Template talk:Braille

Variants
Really, we should keep languages apart from scripts. e.g. "Russian" sho9ulds note "Cyrillic" somehow logically. -DePiep (talk) 22:38, 26 April 2012 (UTC)


 * I'm not sure I understand, but I changed the header from 'languages' to 'scripts'. — kwami (talk) 23:38, 26 April 2012 (UTC)


 * Except there isn't a single Cyrillic braille. Rather, languages have their own braille encoding for Cyrillic, based on the particular subset they use, the phonetic value of the different letters, and the best matches to the French/English phonetic archetypes for unified braille. Russian and Serbo-Croat braille assign different patterns to the same Cyrillic letters, and the same cells to different letters. The Latin script seems to be the only one that assigns braille cells to letters, regardless of the phonology of the language being written. VanIsaacWScontribs 23:49, 26 April 2012 (UTC)
 * This version I prefer. Still, from the sources (including old UNICEF), I get the feeling that "English braille" diffs from "French braille"?, and so in the template. Or: "Russian brallle" diffs from "Ukrainian braille"? (while both Cyrillic). But I am not brave enough, correctly, to edit so.
 * In general: we should describe Brailles by script (punctuation variants expected), not by language. At the moment, the template is mixed in this (worst of all). -DePiep (talk) 23:53, 26 April 2012 (UTC)
 * If we zoom in to the Cyrillic script. There are about a hundred of languages (and dozens of different alphabets too, as we know) Languages written in a Cyrillic alphabet. I'd say: if you can prove that language "Z" uses a diferent Braille-list (other that Russian), you go ahead and distinguish that one language. I expect that all Slavic languages that use Cyrillic &etc., will use the "Russian" Braille. My round up: Braile is by script, but exceptions by language to be expected. -DePiep (talk) 00:02, 27 April 2012 (UTC)
 * The problem is that the presumption of uniformity does not bear up to the evidence. Serbian and Macedonian are Slavic languages that use Cyrillic, and they use a different braille list than Russian. Your expectation is wrong. VanIsaacWScontribs 00:42, 27 April 2012 (UTC)
 * Bulgarian is also different, although minimally so. There is no general Cyrillic braille. It is completely unique to language. VanIsaacWScontribs 00:52, 27 April 2012 (UTC)
 * Even then: group languages within a script. Why not in a good Braille article:

(free examples:) -DePiep (talk) 01:56, 27 April 2012 (UTC)


 * Because it suggests a genetic relationship between those different languages' braille systems that is both false and misleading. The real question is why in the world would you do that? We have a very good example of a single braille system that is used to represent dozens of different written scripts. That's a good reason to have a single braille table. Obviously, written scripts are useless when it comes to braille - it's only about the individual languages. VanIsaacWScontribs 02:12, 27 April 2012 (UTC)
 * No it does not "suggest": it only says what is a fact. If not, then do not write it -- come on, what is "misleading and false"? How many Braille variants are there for languages that use Cyrillic? One per language really? Please. Even better: exactly the page you link to has such a table (albeit pivoted): Braille "a" (by ISO) has multiple (very many) associations. Now after looking at your own example, you still want a different Braille page for every language? -DePiep (talk) 02:36, 27 April 2012 (UTC)
 * No, I want one for every braille system, whether it represents a single script, a single language, dozens of scripts, dozens of languages, or any combination thereof. Of course the page I showed you has such a table. I was trying to show you exactly when doing that sort of thing is justified. If you can't understand the difference between the designed multi-script, multi-language, braille system of Bharati braille and codging together a bunch of independently developed braille systems because they happen to represent written forms that are related, then I don't know what to say. Russian and Serbian braille are no more related than French and Hindi braille. Treating them as a single system with variations is factually wrong and inherently misleading. VanIsaacWScontribs 03:16, 27 April 2012 (UTC)
 * Then, what is a "Braille system"? A script into dots, a language into dots, or something else maybe? Are there a hundred Braille systems for those hundred languages now written in Cyrilic? All developed independently? The just "happen" to use &#x0411;=⠒

all? What I say is: if there is a systematic overlap (whereever from), we should use that. -DePiep (talk) 03:25, 27 April 2012 (UTC)


