Wikipedia talk:Featured picture candidates

From Wikipedia, the free encyclopedia
  (Redirected from Wikipedia talk:FPC)
Jump to: navigation, search


FPCs needing feedback
view · edit


Stop promoting Theora files[edit]

Stop promoting Theora files, or maybe stop evaluating video files, given the audience at FPC does not understand or care about video formats.

Theora is an outdated, its performance is worse than the non-free H.264 standard, and worse than the free VP8 and VP9 codecs. This means that for a Theora file to have the same quality as a WebM (VP8/VP9) file, it must be larger. If the files are the same size, and the source files are transcoded with the same quality settings, the WebM version will be better.

Compare browser support for Theora vs. WebM. WebM is supported on Android browsers, the most popular operating system in the world. Theora has no mobile support. Mobile usage on Wikimedia projects is roughly a third to a half of the total. Every time a Theora file is served to a mobile browser, Wikimedia must serve a WebM transcoded version. That WebM file will be transcoded from the original Theora upload, downgrading the quality further, it may also be limited in resolution as Wikimedia servers automatically transcode files only to a fixed set of resolutions.

Theora has poor hardware support. I have no real idea which devices have dedicated Theora hardware, Xiph.org is no help. I know that lots of devices have WebM hardware, including some of the most popular mobile chips like Qualcomm's Snapdragon. Hardware support means that there are dedicated circuits which will decode WebM video. Formats lacking hardware support means that the CPU will have to use its general purpose circuits to decode video, this increases CPU and battery usage.

None of this is new, Theora is years out of date, it has never been up to date, being noticeably worse than H264 even when it was new. I was uploading WebM files in 2013. I was opposing Theora FPCs in 2013. In that case, after I pointed out the original had been transcoded at a lower resolution and with an outdated format, the nomination was withdrawn. Yet in 2015, a file was promoted that had the exact same problems.

We should not be promoting Theora files as Wikimedia's best work. They should not be featured. At a minimum standard, video files should be WebM, and preferably using the VP9/Opus codecs. - hahnchen 21:43, 3 June 2016 (UTC)

I can't help but notice that the only thing you haven't actually done here is be BOLD and convert the videos yourself. Why should the rest of us have to suffer for your film preference? Isn't it enough that these clips have encyclopedic value for our project, or must those of us who see a neat image and bring it here with little understanding of the photographic process as is be further marginalized in an already uncaring process? TomStar81 (Talk) 23:50, 3 June 2016 (UTC)
TomStar, that strikes me as a pretty bad answer. If what Hahnchen says is correct (and I really don't have the knowledge to make a claim on that) then this should be a concern for all of us. FPC isn't about "neat images", it's supposed to be about the best- if you nominated something at FAC and then said "actually, I don't know anything about this, I just thought it was a pretty cool article" you wouldn't last long. The promotion of videos does worry me slightly, as it's hard to see if we're applying appropriate standards to them; the use of videos just hasn't taken off on WP as much as it could. Josh Milburn (talk) 07:46, 4 June 2016 (UTC)
I'd rather Theora files were not nominated in the first place. There are millions of files on Wikipedia with encyclopedic value, and only the tiniest fraction are featured. I've not edited or retaken the still images I've opposed either, they should just nominate better files. Compared with the minor subjective aesthetic "preferences" in photographs that are suffered over at FPC, Theora files are clearly objectively worse. The "rest of us" can't even view these files on their mobile devices - try opening a featured Theora file on the most popular Android browsers. - hahnchen 07:57, 4 June 2016 (UTC)
So FPC is lost to the rest of us then. Just a matter of time I suppose. Very well, I'll refrain from nominating images and videos here in the future. TomStar81 (Talk) 14:26, 4 June 2016 (UTC)
TomStar, that's a little melodramatic. Hahnchen, it's possible that we could try to make some changes; perhaps some of those who supported that candidate will have a view about the file format? Josh Milburn (talk) 14:47, 4 June 2016 (UTC)
Hahnchen: Why not start an RFC on FP Criteria's talk page about adding a note in the criteria prohibiting nomination of Theora files if a non-Theora source file is available (which means the source file has to be converted to WebM).... Also, is VP9 supported on Commons [1] [2]? Bammesk (talk) 21:39, 4 June 2016 (UTC)
VP9/Opus is supported, here's an example File:Armello - Launch Trailer.webm. Do we need a formal RFC, is this not covered by "If it is considered impossible to find a technically superior image of a given subject, lower quality may sometimes be allowed." - hahnchen 21:55, 4 June 2016 (UTC)
Hahnchen: About do we need an RFC: 1-the visual difference between a Theora video and a WebM video is not obvious, say at 640×480 pixels, 2-not everyone is aware of the latest browser and hardware support technology. Therefore, as long as this is the case, and the FP Criteria is mute about the issue that you raised, then the problem will continue, just my 2 cents. Hopefully others will opine. About VP9/opus: is there a free and user friendly converter? There is one for VP8/vorbis here and as you mentioned one here, but I cannot find one for VP9/opus. Bammesk (talk) 03:42, 5 June 2016 (UTC)
@J Milburn: Actually, its not so much melodramatic as it is a statement of fact related to evolution: way back when I first started literally any image could be a Featured Pictured. We've since gone to size requirements, then to photo restoration, and now to a point where images are judged less on what they add to an article and more on their technical requirements. I have no image restoration skills, nor do I grasp what little difference there is between one video format or another. From my perspective then, with this petition, we've reached a point where I can't really participate as a nominator here since I can not address issues with the images which more or less come up every time an image goes through here. For me and those like me it would likely be a smarter move at this point to seek out an image restoration specialist like Adam to run a nomination since one as skilled as Adam would be able to properly address the issues that seem to come up in nominations these days. Knowing when you can no longer contribute and respecting the need to bow out at that point helps keep the drama down on Wikipedia, which is ultimately a win for everyone on site, don't you think? :) TomStar81 (Talk) 23:52, 4 June 2016 (UTC)

