
    • Reviewers treat both forums the same.
    • Junk images on FP, reviewers on FP also have to review as if QI
    • As FP is voting system, crap questionable images can also be promoted even if they don't meet QI.
    • Newcomers completely somewhat confused how to nominate properly in which forum to nominate.
    • Newcomers completely somewhat confused how to review properly.
    • No clear guidelines in multiple languages for either forum.
    • No clear guideline to define the difference of the two forums.
  • Have I missed anything out?
  • I've provided some suggestions and ideas. not sure what else to do. I've always found that if you disagree with a particular suggeted solution to a problem, you need to provide better suggestion. So p(Red wine talking). Please feel free to make alternative suggestions. because I'm running dry(unnecessary comment). Jnpet 19:29, 25 January 2007 (UTC)

Edited the above a bit, had come home having had a few glasses of red last night and I see now my street language came out. Sorry about that.
Anyway, anyone else have any ideas how we can fix this? I really think something needs to be done to improve the current situation, in order to make this place a friendly and welcoming place for all users new and old. As an experiment, I'm happy to start sending QI promoted images to FP. Shall we see how it works? Then we can run a little statistic on how many of those get promoted to FP as well. I'm thinking a proper ratio should be something like 1 in 5 reach FP. A short experiment for say a couple of weeks, without changing any guidelines. An unofficial auto promotion. Then we can discuss this further see how we felt about it. OK?Jnpet 08:12, 26 January 2007 (UTC)

• Answer to Jnpet

I started working regularly in Commons less than 4 months ago which, technically, makes me a newbie (just like you!). So, I’m not really sure what the main goals of Commons are or if they are being achieved or not. For me the main reason for being here and help is because it is fun. Also I’ve learned more about digital photography in the last 4 months in this forum than in the 4 years before. All this to say that I have no idea whether the number of QI and FP being promoted every month is enough for the people using Commons as a media database or not. When I used the expression “mild crisis” in the POTY 2006 proposal I was referring to the small number of pictures being nominated to the FP candidates’ page. Now I know that was a seasonal effect caused by Xmas holidays. From the first days of January everything came back to normal and we have now more pictures coming in and being promoted than the usual. Now, let’s see some of your points (I like red wine too. Did you know that the Portuguese have the best red wine in the world?...):

Not enough QI or FP nominations: I don’t know, it depends on the objectives. Did you notice that the numbers are increasing in both forums?

Some confusion over the difference between the two, no clarity; reviewers treat both forums the same; no clear guideline to define the difference of the two forums: if the guidelines need improvement, let’s go ahead with it. But I don’t feel comfortable with doing it myself. English is not my mother language (that should be obvious) and I don’t have enough experience as a Wikipedian.

Junk images on FP, reviewers on FP also have to review as if QI: not a real problem, because those images are rejected in an early phase of the process.

As FP is voting system, questionable images can also be promoted even if they don't meet QI: possible, but quite uncommon. Actually that is more likely to happen in the QIC page, because a single opinion is needed for promotion. Newcomers somewhat confused in which forum to nominate: our fault. The pages should be better designed and the guidelines improved.

Newcomers somewhat confused how to review properly: I’m not sure newcomers should be reviewing from the moment they arrive, specially if they are also newcomers to photography...

I agree these forums should be friendly and welcoming places for all users, new and old. It is in our hands to treat people with courtesy and help newcomers to find their way in the world of “creating and recognizing beauty” (that is, IMO, what this place is about). If you go back to the period October-November 2006 you will see that this place was much less friendly than it is now (I remember being badly bitten with my first nomination...). I also agree with your experiment of sending some chosen QI to the FP candidates’ page. Let’s wait and see, we all appreciate your effort and concern Alvesgaspar 10:10, 26 January 2007 (UTC)

Language Template for QIC

čeština | Deutsch | English | español | français | македонски | 日本語 | polski | português | +/−
I created this QIC language template. Anyone able to provide translation to the various languages would be so very kind! Jnpet 08:32, 26 January 2007 (UTC)

Hello Jnpet, with this template i can create a new page for QIC. But the pictures are on the page "Commons:Quality_images_candidates", and if i create a new page there won't bee shown the pictures. I had to translate the page "Commons:Quality_images_candidates/Intro", because this template is included in the english QIC page. Is that right? ---donald- 12:23, 14 March 2007 (UTC)

How we review large images.

I think that we are being a little unreasonable and unfair in how we review the images. It appears that people are looking at the images at 100% magnification for criticisms about sharpness and noise. If all images were the same size, this would be fair, but they are not.

If I take a 5000px image and downsample it to 1600px it will have less noise when viewed at 100% due to downsampling (...almost 10dB less, but since bayer sensors have color noise is tilted towards 1px spatial frequency, I'd expect even more perceptual improvement from downsampling). As such, the 1600px image would have an easier time passing as a quality image, yet the 5000px would always look better when printed at the same size. Both would look identical as a 800px thumbnail on the image page. By doing this we are encouraging people to submit lower resolution images than they might otherwise, these lower resolution images are useful for a smaller number of uses.

To address this issue, we should make always make our quality comparisons on the same size image. So if you are looking at a 4000px image at 100%, when you evaluate a 1600px you should upsample it to 4000px. Doing so might bias us against smaller images, but I don't know that such a bias would be unfair... Alternatively you could evaluate all images at 2000px or some other middle ground size (such as whatever size makes the image, say, 8" wide on your screen) which would involve downsampling some images and upsampling others.

I don't think it matters what people use, so long as each person is consistent and have an unreasonable noise expectations that come from looking at highly downsampled images. --Gmaxwell 19:50, 11 February 2007 (UTC)

• Noise; what gets to me, i look the image first at screen size and then in full view. if the noise alone is the only problem of the image, and if it only apears when in full view . i tipically ignore it and promote the image. but rarely this sort of problems comes alone. LadyofHats 14:32, 13 February 2007 (UTC)

This is the same thing I proposed some time ago further up this page. At the moment it's Alvesgaspar, Digon3, Simonizer, Dschwen and MichaelMaggs against Infrogmation, Gnangarra, MRB, Lestat and me while I'm not able to see a clear statement in the long text of User:LadyofHats. This is a stand-off situation.

In my opinion the statements of the opposers are contradictionary in zwo main points.

1. QI standard is higher than FP in terms of resolution (<0.5 PM pics got featured 2006)
2. people are encouraged to upload downsampled versions

--Ikiwaner 19:35, 13 February 2007 (UTC)

• • There appears to be some confusion about a so-called stand-off. Please don't just mix this issue with the proposed QI rule change. I absolutely support GMaxwells point. As a matter of fact I uttered pretty much the same thing already about a year ago. --Dschwen 20:06, 13 February 2007 (UTC)

Just in case some people are watching and don't quite believe: Here are two images. Don't click on them.

Now click HERE to see the image at 2,304px wide (how it would look on a 24" wide 96dpi print)

Now click HERE to see the image at 2,304px wide (how it would look on a 24" wide 96dpi print)

Now, study them for a moment. One was accepted as a quality image. One was rejected due to sharpness and noise issues. I think at the 400px size above they both look pretty good. Now, click on the 'HERE' on each of the images. Compare the two images. Which one is of better quality? ... Now go click on the actual thumbnails. Look to see which one is a QI.

