Talk:History of Japan/Archive 10

Latest comment: 8 years ago by CurtisNaito in topic The article's sourcing is a mess
Archive 5 Archive 8 Archive 9 Archive 10 Archive 11 Archive 12 Archive 15

The article's sourcing is a mess

Each sentence should have a citation to a specific page or small range of pages in a specific source that specifically backs up what is said. I've been working on what I could just now, but it's hard work.i love penelopy


For example, the text During the Shōwa period Japan became the world's largest manufacturer of automobiles and a leading producer of electronics. is currently cited to sixteen pages of Henshall and six pages of Weston. The Henshall citation is attached to an abundance of other text, all of which I'm sure is related to something somewhere in those sixteen pages, but the Weston citation is only attached to this one sentence. This makes verification very difficult. The most egregious example is "Henshall, 112–138", a 27-page range cited twelve times in the article.

I will be working on this myself in the foreseeable future, but some help would be most appreciated.

While sloppy pagination is not explicitly a violation of WP:V (although changing "over 100,000 dead" to "roughly 100,000 dead" is), it goes against the spirit of V to be this obscure with where information was taken from.

Hijiri 88 (やや) 13:31, 28 October 2015 (UTC)

Obviously the best thing would be to just rewrite the section with a range of reliable sources. There's no deadline etc zzz (talk) 13:45, 28 October 2015 (UTC)
Or we could follow Sturmgewehr's advice to use these overview sources to give us an idea of what should be there, and then use more specialized sources to base our actual wording on and to verify what the general sources say. Curly Turkey 🍁 ¡gobble! 23:15, 28 October 2015 (UTC)
Thing is, I'm not a topic expert on the early Showa period, and I have no reason to believe Henshall doesn't verify 90% of that material -- I'd rather just read Henshall and cite him in a more verifiable format. As an aside, Henshall's coverage of Japanese classical literature is (naturally) very superficial, and virtually everything is cited to Keene and McCullough (translations from Keene's Anthology are cited six times -- it doesn't seem like any non-translated edition of a classical work was consulted); as a tertiary source with nothing of real value to add to our article when it comes to this topic, Henshall should never have been cited for any of this. (Ironically, though, CurtisNaito would have done much better to cite Henshall 36 than "Weston, 135–139" for the information I asked him to add on Yoshitsune.) Hijiri 88 (やや) 14:17, 28 October 2015 (UTC)
I think that we can restore the part on lifetime employment. Henshall argues that it was exaggerated, but the fact that it existed in some areas until the Heisei period is clear enough. Henshall says that corporate workers were given a "promise of security" in their jobs, and the promise in question was "lifetime employment". In general I cited Henshall in blocks based on the period in order to avoid citation clutter.CurtisNaito (talk) 14:41, 28 October 2015 (UTC)
Henshall argues that it was a popular misconception originating in the postwar period and ending in the 1990s. That clearly is not the same thing. The exact quote, in context, reads: Workers, at least white-collar males in major companies, were further mollified by the promise of security in the form of ‘lifetime employment’. This has been greatly exaggerated in terms of both its scale and its history. Though selective scanning of earlier history can reveal some antecedents in the Meiji period and even the Tokugawa period, it is essentially a postwar practice. Moreover, it has only ever applied at most to a quarter of the workforce. Curtis, your clipping this quote down to workers were given a "promise of security" is a gross misrepresentation of the source, and you need to stop saying otherwise. And please don't "avoid citation clutter" in the future; "citation clutter" is significantly more convenient for readers and other editors than what you have done here. Hijiri 88 (やや) 14:52, 28 October 2015 (UTC)
Henshall doesn't say it was merely a popular misconception. He says it applied to a quarter of the workforce (a very substantial institution in other words) and then collapsed in the 1990s. We should restore the information on lifetime employment as it was written. I guess you've never been through a good article review before, because in my experience it's not uncommon for the reviewer to request that multiple citations to similar pages from the same book be consolidated. Someone who wants to verify the citations only needs to read a chapter or a section of a chapter to do so.CurtisNaito (talk) 14:56, 28 October 2015 (UTC)
