Jump to content

Wikipedia talk:Romanization of Russian

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by SilkTork (talk | contribs) at 16:21, 4 September 2011 (Closing straw poll). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

WikiProject iconRussia: Language & literature Project‑class
WikiProject iconThis page is within the scope of WikiProject Russia, a WikiProject dedicated to coverage of Russia on Wikipedia.
To participate: Feel free to edit the article attached to this page, join up at the project page, or contribute to the project discussion.RussiaWikipedia:WikiProject RussiaTemplate:WikiProject RussiaRussia articles
ProjectThis page does not require a rating on Wikipedia's content assessment scale.
Taskforce icon
This page is supported by the language and literature of Russia task force.
Shortcut

Does the wording of this guideline actually reflect consensus?

I'm closing this RfC, because there appears to be consensus that the romanization guidelines need to be reworded / changed, which is the only question this RfC was intended to answer. Hopefully the handful of editors who are engaged in the discussions continue to be make progress towards improved guidelines. Mlm42 (talk) 16:10, 24 June 2011 (UTC)[reply]

The guideline for place names currently reads:

A conventional name of a place is the name listed in major English dictionaries and should be preferred over default romanization at all times. In particular, if major English dictionaries do not list the place, then default romanization should be used.

Some editors have been enforcing this guideline very strictly. Even if there is evidence to demonstrate that a place has a common English name, this policy insists that if the place is not listed in a "major English dictionary" then the default romanization must be used. So I ask: do other editors support or oppose the current guideline for place names? Mlm42 (talk) 17:34, 9 June 2011 (UTC)[reply]

Responses

  • Oppose. The guideline essentially redefines the term "conventional name of a place" to something which is not consistent with WP:COMMONNAME. Only a small number of places are actually listed in major English dictionaries. Mlm42 (talk) 17:34, 9 June 2011 (UTC)[reply]
  • Support — V = IR (Talk • Contribs) 17:47, 9 June 2011 (UTC)[reply]
  • Oppose, per nominator. As near as I can tell, this is the only romanisation guideline on wikipedia that limits placenames to major English dictionaries ignoring reliable sources from other Geographical and Language organisations. This should either be corrected or the whole guideline downgraded to an essay on the subject. Stuart.Jamieson (talk) 17:57, 9 June 2011 (UTC)[reply]
  • Oppose per nom; the present text is absolutely unacceptable. As a relevant example, Orel is much more common in English than Oryol; whether it is quite common enough to squeeze into the few pages an English dictionary can spare for a world-wide gazeteer is another question entirely, which should not decide our article title. Septentrionalis PMAnderson 18:17, 9 June 2011 (UTC)[reply]
    Britannica is hardly a gazeteer, and "quality encyclopedias" is explicitly in the WP:COMMONNAME's list of sources one is supposed to use to establish the common name. As I mentioned below, it's unfortunate that the wording of WP:RUS ended up with "dictionaries" being listed as the only acceptable source. The original intent was to include all major reference works.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 19:13 (UTC)
    That that version of Britannica chooses to use pidgin English, as here is not our fault; it is past time for NCGN reconsider our recommended references, since Encarta is no longer supported. But this text does not even reach to the Britannica, as you yourself admit. Septentrionalis PMAnderson 22:24, 9 June 2011 (UTC)[reply]
    Britannica is using "pidgin English" now? Really? Based just on your observations, no doubt? And, with Oryol/Orel, unrefined google hits is your best evidence? Did you know that the word "oryol" means "eagle", and is bound to pop up in all kinds of contexts which have nothing to do with the city? For that matter, how about finding an example that's not based on the unfortunate fact that the Russian letter "ё" is mostly optional in common nouns (and tends to be omitted even in the proper nouns), which would skew the romanized results even more? That same things also goes for Korolyov, by the way.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 13:48 (UTC)
    Yes, unfortunately Britannica has steadily conceded to the demands of non-anglophone nationalists like those which inflict themselves on Wikipedia. We have the power to ban them and should do so. Septentrionalis PMAnderson 16:43, 10 June 2011 (UTC)[reply]
    Well, that's one opinionated (and unsubstantiated) remark, if I've ever seen one! By all means do everything you can to convince others that Britannica is now a private Nazi nationalistic joint, but please don't act on this opinion unless you can demonstrate that the community agrees with this assessment of yours. Filing an RfC on WP:COMMONNAME would be a great first step. For now, however, let's work with what WP:COMMONNAME currently says, not with what you want it to say.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 17:12 (UTC)
  • As for the usage of oryol as a transliteration of eagle: That is unlikely to be common enough to make a difference; if anything, it would provide false positives for oryol; so removing it would make "Orel" yet more common. For one thing, Google ngrams are case-sensitive. Septentrionalis PMAnderson 16:51, 10 June 2011 (UTC)[reply]
    And you think it "unlikely" based on what, if I may be so curious (sorry, it must be maddening that I keep requesting evidence to back up the sweeping claims you keep making)? As it happens, "Oryol"/"Orel" is also a last name, and the false positives are just as likely for "oryol" as they are for "orel".—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 17:12 (UTC)
  • Support. For those who don't know, when WP:RUS was up for adoption, it was universally supported, mostly by people whose editing was to be affected by it, and who knew firsthand the idiosyncrasies of the Russian romanization (which are many). With all due respect, I'm not seeing this kind of people among those who commented so far. When it comes to romanization, "common usage" cannot possibly be determined by google tests or even by studying the usage by "Geographical and Language organisations". Since there are quite a few systems of romanization of Russian, such tests produce the results which are pretty much random. It is that randomness the wording of WP:RUS is supposed to eliminate—we aren't serving our readers well if we have to make them guess at which spelling any given article is supposed to be found! If a place name is not found in the dictionaries, it is romanized using one of the available romanization systems. That is precisely what the "Geographical and Language organisations" do—WP:RUS borrows the practice, not the end result. If "dictionaries" is seen as too restrictive, let's replace it with "reference works" (which would cover the encyclopedias, maps, and such)—that was the original intent anyway.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 18:34 (UTC)
I believe Ezhiki is referring to this discussion in 2007. But people who broadly supported the guideline raised specific concerns, including the "Tolyatti" example. Mlm42 (talk) 18:42, 9 June 2011 (UTC)[reply]
Yes, that's the one I was referring to. Note, however, that the specific concerns weren't enough for those people to oppose the whole guideline. I very much doubt one can devise a guideline everyone will be happy with—in the current edition of WP:RUS I myself don't like a few points which were results of earlier discussions and compromises.
I do oppose much of the guideline; I am genuinely shocked that it has been permitted to stand in this condition. Most of it is contrary to usage and to policy.Septentrionalis PMAnderson 23:03, 9 June 2011 (UTC)[reply]
Also, could you please comment on my proposed re-wording? Thanks.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 19:08 (UTC)

Discussion

In addition to my support in the section above, I don't quite agree with the premise of this RfC. WP:RUS does on surface seem to override the more general common names provisions, but if you look at it closer, it only filters out the randomness the use of multiple romanization systems introduces. Looking up a place name in a dictionary (or a similar reference work) unambiguously establishes a "common English name" when one is found; when the entry is not in a dictionary, all other cases (the usage of major international organizations, major English-language media outlets, geographic name servers, major scientific bodies and scientific journals) result in a romanized name produced using the romanization system the "major international organizations", "English-language media outlets", etc. happens to standardize on. Case in point: if a certain place name is most often romanized using the ALA-LC system (perhaps because that place has a rich history, and historians tend to use ALA-LC, thus skewing the usage patterns into its favor), and then suddenly there is a disaster there which media outlets start to cover (using, as they usually do, the BGN/PCGN system), the balance would suddenly and very visibly change. Does it mean the "common English name" has just changed as suddenly? Not really. Common names don't change abruptly, so the logical conclusion is that we weren't dealing with the "common name" to begin with. What we were dealing with, of course, is the result of the application of one (random) romanization system to the original Cyrillic. Not really the same, is it? WP:UE, by the way, recognizes this caveat by referring editors directly to the language-specific romanization guidelines when a place name is not originally in the Latin alphabet.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 19:06 (UTC)

Regarding Does it mean the "common English name" has just changed as suddenly?, I'm sorry but yes, it does. Common names really do change abruptly all the time. I understand how that can be frustrating, but that's why we have a UCN policy. UCN refers editors to the language-specific guidelines only as a last resort, when there simply isn't enough English language coverage for a common name to be established. You're meeting resistance here because you're trying to reinvent (or at least reinterpret) fairly well established policy.
— V = IR (Talk • Contribs) 20:33, 9 June 2011 (UTC)[reply]
One possible rewording is:

If the name has a common English-language form, then it should be used (per WP:UE). Otherwise, the default romanization, as defined below, should be used.

The rest of the guideline contains some points on how to decide between multiple English-language forms. I think it's best to leave "common English-language form" up to a certain amount of interpretation, to avoid instruction creep. Mlm42 (talk) 19:13, 9 June 2011 (UTC)[reply]
But WP:UE explicitly refers to the romanization guidelines when the original name is not in the Latin alphabet:

Names not originally in a Latin alphabet, such as Greek, Chinese, or Russian names, must be transliterated. Established systematic transliterations, such as Hanyu Pinyin, are preferred. However, if there is a common English-language form of the name, then use it, even if it is unsystematic (as with Tchaikovsky and Chiang Kai-shek). For a list of transliteration conventions by language, see Wikipedia:Romanization.