Now, in this case, I probably picked one of the worst QIs, but it was hard to resist two images which were so similar. Go look at both at 100%, ... if you do that, you'll see that the larger image suffers when you compare that way. --Gmaxwell 21:40, 13 February 2007 (UTC)

For what it's worth, you have my agreement. Jnpet 04:21, 14 February 2007 (UTC)

And definitely mine as well. It's very sad that we're scaring people away from uploading higher resolution (and higher quality) versions of their images just so they can make it past a featured picture contest! --Cyde Weys 05:11, 14 February 2007 (UTC)

I would caution against judging images at the largest size that fits on your screen. When you view them this way, your browser does the resizing, and it's often not flattering. For example, when I look at Image:Wolf Trap (national park) meadow pavilion.jpg by Gmaxwell on the Image page or at full resolution, it's beautiful, but when I let the browser (Firefox or Internet Explorer) resize it to fit the screen, it looks seriously oversharpened. It's especially noticeable in the light-dark edge of the roof. Fg2 08:23, 14 February 2007 (UTC)

The browser isn't properly downsampling alas, a few browsers do (safari on mac for example).. What firefox does is just dropping pixels. I understand FF 3 is supposted to do better. Until then you can ask mediawiki to size it for you. --15:43, 14 February 2007 (UTC)

I argee with Gmaxwell statment as well. As Dschwen said, don't mix this issue with the proposed QI rule change. --Digon3 02:56, 15 February 2007 (UTC)

Template QICpromoted

Hi guys, there is this template Template:QICpromoted to inform a user that his image has been promoted. You can find this template ten times on my user talk page, even if just two of my pictures were promoted; just because I nominated those ten pictures. Wouldn't it be more useful to use it to inform the author or uploader instead of the nominator? norro 18:59, 13 February 2007 (UTC)

Yes it is usefult to inform primarily the uploader. At the moment usually the uploader and the nominator get the template. --Ikiwaner 19:14, 13 February 2007 (UTC)

We can limit this to notifying just the uploader. Not an issue. However, may I recommend that the nominator identifies the uploader in the short description if it is not their own nomination? --Jnpet 04:25, 14 February 2007 (UTC)

I personally would like it, if the uploader or author would be informed instead of the nominator. I think an additional parameter (uploader) in the Promotion-Template would be okay if it's necessary. norro 13:42, 16 February 2007 (UTC)

I think they should be all informed --Arad 03:29, 22 February 2007 (UTC)

"Purpose" and "Guidelines" in a location to be included in the QIC page

What do you think about moving the intro of the QIC page to another location (e.g. Commons:Quality images candidates/Intro) and to include it with {{Commons:Quality images candidates/Intro}}? Like this the page would become smaller (at the moment it exceeds 32kb) and more to edit. --Leyo 00:06, 6 March 2007 (UTC)

Agree. For nominations and judging single pics you can always use section-editing, but to work on the page big-time the source is too cluttered. --Dschwen 11:24, 6 March 2007 (UTC)

Done. --Leyo 21:01, 6 March 2007 (UTC)

A bot for this page

Hey gang! I'm currently developing a bot to perform some of the dirty work on this page. It is based on the pywikipedia framework and could run regularly from my toolserver account. In its current state it can parse this page and pick out nominations which were not commented on in the last n days (n being 4 currently for testing purposes). Right now it can extract declined nominations and archive them the the propper Archives Monthname YEAR page. Check out the results on my testpage (testarchive) bot at work. If you consent I will continue working on the bot so that it also knows how to process Promotions. That takes some more steps besides the already working archiving:

• Tagging the images (easy)
• Extracting the nominator and notifying him (easy)
• Extracting the original uploader and notifying him (a bit harder)
• putting the picture into QI (a lot harder, read on)

The last point is a bit tricky. First of all I need a way to find out which category to put the picture in. Secondly COM:QI shows the last few promoted pics for each category (but that's just tedious programming work for me...) --Dschwen 17:51, 7 March 2007 (UTC)

Great stuff. You have my admiration! --MichaelMaggs 17:59, 7 March 2007 (UTC)

It will need some testing to make sure I don't screw up the live page. But I need some ideas how to approach the promption issue with the QI categories (merit and subject). Should we suggest them upon nomination somewhere in the template? --Dschwen 18:08, 7 March 2007 (UTC)

Yeah the categories looks nasty, just see how complex animals is. What if the bot just place them in a gallery unsorted recently promoted. Then all that has to be done is that an editor needs to just cut and paste into the sorted galleries. Messy kinda, but this gives a short term solution, while gaining the major bot benefits of tagging, archiving, notifying.

Alternative thought would be to change the page layout from the current date nominated format which is in the signatures anyway to being a category format. Gnangarra 00:56, 8 March 2007 (UTC)

I was going for the unsorted recently promoted solution, but just noticed that the {{QualityImage}} template takes a Topic parameter, which I would not be able to fill in. Is that parameter even used currently (looks like it is not)? Anyways, otherwise usernotification and tagging code are written. They are just awfully hard to test, so I might have to do a live run on this page. Should be ok though. I'll announce it here, so just in the unlikely event that it completely blows up in my face you know about it in advance. --Dschwen 09:05, 8 March 2007 (UTC)

I just put up a bot status request to get this going. Please comment on the request there if you find a minute. --Dschwen 09:38, 8 March 2007 (UTC)

The bot status request is still pending, so we have to be patient. In the meantime here is a little toy which I now also use to browse QIC... --Dschwen 10:03, 13 March 2007 (UTC)

Bot status has been granted. I'll start a testing run later today. Expect things to get screwed up :-). I'll be gone for the whole next week with sporadic internet access. So we'll see later if the bot is ready to run without my supervision... --Dschwen 14:55, 15 March 2007 (UTC)

veedee veedee veedee. does compute! bleep. First bot run completed. bleep! --QICbot 18:06, 15 March 2007 (UTC)

Nicely done QIC bot! Wow sockpuppeting is kind cool ;-). Anywho, guess it's time to revise the archiving instructions on this page to.. ..basically do nothing, except regularly check Commons:Quality_Images/Recently_promoted and sort the images into the approproate categories. If we always append to the bottom, or prepend to the top, I could also have the bot automatically update the COM:QI page with the latest three images that were sorted into the corresponding categories. --Dschwen 18:11, 15 March 2007 (UTC)

Ooops, looks like someone was a bit sloppy and forgot to tag the promoted images... --Dschwen 18:15, 15 March 2007 (UTC)

Fixed. And, yeah, consensual review still has to be taken care of. But we might find a way how to do that bot-assissted as well. --Dschwen 18:43, 15 March 2007 (UTC)

Any progress on automated move to consensual review? Ben Aveling 07:33, 28 July 2007 (UTC)

Move "Nominations" and "Consensual review" to seperate page?

