OpposeAbstain from voting after more thorough explanation has been given by the nominator. -- Slaunger (talk) 20:22, 5 August 2008 (UTC) I think this is very difficult image to review. This image is itself a reproduction from around 1930, long after the creator died. So one could argue that is it actually a fake, others say that special care has been taken in this reproduction to achieve colors closer to the original. There is another version, Image:Tsunami by hokusai 19th century.jpg, which is from the Metropolitan museum. However, on the image talk page it is claimed that the metro version is also a fake presumably made in China. The candidate here gave rise to quite some discussions at COM:FPC,where the origin of this image was discussed. The nomination process ended in a decline, although it seems for me that the issue was never really resolved. Since there is still doubt about how well this image represents the authors original, I cannot support it, although I personally think it is the most striking version we have on Commons. If better evidence is presented on the image page that this image is indeed a good representation of the original (and better than the Metro version) I will reconsider. -- Slaunger (talk) 20:28, 31 July 2008 (UTC)[reply]
Comment I selected this version because, based on the evidence that I saw, it is the most well-preserved and complete of the prints. As far as the claim that this is a fake goes, I poked around the net in both English and Japanese and was unable to find anywhere repeating this claim. As you may know, Japanese kanji underwent a bit of a transformation after World War II involving the replacement of old characters, or kyūjitai, with the modern characters used today. Thus 應 became 応, 寛い became 広い, 學 became 学, and 樂 became 楽. While I haven't been able to nail the character in question down as the definitive ancestor of 裏, there's no doubt in my mind that the extra stroke is simply because Hokusai used an old form of the character. Besides, accepting that this image is a fake means accepting that the Metropolitan Museum of Art does not perform even cursory verification of its collection.
As far as the colors go, the image is a Nishiki-e (錦絵; see [1] for verification), which means that the colors were ostensibly linked with the individual woodblocks. While I have no conclusive proof that this is the case, it seems much more likely to me that the difference in color between the new copy and the old copy is due to advances in coloration and preservation making possible Hokusai's original vision for the piece than it is that the difference is because the artisans copying the piece in the 1930s ignored Hokusai's woodblocks or inserted their own interpretations. This is the reason why I picked this image over the other as the most valuable in the 36 Views scope. --jonny-mt16:33, 5 August 2008 (UTC)[reply]
Thank you for your thorough explanation. I am inclined to support, but the subject matter is so much away from my domain knowledge that I will rather abstain from voting that risking to have made an erroneous judgement based on - ignorance;-) However, I think it would be a pity if this was not settled, so I will contact a few users, who I think will have better competences in performing the review. -- Slaunger (talk) 20:22, 5 August 2008 (UTC)[reply]
Support However, allow me to clarify what's going on. This is an exact copy of the original, restored by modern use of the same exact processes and at that made the original run. This sort of exact reproduction was a skill that was kept up in Japan for many years after Hokusai's death, and is similar to remaking the die for a coin when the first die wears out. For all intents and purposes, this may be considered an extra-labourious restoration. Adam Cuerden (talk) 21:26, 5 August 2008 (UTC)[reply]
Conditional Oppose I've been asked to give an opinion on this image. Unfortunately there's a problem here and I'd really like to see a copy of the original scan. Someone has gone into the upper portion of this image with a crude attempt at restoration. It's particularly visible above the crest of the large wave. I'm not certain which program was used here, but it has the look of a large pixel selection with the healing brush in Photoshop. And what it's done is obliterated the texture of the background surface. If someone's going to restore this image let's do it properly, and rework that area (or preferably start from a new copy of the original file) plus address the dirt specks in other parts of the image and remedy the crease to the right of Mount Fuji. If an unaltered version were also uploaded and linked from the image page, plus a description of the alterations, then I'd make a conditional support. But I really can't endorse this kind of practice. Durova (talk) 21:37, 5 August 2008 (UTC)[reply]
Thanks for the comments--I believe the image on Commons was sorted from here (direct link to full image). From what I can see, though, there's no difference between this scan and the uploaded image, and a checksum gives the same values for both. --jonny-mt15:45, 6 August 2008 (UTC)[reply]
And that was an attempt at cleanup of this. As I suspected, there had been stains where the healing brush was used. The changes are quite extensive--note the tones in the clouds and the crest of the wave. I'd support the original or a less intrusive restoration. Durova (talk) 09:36, 7 August 2008 (UTC)[reply]
I'll admit I hadn't seen that other version until you pointed it out (incidentally, it's interesting how the website gives the same summary for this picture as it did for the image above but still categorizes it as an "old print" as opposed to the "new fascimilies" designation it gave to the other one), but to my beginner's eye the changes look too significant to be accounted to digital cleanup, particularly given the extreme differences in resolutions. For example, the sea spray starts higher on the image being discussed while the white area in the sky to the right of the wave crest extends much further out. At any rate, I fully respect your opinion and am thankful for your insights (I'm in full agreement with Slaunger about this being a great discussion), but I thought it might deserve mentioning. --jonny-mt12:45, 9 August 2008 (UTC)[reply]
Details of ineffective attempts at restoration in nominated version.Alternate without restoration.
I've uploaded two images. First shows the area that caught my eye in the nominated version. The first area circled has rainbow banding. That's usually a result of scanner glass that's wet from rubbing alcohol. The second circled area shows where the attempt at restoration obliterated a region of paper texture. Both of these problems would be easy for me to fix, but intrusive alterations elsewhere and mild .jpg artifacting make it not worth the effort. Basically these are red flags that the original nomination was not worked on with skill and care. I support the alternate version, though. Durova (talk) 07:18, 10 August 2008 (UTC)[reply]
Question Have you got any idea why the alternative edit you are proposing has a lower resolution that the candidate? The candidate does not seem to have been upsampled? -- Slaunger (talk) 09:17, 10 August 2008 (UTC)[reply]
Different scans, probably. And to be clear, to the best of my knowledge the version I uploaded is unedited, as an alternative to a bad edit. Durova (talk) 03:12, 12 August 2008 (UTC)[reply]
Oppose OK. I am convinced, and I agree with you that an unedited version is better than than a bad edit (although I think it is pretty good). I have made a nomination of the alternative and moved the two candidates to the most valued review section: Commons:Valued image candidates/Great Wave unrestored.jpg
SupportConvinced by Durova that the unrestored original is more valued - which is the third and last time I change my mind during this review... -- Slaunger (talk) 20:56, 12 August 2008 (UTC) I find it satisfying to participate in this review! This is what VI is about. Thorough investigations and insightful reviews (except from me, lol)! Anyway, thanks to Adam for the coin-die analogy. that makes sense to me, and I can also see the rather odd looking areas at the crest of the large wave Durova is mentioning, and that this looks suspicious. So, I acknowledge that it may be possible to obtain a better representation of this image, and it sure would be nice if we got an even better one. However, if we step back for a while and cross-check with the criteria we should keep in mind that we should assess the image in review size on the review page, and in that resolution I cannot see the odd looking patches visible in full resolution. Next, a VI does not have to be the best possible illustration of the scope, but the most valued available on Commons. And for me, with the additional comments provided by Adam and Durova it seems clear that it is the best we have. I have therefore again changed my mind from oppose to abstain to support... -- Slaunger (talk) 20:52, 6 August 2008 (UTC)[reply]
Oppose - As one of Commons' restorers, I, like Durova, am made very uncomfortable by the idea of promoting a bad restoration as, not only the best copy of itself, but the best copy of any image in a set of 36. Adam Cuerden (talk) 03:19, 12 August 2008 (UTC)[reply]
