Wikipedia:Featured picture candidates/Lord Kitchener

Lord Kitchener edit

 
Original - British World War I recruitment poster featuring Herbert Kitchener, 1st Earl Kitchener, 1916.
Reason
Per the other day's nomination, this is a subject that does deserve a featured picture. Not quite the same iconic value as the original poster, but much higher technical specs and still pretty good encyclopedic value. Restored version of File:Lord Kitchener duty.jpg.
Articles this image appears in
Herbert_Kitchener,_1st_Earl_Kitchener#Death
Creator
Bassano, Bond St. W. ; Printed by Johnson, Riddle & Co., Ltd., London, S.E.
 
Alternate, color corrected from gray bar only, paper color persists at the level of decomposition that existed at the time when the color photo reproduction was made. Kodak's copyright notice on the color bar and grayscale is 1977. So this poster was already more than 60 years old when the reproduction was taken. (not for voting)
 
Edit 1 Some issues fixed
 
Edit 2 Completely reworked
  • Support as nominator --Durova331 23:55, 18 October 2009 (UTC)[reply]
  • Support Yes, I will do my duties :P ZooFari 01:34, 19 October 2009 (UTC)[reply]
  • Weak Support Looks a little weak in page exposure, despite EV. You all sure this only appears on one page? Get it on some more pages!   Nezzadar    03:57, 19 October 2009 (UTC)[reply]
Weak Oppose Hold the phones again... This has so very little exposure that I cannot support it. It's a nice image, sure, but it appears near the bottom of one artile, and nowhere else. Sorry. Message me on the golden phone when you think you've spread the poster around enough, I'd say you need at least one highly prominent placement, or a half dozen of these lowly ones.   Nezzadar    04:13, 19 October 2009 (UTC)[reply]
    • Prominent placement is not an FP requirement; encyclopedic relevance is. This image is relevant to the section about his death. Durova331 04:23, 19 October 2009 (UTC)[reply]
      • From FP Criteria "A picture's encyclopedic value is given priority over its artistic value." I see prominence as a component to encyclopedic value. It's pretty, but not prominent at all. Respectfully, My decision stands.   Nezzadar    04:38, 19 October 2009 (UTC)[reply]
        • Why is prominence more important than context for encyclopaedic value? The images are there to support the text after all. Noodle snacks (talk) 06:15, 19 October 2009 (UTC)[reply]
          • Perhaps Nezzadar wants FPs to be easier to find on a page. Trouble is, once the FP gets supersized everything else does too. Articles end up looking like this with 60K thumbnails fighting for the same attention as 39MB FPs. Have proposed an alternative at FPC talk to display a featured star in the caption space; that would signal readers more discreetly and effectively. Durova331 06:41, 19 October 2009 (UTC)[reply]
  • Oppose I think you over did the "Extensive brightness and color corrections" the original has color bars, so to color calibrate to the original wouldn't be very difficult. I loaded the original in Photoshop and sampled the black & white on the color calibration bars for a levels and it looks far more natural. Whatever color corrections you did went over-and-beyond restoring it to how the paper actually looks and has lost all color details in his face. — raeky (talk | edits) 06:48, 19 October 2009 (UTC)[reply]
    • Color bars in this sort of restoration are useless. Sometime after 1916 a curator photographed it using color film. That probably happened after 1940 (due to the technology) but the date was not recorded. Lighting was uneven and the document did not lie flat. Durova331 07:01, 19 October 2009 (UTC)[reply]
      • Sure they are, provided the original color bars was accurate black & white and 50% gray, you can produce a very accurate color calibration from it regardless of lighting or film quality. Thats what they're for. — raeky (talk | edits) 08:43, 19 October 2009 (UTC)[reply]
        • I linked a version that is just color corrected from the gray bar, should be pretty accurate to how the original looked when it was photographed. Paper is of course discolored from age, but I think there is more detail in the face. Your version is of course probably closer to how it originally looked when it was produced with white paper. — raeky (talk | edits) 09:12, 19 October 2009 (UTC)[reply]
          • The bottom half is more deficient in blue than the top half, with substantial localized problems such as the yellow band at far left, because chromolithography of this age often loses color in uneven ways. It needs a perspective crop and then half the caption has to be separately rotated because it doesn't lie flat. It needs substantial brightness gradient masking because the edges had dried out and darkened. If it were the goal of restoration to recapture the decomposition as it existed at some unknown moment possibly in the third quarter of the twentieth century then yes those particular color bars would be useful, but the poster already had serious decomposition by that point. Those color bars could have been set there in 1966, but for all we know those markers were placed in 1946 or 1986. What it doesn't tell you is how this looked in 1916. Durova331 14:49, 19 October 2009 (UTC)[reply]