Request for comment: proposal to increase the minimum dimensions for featured pictures and videos[edit]

WITHDRAWN:

After seeing the discussion below, I am withdrawing my proposal. --Pine 20:44, 17 July 2016 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Our current minimum for featured pictures is "a minimum of 1500 pixels in width and height" with some exceptions. Considering the continuing increases in the quality and size of digital photographs even from smartphone cameras, and the trend toward 4K UHD (3840 x 2160) displays, should we make the following changes:

  1. increase the minimum size for still images to "3840 pixels x 2160 pixels (4K UHD), with the height and width interchangeable", and
  2. change the exception of "Animations and video may be somewhat smaller." to "Animations and video must be a minimum of full HD (1920 x 1080), with the height and width interchangeable. Exceptions to the minimum may be made for historical, technically difficult or otherwise unique images, if no higher resolution could realistically be acquired."?

The full set of revised size criteria would read:

"It is of sufficiently high resolution to allow quality print reproduction.

  • Still images should be a minimum of 3840 pixels x 2160 pixels (4K UHD), with the height and width interchangeable; larger sizes are generally preferred. Further information on image size can be found here.
  • Note that vector graphics in SVG format can have a nominal size much smaller than this, as by their nature they can be infinitely scaled without loss of quality.
  • Exceptions to this rule may be made where justified on a case-by-case basis, such as for historical, technically difficult or otherwise unique images, if no higher resolution could realistically be acquired. This should be explained in the nomination so that it can be taken into consideration.
  • Animations and video must be a minimum of full HD (1920 x 1080), with the height and width interchangeable.
  • Exceptions to this rule may be made where justified on a case-by-case basis, such as for historical, technically difficult or otherwise unique images, if no higher resolution could realistically be acquired. This should be explained in the nomination so that it can be taken into consideration."