The new wording would simply create a circular reference and does not address the problem of prevailing romanization variants creating an illusion of having a "common name" where none in fact exists. Having a clarification to that effect is absolutely essential, or we might as well not have a guideline at all. Do you have a reason to think that generic "reference works" (instead of "dictionaries") would not leave room for a certain amount of interpretation?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 19:28 (UTC)
It still seems like needless instruction creep. It's conceivable that the common English usage doesn't agree with every "reference work". What's wrong with just saying "common English usage"? Mlm42 (talk) 19:35, 9 June 2011 (UTC)[reply]
Anyway, my understanding is that the main purpose of this guideline is to spell out the default romanization rules. Mlm42 (talk) 19:37, 9 June 2011 (UTC)[reply]
Just saying "common English usage" is impractical, because we are simply referring the editors to a more generic guideline. The very reason of having a more specific guideline is to provide more specific guidance for a narrowly defined problem set (which in this case is Russian names). Vaguely telling people to use the "common English name" and "romanization" is precisely what we had before WP:RUS had been adopted. In practice, it was a living maintenance hell—articles were being moved left and right and left again, people were inventing their own romanization systems and moved articles in bulk according to their visions, and readers were left wondering just where in hell they are supposed to find the articles they need, and why some places which shared same name in Russian were under different titles in the English Wikipedia (and that one was mostly because they were romanized using different systems).
What WP:RUS does is clarify the "common name" provision with the Russian romanization-specific problems in mind. It very simple to follow in practice (look the name up and if it's not there, use the default romanization; and if the reference works do not agree, use the one that matches the default romanization or is the closest to it) and it just plain works (and even if in 0.01% of cases it doesn't, it's nothing that a well-designed redirects/disambiguation net can't take care of). You are basically proposing to dismantle a working system just to accommodate a handful of odd cases (some of which aren't even that odd).—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 20:03 (UTC)
I understand that for you, someone who oversees thousands of these articles, it makes life easier. But that doesn't mean it's what Wikipedia's wider community wants. Instruction creep is bad. Mlm42 (talk) 20:15, 9 June 2011 (UTC)[reply]
So, are you saying that the instruction creep is so bad that the usability for the readers and "easier life" for editors who can contribute more content in a "convenient" environment should be sacrificed? I would argue that's a small price to pay for reducing the maintenance overhead and making the locations of the articles being sought more predictable! It's not that much a creep even; just a sentence, and one that addresses an actual problem at that. I, for one, (and I assume you, too) would rather contribute content than ward off the folks determined to move hundreds articles to the "official Russian government" spelling (one which most Anglophones have never seen and won't ever see again), because, see, it's "an established transliteration system", as per WP:UE, and we have no other guidelines explaining why it's not the right approach. If you think I'm exaggerating, I am not. This was a common problem before 2007.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 20:49 (UTC)
Official Russian government spelling is not necessarily "common English usage". Common English usage in reliable sources is what it is, I don't see a reason to define it further. Your main point seems to be that it's easier to use default romanization instead of common English usage (unless dictionaries say otherwise), and therefore that's what we should do.. I'm not buying this argument. The "easier" solution is not always the right one. Mlm42 (talk) 02:49, 10 June 2011 (UTC)[reply]
I'd support rewording the text to User:Mlm42's proposal above. The "major English dictionary" qualifier that is currently being used is certainly odd. It looks like User:Ezhiki's attempt to standardize all article names in this area (or at least the vast majority of them) to a narrow system, which is the kind of thing that regularly meets with fairly significant resistance in Wikipedia. I'm surprised that this hasn't come up sooner than now, honestly.
— V = IR (Talk • Contribs) 20:27, 9 June 2011 (UTC)[reply]
It hasn't come before now because the system works fine as is (the approach may seem odd to a person unfamiliar with the idiosyncrasies of romanization, but it is a standard practice followed by the geographic organizations and publications, which is why we have adopted it as well). Is there a reason why you think that replacing overly restrictive "dictionaries" with "reference materials" wouldn't work? The guidelines, after all, are supposed to be useful in practice; it's no help to either readers or editors if we have to follow the guidelines which are overly vague and do not address obvious problems.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 20:49 (UTC)
Other policy simply states "reliable sources". Why should Russian articles be so different from the rest of Wikipedia?
— V = IR (Talk • Contribs) 21:07, 9 June 2011 (UTC)[reply]
It's not very different, really. As I've already explained above, since multiple romanization systems of Russian exist, querying for the most common spelling would produce a pretty much random result for any given place; one that's not even necessarily stable in time. A "common English name" should not be dependent on the random choice of a romanization system by various sources. All those variants are technically correct, of course, but just because one is more common than another doesn't automatically make it the "common English name"; not in the sense WP:COMMONNAME establishes. This is not a problem most place names have, only those which are not originally written in Latin script. WP:RUS addresses this very specific problem by narrowing down the sources to those which can unambiguously be used to establish a common English name—with the general reference materials being the prime choices. All other sources would either use the spellings provided by those dictionaries and encyclopedias, or adhere to one of the existing romanization systems. Hence, we do the same—we either take the variant that can unambiguously be shown to satisfy the commonality test, or we use an established romanization system instead. This is too specific of a problem to be addressed in a general overview guideline such as WP:COMMONNAME. Does this help?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 9, 2011; 21:21 (UTC)
Ezhiki, you seem to be too focused on the technicalities of Romanisation. Doing so is like comparing the reasons for other English common Names; we do not for instance worry about sources that refer to Bill Clinton as "William Clinton" or "Willy Clinton" or "Will J Clinton" although they are technically correct alternative article titles we don't worry about the technical reasons for why each of these variations exist or which is the most technically accurate (in this case "William Jefferson Clinton" is technically most accurate), we choose the one that appears most often in reliable English sources hence "Bill". Stuart.Jamieson (talk) 22:03, 9 June 2011 (UTC)[reply]
The whole point I'm trying to make is that it's not just a technicality; it's a valid concern which is specific to romanization and is not normally a concern in other cases. The Clinton example has nothing to do with what I'm trying to explain here; it's not even a close analogy.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 13:41 (UTC)
No, it's your concern which is not a concern of the encyclopaedia - and if you were right and it's a valid concern specific to romanisation we should see it in all of Wikipedia's other Romanisation guidelines:
  1. Armenian - "However, for transcriptions of proper names, apart from other Armenian text, it is Wikipedia's general convention to follow English usage, where it exists; this may frequently mean using -ian instead of the systematic -yan as a name ending." - Nope doesn't seem concerned enough to specifically check dictionaries.
  2. Cyrillic (including Belarusian, Bulgarian, Macedonian, Mongolian, Montenegrin, Russian, Serbian, Ukrainian.) - "If a name or word has a conventional English spelling, that is used...", "When something has a conventional name in English, use that name instead of transliterating. Conventionally-used names may stem from various sources:[...]They may be anglicized versions[...]They may be transliterated by a different system, or for another language[...]They may be simplified, more familiar-looking, or easier to pronounce for English-speakers[...]They may be names borrowed into Russian from a third language" - no mention of conventional English names being found in dictionaries there either
  3. Chinese - "In general, the titles of Chinese entries should be in Hanyu Pinyin (but without tone marks). Exceptions would include: When there is a more popularly used form in English (such as Yangtze River)" - nope no establishment of dictionaries or any other specific works to decide whether the English name is popularly used.
  4. Greek - " If there is a common anglicization of a Greek proper name, it should be used in an English language context." - I see a trend forming here.
  5. Hebrew - "Some topics may have several common widely-accepted English transliterations (e.g. the name Chaim vs. Haim vs. Hayim), and sometimes it is debated whether there is a standard English transliteration at all (Beersheba vs. Be'er Sheva). In these cases, the context of the article should be taken into account." - This is the closest to what you suggest for Russian, but still defers to contextual English sources rather a formulaic approach.
  6. Mongolian - "When something has a conventional name in English, use that name instead of transliterating."
  7. Ukranian - "It is subordinate to the naming conventions...", "Keep the readers in mind: they read English, but might not be familiar with Ukrainian. Ukrainian words should be used for a reason, not as a substitute for English.[...]An object that has a conventional name in English should be named that way, instead of transliterating"
All of these defer upwards to Wikipedia's English methods of determining a Common or conventional name; WP:RUS stands alone in deciding that we should determine that common name through a very narrow set of parameters. Despite your claims to dislike the current wording, you proposed it and whilst it was accepted by consensus I'm not sure that any of those supporting consensus had the foresight at the time to see what effect it would have on all the articles you have renamed or enforced this guideline on. It is time the shortcomings are addressed. Stuart.Jamieson (talk) 17:43, 10 June 2011 (UTC)[reply]
Thanks for looking these up, but the reasons behind why none of those guidelines defines "common usage" are rather simple. Some of the WikiProjects for the benefit of which these guidelines have been developed are very small and the activity there is low (such is the case with Armenia, Belarus, and Mongolia). There are more pressing concerns they need to address, and having a romanization guideline at all is pretty much just a stopgap measure (you may have noticed how poorly some of these guidelines are written). Once the activity there picks up, they are going to run into the same kind of issues we have with WP:COMMONNAME being too vague and not addressing important problems. Articles of WikiProject Ukraine are a good approximation of the horrors awaiting WP:RUSSIA—the articles on that country's administrative divisions are a complete mess because they all use different names on different occasions, so finding anything is very hard to the point of impossible.
Other languages, like Chinese, Greek, and perhaps Hebrew, don't have the same problems Russian has—for those languages either exists one romanization system which is used in the vast majority of cases (Chinese Pinyin is a good example), or, when there is more than one system, their use is clearly segregated by knowledge area. One can actually trust the search results under those conditions—your hits would include English name and romanization, and deciding which one is "more common" is a snap. For Russian, we have a dozen of different system, most of which are intended to cover everything and the kitchen sink (with scientific transliteration, which is aimed primarily at the linguists, being possibly the only exception). So, you are not choosing between a possible "common name" and romanization; you are choosing between a possible "common name" and half a dozen romanization variants, any of which can "win out" in the end (and be declared "common English name" as a result).
All in all, all those guidelines do not clarify the definition of "common name" not because there is no need, but because there is not enough sufficient activity to stop and think whether this is necessary, or because the generic guidelines work for them just fine. Heck, some WikiProjects (such as Kazakhstan or Kyrgyzstan) don't have any romanization guidelines at all (which, I should note, doesn't make the editors' jobs easy or the end result predictable to the readers). What I'm trying to demonstrate here is that there are problems which are specific not just to romanization in general, but to romanization of Russian in particular, which the generic guidelines do not address well.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 18:45 (UTC)
Ezhiki, I'm still not understanding your basic argument. Your main concern seems to be in the case when there are multiple different spellings used in English-language sources, and no clear "winner". In those cases, I think some of the points in the guideline are good for choosing between the various English-language forms used.
The situation that I think the other editors are trying to debate is when there is a clear choice for a common English name among English language reliable sources. In this case, we should use that common English name, and not give unnecessary extra weight to dictionaries and other "reference works". Could you please make your case more clearly, if you are refuting this point? Mlm42 (talk) 18:55, 10 June 2011 (UTC)[reply]
Actually, for any given place in Russia, there will always be multiple different spellings used in English-language sources (and that includes even such places as Moscow and St. Petersburg!), which is why that's the cause I'm arguing. The universe of those spellings will include the genuine common English names (which, when they exist, can easily be looked up in any major reference works), and a great number of different romanization variants. Now, the definition of romanization is the representation of a written word or spoken speech with the Roman (Latin) alphabet, or a system for doing so, where the original word or language uses a different writing system (or none). The purpose of romanization, as per, for example, this, is establishing standardized Roman-script spellings of those foreign geographical names that are written in non-Roman scripts or in Roman alphabets that contain special letters. The definition makes it pretty obvious that romanization is supposed to be employed to establish a standardized spelling for cases where no such standardized spelling exists, which, in turn, means that no "common English name" exists in the first place. However, if you apply the generic provisions of WP:COMMONNAME, there is no way to make a distinction between what is genuinely a common English name, and what is a "standardized spelling" developed for a particular environment or context, which is not necessarily compatible with the environment and context (or even purpose) of Wikipedia.
All in all, in general when a name is included in a major reference work of some sort, we can be pretty confident that's our "conventional name" (unless, of course, like PMAnderson you believe that so many of the reference works out there are conceding to nationalists' demands, are written in "pidgin English", and are often "illiterate"; let's not go there—WP:COMMONNAME makes it pretty clear that reference works are an acceptable and vital source). When a name is not included in a major reference work, WP:COMMONNAME (and, for place names, WP:NCGN) provides us with other ways to go about the problem, such as gbooks and gscholar analysis (not hit count!), analysis of academic literature in the [knowledge] area in question (i.e., not just any books which happen to mention the name in passing), or analysis of media usage. That works very well when you have to pick between two or several names each of which can honestly contend for the "conventional name" title, but it doesn't work at all when you have to pick between one (or, possibly, zero—there's no way to know beforehand) truly "conventional name" and a bunch of "standardized spellings" used for purposes, most of which are unlikely to meet the organizational needs of an encyclopedia. The results will be truly random, and, as a consequence, truly random will be the locations of the affected articles in Wikipedia. That doesn't help our readers any, don't you think?
Was I able to clarify my position for you?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 19:33 (UTC)
P.S. The most succinct way of putting it I can think of is this. WP:COMMONNAME exists to help us choose the most common name. WP:RUS exists to help us choose the most common spelling of a name. In other words, WP:COMMONNAME's intent is to help resolve situations similar to Gdańsk/Danzig or Nagorno-Karabakh/Dağlıq; WP:RUS' intent is to help us pick the spelling once we figure out what name to use. Not many places in Russia have alternative names to choose from, but all have alternative spellings to choose from.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 20:01 (UTC)
No, I think we're still talking about different things. You are drawing a confusing distinction between the words "name" and "spelling", that I don't fully understand. You said there will always be different English spellings, and appear to be inferring that therefore, we won't be able to choose a common English name without resorting to dictionaries or reference works? Consider the following scenarios: Say usage is split among English reliable sources 40% and 60% between two spellings. Then that's not really enough to declare a common name. On the other hand, if the difference were 10% and 90% between two spellings, then that's a significant difference. As I understand it, the site-wide policy states we should go with the 90% variant, as it is the common name (even if we can't find a "reference work" that agrees). You appear to be claiming that, even in this 10%-90% situation, we should follow the reference works, and disregard the site-wide policy. And you still haven't made a convincing argument as to why. Mlm42 (talk) 20:14, 10 June 2011 (UTC)[reply]
Sorry, I should have been more clear. Let me illustrate. The "name" problem would, for example, be deciding whether to place an article about the town under Baltiysk (Russian) or "Pillau" (German)—the place used to be a part of Germany, see. WP:COMMONNAME takes care of that pretty unambiguously, so we have the articles on all places in Kaliningrad Oblast under their Russian-based names. The "spelling" problem, on the other hand, would be deciding how to spell "Baltiysk"; i.e., to help us choose between "Baltijsk", "Baltiĭsk", "Baltiisk", and who knows what else. Now, some of these spellings are pretty esoteric; however, dictionaries and other reference works tend to stick with more common ones, with general audience in mind, not with the specialists. Since the purpose of the dictionaries and the reference works is very much similar to the purposes of Wikipedia, and since the general reference works is the kind of source the general audience is most exposed to,it makes sense to borrow from there. Other sources (such as those found via gbooks/gscholar/media analysis) would use the spellings which are appropriate in a context which is not necessarily amenable to the purposes of building an encyclopedia; and, as I previously explained, on top of that there is a randomness factor which can affect the end result in unpredictable ways. Is this explanation better?
As for the 10:90 cases you are thinking, I'll bet the shirt I'm wearing that 99.9% of those are caused by use/non-use of the letter "ё" (I recommend you read the article; it's quite interesting even if you aren't into linguistics). I most certainly can't think of a ё-less example that would possible have 90% of sources spelling the name one way with the majority of reference works spelling it differently (and I have a database to query stuff like this up). That, however, is a problem that's worth being addressed separately on its own merits. If we get rid of the requirement to romanize "ё" as "yo" and simplify it as "e", that'll most certainly make the editors' (and my) job easier, even though the quality and accuracy would be somewhat sacrificed.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 20:49 (UTC)
Okay, but the point is that in other areas of Wikipedia there is the same problem, and one could argue that when there is a dispute over a name, that more weight should be given to things like dictionaries and reference works. Indeed this has been argued in the past, and has been rejected by the wider community. This is why Septentrionalis is up in arms about it, because it appears you are blantantly flouting the established consensus. On the other hand, ultimately, the common English name among reliable sources (if it exists) is almost always going to be the common English name among reference works (if it exists), so I don't understand why you are digging your heals in here.
As for the "yo" thing, Septentrionalis brought this up below, and it might be the source of most of the problems. I've suggested a solution in that section. Mlm42 (talk) 18:51, 11 June 2011 (UTC)[reply]
By the way, I don't think policies such as Wikipedia:Naming conventions (geographic names) make a distinction between "name" and "spelling", in the same way you are. Mlm42 (talk) 19:04, 11 June 2011 (UTC)[reply]
That's probably a failure of copyediting. We have been attempting to make clear that our article titles don't habe to be names since WP:Article titles was moved, more than a year ago. Septentrionalis PMAnderson 20:50, 12 June 2011 (UTC)[reply]