As seen on Commons:Featured picture candidates we had to move the nominations and Consensual review to a new subpage called "/candidate list". Only in this way we can translate the intro in other languages and use the same pictures. What do you think about this? ---donald- 12:50, 24 March 2007 (UTC)

Good move to have translations available, for nominations wait until the QICbot is in normal operation to enable User:Dschwen clear any issues. For consensual review as the bot doesnt process them yet I'd say go for it if theres no objections raised. Gnangarra 13:27, 24 March 2007 (UTC)

Are there any new results? -- -donald- 16:02, 4 April 2007 (UTC)

Well, QICbot is in normal operation. The page can be moved anytime, I can easily adjust the bot. I propose a tag or fixed format for consensus closings like {{/consensus|versionwhichwillbepromoted.jpg}}. The bot could parse this and do its magic. --Dschwen 16:17, 4 April 2007 (UTC)

I moved it. I hope i made it right. In the next days i want to translate the intro to german. -- -donald- 10:20, 5 April 2007 (UTC)

QICbot is adjusted and just performed the first run on the new subpage. --Dschwen 12:20, 5 April 2007 (UTC)

Consensual review and bot assistance

• I've just been helping to move images on from consensual review (hope I've done everything right ;-) and there are a lot of entries that'll be very hard for a bot to handle, they are difficult for a humanoid to handle ;-). My problem is what to do with all the nominations that have edited versions added during the discussion - these edits are often not by the author/submitter of the original work, and it is not always clear whether everyone will automatically transfer their votes to the new version anyway.
• Can we formalise how these revisions are handled - probably cleanest to have original author upload new version over top of their original but this would cause problems because original discussion (which will get archived) will now link to new version. So probably best to stop discussion on original with not promoted and have original author submit revised version to consensus review (original can be nominated for speedy deletion if author wishes). --Tony Wills 00:38, 26 March 2007 (UTC)

I see the old version can be linked to so uploading overtop of existing version isn't a problem after all (might have to edit image refs to point to old version) --Tony Wills 21:57, 31 March 2007 (UTC)

Who should get QI credit

In cases where a revised image has been created and promoted should the QI message be added to the original image creators page, rather than (or as well as) the creator of the modified version? (obviously the QI logo only goes on the actually promoted image). --Tony Wills 21:57, 31 March 2007 (UTC)

It is not so much credit, just a mere notification. I'd really prefer to see QI as a system to assemble a subset of high quality images than yet another dangling carrot reward system to please contributors. For the business of QI promotion the notifying the nominatior is sufficient and appropriate IMHO. --Dschwen 08:06, 1 April 2007 (UTC)

I wouldn't want a QI notice for just cropping an image based on discussions at CR, the author/uploader deserves any recognition Gnangarra 10:53, 1 April 2007 (UTC)

The reason I brought it up is that I've noticed that for a few Featured Pics which were submitted by someone other than the initial uploader/author (with some modification, eg white balance adjustment), the original author has never been notified. I'll review what's been happening with QI's. I think rather than just creating a subset of quality images, the purpose of this category is to encourage people to improve their photographic skills and hence submissions to commons. --Tony Wills 12:16, 1 April 2007 (UTC)

Commons:Photography critiques is supposed to help you improve your skills. QI is not the ideal place for that due to the brevety of the judgements by design. --Dschwen 12:33, 1 April 2007 (UTC)

QI started out to recognise and encourage good contributions especially in larger resolutions which would enable usage beyond wikipedia articles. PC compliments and extends beyond the QI/CR ability by being able offer more in depth discussion on the image. Something to think about maybe CR section is removed from QI and images get moved to PC for review. Gnangarra 13:03, 1 April 2007 (UTC)

It would be a bit confusing to remove CR from QIC: After having been yanked out of QIC and put into PC, would the images be resubmitted to QIC or bypass that and go directly to QIs? At the movement it is fairly easy to move images from review to consensual review as it's all on the same page. The only problem I see is that the page gets rather long, mainly because people aren't very explicit about their voting, and don't follow up by voting on revisions so that the the image can be promoted/declined. --Tony Wills 03:32, 2 April 2007 (UTC)

QIC to become superfluous

I'm close to a breakthrough with my QICbot. Using sophisticated image recognition techniques based on SIFT and GIFT, noise quantization and focus checking routines, perspective correction detectors and depth of field evaluator procedures, I created what I call the Advanced Prettyness-Rating Image Library. We are about to close down this page and perform fully-automated promotions, without the erroneous human element. Perfect digital consensus is coming! --Dschwen 09:04, 1 April 2007 (UTC)

Great! Full support!! Lycaon 09:13, 1 April 2007 (UTC)

Hi Dschwen, Lycaon. This is a great idea, but I'm not a friend of letting a bot decide whether a picture is QI/FP or not. Humans must retain control on that. On the other hand I see the necessarity of automating this nomination thing a bit. Because of that I just startet to write a bot analysing the statements of commons users, the Human Opinion Analyser X. This bot will be able to automatically interpret all statements on the nomination page, on discussion pages and so on. I think, that's the way to go. Hopefully I can show first results next week. --norro 10:05, 1 April 2007 (UTC)

That would be wonderful. Will your bot also be able to create comments from users, based upon their previous postings, and add them to the listings here? That would mean we could do away with the random human element entirely. If it's not going to be in version 1, I could have a look adding that feature, but I can't promise it will be ready for 1st April next year. --MichaelMaggs 11:06, 1 April 2007 (UTC)

Marvelous idea your APRIL bot! With a little more effort towards a realistic prameterization of the human randomness, we should be able to get free of the boring work of reviewing pictures, not only in QIC but also in FPC. Kudos for Dschwen! Alvesgaspar 11:50, 1 April 2007 (UTC)

No, all that's redundant, I've created a bot that uses online video-cams to grab high quality pictures of whatever is needed for other wiki projects. It automatically monitors and analyses the images coming in from millions of cameras and does frame grabs of appropriate images. Where necessary it will blend multiple frames to give the quality desired. Obviously there is often some delay until an appropriate image becomes available, so I am working on a caching system that will predict what images will be called for and collects them before a request is even entered. --Tony Wills 11:52, 1 April 2007 (UTC)

Consenual Review

This area is in desperate of an overhaul its not working and images are sitting there for weeks with discussion unresolved. Since I have an image in there and have participated with other discussions I wont close any section for want of being seen gaming the system, can someone do it.

During a discussion yesterday I raised the possibility of taking CR out of QI and sending disputed images to Commons:Photography critiques where indepth discussions can take place, if these discussions decide the image (with or without an edit) is QI then it should just be promoted from there. Gnangarra 01:52, 2 April 2007 (UTC)

See my notes in previous discussion above, basically the problem is that if people don't vote on the revised versions things sit there as undetermined for 15 days - solution is go vote on the revisions. --Tony Wills 03:35, 2 April 2007 (UTC)

• I have processed all that have been 48 hours since their last vote/comment. I've tried to be neutral in assessing peoples votes and have not assumed people will automatically transfer their votes to new revisions. If the results aren't what people expect feel free to nominate the images again :-). To make this process smoother people need to be explicit about whether they're voting or just commenting. Also remember the result is not to be posted until 48 hours after the last vote or comment. I have trialled separating the remaining consensual reviews into votes/comments per image so it is clearer what the status is. --Tony Wills 06:39, 2 April 2007 (UTC)

Potential candidate

en:Image:DownhillJump.jpg uses an interesting technique, but is perhaps not the best composed. I was going to nominate it here, but I noticed it isn't on Commons. Though if someone wants to move it over, it might be worth considering. -- Beland 01:52, 4 April 2007 (UTC)

• I think the image is too small to have a chance in QIC. Please check the guidelines. Alvesgaspar 09:11, 4 April 2007 (UTC)

Reverted move of images to subpage

I reverted the move of the images to a sub-page as it stuffed up editing the page as templates are in the wrong place and no one had attempted to fix it an hour after the move. What is the rational for the move? If it is necessary can someone do it properly ;-). If QIC Bot has been adjusted it might have to be adjusted back :-) --Tony Wills 12:35, 5 April 2007 (UTC)