--Pine 04:29, 24 June 2016 (UTC)

Support[edit]

  1. Support as proposer. --Pine 04:29, 24 June 2016 (UTC)

Oppose[edit]

  1. Yes, it is getting easier to get higher resolution images. No, higher resolution does not mean higher quality. For certain subjects, even by focus stacking with my not-that-old Canon EOS 60D things will come up short. Then there's the fact that a lot of professional photographers are not willing to give higher resolution images of their work; I rather like seeing photographs by (for instance) Kyle Cassidy coming through here, and I'd hate to see these photographs fall by the wayside. Or such wonderful images as File:Florence Earle Coates Platinum Print 3 - Restoration.jpg — Chris Woodrich (talk) 12:39, 24 June 2016 (UTC)
  2. Too soon to do something like this. Doing so incurs unnecessary hardware cost on very good photographers and threatens to reduce the quality per pixel ratio. (Photographers take photos with more pixels that only add artefact at 100% zoom.) We should only enforce such a guideline when there is concern that there is insufficient pixels for high quality presentation. In addition, some cameras use different aspect ratios like 3264×2448 or 3264×2176. —Codename Lisa (talk) 11:55, 25 June 2016 (UTC)
  3. Leaning oppose. I have sympathy with Pine's argument, but I note that we can and do oppose images for being too small even when they meet the minimum requirements. I'm not sure we need hard-and-fast rules about image size at all, to be honest. Josh Milburn (talk) 12:29, 25 June 2016 (UTC)
  4. Oppose If nothing else, a slight cropping can often improve an image. Plus, a scan of a smaller image, like a historic postcard or carte-de-visite, at 300 dpi could easily fall short while being around the highest meaningful resolution. Adam Cuerden (talk) 09:08, 26 June 2016 (UTC)
  5. Oppose - While I too understand where the proposer is coming from, we are limiting our selection of images if we increase the minimum size for still images. Meatsgains (talk) 22:28, 7 July 2016 (UTC)
  6. Oppose - Pictures don't need to be very high resolution to look good on an online encyclopedia page; in many cases a more compact, faster loading image is desirable. When choosing images to feature, other considerations, such as beauty, creativity, impact, relevant subject matter, intended use, etc., should be more important; the resolution needed to present a superior image will vary depending on many factors and I see no need for a fixed minimum.—Anne Delong (talk) 22:11, 11 July 2016 (UTC)
  7. Oppose - resolution doesn't equal quality. The current resolution requirements are adequate to ensure a minimum standard of detail in the image, but for some subjects such as macro images, we are limited by factors other than technology, such as diffraction softening and diminishing depth of field. No matter how good the lens and sensor are, it just isn't always possible to squeeze more detail out of a photo. In those cases, using the highest resolution possible is just a waste of space and it's better to down-sample to something that is sharp at full resolution. Kaldari (talk) 02:57, 12 July 2016 (UTC)

Neutral[edit]

Discussion[edit]

  • As long as you're putting this forward why don't we incorporate the above proposal by Hahnchen (talk · contribs) to disallow .ogg files in favor of WebM? It does solve two problems at once. Also, I would propose a grandfather clause stipulating that any image promoted within the last year should be considered exempt from delisting under these guidelines until this time next year. Don't want to have a rush to jettison the undersized images that just got promoted, so this seems like a nice middle ground. TomStar81 (Talk) 08:35, 24 June 2016 (UTC)
  • Is there some other problem this is trying to fix besides just not keeping up with current trends? Is WP:FP getting overloaded with bad quality/low resolution images and videos? I'm not so sure higher resolution necessarily equates to higher quality, but I suppose it would depend on the content and how detailed it is. -- œ 05:42, 25 June 2016 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Wikipedia:Featured picture candidates/Elsie Leslie by Zaida Ben-Yusuf[edit]

Any chance of a few more eyes on this? It'll be the most annoying of near-misses if it doesn't pass (or at least get voted into a more ambiguous result), given it's just short of quorum because of a disregarded vote. =) Adam Cuerden (talk) 19:29, 13 July 2016 (UTC)