 * I misunderstood you. When you said "script", I assumed you meant the Braille script. That's what the articles are about. Latin, Cyrillic, etc. are nearly irrelevant, because the Braille system is not based on them. Braille is its own family of scripts, and e.g. Russian Braille has its ultimate roots in Latin, not in Cyrillic. — kwami (talk) 03:28, 27 April 2012 (UTC)
 * Aha. Well, to me (and to Unicode) Braille is no script, because there is no definition like b=⠒ (this definition is only defined in English, or maybe more western alphabets). I think we are talking about: English braille or Latin braille? Russian braille or Cyrillic braille? etc. Now I cannot believe that all 1000's of languages have their own separate braille system. And also: I presume braille is based on the (local) written language. Hence my stress for "Cyrillic", not "Russian". But surely I can be wrong in this. -DePiep (talk) 03:36, 27 April 2012 (UTC)
 * Umm, Braille is a script in Unicode. From scripts.txt: . Actually, all of the Latin script languages use the same braille assignments. There was also an effort starting in the 1880s to base other international braille systems on the French and English sounds. This unified international braille is why Russian, Bulgarian, Serbian, and Macedonian braille are easily confused as being related to each other - they were all adapted from French/English phonology as part of the unified international braille system. Bharati braille and many Arabic script language brailles are also unified. In fact, Korean and Japanese (and to some extent, the Chinese brailles) are the only WP braille articles that aren't based on unified braille. VanIsaacWScontribs 03:56, 27 April 2012 (UTC)
 * I think he means that as far as Unicode is concerned, ⠃ is ⠃ no matter which script it belongs to. That is, it is not associated with any script. This is different from English B, Greek Β, and Cyrillic В, which have been assigned to different Unicode blocks because although they look the same they are ordered differently, have different italic and lower-case forms, etc. While such distinctions pertain to braille too, they haven't been incorporated into Unicode. — kwami (talk) 04:11, 27 April 2012 (UTC)
 * Correct (=OK), kwami. Unicode on Braille is in chapter 15.11. Braille does give the ISO 15924 script code (Brai) in Unicode, but Unicode also states (correctly imo) that the "⠃=b" idea is not prescribed. (so: for Unicode Braille is not a writing system. it needs a translating scheme. That scheme differs between languages & even scripts). I say. -DePiep (talk) 21:24, 27 April 2012 (UTC)
 * Yes, there are different Braille scripts. Most Braille alphabets are what I've been calling "Latin" Braille, where b=⠃ because 2=⠃ and b=2, which in general is only true in the Latin script (specifically in the French alphabet). So that is one Braille script, an alphabetic script, and it includes the English, French, Greek, Hebrew, and Russian alphabets. Korean Braille is a separate script, as is Japanese. Bharati, Chinese, and Cantonese, however, are part of the Latin Braille family, though they've been extended with additional letters, and in the case of Chinese and Cantonese, have changed in form to a semi-syllabary: graphically descended from Latin Braille, but conceptually descended from zhuyin and traditional Chinese phonology.
 * When I think of "scripts", this is what I mean: the different families of related Braille alphabets. Whether a braille alphabet is used to transcribe Cyrillic or Tibetan is about as essential as whether a Latin alphabet is used to transcribe Chinese (pinyin) or Hindi (IAST) or whatever.
 * Braille scripts are not distinguished in Unicode because they are irrelevant to computer encoding, because there can be no graphic distinction among them. Now, if there had been an established tradition of alphabetizing by braille, so that French ⠃ were ordered differently than Bharati ⠃, then we might have found it useful to have separate Unicode blocks for French and Bharati braille. — kwami (talk) 03:50, 27 April 2012 (UTC)

As for classifying braille alphabets depending on which linear script they encode, consider Bharati, where one braille script encodes various Indian scripts as well as Latin, and Serbian Cyrillic, which encodes Gaj's Latin—or does Gaj's encode Cyrillic? To me, classifying braille alphabets based on which linear script they encode is like classifying Serbian Cyrillic apart from other Cyrillic alphabets because it encodes Latin (or, equivalently, classifying Gaj's apart from Latin because it encodes Cyrillic). — kwami (talk) 04:22, 27 April 2012 (UTC)