I was part of the consensus that wrote the sentence Ezhiki quotes. It does not mean what he would like it to mean; it was never intended to do so; it merely provides what we do when, as often, the rule of following what English does gives no clear guidance.

That is, of course, more than simply counting google hits, although that is part of it; for more, see WP:NCGN#Widely accepted name. I am changing my !vote and removing the disputed text; it is already plainly not consensus. Septentrionalis PMAnderson 22:14, 9 June 2011 (UTC)[reply]

How about at least giving a semblance of courtesy to the opposing party and letting the discussion run its course before making any changes? Why such hastiness?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 13:41 (UTC)
For two reasons: There is no consensus on the sections that invoke dictionaries - there would be none if every good soul who !voted in 2007 were to be canvassed and appear; where there is no consensus we should be silent. We may well be able to attain consensus on somewhat modified texts; but retaining an alleged "consensus" because a minority (or even a non-consensus majority) still supports it is the demand of our more unscrupulous bullies; I would deeply regret seeing Ezhiki join them.
I am not interested in hearing from those who would propound "It's in Russia; always adopt the Russian spelling." Wikipedia has opposed such nationalism everywhere else: to my knowledge, in Greece, Turkey, Poland, Germany, and Iran. No reputable editor would suggest we do so for Russia. Septentrionalis PMAnderson 16:17, 10 June 2011 (UTC)[reply]
You still haven't answered my question about why all these changes have to be done in such haste. The very reason this RfC has been filed is to gauge the consensus and to determine whether it has changed (which it very well may have). RfCs are recommended to be kept open for thirty days so all interested parties have a chance to comment; and you are ready to call it a done deal the day after it has been open? What's going on?
On your second remark, I have no idea where you got the impression that the guidelines advocate to "always adopt the Russian spelling". In fact, even if you read WP:RUS in its present form, it's very hard to miss all those "use English" pleas, and an editor is actually required to jump through all sorts of hoops to check usage before s/he even gets to the "default romanization table", which is supposed to take care only of the cases for which English usage can't unambiguously be determined or simply does not exist. It seems you are fighting a problem which you yourself have created, and to call the intentions of the people who !voted in 2007 as "nationalism" is just a low thing to do. What ever happened to assuming good faith? Why in the world would you think that the only reason WP:RUS is worded the way it is is to uphold someone's nationalistic attitudes?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 10, 2011; 16:27 (UTC)
Is this supposed to be good faith? "Names of places located in Russia must be romanized from Russian" is "never use English" unless it happens to be the romanization of Russian (or perhaps unless it falls into some other exception). But this never allows for the possibility that English has adopted the German or Polish or Yakut spelling. Septentrionalis PMAnderson 01:39, 11 June 2011 (UTC)[reply]
I agree the wording of this point isn't good, but I think it could be reworded to be helpful. I think the point is that some places in Russia may have local languages that aren't Russian, so one might be tempted to romanize the name from the local language instead of from Russian. So this point is saying that if we are going to romanize the name (i.e. it doesn't have a common English name), then we should romanize from Russian and not from some other language. Makes sense? Mlm42 (talk) 16:53, 11 June 2011 (UTC)[reply]
You have the nub of sense that was behind this; but that's not what the text says.
Even so, this should be a rule of thumb, rather than never or always. Most names will be romanized from Russian, because anglophones encountered them through Russian; but suppose that most names in the Outer Foolander Autonomous Republic represent the Fooland variant; shouldn't the stubs on the small villages do so too? Septentrionalis PMAnderson 18:02, 11 June 2011 (UTC)[reply]

Problems with default romanization

The table reads: –ий endings -> -iy or -y. Why is there an OR-directive? Random choice of romanization? Bogdan Nagachop (talk) 13:19, 14 June 2011 (UTC)[reply]

I had also wondered about this.. perhaps we should just choose one? Mlm42 (talk) 16:33, 14 June 2011 (UTC)[reply]
I think it's the same as yo or e above: English usage is -y, but -iy is the "scientific" transcription. Thus we should say
Vissarion Grigoryevich Belinsky (Russian: Виссарио́н Григо́рьевич Бели́нский);

and we should add Belinskiy, although some non-anglophone pedant has changed the rest of the article to Belinskii. Fortunately Joseph Brodsky is read enough in English to escape this nonsense. Septentrionalis PMAnderson 17:02, 14 June 2011 (UTC)[reply]

It's an OR-directive because that's what some people felt strongly about before the 2007 proposal had been initiated, so it got added to the proposal ("-iy" is strict BGN/PCGN, while "-y" is a simplified approach). In reality, most of our articles use the "-y" ending, which, as PMA correctly noted, tends to be more common in the English language publications despite technically not being as accurate. I'd be all for sacking the "-iy" part and leaving just "-y". It's what we mostly do anyway.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 14, 2011; 18:01 (UTC)
As Bogdan points out, it's better to just choose one.. so maybe we should stick with the more common -y ending. Mlm42 (talk) 18:37, 14 June 2011 (UTC)[reply]
Done. Changes of wording are acceptable. Should yo be treated in the same manner? Septentrionalis PMAnderson 21:25, 14 June 2011 (UTC)[reply]
I have finished, I think, copyediting to make this consistent. Again, why not replace yo with e in romanizations? Septentrionalis PMAnderson 23:43, 15 June 2011 (UTC)[reply]

Problems with I and Y - Text vs Table vs Talk

I reduced the table to show only the differences from BGN/PCGN. By doing so I found some inconsistency. Text says

  • -ый endings become -y;
  • -ий endings in adjectives [bold by me] of Russian origin become -y.

Table does not say about adjectives.