Yes check.svg Done...-The Herald (Benison)the joy of the LORDmy strength 01:57, 14 July 2016 (UTC)
Thanks! Adam Cuerden (talk) 02:39, 14 July 2016 (UTC)
Ah, no wait, ten minutes too late, so the vote got deleted and the nomination closed as a failure, there to remain until such time as it can be renominated - probably at least a month in future. I really do wish Armbrust wouldn't just rollback votes. At least cross them out as "past time" so that when they're renominated - and any nomination in that state should be renominated - it has a good justification for the renom. Adam Cuerden (talk) 04:03, 14 July 2016 (UTC)
Sorry, I didn't realize that you had one short a vote otherwise I'd have added mine. I hate it when that happens. TomStar81 (Talk) 05:13, 14 July 2016 (UTC)

Featured picture criteria[edit]

If a picture is not used in any article, then it can't be a featured picture? This rule is not good, if the picture is of best quality, is in public domain and identifies the subject properly. --Rainbow Archer (talk) 16:37, 21 July 2016 (UTC)

@Rainbow Archer: Well, remember you can add things to an article, and we also need to distinguish ourselves from Commons:Featured pictures. If an image won't "stay" in an article, it's usually not as useful as you think it is. Adam Cuerden (talk) 17:12, 21 July 2016 (UTC)

Last decision[edit]

Can anybody take final decision to promote a picture or only administrators decide after voting is over? --Rainbow Archer (talk) 03:17, 23 July 2016 (UTC)

Anyone can decide, but we usually leave it to Armbrust, who does a consistent, fair job. Adam Cuerden (talk) 21:06, 24 July 2016 (UTC)

Once again, we're stalling badly at three to four votes[edit]

Currently stalled candidates include:


Adam Cuerden (talk) 21:06, 24 July 2016 (UTC)

I think that users have to nominate beautiful, interesting pictures of different subjects, variety of topics as Animals, flowers, landscapes, insects, aquatic animals, orchids, fruits, nature ...... --Rainbow Archer (talk) 05:45, 25 July 2016 (UTC)
That basically ignores the actual skills of the person. Restoring a historic photo of a flower is probably not very useful, unless the flower no longer exists, after all. Adam Cuerden (talk) 11:13, 25 July 2016 (UTC)
And we seem to be at least mostly out of the issue again. Two images at four votes is a lot more reasonable. Adam Cuerden (talk) 11:12, 28 July 2016 (UTC)

Commons images[edit]

Can pictures uploaded in commons be nominated or only those pictures uploaded in English Wikipedia will be nominated. Another thing is that if a picture in featured picture in Wikimedia commons then is it considered featured picture here, or any separate nomination is done here? --Rainbow Archer (talk) 11:31, 28 July 2016 (UTC)

In FPC here you must have the picture in free copyright, i.e. CC-BY-SA or in public domain. Per Wikipedia file policy, those pictures are moved to Commons, which is a file repository by Wikimedia. Pictures featured in Commons are not considered featured in (English) Wikipedia, and hence they must be nominated for FPC separately. -The Herald (Benison)the joy of the LORDmy strength 11:47, 28 July 2016 (UTC)
@The Herald: That's not entirely accurate - there are a few cases where Wikipedia file policy either permits or requires the image to stay here, sometimes with a copy on Commons. {{KeepLocal}} and {{PD-US-1923-abroad}}, for instance. Adam Cuerden (talk) 11:54, 28 July 2016 (UTC)

[Edit Conflict:]

@Rainbow Archer: In order:

  • They can be uploaded to either here or Commons; uploading here, however, is generally for something like the Käsebier image, where it's out of copyright in America (which is sufficient under English Wikipedia rules) but not in its home country until next year, so can't yet be moved to Commons.
  • Commons' Featured picture process is its own thing, with its own rules, and any featured picture status there does not affect here, and vice versa. Commons generally ignores its value to the encyclopedia, English Wikipedia's featured picture process makes that arguably the most important thing. For example, see Commons:Featured picture candidates/File:Liège-Guillemins Station, Calatrava.jpg/2, which would never even have a chance here because it's been manipulated in a very misleading way, and Wikipedia:Featured picture candidates/Vivian Malone - an important historical image that would have no chance on Commons due to the compositional issues. Adam Cuerden (talk) 11:53, 28 July 2016 (UTC)

Hope that helps! Adam Cuerden (talk) 11:53, 28 July 2016 (UTC)