Strictly because I do not see anyone directly pointing this out, "at most to a quarter" is substantially different from "a quarter" (assuming that the italicized section above is a quote). That's me done. Evangeliman (talk) 20:16, 30 October 2015 (UTC)
@Evangeliman: I know. If I was writing this article from scratch based on what the source said, I would probably take that into account. But unfortunately the nominator makes it a consistent habit of distorting sources' wording in this manner ("over 100,000 dead" becomes "roughly 100,000 dead", and so on). Hijiri 88 (やや) 00:55, 31 October 2015 (UTC)
No, I never distorted any wording from the source. The figure of one quarter was never in the article, though Henshall does indicate that a quarter of the workforce were covered by the lifetime employment system at one time. The point Henshall makes is that lifetime employment did not cover the whole workforce, nor did it exist prior to the postwar period. However, what I put into the article was only the basic facts Henshall gave. I didn't bother delving into the areas which Henshall labeled as misconceptions.CurtisNaito (talk) 01:00, 31 October 2015 (UTC)
Curtis, using your own faulty interpretation of the source's wording to argue on the talk page that your interpretation of the source was correct and that your distorted wording should be restored to the article is the same as misquoting the source in the article text. And what about changing "over 100,000 dead" to "roughly 100,000 dead"? The fact that you wrote an entire article about death toll estimates makes this very concerning -- how many of that article's uses of the word "roughly" meant "over" and "over" meant "roughly"? You say above "I never distorted any wording from the source" (my emphasis) but you very clearly changed "over 100,000" to "roughly 100,000". Stop it now. Hijiri 88 (やや) 01:10, 31 October 2015 (UTC)
100,000 is only a rough figure for the death toll. Henshall evidently rounded his estimate to the nearest round number. At the time when I consulted with the Wikipedia article on the 1923 earthquake, it gave a death toll of 105,385, so I didn't think there was anything wrong with using the word "roughly". It's fine to say over 100,000, or else to say 105,385. However, you can't say that I misrepresented the source which only gave a rounded estimate of the total, as one would expect from a summary history.CurtisNaito (talk) 01:16, 31 October 2015 (UTC)
100,000 is only a rough figure for the death toll No, it is not a "rough figure" - it is a minimum estimate. When sources give rough figures for the death toll of that earthquake they usually say "roughly 140,000". Admittedly this does include 行方不明者, but the standard practice is to include them, since why else would they go missing? You would know this if you actually read our article on the disaster rather than glancing at its poorly formatted infobox. But you shouldn't be consulting Wikipedia to begin with -- you should just be taking your reliable sources at their word and saying what they do, not trying to synthesize what they say with what a Wikipedia article quotes two employees of Kajima Corporation as saying over and above what has been historically given as the death toll estimate. Your source said "over 100,000", which does not contradict this historical death toll estimate, and you should not have distorted it so that it does contradict the historical estimate so that it matches up with a source that says something different. Hijiri 88 (やや) 01:33, 31 October 2015 (UTC)
100,000 was a rounded estimate. We shouldn't belabor this point though. The previous version was fine, but obviously I'm not opposed to using a direct quote from the original book. If we have no further disagreement over the actual text of the article, we can move on to other issues. As I said before, it's best to keep talk page discussion limited to the text of the article. Bringing up death toll estimates of the Nanking Massacre, for instance, is irrelevant. If the issues relating to article content are dealt with, then we can move on.CurtisNaito (talk) 01:40, 31 October 2015 (UTC)
Curtis, we are going to continue disagreeing over every tiny point unless you show a little more humility. When you are wrong you are wrong, and should admit it and apologize. This is how I work, this is how Nishidani, Sturmgewehr88, Curly Turkey and everyone else who abides by Wikipedia's conduct policies work, but it is not what you have been doing. Every time someone finds something wrong with something you wrote, the result is an extended discussion involving you claiming either that you were not wrong in the first place, or that you "clearly meant" what your wording was changed to rather than what you actually wrote. This is not how talk page discussion should work, and it is extremely frustrating for everyone who has to work with you on this article. It is why Dennis Brown explicitly said that if you ever did this again you would be blocked for a minimum of 72 hours, and it is why Prhartcom made a compromise proposal that said that you would never do this again. You said that you accepted this proposal, but you immediately violated it, and virtually every edit you have made since then has violated it. Hijiri 88 (やや) 02:05, 31 October 2015 (UTC)