The reason is to enable the translation for other languages. It is the same as in FP. ---donald- 13:17, 5 April 2007 (UTC)

Great idea, but presumably all languages are going to be nominating/promoting on the same page (maybe translations needed in the 'consensual' section (otherwise do we promote/decline images independently so one language could promote an image another declines?). So the intro section at the top is the main bit that varies with language, the main body is the same for everyone? --Tony Wills 13:33, 5 April 2007 (UTC)

The Consensual review rules section could be moved to a subpage for the different language versions. --Tony Wills 13:36, 5 April 2007 (UTC)

Please restore the moved version. Take a look at how FPC works. Apparently you misunderstood the way the translation works. --Dschwen 13:45, 5 April 2007 (UTC)

The problem is how the templates used on the QIC page are referenced. At present you add {{/Nomination etc to new entries - referencing the Nominations template attached to that page. At present this needs to be editted to {{../Nomination when entries are archived to the archive subpage.

When you move the Images section to a subpage, the template works when viewed from the main page, but when you're editing the subpage (eg adding nominations or reviews) the preview tries to use a template that is attached to the subpage - a template that doesn't exist. If you move the template to the subpage then the main page and archives won't find it. You could duplicate it but that creates mainetance problems (multiple versions of the same thing get out of sync). We could move it to the main template namespace then just reference it as {{Nomination but that would break all the archives (perhaps a bot could edit all the archive and history page references to it).

I think a solution would be to change all references to {{COM:QIC/Nomination which is a bit of a pain but would mean that it wouldn't need to be changed when images are archived and doesn't break anything.

Another problem is that when editing entries one seems to end up viewing only the subpage, and not the main page so suddenly the intro etc are gone ... a bit disconcerting.

• Perhaps experiments should be carried out in a sandbox rather than a live page! :-) --Tony Wills 00:11, 6 April 2007 (UTC)

Alternative is to leave the current template as for a historical record, and then change to a new template once an agreed solution is created. Gnangarra 05:10, 6 April 2007 (UTC)

Or add a copy/redirect of the templates to the candidate list subpage. And geez, how are you supposed to test such a move on a sandbox? Plus this is not an experiment, but a standard procedure on other pages! --Dschwen 07:10, 6 April 2007 (UTC)

Just tested it, inserting a redirect page inserts the target template. I will redo the move and create the template redirects later today. Then we can also create Commons:Kandidaten für Qualitätsbilder --Dschwen 08:52, 6 April 2007 (UTC)

• Good work guys :-). --Tony Wills 10:02, 6 April 2007 (UTC)

I translated it to german. It is not 100% completed yet. Feel free to finish. :-) ---donald- 21:56, 6 April 2007 (UTC)

Consensual review again

The earlier suggestion of separating consensual reviews into their own page makes more sense to me now, as I don't see how to have a translation of the (CR) consensual review rules otherwise. But that will make it a pain for people to move their nominations to CR especially if they're new here. Can we change to doing CR inline, just changing the template from /Decline to /Discuss say (a wide box without colour), discussion can all be carried out within the box. Any new image versions can also be added with further /Discuss boxes and the version accepted changed to a /Consensus as suggested earlier by Dschwen when discussing the QICbot. (still have to flag the non promoted discussions, perhaps /Dissension ;-) --Tony Wills 10:02, 6 April 2007 (UTC)

Question

How many people are working on this? Why are my pictures (mostly) not being rewieved? Thank you, --Orlovic 11:30, 9 April 2007 (UTC)

also, isn't it too uninformal to have only one opinion given about the photo --Orlovic 12:19, 9 April 2007 (UTC)

Have a little patience. Usually pics which are not clear cut take a litte longer. Also the process is geared towards throughput. One opinion is enough if nobody disagrees with it. If someone does there is consensual review. --Dschwen 12:52, 9 April 2007 (UTC)

• As for which images are reviewed, it tends to be what interests individuals, if the subject is uninteresting to them they'll probably skip to the next. Often there seems to be a hesitation in reviewing when something is marginal - not a clear pass or fail. I have three images awaiting review too :-), plus one in consensual review that has a consensus but I'm waiting for someone else to check the votes and close it.
• And yes there aren't enough people reviewing, read the criteria and other reviews and feel free to carefully review other nominations :-) --Tony Wills 13:27, 9 April 2007 (UTC)

---

Another thing, are criteria for QI less strict for featured images? (just to be sure), and can a photo be QI and featured at the same time? Thank you, again. --Orlovic 18:26, 9 April 2007 (UTC)

• Yes and yes. A featured picture has, in principle, to be technically flawless. But that is not enough, it also should be "valuable" for the Commons project and have a "touch of magic" (the so-called "wow" factor). Formally (i.e. according to the agreed guidelines) there is nothing against a FP being also promoted to QI. But it is useless IMO. - Alvesgaspar 19:02, 9 April 2007 (UTC)
  • Aren't the technical criteria for QI stricter than for FP? FP lets some low res pics slide sometimes if they score in other areas. --Dschwen 20:21, 9 April 2007 (UTC)
    • Ack Dschwen. In my opinion FP should be more about "wow"-factor, while the required technical quality wouldn't necessary need to be "flawless". On the other hand, it certainly helps is FP candidate is well executed in technical sense. --Thermos 14:02, 10 April 2007 (UTC)

---

I need to know the meanings of some terms you use in order to create better quality photos in the future.

• Poor framing = ?
• Bad composition = ?

Please check w:Composition_(visual_arts). --Dschwen 13:35, 10 April 2007 (UTC)

Thank you, --Orlovic (talk) 13:23, 10 April 2007 (UTC)

• I believe that most of the answers you seek are in here or here. Composition is the arrangement of the various elements of your subject within the image. A composition is poor when, for example, the image is cluttered, resulting in an unclear subject; or when the distribution of the elements is trivial or too symmetrical, resulting in an uninteresting arrangement. Framing is the choice of the photographer regarding the part of the subject being captured. The framing is bad when, for example, important parts of the subject were left outside of the image. Alvesgaspar 13:48, 10 April 2007 (UTC)

Example of bad framing

• • Bad framing or composition can appear in many forms.

• • For the purposes of your question, bad framing (in my understanding) means that a picture that would be otherwise acceptable or good (perhaps even for QI or FP) is framed so, that part of the main subject is in fact not in the picture at all. That would be clearly very bad framing (there are exceptions to that tule, like a picture where part of the subject is intentionally left to form a frame in the picture or something comparable). However, bad framing may be less drastic. Perhaps more common form of bad framing is a situation, where the main subject is cropped too tightly in the picture without "breathing space". As an example, you can use the picture on the right. Although the picture should be of reasonable quality otherwise (it is non-cropped with just elementary post processing), the main subject is roughly on the "bottom horizontal line" of "rule of thirds", the right wing is clearly too close to right edge of the picture. That would make the example obvioulsy badly framed (Own picture - wouldn't be too kind to use other people's pictures as an example of bad photos). --Thermos 14:02, 10 April 2007 (UTC)

QICbot after 4 or 5 days?