Furthermore Talk:BGN/PCGN romanization of Russian#Simplified system refers to proper nouns. So what is wanted? Bogdan Nagachop (talk) 09:32, 19 June 2011 (UTC)[reply]

I would simply have -y as the default, as in Belinsky and Brodsky. It may be that being a loanword is a reason not to use the default. Septentrionalis PMAnderson 22:25, 20 June 2011 (UTC)[reply]
Belinsky and Brodsky aren't exactly nouns, as Russian last names have an adjective form, so using "-y" is warranted. Endings of nouns are usually left as "-iy" (not that there would be terribly many of such nouns; "гений"/geniy is one example), and names of non-Russian origin usually retain "-iy" as well (example). The last one is important.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 21, 2011; 13:01 (UTC)
That would be a reason to depart from any default; a stronger reason is that the place is called Ryrkaypiy in English, not Ryrkaypy. How is it spelled in Chukchi? Septentrionalis PMAnderson 21:15, 22 June 2011 (UTC)[reply]
I assume you mean the names of non-Russian origin. There are, as far as I can tell, only about a couple dozen such places (out of ~150,000 of all places in Russia), and neither is high-profile enough to have a common English name; they are simply romanized names. Ryrkaypiy is probably the most notable of them, which says something about the obscurity of the rest. There is also another dozen or so places which are nouns of Russian origin. The only nouns I can confidently say I've often seen romanized via "-y" and not "-iy" are the male names (such as Vasily or Dmitry); all other nouns, and especially toponyms which are nouns, tend to use "-iy" almost exclusively. How to describe all this in a guideline without too much creep and without sacrificing the accuracy I am not exactly sure. Considering that only a handful of obscure places is going to be affected, perhaps this doesn't deserve a mention at all.
As for Chukchi, I don't know how it's spelled. What does it matter?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 22, 2011; 21:46 (UTC)
It matters because it is entirely possible that the place is known in English in a spelling direct from the Chukchi, more than through Russian; but some Chukchi activist will have to pursue that, not me. But I agree that the proper course is not to mention so rare a case in mentioning what is default. Septentrionalis PMAnderson 22:15, 22 June 2011 (UTC)[reply]
Hm, I'd say that chances of a place name being know in English primarily through Chukchi (or any other indigenous language of Russia) are just as slim as the chances of a place name being a noun ending in "-ий".—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 23, 2011; 14:31 (UTC)

Problems with hard sign - Text vs Table

Text says : omitted, Table says : omitted when followed by a vowel. Bogdan Nagachop (talk) 09:39, 19 June 2011 (UTC)[reply]

Default redirect

Instruction creep

  • Places: The variant produced using the default romanization must be a redirect to the main article.
  • Persons: The default romanization variant must be a redirect to the main article.

Please delete or change. It can also be a DAB. No specific guideline needed, general rules have to be applied. Bogdan Nagachop (talk) 02:02, 16 June 2011 (UTC)[reply]

Thank you; that problem had not occurred to me. Is there argument against? Septentrionalis PMAnderson 21:40, 16 June 2011 (UTC)[reply]

Reducing section Use conventional names

Could this section please be reduced and point to general guidelines? Then we are almost only left with actual Romanization of Russian and it that area except for the yo there seems to be wide consensus. So it should be easy to re-establish the page as guideline. Bogdan Nagachop (talk) 21:37, 16 June 2011 (UTC)[reply]

That would be acceptable. We can then discuess yo. Septentrionalis PMAnderson 21:41, 16 June 2011 (UTC)[reply]

Also the numbered rules are for Places listed as "Clarification" and for People they are not, with People having extra "Clarification" attached to it. I joined the rules that apply to People and Places into one section, leaving the examples, so there is always one for People and one for Places. I think the main problem left is the WP:UE part. But I think Ezhiki already agreed that "dictionary" is not the best wording. Bogdan Nagachop (talk) 02:12, 17 June 2011 (UTC)[reply]

Two rules

  • Selecting the most frequently used variant based on a search engine test is not acceptable.
  • When in doubt, use the default romanization guidelines.

did only exist in the People section, I think if, then they should apply to People and Places. I will change this, since I think this was only a mistake. Please revert me if I am wrong. Bogdan Nagachop (talk) 02:21, 17 June 2011 (UTC)[reply]

The search engine test limitation should be removed altogether - it is acceptable in the rest of Wikipedia to search among english language reliable sources to see which spelling is most frequently used by those sources. The inclusion of the clause here seems to be the same as limiting "Common Usage" to only dictionaries or reference works rather than to all reliable sources.Stuart.Jamieson (talk) 05:44, 17 June 2011 (UTC)[reply]
I think simple SET usage is problematic. With small majorities there may be different Romanization outcomes, with the effect that editors and readers never can predict under which name an entry is to be found. In listings it may look strange to have several Romanization systems mixed together. I would like to know what other encyclopedias do, are they mixing different Romanizations on a large scale or are they sticking mostly to one system. I would prefer sticking mostly to one system. So it would mean WP:RUS specifies the WP:SET how-to-guide, asking for 75% qualified majority or so. Bogdan Nagachop (talk) 07:52, 17 June 2011 (UTC)[reply]
I don't agree, we can redirect from less common romanisations, paper encyclopeadias cannot. If the majority is extremely slim, then we look at the quality of the sources but the SET remains as an aspect in weighing up commonality. As Pmanderson appears to imply above (with the Pidgin comment) Britannica follows sources in some cases and applies strict romanisation in others. For instance it's article on korolyov/korolev is strictly titled korolyov, yet discussing the city in other articles it follows sources on the subject in hand and uses the common name Korolev. As I discovered recently in an (all English) common naming discussion about a series of related articles - consensus on WP tends to be to name based on most recent common English usage rather than enforcing a consistent system of naming that doesn't represent current common usage of individual names. Stuart.Jamieson (talk) 09:51, 17 June 2011 (UTC)[reply]
Well, WP:RUS' intent is not to tell editors how to spell something in a context; it is strictly to determine what spelling to use for the article's title (since there can only be one). How to link to that title from other articles is not in the scope of this guideline, nor should it be.
As for SET, WP:RUS warns against using a basic search test as the sole criterion for making a decision. Googling up one variant, then another, then comparing the number of raw hits with no regards to what the results include is not an acceptable approach. An approach outlined in WP:NCGN#Widely accepted name, on the other hand, is perfectly fine (although, as I commented above, it only works well for choosing between alternative names, not between alternative spellings of the same name). I would support re-wording the SET clause accordingly, but I would oppose its removal. Attempts to move articles around based solely on the raw ghits counts are, sadly, not uncommon.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 17, 2011; 14:06 (UTC)
A red herring. No guidance on Wikipedia recommends following raw google hits; most of the sections to which this could be redirected expressly recommend against doing so. (Engines like Google Books, which can be set to sample published and copy-edited texts in English, are another question.) Therefore the simplest way to discourage appeal to www.google.com is to install the proposed redirect (for example, to WP:UE and WP:NCGN#Widely accepted name, both of which do), and perhaps to repeat the warning as summary text. Is there a non-figmentary reason to oppose? Septentrionalis PMAnderson 17:34, 17 June 2011 (UTC)[reply]
How's this conclusion different from what I said???—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 17, 2011; 17:36 (UTC)
The widely available and consensus judgment against using raw google alone is no argument against any of the fixes proposed in this section; none of them would endorse raw google. Is there an argument against, or shall we continue? Septentrionalis PMAnderson 17:43, 17 June 2011 (UTC)[reply]
I don't understand what you mean; could you re-word your statement more clearly, please? I was not using "the consensus judgement against using raw google hits alone" as an argument to oppose "any of the fixes proposed in this section". What I said is that despite that consensus judgement, attempts to act based on using raw ghits alone are, unfortunately, not uncommon. Some can be found on this very page.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 17, 2011; 18:02 (UTC)
Some can be found on this very paqe. Where, pray? Do look closely, and do distinguish (as other guidelines do) between raw Google and more reliable search engines. Septentrionalis PMAnderson 18:27, 17 June 2011 (UTC)[reply]
Did someone mention "ngrams"?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 17, 2011; 18:59 (UTC)
As Ezhiki would have seen if he'd looked, ngrams are based on Google Books: printed, usually copy-edited, off-line sources; one of the reasons we say "raw Google" is to differentiate such searches from www.google.com. If would be nice if this discussion had some contact with reality. Septentrionalis PMAnderson 19:18, 17 June 2011 (UTC)[reply]
Yeah, you have me convinced. Counting raw ghits without looking at the results is very much different from counting raw gbooks hits without looking at the results. Not at all the same. I stand corrected.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 17, 2011; 19:32 (UTC)
Well, good. Now that Ezhiki is "convinced" (even though he continues to ignore the difference between two different search engines which have only a manufacturer) maybe he will present an objection with some merit or an altetnative text. Septentrionalis PMAnderson 19:53, 17 June 2011 (UTC)[reply]
I'm sorry, but which part of my comments prompted you to conclude that I have an objection or am about to propose an alternative text? Mine was a clarifying remark pointing out that the wording might use some tweaking. Do you have a suggestion, perhaps? You did mention a possibility of replacing this part with a "summary text", which sounds fairly close to what I had in mind.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 17, 2011; 20:53 (UTC)
I can't think why I would oppose its removal would suggest opposition to the proposal discussed, so I guess I'll just have to do a draft. Septentrionalis PMAnderson 21:09, 17 June 2011 (UTC)[reply]
Ezhiki, How is your conclusion any different from mine where I said: " If the majority is extremely slim, then we look at the quality of the sources but the SET remains as an aspect in weighing up commonality." WP:NCGN#Widely accepted name, sets the bar as anything less than a 10% majority and I'm quite happy to accept that in my defenition of "extemely slim". I also agree that only searches of reliable sources (News, Scholarly, Biblio) should be accepted and that those results should be filtered. Stuart.Jamieson (talk) 17:51, 17 June 2011 (UTC)[reply]
In that aspect, it is not different. I was merely implying that having a short reminder on the unacceptability of using raw google hits for deciding anything has practical value. How to best word that reminder is a different matter.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 17, 2011; 18:02 (UTC)

Draft

Let's keep this simple. Replace the entire section on conventional names with:

See WP:COMMONNAME
Wikipedia generally uses titles which represent what the subject of the article is most commonly called in English reliable sources; several ways of determining what is most common for place names can be found at WP:NCGN#Widely accepted name and the following sections; similar methods apply to articles on people.
Search engine results should be treated with care, and hits on Internet searchers like raw Google (www.google.com) should be ignored unless they are overwhelming; the internet is neither reliable nor representative. For more, see WP:NCGN#Search engine issues.

I have not bothered to pipe the full names of the alphabet soup. Septentrionalis PMAnderson 21:09, 17 June 2011 (UTC)[reply]

There is a point (Moscow vs. Moskba?) where all the noise and imperfections of www.google.com are drowned out by the signal; that's what this means by "overwhelming". Septentrionalis PMAnderson 21:13, 17 June 2011 (UTC)[reply]

WP:NCRUS

Shouldn't we go even further and call this thing "Wikipedia:Naming conventions (Russia)", like so many others in Wikipedia:Naming conventions (geographic names). At the end, this is not about Romanization, but about article title naming. See also Wikipedia:Naming conventions (Cyrillic).