Prhartcom sais "Drop all fierce objections. Change the tone." As I said, the best way to move forward is to keep discussion limited to article content. I have never asked you to apologize for the many errors and mistakes you have made which I pointed out here and elsewhere. The reason why I never asked you to apologize for your errors is because it has no direct bearing on article content. If we disagree over article content, a discussion can take place, but there's no reason why you should apologize over issues which are not actually directly related to the text of the article itself. Look at the lengthy post you wrote above. It includes not one word about the content of the article itself. My idea is that from now on we should only discuss article content.CurtisNaito (talk) 02:12, 31 October 2015 (UTC)
So you only recognize the part of the agreement that is binding on your opponents, and refuse to accept or even acknowledge the part of the agreement that says you will never engage in IDHT behaviour again. That's nice -- so the agreement is moot, as you have refused to accept. So why should anyone else abide by it if you are not willing to? And what "fierce objections" have I made? What "tone" needs to be changed here? I have been limiting my discussions to article content -- you are refusing to just let it go when you lave lost on a content issue like how many were killed in the earthquake. Furthermore, you promised above that you would start improving the sourcing by gradually removing the citations of huge chunks of pages. You have not been doing this: this string of edits added another citation to one of your earlier large-range references ("kofun") and consolidating a reference to the end of page 181 and the beginning of page 182, and another reference to the end of page 182 and the top of page 183, into "181-183". This is making the problem worse, not better. Please actually keep your promises. Stop engaging in IDHT behaviour, and start doing the work you promised to do (or, rather, helping do the work, as so far others have done far more than you). Hijiri 88 (やや) 02:22, 31 October 2015 (UTC)
I removed the good article nomination and have listened to all concerns expressed on this talk page. Therefore, I have fully abided by the agreement. As long as we keep discussion limited to article content, I won't ask you to apologize for your IDHT behavior or anything else. In my recent edits, I reduced one of the Henshall citations to only one page in the Kofun period section. In the Heian period, I didn't think a range of only three pages was a big deal, but if you think three pages is so many, I will reduce it. However, your citation to Keene, "Keene 1999 : 33, 65, 67–69, 74, 89", included 7 pages, and no one objected to that. I don't think you can say my edits have made the article worse, when your edits to the article have used a much longer page range than any of my recent edits.CurtisNaito (talk) 02:32, 31 October 2015 (UTC)
No, you removed the GA nomination but you have spent all the rest of your efforts undermining attempts to improve the article by posting an endless string of IDHT remarks claiming that your version more accurately reflected the sources than what it was replaced with, despite everyone disagreeing with you. As for your reference to my Keene citation, it is obvious that you have checked neither the source nor even the article text to which it is attached. The article lists four major poets who are referred to as the representative poets of their particular eras, and one more who is described as the last waka poet of note; the ref is formatted to meet the the WP:V criterion that all material be explicitly supported by the reference, and so five separate page ranges (all brief) for the four poets are cited. To compare this to your citing a chapter-length discussion of an entire era and adding it to every statement in the article related to that era, and doing this for every section in the article, is both absurd and deeply hypocritical. Hijiri 88 (やや) 04:11, 31 October 2015 (UTC)
No, that's not true. If a range of seven pages is acceptable for poets, there is no reason why three pages aren't acceptable for a single topic relating to the economic difficulties of the Heian period. Again, you're just casting aspersions instead of focusing on article content, though you have been told by many other users that doing this is not useful for improving the article. As I said, from now on we should stick to article content.CurtisNaito (talk) 04:18, 31 October 2015 (UTC)