It looks to me as though QICbot moves images after 5 days not 4, eg entry

Image:2062 series locomotive (2).JPG/Promotion|EMD G26 locomotive in Pula, Croatia--
Orlovic (talk) 01:12, 13 April 2007 (UTC)|Good composition. ---donald- 15:45, 14 April 2007 (UTC)

should have gone on the next run after 15:45 on the 18th - ie the run on the 19th. --Tony Wills 10:26, 19 April 2007 (UTC)

Yes, but that run didn't happen yet. The bot is scheduled for 12:12 UTC. --Dschwen 10:59, 19 April 2007 (UTC)

• • Hmmm, yes my confusion was that the page history and watchlists display times in my local timezone which showed it running at: 01:12, 19 April 2007 QICbot (Talk | contribs) m (extract processed nominations older than 4 days), but the QI candidates are timestamped UTC ... all very confusing really ... :-) --Tony Wills 13:26, 19 April 2007 (UTC)

Improving consensual review process

I've been looking at the idea of putting consensual review items inline with normal reviews with these aims in mind:

• Creating a standard format so that QICbot can easily handle them
• Extracting the consensual review rules so they can go on the main page and be translated with the rest
• Easing move to C.R. - just move down a little on same page
• Getting greater participation in consensual review by having them up-front

I have created a Commons:Quality images candidates/Discuss template.
So basically the procedure for consensual review is:

1. Move nomination from gallery to immediately after gallery
2. Change to [[Image:thepic.jpg|left|180px]] (could use thumb but that adds unwanted space between it and discussion)
3. Change /Nomination,/Decline or /Promotion to /Discuss
4. Discussion carried on as now, but inside template, like adding more comments to review. (NB for template related reasons the "=" character within a discussion causes problems)
5. When decision made, or 15 days elapsed, post result on end of discussion and change /Discuss as appropriate (change to /Nomination if 15 days elapsed and no decision)

• Slightly modified bot can ignore /Discuss templates and handle /Promote, /Decline, /Nomination (time expired) as for other nominations.

(I note that all the templates have coloured backgrounds but they don't show up when used in galleries. I think the clear background looks better, perhaps edit all 4 templates to remove background colour and just have coloured boxes? --Tony Wills 11:50, 20 April 2007 (UTC))

I have removed the background colour in the /Discuss template --Tony Wills 07:47, 21 April 2007 (UTC)

See example at User:Tony Wills/sandbox (C.R.s near bottom, feel free to tinker :-)

Thoughts? --Tony Wills 11:38, 20 April 2007 (UTC)

I like the idea and have no concerns, the box without background colour looks fine, providing dschwen doesnt have any problems with adjusting the bot go for it. Gnangarra 23:54, 24 April 2007 (UTC)

I think the picture should be placed inside the template, and limit the length of the red templates to be the same as the yellow (see the crickets template in the User:Tony Wills/sandbox for what I mean). Don't know if it can be done but it would be nice. --Digon3 00:37, 25 April 2007 (UTC)

• I experimented with putting the picture inside the template and it works ok, and certainly make it clear which image is being discussed. But one of the things I was trying to do was make the instructions for moving things to consensual review no more complicated for new users than they are at the moment and retain a layout similar to the ordinary reviews.
• The template boxes automatically adjust themselves to the longest line width (an effect sometimes seen in the ordinary reviews when something forces the box too wide and nominations overlap). I don't think different size boxes will be so much of a problem when (if) we remove the background colour. But I'd be happy if someone could suggest a way of making fixed width boxes work for review & consensual review. --Tony Wills 01:00, 25 April 2007 (UTC)

• Your template solution is good for the QICbot, but I fear that it makes editing for the human contributors more complicated. The format with the whole discussion in one template is easy to break (either forget to close or use a pipe symbol) and most newbies won't know what to fix. It would be just as much work for me to adjust the bot in a way that it works it the current format, with just a closing template. That closing template would replace the text in the dotted box (which lately is mostly added by you, Tony) and should contain the filename of the promoted picture. That way only the closing folks have to learn something new. Either way, the discussion shouldn't be influenced by what's easy for the bot too much. Rather think of the children!! ;-) --Dschwen 06:03, 25 April 2007 (UTC)

• If people can handle doing reviews within the template box, they can handle doing an extended discussion in the same way, the only difference is they should put comments on a new line preceded by a colon or asterisk (and it won't break if they don't). They don't need to close it or anything, just insert their comments in much the same way that some already do tacked onto other's reviews. Anything that will break a review, will also break this discuss box, but we don't have too many problems at present. Basically the minimum people will need to do, to move to consensual review, will be to change the nomination, decline etc to discuss and that will work. Anyone else can then move the image and template outside the gallery so that the discussion will be more readable.
• When the discussion is closed, add a summary of voting in any format you like at the end, and change the discuss to promote/decline as appropriate - very clean and simple :-) (thinking of the children) --Tony Wills 07:33, 25 April 2007 (UTC)

• Hm, yeah, ok. The main difference I see between short and consensual review in templates is the length. People are just not used to putting a whole list into a template, and as far as I know there is no precedence for that anywhere on wikipedia, is there? --Dschwen 08:04, 25 April 2007 (UTC)

• Is there a limit on the size of template arguments? The number of C.R.s is pretty small so I don't suppose excessively long discussions are going to add significant load. Shall we do a trial run til the end of the month but leave them in the separate C.R. section for now and see if there are any problems (I'm happy to convert the existing C.R.s and revert them again if it causes problems). --Tony Wills 10:16, 25 April 2007 (UTC)

• Trial is fine by me, but I most likely won't have time to adapt the QICbot till early may. --Dschwen 20:37, 26 April 2007 (UTC)
  • Ok implemented trial, a couple of conversion problems, but I think it's all straightened out - let me know if there's any residual problems. I will tidy up and move the entries due for archiving tomorrow. --Tony Wills

Template background colour

I have removed the template-box background colour (see preceding discussion), I see this has affected the rendering of the box outlines when viewed in galleries (effectively changed the border style). Let me know if this is a problem (or of alternatives) --Tony Wills 11:26, 25 April 2007 (UTC)

Image:Lars-Hendrik_Roeller-portrait.jpg in consensual review

If anyone else agrees that this image isn't authored by a wikipedian and therefore it is ineligible then they'd better vote against it as current voting means it'll actually be promoted to QI. --Tony Wills 13:46, 27 April 2007 (UTC)

Image: Hakone open air museum (16a)

I was surprised to see this image nominated 0n May 9, 2007. This photo was not taken by a Wikipedian, but uploaded from Flickr on the same day as I myself uploaded Hakone open air museum (16) only a few hours earlier to Wikimedia Commons. Beleive it or not it is the same photo and both images are to be seen next to each other in Category: Hakone Open-air Museum. What must I think of that?--Gerardus 10:02, 10 May 2007 (UTC)

The second version is a crop of your upload to remove the person in the foreground. The editor appears to have saved it at 100% as the file is 3 times bigger than the original --Tony Wills 01:17, 14 June 2007 (UTC)

Can we please reformat this page?