The default Romanization can be deleted alltogether after being incorporated into the article Romanization of Russian. Bogdan Nagachop (talk) 16:44, 18 June 2011 (UTC)[reply]

Except that the "default Romanization" is unsourceable. While it is another not unreasonable convention, like any of the other conventions in that article, it is something we made up.
You are probably right on what the title should be; if there's any of this page left after your suggestions are implemented. ;-> Septentrionalis PMAnderson 21:16, 18 June 2011 (UTC)[reply]
Maybe at least the table showing the default Romanization can be reduced to the rows that are not identical with BGN/PCGN. I think I read somewhere that Ezhiki or someone else said, this default Romanization or simplified BGN/PCGN is not made up, but is wide spread use. It should be possible to document that in the article? Bogdan Nagachop (talk) 08:59, 19 June 2011 (UTC)[reply]
I'm perfectly willing to look at sources - preferably in English. Septentrionalis PMAnderson 19:47, 19 June 2011 (UTC)[reply]

Splitting the page into WP:NCRUS and WP:ROMRUS

I think this page really needs to be split. There is one part dealing with a self-made romanization system and one part dealing with how to apply the general WP naming conventions.

The shortcut WP:RUS could be a disambiguator:

Bogdan Nagachop (talk) 16:55, 19 June 2011 (UTC)[reply]

  • Quite reasonable; my qualms are: The first will be really short, if we adopt your suggestion of making that material a summary section; the second sounds more like a section redirect. I'm waiting for a week's silence or third party comment until acting; but if you think this reasonable, do try a draft. (It might be easier if the draft began as two separate sections on the same page.) Septentrionalis PMAnderson 19:50, 19 June 2011 (UTC)[reply]
Wikipedia:Naming conventions (Cyrillic) has a section on Russian (which is basically empty).. this may be an appropriate location for a naming conventions section, if these are divided up as Bogdan suggests (which I wouldn't be against). Mlm42 (talk) 22:47, 19 June 2011 (UTC)[reply]

I made a first version of the split page Wikipedia:Naming conventions (Russia), and added all guidelines important for article naming that I could find. I also imported the rules from WP:NCGN#Russia to Wikipedia:Naming conventions (Russia)#Inhabited localities. I think it should be deleted from WP:NCGN#Russia, and a reference should be made to NCRUS similar to NC-Canada and NC-New Zealand, two NCs only referenced from NCGN. Bogdan Nagachop (talk) 18:04, 21 June 2011 (UTC)[reply]

There are now several attempts to update these guidelines, and someone needs to declare how the issues are to be progressed through discussion. But I suggest your draft page should not retain the use of the word "dictionary" per the discussion above; my view is that general WP principles require that where there is a widely used English form, that should be used for the title. Widely used means mainstream media such as newspapers or travel guides, not scholarly publications which likely have their own romanization systems. If the word dictionary is interpreted to mean a reference book (Britannica, The Times World Atlas etc) we would effectively be inviting article creators to adopt another institution's romanization system. Secondly, please do not use the Togliatti example. It is an exceptional case being derived from a non Russian source. Better examples where there is a variety of commonly used spellings would be Kruschev (prefer Khrushchev) and Ekaterinburg (prefer Yekaterinburg) where in both cases the spelling closer to WP standard is chosen. Sussexonian (talk) 07:32, 22 June 2011 (UTC)[reply]

Splitting is done. I think WP:ROMRUS can be activated as official policy for a WP specific Romanization method. WP:NCRUS needs more work. WP:ROMRUS is now only about one Romanization method for Russian, namely the WP method. When to use this method is covered by NCRUS, which is mostly a collection of general WP rules. Bogdan Nagachop (talk) 03:12, 23 June 2011 (UTC)[reply]

Convenience header

Hard sign

  • Text says : omitted,
  • Table says : omitted when followed by a vowel.

If table is followed, it would mean using BGN/PCGN romanization of Russian when not followed by a vowel, i.e. using "”". Is this really desired? Bogdan Nagachop (talk) 02:45, 23 June 2011 (UTC)[reply]

I put " When followed by a vowel " into comment. The comment now also says "...in text version of rule not mentioned. In modern Russian, is it ever not followed by a vowel?" Bogdan Nagachop (talk) 03:01, 23 June 2011 (UTC)[reply]

I think this was an oversight; it was meant to be "iotated vowel", not just "vowel". When the hard sign is followed by a non-iotated vowel, "y" or ” is used to indicate the hard sign (Мусийкъонгийкоте→Musiykyongiykote or Musiyk”ongiykote; a tiny rural locality in Ingushetia), but such cases are so far and few between that the "always omit the hard sign" rule works for pretty much all practical intents and purposes.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 23, 2011; 13:38 (UTC)
I added that special rule so that the Ingushetia locality is covered. Bogdan Nagachop (talk) 17:35, 23 June 2011 (UTC)[reply]

Double letters

The only reason to discuss the Cyrillic е at all is to say that the combination -йе- is to be transliterated -ye- not -yye-. By changing this, we can get rid of several lines in the table which say effectively "follow the BGN transliteration". Similarly, the lines which explicitly say "follow BGN except" are unnecessary; all we need are the cases where it doesn't. Doing this will not change guidance atr all; it will just make it much easier to comprehend. Septentrionalis PMAnderson 14:37, 23 June 2011 (UTC)[reply]

Can you please clarify if you are suggesting adding a section dealing with the -йе- combination in addition to simplifying the "e" section, or if you are suggesting simplifying the "e" section and not mentioning -йе- at all? This combination is very rare, so it's probably not worth being covered separately.
Another instance of double letters the guideline fails to mention in its current form is the "-ые" ending. The current practice is to romanize it as "-ye" ("y" for "ы" and "e" for "е"), not "-yye" ("y" for "ы" and "ye" for "е")—see, for example, Naberezhnye Chelny. Unlike -йе-, this combination is very common and is worth being covered, but (things are never easy, are they) it is seldom romanized as "ye" in the middle of the words, only when it's an ending. Otherwise it would be impossible to distinguish "въезд" from "выезд" and their numerous derivatives.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 23, 2011; 15:15 (UTC)
I'm proposing to replace the e section with a -йе- section. I would have no objection to a -ые section in parallel. But the rest of e only repeats BGN. Septentrionalis PMAnderson 20:19, 23 June 2011 (UTC)[reply]
Done Septentrionalis PMAnderson 15:12, 24 June 2011 (UTC)[reply]
I added the rule for "-ые" [3] Bogdan Nagachop (talk) 09:56, 27 June 2011 (UTC)[reply]

Ё (ё)

Yo-yo

Our actual practice, independent of this guideline, appears to be to Romanize ё as e and to transcribe it as yo; that's also what most Romanizations do:

  • Mikhail Sergeyevich Gorbachev (Russian: Михаил Сергеевич Горбачёв, romanized: Mikhail Sergeyevich Gorbachyov, IPA: [mʲɪxɐˈil sʲɪrˈɡʲejəvʲɪtɕ ɡərbɐˈtɕof] ⓘ;

Let us take this as default. As usual, we can vary from the default for good reasons. Septentrionalis PMAnderson 14:37, 23 June 2011 (UTC)[reply]

Gorbachev isn't a good example. His last name is spelled with "e" not because it's an exception to the default romanization table, but because that's what you end up with if you go through the steps outlined in the "People" section. Do you have a better example?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 23, 2011; 15:24 (UTC)
And so do most other people and places. Septentrionalis PMAnderson 20:26, 23 June 2011 (UTC)[reply]
People, maybe. At least those who have coverage in the English-language sources. There are also plenty of people who meet our notability criteria, but the coverage is in the sources which are not in English. The omnipresent soccer players who, unfortunately, manage to pass our notability tests, are the best illustration.
Places, not so much. The easier the use of "ё" can be documented, the more likely we are to see a push to using "yo" and not "e". From what I've observed, our articles gradually migrate to the variants with "yo", not the other way around. (And no, it wasn't me who moved them all :))—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 23, 2011; 20:46 (UTC)
Because you move them, relying on the former text of this guideline? Septentrionalis PMAnderson 20:49, 23 June 2011 (UTC)[reply]
See the fine print above.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 23, 2011; 20:56 (UTC)

Instead of examples we would need statistics. WP:ROMRUS is meant to document most common deviations from BGN/PCGN found in English language publications. Maybe this is a case where people are treated different to places. For people a lot of "-ёв" -> "-ev" seems to be documented. What about place names, what about non-endings? Bogdan Nagachop (talk) 17:54, 23 June 2011 (UTC)[reply]

Where does it say ROMRUS is meant to document most common deviations? This page is meant to guide WP editors on how to provide 'English' versions of Russian names to a largely non Russian-speaking audience. I think the purpose of choosing variations from the over-academic system is to make the spelling less intimidating to that audience while continuing to give an idea of the pronunciation. Anyway we could not easily find statistics to answer some of the questions we might pose.
I can understand the idea of a specific rule that says "-ёв" -> "-ev" (and "-ёва" -> "-eva" ) but if that is adopted it should be an exception to a general rule of ё -> yo. I expect there are more cases of Fyodor than Fedor, for example, and the "yo" indicates how the name/word is pronounced (so I prefer Oryol to stay put). Sussexonian (talk) 19:52, 23 June 2011 (UTC)[reply]
One of the ways in which this rule "intimidates" is by providing our readers with versions of well-known people and places which they have not seen before. It is not necessary for Romanizations to indicate pronunciations, any more than it is necessary for native English spellings to do so (Worcestershire, anybody?). Septentrionalis PMAnderson 20:26, 23 June 2011 (UTC)[reply]
My reference to "intimidating" is the same as yours: unfamiliar strings like "yy" in Sosnovyy Bor mean nothing to a non Russophone reader and if we can do without them so much the better. The average reader will assume that a name that has been rendered into script readable in English conveys the rough pronunciation. There is nothing we can do with Worcestershire or Łódź but we can help by using Oryol and Fyodor rather than the alternatives. Subject always to the common English name rule. Sussexonian (talk) 12:03, 24 June 2011 (UTC)[reply]
I don't feel helped by reading a page about Oryol, which means nothing to me. I've heard of Orel; I know something about it. What we should do is to indicate that Orel, like Gorbachev, is not pronounced as it is spelt.
Fyodor is another question; it is now the conventional and customary spelling, understood by most readers. Therefore one necessary step is to indicate that the default spelling (whichever it is, even the BGN ë) should rarely be used. Septentrionalis PMAnderson 13:02, 24 June 2011 (UTC)[reply]
@Sussexonian "Where does it say ROMRUS is meant to document most common deviations?" - My understanding was, that WP:ROMRUS is a deviation from BGN/PCGN and that the differences to BGN/PCGN that WP:ROMRUS contains are commonly found. Ezhiki wrote: The "simplifications" were the observations of real-life usage, both indirect and direct [4]. Bogdan Nagachop (talk) 13:26, 25 June 2011 (UTC)[reply]

WP:ROMRUS Ё rule has been removed

The WP:ROMRUS Ё rule has been removed [5], that would mean articles need to be moved now? Bogdan Nagachop (talk) 13:56, 25 June 2011 (UTC)[reply]

Not until we have a consensus guideline (which we don't), that requires their being moved. Even if we did, that would only apply to articles which don't have a customary English title. Septentrionalis PMAnderson 23:27, 25 June 2011 (UTC)[reply]
@customary English title - of course. So how can we get Ё out of the way. Maybe even without consensus, I mean, some people may just stay with different opinions. Shall there be a vote about Ё treatment? Bogdan Nagachop (talk) 18:41, 26 June 2011 (UTC)[reply]

Ё summary 2011-June

As far as I can see there are

  • BGN/PCGN romanization of Russian: Yë (yë) + Ë (ë)
    • supported by:
      • indirectly via the removal of the old WP:ROMRUS rule [6]
  • Old WP:ROMRUS, from start of the page in 2005-12-14 [7], modified 2006-01-02 [8], removed 2011-06-24 [9] : Yo (yo)
    • supported by:
      1. Ezhiki
      2. Sussexonian
      3. Bogdan (only for procedural reasons, no opinion on the rule itself)
  • Personal name based suggestion: E (e)
    • supported by:
      1. Pmanderson

Bogdan Nagachop (talk) 10:44, 27 June 2011 (UTC)[reply]

Voting on Yo rule re-introduction for procedural reasons

The Yo rule that existed since 2005/2006 should be added back to WP:ROMRUS for procedural reasons and to allow the WP romanization system to be back in policy status. The Yo rule existed since at least 2005/2006. Since the rule affects a lot of article names, its removal or change should be discussed and if people cannot get a consensus a voting should be done.