First off, I apologize for my misremembering. The Keene ref you referred to had nothing to do with Kamakura poetry, but rather was for the literature of the Nara period. You or someone else appears to have now removed the citation in question and replaced it with a hodgepodge of unreliable, non-specialist sources like Rhee, Henshall and Totman. Please explain this. If you do not, I will reinsert the far-better citation of Keene. Furthermore, your analogy is deeply flawed, as what you did and are still defending is not equivalent to citing "Keene 1999 : 33, 65, 67–69, 74, 89" but more like citing the same pages in the format "Keene 1999 : 33, 65-89". When CT, Prhartcom and I asked you to fix the messy refs in the article, we meant the ones you had added to sources you at least claim to have read. We did not mean that we wanted you to track down any kinda-sorta-maybe similar refs added by other users to books you haven't read and remove those. Hijiri 88 (やや) 04:39, 31 October 2015 (UTC)Edit: Sorry. Ctrl+Fed "nihon" and misread the results. Curtis, when you make a mistake or say something wrong, maybe you too should own up and admit it so we can all move on? How about we try that? Hijiri 88 (やや) 04:47, 31 October 2015 (UTC)

I was told to fix the citations in general, not specifically the ones I added. You were criticizing me for including a page range of three pages in one of my citations, and so I fixed it. However telling me that I was making the article worse was clearly wrong, otherwise you were making the article disastrously worse by citing "Keene 1999 : 320–324, 650–651, 674, 676, 680–681, 700–705, 735–736", which goes through nineteen pages. As I said, stick to article content and constructive criticism. I don't think that you should criticize me so strongly for citing sources in the article in the exact same way you have been citing sources.CurtisNaito (talk) 04:49, 31 October 2015 (UTC)
Again, you're in a glass house regarding "Keene 1999 : 320–324, 650–651, 674, 676, 680–681, 700–705, 735–736", because it may be nineteen pages but it's the smallest range possible to support the material to which it is attached. If it was formatted like the refs you added in August it would look like "Keene 1999 : 320–324, 650–736" -- that's ninety-two pages, seventy-three of which are unrelated to what our article says.
Admittedly, at the time I was trying to mirror your one or two refs per paragraph style, so I consolidated it into one ref where my normal preferred style would be to have one or two facts per sentence and one ref per sentence. I will break it down a bit better later.
But this doesn't change the fact that, rather than admitting that one of the article's main problems is a result of your obscure and barely verifiable sourcing style, you are grabbing at straws trying to place equal blame for this problem on me.
Hijiri 88 (やや) 05:37, 31 October 2015 (UTC)
I used the sourcing style which was recommended to me before I wrote the article. Naturally, I use a different style depending on what is asked of me. However, none of the sources I have cited are obscure. All of them are high quality, reliable histories.CurtisNaito (talk) 05:42, 31 October 2015 (UTC)
Curtis, please learn to read English. When I said "your obscure and barely verifiable sourcing style" I was not making a general comment about the quality of the sources (though none of them are as good for classical literature as Keene). I meant what I said: your sourcing style is "obscure" ("hard to make out or define; vague") and so is barely verifiable. And no one requested you to do that here: if you had evidence of such you would have presented it already. Hijiri 88 (やや) 06:24, 31 October 2015 (UTC)
Well as I said, your sourcing was the exact same as mine until recently. However, I am now changing my sourcing style to a different format. I already quoted above the fact that I was told by a good article reviewer that the original citation style I was using was the appropriate format for good level articles. I was using the citation style recommended to me by users more experienced than either of us.CurtisNaito (talk) 06:29, 31 October 2015 (UTC)
No, even when I was trying to make my refs look like yours I still only listed the pages that either verified the article material or provided relevant information closely related to the article material (my ref not only verified that Shunzei was the representative poet of his age but also mentioned that he was commissioned to compile an imperial anthology and that he was the most eminent such compiler in three centuries). Several of your "exact same" refs consist of random chunks of twenty or more pages attached to about a dozen simple factual statements that may or may not be verifiable somewhere in those twenty pages. Further: Which user told you that "good level articles" have some kind of uniform citation style? If that is what they said they are wrong, whether or not they are "more experienced" than me. And I seem to recall the Iwane Matsui article being promoted in spite of your refusing to even name the authors of your sources and needlessly listing the names of the same books dozens of times in the references. Wikipedia does not have a standard citation style: each article uses a style best-suited to it, with more verifiable styles generally being preferred. Hijiri 88 (やや) 08:21, 31 October 2015 (UTC)