 * Okay, so it looks like DePiep's opinion is that braille in not a script, so it only makes sense to classify it according to the scripts it encodes. VanIsaac's opinion is that it's a family of scripts that can be classified according to their inherent features. IMO both of you have a point, but only the latter is interesting. For DePiep's POV, we would simply have a table for braille codes, with a row of letter assignments in different alphabets for each one. With VanIsaac's POV, we would treat different braille encodings as alphabets in their own right. Since many of them have internal structure, I find the latter to be not just more interesting but more informative. Also, this template links to articles for different alphabets, so IMO it makes sense to classify them as different alphabets. — kwami (talk) 03:00, 28 April 2012 (UTC)
 * Thanks for the overview. I rest my POV. Please go ahead & improve. (Worst case: I and others do not understand the outcome). -DePiep (talk) 00:01, 29 April 2012 (UTC)

Korean
Alphabetic, syllabic, or some weird hybrid? How do we indicate it? Should we indicate it? Stick your opinions below mine.


 * It builds identifiable syllabic units with positional alphabetic units: hybrid. Maybe its best if we eliminate the abugida/semi-syllabry/alphabet distinction, and just call them other scripts. VanIsaacWScontribs 00:03, 27 April 2012 (UTC)
 * If I can help you: don't mix up language with script, not even in Braille. But do keep the writing system that a Braille describes (like Korean). If a Braille scheme describes a syllabic language/writing system, who are we to alter that? -DePiep (talk) 00:08, 27 April 2012 (UTC)

Korean Braille is an alphabet which marks syllable boundaries. Many traditions of romanized Chinese separate all syllables with hyphens, yet we don't argue that the Latin alphabet is a syllabary when used that way, nor some weird hybrid. Similarly with many North American languages: Osage, for example, where periods are placed between syllables despite it being otherwise a 'true' alphabet. Now, Chinese Braille–that's a hybrid, partially alphabetic and partially syllabic. — kwami (talk) 03:39, 27 April 2012 (UTC)

Braille Cell upgrades
The Braille Cell template has a new upgrade. You can now indicate multiple-cell braille patterns with a single call of, up to 20 cells. It still takes the exact same named parameters, ie type=, size=, alt=, text=, which are applied to all cells. I may enable cell specific parameters if anybody wants them. Please post any comments at the template talk page, but I thought the people watching this page would want to know about it as well. VanIsaacWScontribs 06:33, 29 April 2012 (UTC)

Tibetan braille
So, I got ahold of Braille without borders, and they sent me a map of the Tibetan braille script. Although it doesn't have any dependent vowel marks, or subjoined consonant indications, it gives a basic overview. I'll be putting it up as an article some time in the next couple of days, but from what I can see, it has some correspondence to other braille systems, but does not appear to assign codes in the systematic manner of Unified international braille systems. It seems to be along the lines of the Chinese brailles - some effort was made at international standardization, but it was created with some other goals in mind. It departs significantly from Bharati. VanIsaacWScontribs 01:23, 3 May 2012 (UTC)
 * I think the major aspects that are non-unified are: Tibetan ca/cha/ja/nya are unified th/y/q/a2, Tibetan pha/tsa/tsha/dza are s2/x/z/dh2, and Tibetan za/'a are unified sa/d2. Sa and Nga are outside the unified braille suggestions, despite many appropriate patterns, although Nga is analogous to Ga. That's 10 to 12 signs that are contrary to Unified braille assignments, out of the 34 base letters. Compare with Bharati's non-necessary deviation of only bha, out of 50 single-cell, non-conjunct letters. That's a significant statistical difference between the two, and I think it's the difference between a system designed as part of the unified international braille, and one that just borrows and invents from other systems at-will. VanIsaacWScontribs 11:20, 3 May 2012 (UTC)


 * I don't think UB is a set convention. Or at least I haven't seen any pubs which indicate that. The UNESCO doc just listed common letters, and gave approximate IPA values for them. Many of them are listed because they occur in Bharati, so it's tautologous to argue that Tibetan is not UB because it's not as close to UB as Bharati is: all that tells us is that Tibetan is not Bharati. But it sounds like the basic Latin 26 are unified, more or less, which would mean that it's still part of the UB declaration that Braille alphabets should follow the French Braille order. If you look at Tibetan pinyin you'll see similar disparities from 'unified Latin', but it's still a Latin alphabet. — kwami (talk) 22:04, 3 May 2012 (UTC)