Support re-introduction of the 2005/2006 Yo rule for procedural reasons
Oppose re-introduction of the 2005/2006 Yo rule for procedural reasons
Neutral

Voting on Yo rule re-introduction for substantive reasons

Oppose Yo rule re-introduction for substantive reasons
Neutral to Yo rule re-introduction for substantive reasons

WP:ROMRUS Ё rule 2011-June re-introduction

I re-inserted the rule. Ezhiki and Pmanderson, two opposing parties (on Ё treatment, this is not a statement about the procedural matters) did edit in WP in the last 24 h, but didn't come here two vote. I just want to accelerate the process to get this page back to policy status. The current Ё rule documents current usage in en WP. If policy change is wanted, please try to get a majority via the talk page. Bogdan Nagachop (talk) 16:52, 28 June 2011 (UTC)[reply]

-его -ого

Shouldn't there be a note to require these to be -vo rather than -go? In letter-for-letter bibliographic transcription -go might be acceptable, but for general rendering of Russian it should be -vo per the pronunciation. Sussexonian (talk) 20:02, 23 June 2011 (UTC)[reply]

That would be a transcription element, not transliteration. No existing transliteration system uses this convention, and we shouldn't either. For other proposed exceptions above one can easily find an abundance of real-life examples; as for this one, while I've seen this convention used occasionally in the English-language sources, I can't say it's very common. Do you perhaps have a specific example we can discuss?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 23, 2011; 20:17 (UTC)
This page is headed "Romanization". Can you write an accessible description of the distinction you are making between transliteration and transcription that would make sense to include in this guideline? In the Gorbachev example, which word describes the text in italics after the Cyrillic? Is that text intended to convey the pronunciation or the scholarly transliteration(?) that might be found in library catalogues? Sussexonian (talk) 12:09, 24 June 2011 (UTC)[reply]
I expounded the difference at some length in the threads above. If you do an inline search for the terms, you should be able to find my explanations easily. You can also refer to our transliteration, transcription (linguistics), and romanization articles, which are good enough to understand the basic differences. Additionally, our romanization also has a standardization aspect, as defined here (the purpose of romanization is establishing standardized Roman-script spellings of those foreign geographical names that are written in non-Roman scripts or in Roman alphabets that contain special letters).
With Gorbachev, the text in italics is a WP:RUS romanization, which, in turn, is a slightly simplified version of the BGN/PCGN romanization of Russian system, which, in turn, is a system of transliteration with a limited use of transcription elements. It's not the only transliteration possible, but it would be the variant most likely to be encountered if "Gorbachev" hadn't already been established as a common name (with "common name" being defined per the "People" section criteria in the pre-RfC WP:RUS).
As a side note, it's impractical to include all possible transliterations of a name in an article's lead simply because so many systems exist, which is why only one is selected. WP:RUS/BGN-PCGN spelling is not what you'll find in library catalogs (which use ALA-LC) or in the scholarly works of linguistics (which use scholarly transliteration), but it is what you are more likely to see in the works of general reference, books, or media, and those are what matter the most to us.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 24, 2011; 14:11 (UTC)
Then we should abide by it, insofar as we are likely to find it. We should alter it only in ways that make it more likely to be found: Veliky, not Velikiy. Septentrionalis PMAnderson 15:03, 24 June 2011 (UTC)[reply]

Adjective?

-ий endings in adjectives of Russian origin become -y.

Why restrict this? It's one of the most common male endings in surnames, and as such is almost always transliterated -y. Septentrionalis PMAnderson 15:17, 24 June 2011 (UTC)[reply]

Last names ending in "-ий" follow the same rules as the adjectives, which is why they tend to be treated the same way as the adjectives for the purposes of transliteration/romanization. So the current wording does not necessarily exclude the last names. But perhaps changing "adjectives" to "adjectives and human names" (with the latter also covering first names such as Dmitry, Vasily, etc.) would be more helpful. Such wording would be more representative of general usage, yet still allow treating nouns ending in "-ий" (such as "sanatoriy") differently.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 24, 2011; 15:43 (UTC)
It would be more helpful; English (and therefore many of those who will be reading this table) doesn't group that way.
But is this a common ending for nouns, or is restricted to Latin loan-words? Septentrionalis PMAnderson 17:03, 24 June 2011 (UTC)[reply]
A good number of such nouns are indeed Latin loanwords, but there are also some Russian words (such as "lesnichiy"). The ending is also very common in plural genitive. I do not, of course, expect us to use romanized nouns in plural genitive as article titles, but such nouns do occur in compound toponyms.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 24, 2011; 17:28 (UTC)

Where we are

OK, now that Mlm42 has closed his RfC, can we do something about the mess the rest of the discussion has become? I've been involved with this from day one, and now even I can't make out what's going on and where. I previously attempted to keep track of all the proposals voiced so far, and more suggestions came in since I had last updated that list. The list is obviously too long and contains some of the proposals people to whom they are attributed to probably no longer wish to pursue, but there is no indication on this page that they have been retracted. Can we perhaps do a shorter version of it and tally the supports/opposes to each item? It would be pretty similar to the last several threads on this page, except it should also account for the proposals and opposition voiced before those threads started to materialize. I just can't think of any other ways to organize the discussion without dismissing all points of view, although I am, of course, open to alternative suggestions. Change-as-we go approach exercised so far is obviously not working—can you imagine someone who's just joined being able to figure out what's going on here?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 24, 2011; 16:54 (UTC)

This is one advantage of a divide: there's only one pressing question on this page: the default Romanization of ё. All the rest are tweaks. Septentrionalis PMAnderson 17:06, 24 June 2011 (UTC)[reply]
Not exactly. The fate of the conventionality cause hasn't been decided yet either. You, I take, are for removing it altogether, while me and some other participants are for amending it, although each has his/her own opinion as to how exactly the final version should look like. There's no consensus on this by a long shot.
With the rest, even though they are mostly tweaks, it's still helpful to have them documented in one place rather than to hunt them down all over this page.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 24, 2011; 17:13 (UTC)
What do you mesn by conventionality case? If you mean
•In absence of documentation supporting one of the criteria of conventionality, articles can be moved to the spelling produced using the WP romanization of Russian.
I'm for banning any editor who acts on it; but it's at WP:NCRUS in all its bossy and unidiomatic splendor; not here. How strongly any default spelling should be enforced is separate from what the default should be. Septentrionalis PMAnderson 17:19, 24 June 2011 (UTC)[reply]
WP:NCRUS is itself only one proposal as to how to go about that clause. Whether the split is even necessary is debatable.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 24, 2011; 17:38 (UTC)
We have 1) Wikipedia's "default" romanization of Russian, and 2) Wikipedia's naming conventions for Russian articles. Before, they were split as two sections of the same page, and now they are split into two different pages. I think having two different pages makes more sense because it more clearly separates the talk page discussions. Although the two are obviously related, I think it's helpful to keep the discussions (and arguments) separate whenever possible. Mlm42 (talk) 00:38, 26 June 2011 (UTC)[reply]
Not to forget that some WP:RUSSIA related naming conventions were only contained in WP:NCGN#Russia, so in fact there were two pages before. The WP:NCGN content is now copied to WP:NCRUS. WP:ROMRUS falls into Category:Wikipedia romanization systems and WP:NCRUS falls into Category:Wikipedia naming conventions (regional). Bogdan Nagachop (talk) 07:25, 26 June 2011 (UTC)[reply]
Mlm says ... it more clearly separates the talk page discussions. The bulk of this talk page concerns matters that are in NCRUS, and currently there is very little talk over there, although there are more live issues to be sorted there than here.
The discussion about gazeteers and dictionaries overlaps with the one about 'iy' and 'y' and other rules. If any English source such as an atlas can be used to claim the existence of a "common" English spelling, then we will rarely use ROMRUS at all for place names. And someone would have to change thousands of article names to insert an apostrophe for every soft sign etc. We would be better making ROMRUS simply mirror BGN/PCGN for consistency.
We need to know, is ROMRUS only going to be used for "obscure" people and places, which have no coverage in English outside reference books, or will it be used frequently, with only the most common cases like Tchaikovsky etc standing as exceptions? In the first case we should go for a scholarly ROMRUS and in the latter case we should go for something user-friendly for the casual Wikipedia reader.
Or we could just assume that the vast majority of articles are properly titled already and put the pages back how they were. Sussexonian (talk) 14:40, 26 June 2011 (UTC)[reply]

Where we are? Except for Ё treatment the page is ready to be tagged as guideline or policy again. Or is there any other issue with the WP romanization? Maybe the easiest is to document current use in the guideline, then no pages need to be moved. After that, the people that want to change Ё treatment can gather support and try to change the longstanding Ё rule. Bogdan Nagachop (talk) 18:45, 26 June 2011 (UTC)[reply]

Romanization table

I am in favour of restoring the whole table in some form, rather than only include the "List of differences from BGN/PCGN". This could simply be in the form "In all cases except as listed below Аа -> a; Бб -> b ..." with no need for examples, but it will increase the likelihood of this guideline being used by non experts as well as Russian experts, and non-regulars will already have been passed from one page to another before arriving here (or at NCRUS if that split is kept). I don't see the point in sending the user to BGN/PCGN romanization of Russian when we can provide all the information here. Sussexonian (talk) 18:02, 24 June 2011 (UTC)[reply]