I think QI is a great idea, but trying to read any amount of text in <gallery>...</gallery> is absolutely horrendous. The "consensual review" section has a sane layout; can we please please please convert the top half of the page to use that as well? --bdesham 22:17, 11 May 2007 (UTC)

By using galleries with the review as the image title, it all has to be one line, can't really break it up to make it readable. The idea was, I think, that people review those oldest first (and don't pick and choose), so all the unreviewed ones were short lines at the top of the list - it was easy to find the next one that needed reviewing. Unfortunately people (like me) pick & choose those images that interest them, so you have to delve into the midst of other reviews to find the image you want to review. We could perhaps put a blank line between image nominations (galleries can handle it, I expect QICbot can too). --Tony Wills 08:03, 13 May 2007 (UTC)

Affirmative. Empty lines are no problem. BTW I'd prefer the gallery format to otherwise stay like it is. Votes should be brief if possible. --Dschwen 11:55, 13 May 2007 (UTC)

Ok added blank lines between nom.s, hope that helps, if no unforeseen problems I will add that to the instructions. I have also trialed leaving a /Discuss item directly after the gallery rather than moving to CR (as I suggested in my old sandbox example). I again assume this is no problem for QICbot as it is outside the gallery structures. What do people think, is it a more useful format? --Tony Wills 13:18, 13 May 2007 (UTC)

I was mostly referring to the fact that running text is being presented within <gallery>...</gallery>. Given that something as short as a ~~~~ date stamp takes two entire lines, I think it's really unreasonable to expect people to be able to read text in such a format. --bdesham 05:10, 17 May 2007 (UTC)

Oh, sorry you mean on the page as displayed, rather than when editing it. Well it's just like reading narrow newspaper columns, I think people will survive the experience :-). The image description and review are meant to be short. The whole process is meant to be short and snappy. There are lots of images, don't want to have too many pages to scroll through. Anyone else find it a problem? --Tony Wills 09:02, 17 May 2007 (UTC)

FWIW, I hate the format of this page. The FP process works so much better. I'm technically saavy and I had trouble figuring it out at first. -- Ram-Man 20:17, 22 May 2007 (UTC)

Ditto Ram-Man. It is not at all intuative. If the purpose is to limit the number of reviewers by making it complicated, then the current system is working wonderfully. Cacophony 01:18, 26 May 2007 (UTC)

Mind rapidly failing :-(

I'm sure I noticed [:Image:Granny_Smith_Apples.jpg] was reviewed by someone (Alves?) but it is no longer (still showing as blue nomination) and the edit history seems to show nothing. Maybe there was an edit conflict and I saw it in a preview while editing the CR section? Did anyone promote it, or see it promoted, or has my mind finally left me :-( --Tony Wills 07:53, 13 May 2007 (UTC)

Norro has now promoted it (again?) ... --Tony Wills 13:07, 13 May 2007 (UTC)

Statistics

After a record number of Featured Pictures last month, I thought I'd check out the number of Quality Images. Looking at the archive page I get, 235 submissions, 128 Promoted, 104 declined, 2 un-accessed and one already QI. 54% promotion rate.
Of the 235 images submitted, 39 went to the consensual review process where 18 were promoted and 21 declined. --Tony Wills 10:47, 6 June 2007 (UTC)

QI vs FP

•  Question Under what circumstances, if any, would an Featured picture not also be a Quality image? (Assuming no mistakes made by voters, etc) Aren't FP standards at least as strict in every measure? Ben Aveling 12:56, 11 June 2007 (UTC)
  • Any picture that does not have sufficient technical quality but has a sufficient wow factor to compensate. A noisy picture of great subject would not be a QI, but could be good enough for a FP. A perfect example is Image:Humanitarian aid OCPA-2005-10-28-090517a.jpg which is very noisy and has relatively poor exposure but is special enough to be a FP. It is true that most FPs could easily be QIs. Another current FP nomination is garnering a number of support votes for its beauty despite somewhat low technical quality. -- Ram-Man 14:12, 11 June 2007 (UTC)
    • Image:Freudenberg sg Switzerland.jpg is not really a good example, given that you don't think it deserves to be an FP.  :-) Otherwise, what you say is true, and yet, wow factor does count for QI. It isn't always necessary, it certainly helps, and pictures do get declined as "nothing special", another way of saying "no Wow factor". Perhaps wow factor counts for more at FP, but the bar is a lot higher there to begin with. I have a hard time imagining a picture that could pass FP but fail QI. Consider Image:Passchendaele aerial view.jpg. People were prepared to lower the quality bar a little, but even for something with its emotional impact, that they still wanted "striking" quality, "perfectly" scanned. For a contemporary photo it would have to be Rodney King on a cell phone before it could pass FP yet fail QI, and not even then, I suggest. Regards, Ben Aveling 04:51, 13 June 2007 (UTC)
      • When did this become about me? Just because I don't like the photo doesn't mean it won't become an FP and then later fail a QI. Sometimes a QI should fail for compositional reasons, but those can be judged somewhat objectively, by using principles like the "rule of thirds". I've promoted all manner of boring subject matter for QI that had no wow. Half of my QI images fit that category. -- Ram-Man 12:05, 13 June 2007 (UTC)
        • Sorry, I didn't mean to make it about you. :-) I was just pointing out that a picture that shouldn't be an FP isn't a good example of a picture that should be an FP. I wonder if Image:Humanitarian aid OCPA-2005-10-28-090517a.jpg really would fail QI? For all its faults, I think people would still pass it. Regards, Ben Aveling 12:45, 13 June 2007 (UTC)
          • In that case, what is the point of QI? Just lesser standards? If QI is really just the same as FP, only the standards are lower, then it should be declared as such. The problem we have now is that some people think that QI is as described above, and others like myself feel that QI is about the technical quality of images while ignoring artistic and emotional impression, for the most part. This leads to conflict in determinig which pictures should be promoted. I think we should clarify the issue. -- Ram-Man 13:16, 13 June 2007 (UTC)
            • That's what I'm seeing. I'm not sure that it's a bad thing. If we said that FP is about wow factor, and QI is about technical requirements, that would make sense. It would mean that FP and QI are sort of equal but different. We could do that, but we clearly don't. If we said that QI are our best pictures, and FP are the best of the best, that would also make sense. We sort of do that. We require an FP to meet a higher technical standard than a QI, and have a wow factor as well. We allow a very few pictures to slip through FP on wow factor, but that happens at QI as well. We imply that there's a possibility for a picture to be FP but not QI, but I'm not sure that's actually the case. Regards, Ben Aveling 21:30, 13 June 2007 (UTC)
              • • I don't think there was an intention for QI to be a lesser technical standard than FP --Tony Wills 01:22, 14 June 2007 (UTC)
              • This only makes things more muddy in my mind. -- Ram-Man 23:13, 13 June 2007 (UTC)
              • QI is meant to be images of high technical quality regardless of whether it was the 1000th sunset, or otherwise boring. The emphasis is on quality. There was not a suggestion that FP images should be of higher technical quality at all, in fact I think the FP quality requirements have increased since QI was started. It is understood that FP require some popular support factor ('wow' factor), FP is a popularity vote. The problem with QI is consistency, because it can take just one vote to promote an image, it is easier to let things slip through - occasionally a promotion is queried and taken to the CR process which helps maintain standards. The decisions of new reviewers, especially, needs to be monitored. --Tony Wills 00:35, 14 June 2007 (UTC)

• (resetting indent) Of course the other reason an image can be FP, but not QI is if it was not created by a Wikipedian (ie QI is to promote good images created by Wikipedians rather than found images (eg Nasa)). --Tony Wills 01:22, 14 June 2007 (UTC)

Comparison of guidelines

	QI guidelines	FP guidelines
Guidelines	Qualified as 'only guidelines'. Includes size, jpeg compression, noise, exposure, color, focus, editing, blur, composition, distortion and stitching	Differently worded but a lot of overlap. Requires "high technical quality" and "...most valuable pictures from all others. Pictures should be in some way special...". Includes exposure composition, movement control, DOF, shape volume, color, texture, perspective, balance, proportion and "Symbolic meaning or relevance".
Resolution (the least unquantifiable quality)	1.92 megapixels, expressed on the candidate page as "largest dimension exceeding 1600px"	2 megapixels
Similar images	Allowed - 1000 sunsets are fine	Not allowed - one should be defeatured (but not always followed)
Usable as POTD	No. (Unless all FP already used, not likely at the moment.)	Yes
Qualifies for Meet our photographers	No	Yes (Current threshold is 10 FP)

In short, QI isn't exactly FP-lite, but it comes close. If QI was about technical quality and FP about wow factor, then I would expect images like to pass FP but not QI. Instead, it was the other way around.