 * There is a degree of idiosyncrasy, but y=j and j=q are common correspondences. Also, the English th sounds are assigned to Tibetan sibilants, which is how a lot of people who haven't mastered English pronounce them. So what's really off? ny = a2, ts = x, ' = d2, ph = s2. But the last is also p + dot 6; in Tibetan script (as in almost all Brahmic scripts), ph is derived by adding a stroke to p. Also, given that x is a sibilant in pinyin, that's not so far off either. (Or maybe it was just a leftover letter.) So the only completely arbitrary assignments seem to be ny=a2 and '=d2. That's still the French braille assignment which defines UB. German Braille is further from our UNESCO assignments than Tibetan Braille is. — kwami (talk) 22:29, 3 May 2012 (UTC)


 * Do you have access to the '54 Unesco World Braille document, because it is significantly more pointed than the '90 document on the origin, development, and purpose of unified braille; namely that it is a set convention on braille assignments. It doesn't just list common letters with IPA values, it spells out the correspondences between braille systems, and recommended changes to already existing braille systems. Furthermore, it is pretty clear that unified braille was developed at the conferences of 1878, 1902, and 1911, while Bharati was developed in 1952 on the basis of the already existing international standard - clearly, Bharati had no input on the structure of unified braille. The correspondences of standard patterns for extra related sounds is established well beyond just Indic scripts, and the correspondences are specifically designed for Bharati to match unified braille, not the other way around. Those correspondences are intentional, not tautological.


 * The idiosyncrasies of braille systems are exactly what unified braille was designed to eliminate. I also never said that the non-standard symbols were arbitrary; just that they're not unified braille. It may be completely logical (I believe it is) to associate the symbol of Ng to G, but it's not unified braille. Basing the Ph braille sign on the P might, in fact, be perfectly justifiable, but it's not unified with the international standard for braille assignments. This is not a value judgement, and it's not a game of spotting motivations, it's simply whether the system adheres to the international standard. Having a third of your braille assignments contrary to that standard means you didn't follow it - you used other criteria and associations to develop it. That means it's not unified. VanIsaacWScontribs 14:01, 4 May 2012 (UTC)


 * I only have the 1990 edition.
 * Take German. We have:
 * äu = i², au = ch, ch = th, ei = sh, eu = gh, ie = ng, ll = q, mm = x [typo? x also = x], q = r² [typo?], ss = dh [like Tibetan], sch = h² [like Tibetan], st = t²
 * Should we say that German braille is not part of UB? Or French:
 * é = dh², à = apostrophe, è = dh, ù = t², â = ch, ê = gh, î = s², ô = th, û = sh², ë = d², ï = r²
 * Even Bharati braille has multiple irregularities, such as Q for /kṣ/ rather than for /q/. And Sinhalese has all sorts of odd assignments. If we expect a 100% agreement, I wonder if there would be any extended UB alphabets? We'd also need to correct the UB article. — kwami (talk) 22:48, 4 May 2012 (UTC)


 * Well, I would suggest two things: One, the unified braille is not a limitation on what letters/sounds can be assigned to a braille pattern, but rather a prescription for what braille patterns should be assigned if those letters/sounds occur in a particular language. Notice the direction of assignment there - if you have a second "r", it should be assigned the second R braille, but that doesn't mean that the only acceptable value for the second R braille is an "r" sound. The fact is that the only anomalies I can find in the German standard are "sch" and "ß", which would probably be a best fit at Sh and 2nd S. That having been said, I've never seen anything in either Unesco document that suggests that the world braille is intended to say anything about digraphs/ligatures or accented letters ("ß" is an "sz" ligature). Even the one that looks really bad in German, "ch" = Th, not Ch, is blunted by the fact that the world braille value is for the /tʃ/ "ch" sound, not the /x/ sound that "ch" has in German. The German "ch" would be a best fit at X, except that its already taken by that letter of the Latin alphabet, as is standard to all of the Latin script brailles, no matter the phonetic value. Furthermore, I would actually be prepared to argue that French braille may not best be considered a form of unified braille - it was developed and standardized decades before there even was an international braille standard, and the international braille has many assignments in its vowel inventory that are reflections of English braille, not French. However, I don't have a comprehensive mapping of all 63 French braille patterns to say such a thing definitively, so the second I braille may actually have an "i" sound that I don't know about, and the assignment of "î" and "ï" to Sh and 2nd R are just fallbacks. Like I said before, I'm not entirely sure that the international braille was even intended for digraphs, ligatures, or accented letters, so I can't even argue that point to any extent. All I can say is that I don't think it's justifiable to consider Tibetan braille just an idiosyncratic system developed as a unified international braille; It has far too many anomalous assignments of its basic alphabet for them to be aiming for a unified braille and fall so short of the mark. VanIsaacWScontribs 00:18, 5 May 2012 (UTC)
 * I just realized: you developed the Unified international braille article from the "Phonetic Symbols and Braille Signs" chapter from the '90 Unesco document. I've cited the '54 report on that page now - you really need to get ahold of it, as it is far more specific about the world braille standard. VanIsaacWScontribs 00:39, 5 May 2012 (UTC)