Yes, citation style can vary from article to article. The Iwane Matsui article was promoted because that particular citation style was fine for the purposes of that article. The original citation style I used in this article was also not wrong. It was recommended to me as a fitting style to use for good level articles, and it was fairly easy to verify the citations since they all referred back to same chapter dealing with the same topic. It was no different at all from citing nineteen pages of Keene as you did.CurtisNaito (talk) 08:29, 31 October 2015 (UTC)
Curtis, both our article and Henshall's book are narrated in roughly chronological order, so your citation style of citing essentially an entire chapter of Henshall for each section of our article is about as useless for V purposes as not giving page numbers at all. The only reason Sturmvogel and Calvin didn't tell you the same thing is that neither of them tried to check any of your sources (hence why this article was delisted and Iwane Matsui probably should be delisted) and so likely didn't notice the problem. You need to tell us exactly what page you got your information so the rest of us can verify that no more misrepresentation of sources has happened than that which has already been rooted out. Expecting us to read all of the books from start to finish is asking too much. Hijiri 88 (やや) 09:09, 31 October 2015 (UTC)
You haven't found any instances of misrepresentation yet. The reason why this article was delisted had nothing to do with sourcing. Prhartcom clearly told you that the sourcing issue had already been resolved before the review was over. Both the articles you mentioned were thoroughly spotchecked. It's strange that you would accuse me of IDHT behavior, even though you constantly repeat yourself over this issue, even though you have been informed of the situation many times. At any rate, I am changing the citation style to a different format, though you can hardly criticize me for previously using the same citation style that you also used.CurtisNaito (talk) 09:15, 31 October 2015 (UTC)
Just to briefly chime in (and then leave again): I agree completely with this observation. It was difficult to verify the sources for this reason stated. A typical footnote of this article inexplicably points to multiple pages of text, occasionally more than one book even, and often shared with other footnotes. This should not be the case. If a user wishes to learn more, they should be able to click on the footnote and get the exact page number of the very book that contains the cited information. The footnote should usually not be shared. Anyone doing a source review or wishing to learn more should not be presented with what the article is currently presenting. For good examples of how to do it, see any article I have written or any article Curly Turkey has written. Prhartcom (talk) 18:10, 28 October 2015 (UTC)
Alright I'll gradually work towards splitting the citations. Incidentally, this is contrary to what I've been repeatedly told during good reviews up to now. During one recent good article review I was told "You've got cites splattered all over the article that can be profitably consolidated, so go through the article thoroughly and get rid of them." However, for the purposes of this article, I will de-consolidate the citations.CurtisNaito (talk) 18:50, 28 October 2015 (UTC)
For verifiability purposes, the above quote is from Sturmvogel 66, on May 31 of this year, on the Iwane Matsui GA review.[1] It appears to refer only to cases where "every fact in a paragraph is derived from the same source", and obviously could not when multiple sources are used for a single paragraph. Hijiri 88 (やや) 08:36, 31 October 2015 (UTC)
As I quoted above, he told me to consolidate my citations to eliminate clutter and told me that by doing so I could avoid needing to cite every sentence and possibly even avoid citing every paragraph. It's common, however, for other good article reviewers to say the same sorts of things. This was the style I used when consolidating citations to Henshall, so that several sentences of text were all verifiable to the same section of the same source. I usually write up articles in the way which is mostly likely to please a future good article reviewer, though it's not always easy to tell because different people prefer different citation styles.CurtisNaito (talk) 08:46, 31 October 2015 (UTC)
Just to ensure I am not leading you in the wrong direction (or to ensure that GA reviewer does not), let's check WP:CITE. Prhartcom (talk) 20:55, 28 October 2015 (UTC)