            • Kodak's copyright imprint on the gray scale and color control patches is 1977. Durova331 15:12, 19 October 2009 (UTC)[reply]
              • Like I said your version probably is more accurate to how it looked in 1916, I just did mine to see what detail was lost if any in areas like the face since whiting the paper beyond how it was when the picture was taken likely will loose detail. I also didn't put nearly as much restoration work in it as you did to correct some weird issues you listed, since my goal wasn't to create something to be voted on (thus the not for voting tag). — raeky (talk | edits) 15:51, 19 October 2009 (UTC)[reply]
              • Neutral After comparing yours to my version my original reservation that detail was lost in the face doesn't appear to be true. I think it's POSSIBLE the colors are wrong now on your version due to the heavy tweaking to get the paper looking whiter, but detail wasn't noticeably lost. Changing to neutral. — raeky (talk | edits) 15:59, 19 October 2009 (UTC)[reply]
  • Oppose It's actually highly unlikely white paper was used: You don't use white paper with white ink, and the colourboxed version pretty clearly shows white ink was used. There were a limited number of cheap inks, so they'd often use some form of unbleached or tinted paper and white ink to get an extra colour. Shoemaker's Holiday Over 213 FCs served 12:46, 19 October 2009 (UTC)[reply]
    • Actually it appears that white ink was not used: there was a mild printer error in blue and red across the flag; the flag's white was the same color as the paper. Durova331 15:19, 19 October 2009 (UTC)[reply]
      • It might have been grey paper, like on the Ulysses S. Grant one, but I find the contrast a bit high to believe no white ink was used. Shoemaker's Holiday Over 213 FCs served 16:00, 19 October 2009 (UTC)[reply]
        • Look in close for yourself. What they did was add a small amount of blue ink to suggest shadowing, but the "white" itself is negative space. Durova331 16:10, 19 October 2009 (UTC)[reply]
          • Hmm. Poking at it, it does appear you're right, or at least could be, given rather severe vignetting. The image looks underexposed, though, so I'm going to upload an edit. Shoemaker's Holiday Over 213 FCs served 16:15, 19 October 2009 (UTC)[reply]
  • Support edit1, weak oppose original Shoemaker's Holiday Over 213 FCs served 16:28, 19 October 2009 (UTC)[reply]
  • Comment I like edit1. Why isn't this at Military recruitment? Staxringold talkcontribs 14:40, 19 October 2009 (UTC)[reply]
  • Support Shoemaker's edit. Good job. :) Durova331 16:56, 19 October 2009 (UTC)[reply]
  • Oppose edit 1 Colors are better but it's blown out some of the lighter areas loosing finer details. — raeky (talk | edits) 16:59, 19 October 2009 (UTC)[reply]
    • I used a curve, so nothing's blown. Some parts may be a little more subtle (mainly in thumbnail), but I don't think anything that represents artistic intent is hidden. Shoemaker's Holiday Over 213 FCs served 17:58, 19 October 2009 (UTC)[reply]
      • If you look closely (zoomed) on the face of your version from other two versions the finer detail in the face is missing and has gone to all white. — raeky (talk | edits) 18:11, 19 October 2009 (UTC)[reply]
        • I agree; compare for instance the wrinkles on his brow. TotientDragooned (talk) 15:35, 20 October 2009 (UTC)[reply]
          • Well, I can still see them, but I'll do a second edit with the face masked, so I can treat it a little differently. I'll just upload over mine. Give me a couple hours. Shoemaker's Holiday Over 213 FCs served 12:39, 21 October 2009 (UTC)[reply]
  • Question. Since there appears to be no reliable source for the original ink and paper colors or the effects of aging on the different combinations of inks and paper, how does "restoration" beyond removal of obvious digitization artefacts constitute anything other than original research? 82.251.140.156 (talk) 09:42, 21 October 2009 (UTC)[reply]
  • Support edit 1 edit 2 Excellent image with lots of EV Nick-D (talk) 10:35, 21 October 2009 (UTC)[reply]
  • Edit 2 uploaded, shift support to Edit 2. I've dealt with the face issues. Shoemaker's Holiday Over 213 FCs served 13:24, 21 October 2009 (UTC)[reply]
    • For the record, the upload glitched the first time. Got it now. Shoemaker's Holiday Over 213 FCs served 15:51, 21 October 2009 (UTC)[reply]
  • Support Edit 2 I think Edit 2 resolves my concerns, still think the colors a little funny in the face (yellows with the whites) but it's likely the original colors. — raeky (talk | edits) 16:53, 21 October 2009 (UTC)[reply]
  • Support edit 2. TotientDragooned (talk) 05:00, 23 October 2009 (UTC)[reply]
  • Support GerardM (talk) 07:37, 23 October 2009 (UTC)[reply]
  • SupportJake Wartenberg 03:27, 24 October 2009 (UTC)[reply]

Promoted File:Lord_Kitchener_duty5.jpg --Makeemlighter (talk) 00:22, 27 October 2009 (UTC)[reply]

Post Nomination Update edit

It has been determined that this is Frederick Roberts, 1st Earl Roberts, not Lord Kitchener, on December 2nd, 2009. Nezzadar [SPEAK] 05:21, 3 November 2009 (UTC)[reply]