If digraphs are not included in Unified Braille, then Tibetan is not a problem, because ng etc. are digraphs in romanization. Also, Chinese is not an exception, because the rimes are not single letters in romanization either. (Chinese is an exception because it is not alphabetic, but it's still part of the UB family.) The 1954 doc quotes an early conference as defining universal braille as French braille, and also saying that "full European uniformity in uncontracted Braille was achieved in due course" (p. 26). Full uniformity was *not* achieved; the only uniformity was in the original 25 or 26 letters of the alphabet. That is, in the basic Latin alphabet. Assignments beyond that would seem to not be part of UB. Unless we have a pub that actually defines UB as more than those 26, and lists what they are, so we can check whether an assignment is UB or not? — kwami (talk) 03:18, 5 May 2012 (UTC)
 * I'm sorry, but Nga is most definitely not a digraph or ligature in Tibetan. Apart from Chinese brailles, none of these braille systems encode an intermediary romanization of the languages, just the actual script and how it is analysed and segmented by the language community. If you want to check on whether an assignment is UB or not, you go to page 74, the World Braille Chart, and it seems abundantly clear from the dozens of pages between page 26 and page 74, not to mention the braille charts from page 80 on, that the phonetic associations in 74-79 were used as the basis for unifying the scripts of Asia, Africa, India/Ceylon, Perso-Arabic, and Urdu. Not to mention, even if your basis for evaluating braille systems as unified is based merely on romanization, Tibetan braille fails quite badly at that task too - there's a C, J, Z, Y, and S that contradict the standard Latin script braille. VanIsaacWScontribs 04:20, 5 May 2012 (UTC)
 * They all approximately fit. Your "abundantly clear" is your opinion, not a demonstrable fact; the opposite is just as abundantly clear to me. Pinyin is just as bad a fit to European standards of Latin, but we don't therefore claim it's not Latin. Tibetan is part of the same family of alphabets as most of the rest of Braille. It is not a distinct script the way Korean is, nor is it a reordering the way Algerian is. — kwami (talk) 05:16, 11 May 2012 (UTC)


 * Obviously, I was talking about romanization from the standpoint of whether something should be considered a digraph or not. And the problem with your perspective is that the unified brailles, like Arabic, Hebrew, Bharati, and Armenian don't just approximately fit, like you said - they really fit. If you have a letter in one of those scripts that matches the sound archetype for the braille pattern, it's either assigned to that pattern, or another matching letter is. I look through those braille systems, and there is just no variation in that rule. Tibetan braille just plain fails at that basic litmus.


 * For the record, this is my take on whether something should be considered unified braille: Take the native script (pinyin/kana/hangul for Han ideographs). See if any basic letters are assigned to "wrong" patterns. If they are, see if all the best-fit patterns are taken by other good-fit letter/pattern combinations. If you have multiple wrong patterns, and there aren't good-fit letters interfering with those assignments, I think that means that it was not unified - it was designed from another perspective. It seems clear from Bharati that ligatures are not assigned on the same basis - क्ष "kṣa" / is probably a best fit for X -, but ऒ "o" was randomly assigned instead. Likewise, digraphs and accented letters in those Latin script brailles that we have documentation of do not seem to follow the unified braille assignments in any systematic way either, so native script (not transliteration, just the native script) digraphs, ligatures, and letters+accent pattern matches are not considered. So my question to you is what is your litmus for classification? VanIsaacWScontribs 01:10, 12 May 2012 (UTC)

I'm not sure I understand the question. But let's see. Looking at each braille letter and comparing its value with the one assigned to Unified Braille: So Tibetan is about as regular as Hindi or French, and not as far off as Sinhalese. (BTW, Tibetan ng is not a fail, because 2356 does not have an assigned value in Unified Braille, and so cannot be a mismatch. Unless you want to include punctuation, but then the counts for the others will go up too.)
 * Greek: fails 8
 * Russian: fails 8
 * Hebrew: fails 8
 * Arabic: fails 8
 * German: fails 8
 * French: fails 10
 * Hindi: fails 10
 * Sinhala: as Hindi +5
 * Tibetan: fails 11