Curtis, while I don't doubt what you are saying is true, that was one reviewer's opinion. Calvin999 expressed the opinion in his review of this very page that the citations should perhaps be attached to each sentence. That was his only reference to the sourcing in his very short and superficial review, but he did essentially agree with me on this point; he just didn't think it was the article-breaking issue that I do. Hijiri 88 (やや) 21:47, 28 October 2015 (UTC)
Calvin999 had told me in a previous review that he wanted citations attached to every sentence, but I was subsequently told in multiple good and featured article reviews by other users to not do that any more. When Calvin reviewed this article, he clarified that one citation per sentence was actually just his personal preference, not a rule. Therefore, I wrote this article according to Wikipedia guidelines.CurtisNaito (talk) 21:52, 28 October 2015 (UTC)
I'd like to see what those reviewers actually said. I've been through nearly fifty GA reviews and have never been told to do something so ridiculous. Curly Turkey 🍁 ¡gobble! 23:11, 28 October 2015 (UTC)
Calvin999 informed me that putting a citation at the end of every sentence was his personal preference, stating "Personally, I cite every sentence. It makes it easier to verify information. You don't have to cite every sentence". As opposed to Calvin's personal preference, most reviewers have a strong opinion that, "You can read the policy for yourself at WP:CITE; nowhere does it state that every sentence needs to be cited and I've even had objections from some other reviewers that once per paragraph isn't strictly required either." I myself once believed that a citation at the end of every sentence was generally preferable, but by now I've been informed by a good number of reviewers that, according to policy, even one per paragraph is not required. Since then, I have made great effort to consolidate my citations into as few as possible at the end of paragraphs.CurtisNaito (talk) 01:29, 29 October 2015 (UTC)
You've had some very bad advice. Of course every paragraph needs references - that is fundamental. For example, the folks at ITN won't link any article to the main page otherwise. Beyond that, common sense should be followed. If every sentence in a paragraph is going to require 2 or 3 (or more) cites, some repeated in alternate following sentences, then just dump them at the end of the paragraph. That's not likely to be the case in this article, obviously. zzz (talk) 01:50, 29 October 2015 (UTC)
At the very least, you're misunderstanding the advice. Every word in the article needs to be cited, but not every sentence requires an inline citation. If two or more sentences in a row in a paragraph are taken from the same source (the same page or short range of pages), then you are only required to put the inline citation after the last of those sentences. There is no rule against doing it for every single single, though, and many editors prefer to do that for a variety of reasons, but unless you have a good reason it's best to avoid it, as it makes for aesthetically ugly clutter. Globbing everything together at the end of the paragraph covering a huge page range is a terrible thing to do (though not technically against any rule). Curly Turkey 🍁 ¡gobble! 02:23, 29 October 2015 (UTC)

The problem still isn't solved

The huge, almost useless chunks of literature being cited for each bit of our text appear to have been mostly removed, but the citation style still isn't ideal. In my opinion, each sentence of the body should have a citation, unless that sentence is a caption on a picture (The reputed site of Narihira's residence, near Karasuma Oike Station) or a summary of the sourced material either immediately before or immediately after (The location of Narihira's grave is uncertain). This is just my opinion, and I'm sure others disagree and think that if an entire paragraph's text is directly based on a single source that source can be cited once at the end of the paragraph. But right now a large portion of this article consists of paragraphs where the first several sentences are cited to a single citation, and the next several are cited to another. This makes it difficult to know whether a datum is derived from the source immediately before it or the source cited several lines down. Can I assume that in all cases the information is taken from the "next" source cited? I guess I can fix it myself with a copy-paste job if I can assume this, but I really don't like doing that... Hijiri 88 (やや) 12:53, 5 November 2015 (UTC)

Yes, each citation applies to the information and sentences preceding it. As I noted, the idea that every sentence should be cited is not accepted by most Wikipedia users. Rather, a citation is expected to apply to all the text preceding the citation.CurtisNaito (talk) 10:10, 6 November 2015 (UTC)