What I'd like to see is the two sets of guidelines merged into a single page, with the differences made explicit. That would make clear the differences between the two. I think that's necessary before we can make an informed decision about making any changes to policy. Regards, Ben Aveling 22:52, 14 June 2007 (UTC)

• Not a good example, Image:Cygnus_olor_flirt_edit.jpg was made a featured picture, but failed QI, an edited version Image:Cygnus olor flirt edit 20070609.jpg made QI. So your example actually confirms QI is more stringent than FP :-) :-) --Tony Wills 03:55, 15 June 2007 (UTC)
  • Oops. Bad example indeed.  :-) I really thought that one failed FP, my mistake. And the one that passed FP was solidly on track to pass QI until I pointed out it was tilted. (We probably should go back to FP and get the FP status moved to the rotated version.) I'll have to have a bit more of a look around. Despite these examples, I still have the sense that FP is more technically demanding, and not just on size, and not just on pass/fail threshold and quorum. Looking at the guidelines, that isn't explicitly so (except for size), but going by people's comments, I think I'm not the only one who feels this way, which is of course circular. Regards, Ben Aveling 07:20, 15 June 2007 (UTC)

• There is of course no significant difference in size (just rounded up by someone for FP). --Tony Wills 09:11, 15 June 2007 (UTC)

• Technically, about 4%. Is that significant? Not usually, but maybe sometimes. But I think QI images sometimes slip through with even less, eg 1600*1000, Because the main page only says "largest side must be 1600". Regards, Ben Aveling 11:55, 15 June 2007 (UTC)

• Basically the technical guidelines are the same (or are meant to be), the difference boils down to FP wants "most valuable pictures" and "symbolic meaning or relevance". QI is more an internal wiki thing to promote quality and help contributors improve quality. Hopefully the judgement is objective but there is of course different emphasis by different reviewers (eg some object to anything 'centred', others to anything with the smallest part of the subject cropped off, others about 'noise' others about 'DOF') so images tend to get promoted to QI if the images quality is generally good and the reviewer's pet area is great :-). In FP negative reviews carry twice as much weight as positive ones which has the effect of demanding 'great' rather than 'good' in multiple aspects to get through. FP is not a subset of QI, so merging guidelines doesn't quite work, but I see no reason they can't share a 'technical quality' section. --Tony Wills 04:18, 15 June 2007 (UTC)
  • That all makes sense. But I do get the sense that people at QI are more prepared to overlook an image which hits most of the right notes and misses just one. Regards, Ben Aveling 07:20, 15 June 2007 (UTC)

• You are right, that has often been the practice - if an image falls slightly short in one area it often gets through (this is sometimes said explicitly). But, from reading the archives of QI (ancient history from a whole year ago :-), I don't think that was originally the intention. And yes there is a strong view, especially from those more familiar with FP, that QI is a lesser standard - 'not good enough for FP but good enough for QI'. But I think these attitudes are a matter of assumptions carried over into practice by new QI reviewers. I think QI is meant to be : tick the boxes, stamp that as QI, go on to the next image. --Tony Wills 09:11, 15 June 2007 (UTC)

FP/QI Guidelines Proposal

Agreed, I think there's been a drift happening, and that it should be stopped. How does the following sound?

We unify the technical guidelines into a single page, link to it from FPC and QIC, and then add at the bottom something like:

"A QI must have all of the above, and must have been created by a registered Commoner. An FP should have all of the above, must have a 'wow factor', and may or may not have been created by a registered Commoner. Given sufficient 'wow factor', an FP may sometimes be permitted to fall short on technical quality. There should never be two basically similar FPs. If two FP are basically similar, one should be delisted."

Do you think that would stop the drift without upsetting anyone or anything? Regards, Ben Aveling 11:55, 15 June 2007 (UTC)

• LOL :-) :-) 'without upsetting anyone' !! Couldn't possibly propose anything without upsetting someone ;-). You might want to define 'wow' factor as being outstanding in terms of quality, general interest or significance. I don't know that the uniqueness of FP images needs to be codified, there's no reason not to have multiple images of something sufficiently interesting, it's just that usually it becomes less interesting when it's been seen before - so it is self correcting. Considering current practice others are sure to disagree (and most of my QI images might get demoted ;-), probably worth making sure the regular QI reviewers are aware of this discussion first.--Tony Wills 12:33, 15 June 2007 (UTC)

• Well, without upsetting too many people too badly. And no, I don't want to define wow factor in terms of quality. I want to define wow factor in terms of emotional impact and raw beauty.  :-) But I agree, the importance of the subject is, well, important. I don't think this proposal changes any of the rules as currently written? The rule about not having 2 similar images isn't anything I've made up, nor is any of the rest of it. This just makes much more explicit the ways in which QI is, and is not, FP-lite. Perhaps we should add "There is no formal mechanism for delisting QI. The purpose of FP is to recognize that an image is currently among the most valuable images, something which is sometimes changed by the creation of newer, more valuable images. The purpose of QI is to recognize that at the moment of creation, a commoner exercised sufficient skill to achieve a desirable level of quality, recognition that is not erased by later advances." Regards, Ben Aveling 13:34, 15 June 2007 (UTC)

• I like this last wording by Ben except maybe drop the word "sufficient". Some history when QI was started FP targeted the top 1% percent of all images, QI was targeted at the 10% of images just below that, this was when NASA type images were dominant at FP and the normal image size being upload was still below the 1000px. The intent was to encourage Commoners to increase the size and quality of what they uploaded by having a recognition system that was dedicate to them alone. What goes with that is as the volume of size and quality of images increases the criteria will need to become more stringent, which it has. There was a lot of early discussion that QI could become a requirement for FP nomination as such most if not all (commoner created) FP should be QI as well. Gnangarra 23:35, 15 June 2007 (UTC)

• I've dropped the word sufficient. I think I've made clear that most but not all FP will have the same technical standard as QI. Can you san-check it for me? Thanks, Ben Aveling 02:51, 16 June 2007 (UTC)

I've merged Commons:Featured picture guidelines into Commons:Quality images guidelines. Tony, can you review my changes? If I haven't made a mess of anything, I suggest we move Commons:Quality images guidelines to Commons:Images guidelines (after all, these guidelines apply to all images, not only to glory seeking desperados) and that we redirect Commons:Quality images guidelines and Commons:Featured picture guidelines to Commons:Images guidelines. Thanks, Ben Aveling 02:49, 16 June 2007 (UTC)