My litmus is: does the alphabet follow the French pattern? For Tibetan, the answer is obviously yes. — kwami (talk) 04:29, 12 May 2012 (UTC)

Here's a comparison with French Braille: a-(a) b-p(*b)  k-kʰ  d-t(*d)  g-k(*g)  zero-h ʒ-j (same as German, Hebrew) k-k l-l  m-m  n-n  p-p q-tʃ(*dʒ) (same as Russian, close to Chinese, Hindi) r-r s-s(*z)  t-t ks-tʰ for x (French, English, Russian, Hindi also fail) j-tʃʰ (unique to Tibetan) z-tsʰ (close to German, Chinese) s-pʰ (unique to Tibetan) ɛ-s for th (better than French: s ~ th) y-tʰ for 2nd T (Tibetan matches, French fails) i-ʃ for sh (Tibetan matches, French fails) o-ts for th (better than French) y-ʃ for 2nd H (same as Russian, Hebrew, better than English, French, or Hindi) e-h for 2nd D (Russian, Hebrew also fail) w-w poetry-ɲ for 2nd A (Greek, Hebrew also fail)


 * I'm sorry, but your litmus is "does the alphabet follow the French pattern" and French fails 16% of the time? I'm completely stuck if you don't recognize how absurd that is. VanIsaacWScontribs 10:26, 12 May 2012 (UTC)

Requesting third opinion
I'm tired of fighting this thing, and you just reverted it again so I'm calling for a third opinion rather than edit warring over it.

Our problem is that the most recent source we have that explicitly lists unified braille systems is the '54 Unesco document (http://unesdoc.unesco.org/images/0007/000711/071103eb.pdf), and it has only French, English, Greek, Russian, Arabic, Hebrew, Devanagari (Bharati), Swahili, and Indonesian brailles in the World Braille chart. Any other braille system listed as unified braille is WP:OR, whether it is controversial (Tibetan braille), or not (Yugoslav braille). I offered a geographic-based compromise by which we could put the dispute behind us, and it was rejected without an offer for moving forward. It just got reverted again, so I'm asking for help. VanIsaacWS<sup style="margin-left:-3.0ex">contribs 03:33, 18 June 2012 (UTC)


 * Perhaps we should abandon the term "unified Braille", then. Tibetan is part of the French braille family, whatever you want to call it. — kwami (talk) 06:43, 18 June 2012 (UTC)


 * Whether you call it French or unified doesn't change that Tibetan does not seem to be based on it. VanIsaacWS<sup style="margin-left:-3.0ex">contribs 07:48, 18 June 2012 (UTC)


 * Of course it's based on it. You've said so yourself. I don't understand your argument. — kwami (talk) 08:33, 18 June 2012 (UTC)
 * Hi there. What does it mean to be part of the Unified Braille System? The WP page says that the standard was last updated 101 years ago, so I'm unclear on whether a unified system even exists anymore. — Bdb484 (talk) 14:10, 18 June 2012 (UTC)


 * Yeah, that may be a little misleading. They developed the unified braille back then, getting English and other European brailles unified, but we know that the process was continuing up through the 1950s (our last source of explicit listing of unified systems), with Bharati and Arabic. The "standard" was agreed way back in the 1800s, but the world braille council, who developed the system, has presumably been unifying braille systems ever since. VanIsaacWS<sup style="margin-left:-3.0ex">contribs 21:38, 18 June 2012 (UTC)