Hear, hear.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 24, 2011; 18:06 (UTC)
That would be reasonable, if it is kept short, so that people don't have to search the table. For example, I commend "Ее -> e or ye" rather than specifying at length which is which; those who aren't sure can follow the link. Septentrionalis PMAnderson 22:38, 24 June 2011 (UTC)[reply]
There were inconsistencies regarding the WP:ROMRUS deviations from BGN/PCGN mentioned in the text and those mentioned in the table. A good way to avoid inconsistencies is to always have only one statement about a particular difference between BGN/PCGN and WP:ROMRUS. In that case one would have to choose between text and table. Having it only in text, would lead to no table or to a pure BGN/PCGN table. Having it in a full alphabet table would lead to the deviations being dispersed. Maybe a full table can be restored having a column that allows sorting on the deviation, so anyone who wants to see the deviations can have them in one place. Another and maybe more elegant solution is to have a complete list of vowels and the hard and soft sign. No need for sorting and it still will be quite compact. But with E and Ё removed [10] the number of letter having differences in treatment is now lower. Under the logic that led to the removal of E and Ё, one could also remove Ы, leaving only hard and soft sign and vowel combinations. Bogdan Nagachop (talk) 13:43, 25 June 2011 (UTC)[reply]
Please do remove Ы. Septentrionalis PMAnderson 00:14, 26 June 2011 (UTC)[reply]
It would depend on the logic of the table. It could list all letters that under some situations are treated differently. If done that way, the corresponding rows could be inserted into the BGN/PCGN table as replacement. Since there are some situations where Ы is treated different from BGN/PCGN the row could be kept. But since и and й have been removed it is inconsistent to keep ы. What is nice, is that with these removals the table is easier to read. When ы is removed only hard and soft sign and vowel digraphs are contained. Bogdan Nagachop (talk) 07:18, 26 June 2011 (UTC)[reply]
It may be easier to read but it is harder to implement. I have no idea why sections of the table have been removed as there has not been debate about changing WP practice, (other than some debate, but certainly not agreement, about Ё). If anyone is going to suggest changes they need to state whether the changes are to bring the table into line with WP practice or proposing to change WP practice, if so why and how many articles they think would be disrupted. Sussexonian (talk) 09:28, 26 June 2011 (UTC)[reply]
I added examples for Ы in BGN/PCGN. Now it is visible that the romanizations differ, i.e. the row has to stay anyway. Bogdan Nagachop (talk) 11:29, 26 June 2011 (UTC)[reply]

Abbreviations

"Abbreviations are usually romanized with capitalization as indicated" - what does "usually" mean? When not? This is one more unclear rule. Bogdan Nagachop (talk) 13:08, 25 June 2011 (UTC)[reply]

I presume it means "unless there is a common English rendering". But if a Russian entity is not normally abbreviated in English, I'm not sure why we would want to do so in an article.
Another related case is academic citations such as "Yu. A. Ivanova", but they too vary in English usage and we would want to keep the style found in the source. I guess the main reason for the sentence is to clarify that the style shown is acceptable in Wikipedia. Sussexonian (talk) 09:49, 26 June 2011 (UTC)[reply]
I will change to "unless there is a common English rendering". But maybe for consistency it could be removed completely, since if something is established it will be mentioned anyway. This part would belong to WP:NCRUS. Bogdan Nagachop (talk) 11:04, 26 June 2011 (UTC)[reply]

Й (й)

I re-inserted [11] Й (й) since BGN/PCGN differs, I added an example for the difference to make this fact more obvious. Bogdan Nagachop (talk) 18:23, 26 June 2011 (UTC)[reply]

Bringing WP:ROMRUS to guideline or policy level

If the Ё rules are added back, then WP:ROMRUS could be taken to guideline level? It can then be discussed how and if the Ё rules need to be changed. I think this is the only pending dispute on the enWP romanization rules? After that, we can look how to organize WP:NCRUS. Bogdan Nagachop (talk) 18:27, 26 June 2011 (UTC)[reply]

Now the Ё rule is back. I would like to tag WP:ROMRUS as policy now. Any opposition? Bogdan Nagachop (talk) 16:55, 28 June 2011 (UTC)[reply]

For the record, I'm generally content with the current wording, although I'd say to wait until July 9 to mark the page as a standing guideline (July 9 will be one full month since the original RfC). Of course, if there's additional input in the meanwhile, that can always be extended as necessary.
Also, from what I gather, there is no opposition to reinstating the full romanization table for ease of reference (do correct me if I've missed something).—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 28, 2011; 17:27 (UTC)
I object to reinstate the full table. It makes the differences to BGN/PCGN harder to find. All consonants are equal, most vowels too. Mostly only certain two letter combinations are treated differently. "ease of reference" could also be claimed by those that want only a diff table. Bogdan Nagachop (talk) 01:08, 29 June 2011 (UTC)[reply]
Sensible. Septentrionalis PMAnderson 14:40, 29 June 2011 (UTC)[reply]
I don't understand how having a full table would make locating the differences from BGN/PCGN any more difficult. The differences are documented in a separate column, and on the lines where there is nothing to document (which is most of the lines) that column's cells will be empty. I'd argue that such a setup makes the differences easier to find (they really stand out among all those empty cells), not harder. Besides, it's not the differences this table is going to be consulted for most often; it's to find out how to romanized something properly according to our guidelines. As Sussexonian previously mentioned, that'd require jumping between several pages, making the guideline less usable.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 29, 2011; 14:49 (UTC)
Currently I can see all rules on one screen. Otherwise I have to scroll. Adding more than 20 lines of non differences in between the differences of course makes locating the differences harder. That's simple logic. If the table only has the differences, one knows that each line documents a difference. Only in your setup one would need to look for specific lines.
Maybe you remember that the old version of ROMRUS had a listing outside the table to summarize the differences. Exactly because they were not easy to find. And these two places of documenting differences were not in sync.
"require jumping between several pages" You are again diving into drama, it's only two pages, BGN/PCGN linked in the intro sentence, very easy to go to. Bogdan Nagachop (talk) 11:03, 30 June 2011 (UTC)[reply]
Well, the intent of the guidelines is to guide. The page will be used most often not by you or me (or by people curious how exactly our guideline is different from the BGN/PCGN's), but by those who don't remember the romanization rules by heart and need help. Those folks are better served by seeing the whole conversion table. It's not exactly an easy task to reconcile what BGN/PCGN romanization of Russian says with what this page says, when all you have to work with is the differences. If ease of locating the differences is your only concern, then I suggest we use two tables—one as a letter-to-letter romanization guide (for people like Sussexonian), and the other one to document the differences (for you and other curious folk).—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 30, 2011; 18:37 (UTC)
Two tables - fine with me. But please list the diffs first and make the diff list the normative one, while the full one is only descriptive, or how ever one would call that. Bogdan Nagachop (talk) 20:50, 30 June 2011 (UTC)[reply]

-ev, -yov

I object to the rule which makes normative a transliteration of Ё as yo. When this page was discussed together with WP:NCRUS others did also. It seems obvious that its treatment should be context dependent (spell Fyodor as Fyodor, as default, unless there is some reason not to; but spell -ёв endings as -ev (as most of them are). Septentrionalis PMAnderson 20:30, 28 June 2011 (UTC)[reply]

Who beside you? Maybe the user can be added to Wikipedia_talk:Romanization_of_Russian#.D0.81_summary_2011-June and maybe you can vote at Wikipedia talk:Romanization of Russian#Voting on Yo rule re-introduction. I personally have no opinion, but for procedural reasons favor reinsertion. Maybe a rule "-ev" for names would be ok for all? I would be fine with that. Bogdan Nagachop (talk) 01:13, 29 June 2011 (UTC)[reply]
That would be a great improvement; add a sentence that other exceptions may exist, and I would join the consensus. Septentrionalis PMAnderson 14:30, 29 June 2011 (UTC)[reply]
I suppose one could athe question the other way: Is there something (a person, place, etc) which has a -ёв ending, and has a widely accepted English-language name ending in -yov? Because if not, then we should seriously be questioning it's use as the default.. because there are quite a few -ev endings that are widely accepted. Mlm42 (talk) 01:39, 29 June 2011 (UTC)[reply]
At least Britannica is using http://www.britannica.com/EBchecked/topic/433662/Oryol UPDATE: sorry this is not answering the question by Mlm24, since it does not end in -yov. Bogdan Nagachop (talk) 09:13, 29 June 2011 (UTC)[reply]
Not just Britannica. I should have mentioned this before, but it completely slipped my mind: transliterating "ё" as "yo" is also the recommendation of the Oxford Style Manual (section 11.41.2 "Transliteration"), although at the same time they recommend to transliterate it as "o" after "ж", "ч", "ш", and "щ", which isn't something often seen in use in real life. In practice, this prescribes to transliterate the "-ёв" ending as "-yov" except in "-чёв" and "-щёв", which would become "-chov" and "-shchov" ("-жёв" and "-шёв" endings are invalid or very rarely encountered in last names). I suspect Britannica follows that style guide—while their Gorbachev entry is under "Gorbachev" (presumably because that's the most common one), their entry on Pugachyov is under "Pugachov" and they routinely use "Chorny" when they mean "Chyorny".
Incidentally, the "-ий" and "-ый" endings are also covered (the recommendation is to use "-y" in proper nouns or titles).—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 29, 2011; 14:42 (UTC)
That's perfectly sensible - as transliteration. Septentrionalis PMAnderson 14:45, 29 June 2011 (UTC)[reply]
Hmm.. but Oryol doesn't end in -yov. My question was to find a widely accepted English language name whose ending is "-yov", as romanized from "-ёв".. are there any? The Oxford Style Manual is a single source, and isn't itself enough to determine what is widely accepted.. Mlm42 (talk) 17:06, 29 June 2011 (UTC)[reply]
I am sorry, for my mistake that is does not end in -yov. Bogdan Nagachop (talk) 10:46, 30 June 2011 (UTC)[reply]
But why is this important? We are supposed to research the spelling of each individual person's name anyway (either per WP:COMMONNAME or per the conventionality criteria for people in the pre-RfC WP:RUS) and use that. The default "-yov" is only going to be used for people who meet the notability criteria, but have no coverage in English sources. Including a note about the last names ending in "-ёв" seems completely redundant to me.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 29, 2011; 17:19 (UTC)
Even if such people do have notability, why should they be spelled differently than 99 out of 100 people with the same ending who are mentioned once in English sources? Septentrionalis PMAnderson 21:32, 29 June 2011 (UTC)[reply]
I'm merely questioning the wisdom of setting something as the "default", when there appears to be no individual case where the "-yov" ending is widely used.. if it's not actually used, then in my opinion, it shouldn't be the default.
And the default shouldn't only be when we have no English coverage, but also when a significant proportion of English sources disagree. Mlm42 (talk) 22:12, 29 June 2011 (UTC)[reply]
Using "-ev" for personal names would make the naming of people more consistent within WP. But I would like to know, how many articles are affected. Is it 10 000 articles VS 10 like Gorbachev, Khrushchev ....? Maybe in sum "-yov" is widely accepted?
Also I think Ezhiki is wrong by saying The default "-yov" is only going to be used for people who meet the notability criteria, but have no coverage in English sources. - because notability is not needed for being mentioned in articles. Bogdan Nagachop (talk) 11:13, 30 June 2011 (UTC)[reply]
@Mlm if sources disagree should we ever use a method that isn't supported by any (or a minority) of sources. Perhaps we need a clause that says in the event of disagreement consider in order;
  1. Method WP
  2. Method BGN
  3. Method X
  4. Method Y
  5. Method Z
And use the first method that is used by the conflicting sources? Stuart.Jamieson (talk) 12:44, 30 June 2011 (UTC)[reply]
This belongs to WP:NCRUS or WP:COMMONNAME. The talk here is about "Method WP" (WP:ROMRUS). Bogdan Nagachop (talk) 15:33, 30 June 2011 (UTC)[reply]
Yes, but the only reason "-yov" is being used on so many Wikipedia articles is because the old WP:RUS has been enforced for several years, with some editors following it very strictly.. so we can't really use that as evidence that it's "widely accepted". And of course, we're at a stage now that reliable sources may be inclined to use Wikipedia's spelling, so it becomes a kind of self-fulfilling prophecy. Mlm42 (talk) 16:38, 30 June 2011 (UTC)[reply]
The "-yov" ending is used in so many Wikipedia articles also because it is far from being so uncommon as the editors in this thread seem to imply. Also, a part of the reason of why the "-ev" spelling is more prevalent is because there exist quite a few last names for which (in Russian) only the ending is different (and, consequently, the stress). "Лещёв" is a common last name, but "Ле́щев" is also valid. Spelling the latter via "-yov" would, of course, be absolutely incorrect, so the bias is very much one-sided. In all, I'm pretty confident that for any Russian last name ending in "-ёв", one could find English sources which uses "-yov", and the number of the said sources will not be negligible.
As I stated above, I'm against a separate provision for the "-ёв" endings simply because no manual of style I am familiar with has such a provision. Spelling of the last names via "-ev" is predominantly caused by the optional nature of the letter "ё" in Russian, not per any style guide or as a conscious effort. If we fancy ourselves being accurate, we should make an effort to ensure that the transliterations we use are accurate as well, not go with the flow and codify a common mistake just because it happens to be so common. Every other exception to BGN/PCGN we have so far documented on this page can be traced back to at least one formal recommendation; the "-ev" thing is the only one we are homebrewing on our own (whereas "-yov" flows from the general rule of transliterating "ё" as "yo"). I tend to side with Stuart on this—identifying a list of methods to apply in questionable situations would a better approach than trying to re-invent the wheel.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); June 30, 2011; 18:19 (UTC)