I've copy edited the lead area, removed a couple of redundancies. Note that the page is already translated into Polish so after everyones happy with the wording it'll need to be translated into other languages. Gnangarra 03:34, 16 June 2007 (UTC)

I've reverted 'requirements' to 'guidelines' and I can't find anything in FP 'rules' that confirms "The rule about not having 2 similar images". --Tony Wills 12:38, 16 June 2007 (UTC)

It's looking good to me, good changes both of you. Regarding disallowing similar images, you're right, I can't see anything explicit in the rules either. I'm sure I've seen people say it in conversation. It's sort of implied under Value when it says "Pictures should be in some way special, so please be aware that: almost all sunsets are pretty, and most such pictures are not [in] essence different from others". But if it's made explicit somewhere then I don't know where. Anyway, I've made a minor grammatical fix and I'm going to leave it until some of the other regulars have had a chance to look at it. Thanks, Ben Aveling 13:48, 16 June 2007 (UTC)

Credit received by submitter and not photographer

Hello, I came across Image:Powerbook 100 pose.jpg and submitted it as QI. Now I have found the credit in my talk page. Thank you for letting me know, but credit where credit due, I think the photographer should get the laurels. Now he has. --Klaus with K 15:04, 18 June 2007 (UTC)

I see there has been some prior discussion. Looking at the templates, the FP promotion states "the image that you nominated" whereas QI promotion says "your image", hence my posting. And I agree that both nominator and uploader should be notified, maybe with slightly different wordings. --Klaus with K 21:57, 18 June 2007 (UTC)

That would make the whole procedure unnecessarily complicated. As I said before, just mention the creator of the pic in the nomination and he gets the promotion template. --Dschwen 22:16, 18 June 2007 (UTC)

Is such "created/uploaded by X, nominated by Y" bot-readable? It would work for me. --Klaus with K 22:31, 18 June 2007 (UTC)

Why the obsession with giving or getting credit? If you nominate a pic and you care for the outcome of the vote you ought to watch the page, otherwise you'd have no chance in participating in a consensual review anyways. The notification is just a convenience and should be given to the original uploader as he might not even know about the nomination. --Dschwen 09:16, 19 June 2007 (UTC)

Why the obsession with giving or getting credit? Because it makes us feel good.  :-) Is that so bad? Regards, Ben Aveling 11:53, 19 June 2007 (UTC)

Dumping pictures on QIC - a bad idea

I'm sorry but I don't like the idea of dumping a pile of pictures in QIC as if they were some documents to stamp. This way it looks more like a job and loses all fun. There is no need, in my opinion, that all the pictures we upload to Commons are promoted, even when they have good quality. QI status shouldn't work like a certificate. IMO it is more important and useful that QI serve as examples of quality in the various themes and technical aspects, like it used to be - Alvesgaspar 19:37, 21 June 2007 (UTC)

Good that you bring up that topic...i was also about to write something about this flooding of the candidates list. People should choose carefully which photos they think are of high quality or have a good composition and don't just propose all to wait and see which photo gets promoted. Maybe we should limit the amount of photos someone can propose per day. What do you think? --AngMoKio 19:46, 21 June 2007 (UTC)

• I we ever come to such an extreme measure ... than I would propose just 1 or 2. Alvesgaspar 20:17, 21 June 2007 (UTC)

well for the beginning we can come up with a recommendation that people should chose carefully and that floodings are not really welcomed.--AngMoKio 20:34, 21 June 2007 (UTC)

• We sort of have a throttle because if someone puts up lots of pictures they tend not to get voted on. I like your recommendation, and I'll add it. I don't think we need a rule, people are mostly sensible and it doesn't really matter once in a while. Regards, Ben Aveling 21:43, 21 June 2007 (UTC)
• Agree on the recommendation and hope that it will work. I voted in two or three of the new QIC and I don't intend to vote on more (one single user nominated 44 images!). Maybe what's happening now was inspired be the unfortunate phrase of the new guidelines: "There is no restriction on the number of similar Quality Images". I don't agree with it since it reduces QI status to some sort of automatic certificate neglecting what, in my opinion, is the most important role of a QI (I repeat myself): an example of what a high quality picture should be considering the various asthetical and technical factores involved. - Alvesgaspar 22:03, 21 June 2007 (UTC)
• I had the impression from reading the original discussions for QI that is was a lot about stamping images of quality which was why the process was designed to be quick - to handle volume. I agree that it is somewhat self limiting in that a flood of images will just drop off the end into unassessed images, but it is of course annoying that they clog things up. I also agree a guideline to point this out would be a good idea. --Tony Wills 22:33, 21 June 2007 (UTC)
• Tony is right QI was a lot about identifying good volume of high quality images, at the time 5mp cameras were at the top end of the home market now 5mp camera is standard with cell phones, it was also common to only see 1000px or less images being uploaded. QI needs to evolve with technology advances so limiting the number of self nominations is a way of reducing over load. Maybe its time consider higher demands on image size for QI to reflect this say move from 1600x1200 to 2500x1900(5mpx) Gnangarra 06:05, 22 June 2007 (UTC)
• I have not been around here for long. My opinion on this issue follows Tony Wills' (although I must admit I have not read the original discussion about QI/QIC). It is my interpretation of the presentation of QI in general that it is a place for every image that is valuable and of a high technical standard. "Valuable" means that is can be useful for an encyclopedia such as Wikipedia or another "useful" project (a photo that can be used on Wikinews might not be useful for Wikipedia, for example). "High technical standard" refers to a set of ideals that we need to define - I think the current guidelines are just OK but not better than that. I think we need to improve or tweak the guidelines. One way to do this is by raising the minimum image size as Gnangarra suggested. I support that idea. What I think should not a criterion for QI is uniqueness. That means that pictures of everyday things and occurrences are acceptable. Uniqueness can be found in many QIs because of their high technical standard. The dumping of too many pictures on the candidate list, as has happened recently, is bad. It is not bad, however, because it somehow "devaluates" QI - it is only bad from a logistical and organizational viewpoint, because it overloads the system - images might get lost and forgotten and the participants frustrated. On this issue, I agree with Tony Wills and Ben Aveling in that we probably would not need a strict rule. A guideline should be enough, especially since we always have the option of instating stricter rules. --Florian Prischl 16:53, 23 June 2007 (UTC)
• I've given reviewers the option of rejecting 'easy to take' images as too small even if they're larger than 2MP. I think that makes more sense than a hard limit. The right 2MP are worth a lot more than just any 10MP. (And remember that a lot of cameras that create images with 5 or 6 or 10 megapixels do it by squeezing in the sensors until they are so small that they produce no more real information.) Regards, Ben Aveling 01:16, 24 June 2007 (UTC)

FP from pl wiki

I nominated yesterday more than 40 pictures. These pictures are FP on pl wiki. I'm sorry, that you must looking at them, but we need your opinion (cause on pl wiki on FPC people often use "personal arguments") to removal some FP - templates ({{gmedal}} here on Commons). Maybe better will be removal all of them (on QI) and nominate day per day - 5-10 pictures? Of course without these which are red now ;) What do you think about? This way maybe was not the best - sorry. I will try to talk with pl users - "If you nominate picture on pl wiki - nominate it on QI too". Przykuta 05:37, 22 June 2007 (UTC)