I think the burden of proof lies with editors who claim that one or the other variant is more common. I'm no expert, so I don't know, but given the evidence that has been provided so far on this page, it seems to me that "-ev" is the more common and widely accepted ending in English than "-yov". I think this discussion can only proceed effectively if editors bring forward more evidence to support their claims. This applies not only to this one ending, but the letter "ё" in general (or any other disputed romanization, for that matter).

Also, Stuart's comment acknowledges that we should probably still have a "Wikipedia default" set of rules, which I think is the point of all this discussion. Mlm42 (talk) 21:24, 30 June 2011 (UTC)[reply]

I'm not disagreeing with you about the burden of proof at all; in fact, as far as people names go, the pre-RfC WP:RUS heavily emphasized just that. The problem only exists when no English sources are available for an otherwise notable person, in which case our default is as good of a choice as any. My argument is to stick with "-yov" as a default because that's a more accurate approach. The "-ev" ending can be observed so often because when people transliterate Russian last names, they rarely bother checking whether the letter "е" is indeed a "е" or if it's in fact a "ё". If we strive to be a reliable encyclopedia, we should do better than that. There will, of course, be cases where making 100% sure whether a letter is a "е" or a "ё" is difficult to accomplish, in which case using "e" is perfectly fine (at least until more information is available), but to consciously transliterate "ё" as "e" when we know for sure it's a "ё" is just sloppy. That "everyone else is doing it" is not a good excuse, in my opinion. A commonly made mistake is still a mistake.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); July 1, 2011; 15:09 (UTC)
I agree with all this; although making an exception in favour of "-ev" looks reasonable, it could lead to a whole list of other exceptions: every Мария to be Maria not Mariya, for example, and the list of exceptions would be continually up for debate. There will always be cases where the Wikipedia default will produce a different spelling from the commonly accepted English version, and that can't be avoided. Sussexonian (talk) 20:38, 1 July 2011 (UTC)[reply]
Thank you. Those aiming to codify every little exception often forget about this.
It's important to remember that if we have to resort to any kind of default to resolve the situations which consulting the sources can't resolve, it matters very little what that default is, as long as it follows some reputable and reasonable guidelines. Thus it makes sense that the defaults should have a broad scope, or we'll keep spending most of our time trying to pin down more and more possible exceptions instead of doing actual productive work.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); July 1, 2011; 20:52 (UTC)

I will do a draft of the sort of thing I would prefer: when there is no English writing on a subject, and no other strong reason for a particular transliteration, consider how other similar terms are transliterated; then use the default. This is, I hope, where this list comes from in the first place: noticing that ь is normally omitted in Romanization, and so on. Septentrionalis PMAnderson 21:55, 1 July 2011 (UTC)[reply]

Added a short paragraph on method. Feel free to amend, but with this addition I would have no trouble joining the consensus, no matter what was done with Ё in the table. Septentrionalis PMAnderson 22:08, 1 July 2011 (UTC)[reply]
I took out the portion on "the usage of the subject in other Western European languages" from the "some other clear reason to use a particular spelling" line. To me, the whole line duplicates the intent of the line above (if there is some clear reason to use a particular spelling, then English already uses it), but the example itself I just don't understand. We are romanizing a Russian name to use the result in a text in English, so what do we care what other Western European (or, for that matter, any other) languages use? If a place is not covered in any English sources but happens to be covered a lot in, say, Swedish (perhaps because some Swedish researcher's dissertation is about villages in Central Russia), why would we use the Swedish name instead of romanizing the Russian one? Such approach may sometimes be appropriate for historical references (which are mostly outside the scope of this guideline), but why use this as a title of an article about a modern place?
I also find the third bullet point about analogies too vague to be useful. If a famous person happens to have his/her last name spelled a certain way, why should it matter when choosing a spelling for the articles about completely unrelated people? Would we move someone like Ihor Chaykovskyi to "Tchaikovsky" because the last name is the same? I realize Ihor is Ukrainian, but am using him just for the sake of illustration—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); July 5, 2011; 17:29 (UTC)
I think that is right regarding "Western European" spellings, although there may be occasions, for example, where a Russian footballer signs for a German club and his name is published according to German romanization, which probably we should accept until the name becomes used in English media. But more importantly, these "method" rules should be in NCRUS for as long as the split is accepted, with this page documenting the WP romanization system and the other page defining the circumstances when it is to be used. Sussexonian (talk) 23:22, 7 July 2011 (UTC)[reply]
I still don't understand. I can see how we could occasionally have situations where such a player is not covered by any English sources but is covered by many more sources in German than in Russian, in which case using the German spelling makes sense, but if there are more reliable sources in Russian than in German (perhaps if the player only signed up for one season and didn't do much beyond the ordinary playing), why would we still stick to German? Alternatively, why is the spelling in Western European languages more important than the spelling in, say, Eastern European languages (many of which also use the Latin alphabet)? What if the said player signs up for a Romanian or Polish club?
I do agree on the split, though.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); July 8, 2011; 13:43 (UTC)
There is no "we". I have no interest in the footballer question, but if an editor specializing in footballers creates such an entry I don't think "we" should spend time changing the spelling of that person when there is no good reason to do so. But I don't suggest placing such a rule on the page. In general if a Russian person has no coverage in English language, any article would probably be created from Russian sources so WPROM would be used, no question. Sussexonian (talk) 09:01, 9 July 2011 (UTC)[reply]
I don't have any interest in soccer players either, and what worries me isn't an occasional article that an editor specializing in soccer might put up under a (say) German spelling for no obvious reason. However, I'd be concerned if such an approach became systematic with this guideline serving as substantiation ("...but romrus says we are supposed to use the spelling in 'Western European languages'"; that kind of stuff). It'd be good to have an explanation of the reasoning behind the "Western European languages" choice, but I can see it's not forthcoming.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); July 11, 2011; 14:02 (UTC)
I agree with Ezhiki. I would go even further and say that an article not named according to the NC can be renamed. A systematic approach helps to reduce errors, like duplicated articles, red links etc. E.g. if an article is at the German spelling then there is the possibility that there are red links due to missing redirects. There is also a higher chance of false links, because it gets undetected that there are two articles with persons having the same name in Russian, but residing around under different Romanized names. I would favor having a dab in these cases under the ROMRUS romanized name, not sure this is already mentioned in WP:NCRUS. Bogdan Nagachop (talk) 13:58, 28 July 2011 (UTC)[reply]

Join WT:NCRUS - Wikipedia talk:Naming conventions (Russia)

Invitation to join the talks at

They result from Russia related text imported to WP:NCRUS from WP:NCGN Bogdan Nagachop (talk) 09:25, 29 June 2011 (UTC)[reply]

I've posted a courtesy notice at WT:RUSSIA to inform about the continuing discussion both here and on the NCRUS talk page.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); July 1, 2011; 15:16 (UTC)

Motion to close

Apart from Bogdan's remarks on July 28, it has now been three weeks since the last substantial comment. I motion to close this discussion.

Having reviewed the comments so far, I would say that the new wording is mostly supported by all participants. The only point where some disagreement can still be seen is regarding the way we should treat the letter "ё". To that effect we have PMA's opinion that it should be romanized as "e" in most, if not all, cases, Sussexonian's opinion that it should be romanized as "yo" in all cases with the exception of the endings of the last names, Bogdan's and my procedural opinions that it should be always (where "always"="when this guideline is being applied directly", not "regardless of usage") be romanized as "yo" (as per the 2005/2006 rules), Greyhood's support for the 2005/2006 rule, and Mlm42's neutral !vote. I would say this is leaning to re-instating the 2005/2006 rule without any special provisions such as the endings of the last names. If anyone disagrees with this assessment, we can invite an uninvolved admin to re-evaluate the status of this particular item; otherwise I motion to remove the "-ёв" ending provision, remove the "under discussion" tag from the "ё" provision, move the Usage section to WP:NCRUS where it belongs and can be discussed in more detail, and change the status of this page to "active guideline". Note also that anyone feeling strongly about the "yo" clause is welcome to initiate a new discussion regarding just that clause—there's little point in holding up the whole guideline because of one item. Comments, clarifications, seconds?—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); August 2, 2011; 14:14 (UTC)

Closing straw poll

I took a look to close, but the consensus is not clear as there has been a lot of conversation, and there are still {{under discussion}} tags on the page. So before closing I'd like a show of hands. I'll keep the poll open for seven days at least - if there is insufficient response I'll close this no consensus, though if there is active discussion/polling I'll keep the poll open. SilkTork ✔Tea time 16:21, 4 September 2011 (UTC)[reply]

  • Simply number and sign under the appropriate statement(s):-
I support applying {{WikiProject style advice}} to this page


I support asking the community to discuss making this a WP:GUIDELINE via a notice on WP:CENT


I support working further on the page to iron out the remaining issues


Additional comments on the closing straw poll

To keep matters simple, use this section for any additional comments. SilkTork ✔Tea time 16:21, 4 September 2011 (UTC)[reply]

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Romanization_of_Russian&oldid=448425876"