 * Placing it in 'other' would also be OR, since it would be a claim that it's not part of one of the existing categories. Not only that, but we know it's wrong. We're allowed to use common sense when organizing WP.
 * In the beginning, there were two systems: adopting the French order, modified to fit the new language, or reordering per the alphabetical order of the new language. Since then two other systems have been developed, Japanese and Korean, which are not based on sorting order. AFAICT, Algerian is the only re-ordered sorting-order system still in use; all others are based on the French order, though in some, such as Chinese, new letters outnumber the ones taken from French. — kwami (talk) 23:21, 18 June 2012 (UTC)
 * The thread above is not a debate about how to apply WP policy. It is a debate over your own understanding of the braille systems, e.g., the major aspects that are non-unified are: Tibetan ca/cha/ja/nya are unified th/y/q/a2, Tibetan pha/tsa/tsha/dza are s2/x/z/dh2, and Tibetan za/'a are unified sa/d2 [emphasis added]. Because this is your own personal interpretation, it is OR and should not be used to classify Tibetan braille.  You need to find a third party source that says how to classify Tibetan braille and use that in the nav box.  If you don't have such a source, then you don't have a NPOV way to do it.  Debating amongst yourselves without a reliable third party source isn't going to change that. Coastside (talk) 13:49, 19 June 2012 (UTC)

script classifications
How about this as an option?

It gets rid of the POV/OR on whether a particular system is unified or not, and is probably more useful for navigational purposes anyway. VanIsaacWS<sup style="margin-left:-2.9ex">contribs 00:27, 12 May 2012 (UTC)


 * I don't find it helpful. For me, the interesting thing about "X Braille" is not that it's used in country X, but what kind of system it is. In this table, the different systems are all scrambled up. I mean, imagine if we had a template for linear scripts where Lakhota (Latin), Cree (Syllabics), and Cherokee were all on the same line because they're all in North America, while Vietnamese (Latin), Japanese (kana/kanji), Hindi (Nagari), and Urdu (Arabic) were on another line, and Turkish (Latin), Arabic (Arabic), and Berber were on a third: how would that be helpful?


 * Most of the world's braille systems follow French Braille; we're at "Unified" because you didn't like "Latin", and now you have objections to "Unified", but whichever name we choose, they form an clear family that is reflected in historical accounts of our sources. Then there are reordering systems like Algerian, which are simply the numeric code allocated to the alphabet. I think that deserves a separate section, actually: there were more of them at one time, and we might expand some day. Then there are novel systems like Korean and Japanese, which are neither of the above. Orthogonal to that are systems which are not alphabetic, such as Chinese, though the non-alphabetic part is similar to the contractions in French Braille.


 * Maybe we can called the section on French-derived braille "French-derived"? — kwami (talk) 01:01, 12 May 2012 (UTC)


 * Except that this is a navigation template, not an "I think this is neat" template - that's what articles are for. So are we arguing about a classification that is irrelevant to the average user, eschewing that which is actually helpful? I'm trying to throw you a bone so we can get past the argument above, and it's disappointing that you aren't willing to help out. So the question I have for you is simple: if you had no a priori knowledge, and you were looking for a particular braille system, how would you organize the template; because that's what navboxes are for - helping people find information. If you were looking for the braille for a language that we don't have an article for, how would you organize it so that you don't have to read through the entire list a couple times to make sure it's not there? VanIsaacWS<sup style="margin-left:-2.9ex">contribs 10:15, 12 May 2012 (UTC)
 * I don't think the writing system shoud be leading here. First association should be: "Language X is written in Braille Y". The writing system is a descriptive property, not a defining property.Clearly there are under two dozen Braille languages (here at WP), they cold be together, sorted alphabetically. A clarifying extension for "Unified" will benecessary. -DePiep (talk) 10:35, 18 June 2012 (UTC)

Per DePiep's objections (here and elsewhere), I removed the typological classification (alphabet, abugida, semi-syllabary), and integrated English as just another language. Currently then we list three classes: ABC-ordered braille, re-ordered braille (only found today in Algerian, AFAIK), and non-numerically ordered systems (Japanese & Korean). — kwami (talk) 23:53, 21 July 2012 (UTC)

BTW, I can understand how "French-based" may not be the best way to convey "ABC order", since many are not directly based on French. That's just the best that I was able to do: I've repeatedly said that I'm open to suggestions for better wording. How's "International order"? Or "French-order-based scripts"? — kwami (talk) 03:11, 29 July 2012 (UTC)


 * The name really isn't the crux of the problem, even if "French based" is lacking. The problem is trying to shoehorn braille systems that are associated with each other by the common Latin letters with systems that are associated with each other - including a few of the Latin script systems - by phonetic values, and then trying to cram in other systems that aren't Latin script languages, but kind of fit with some of the Latin letter values - it doesn't hold. VanIsaacWS<sup style="margin-left:-3.0ex">contribs 03:44, 29 July 2012 (UTC)


 * So, you're saying that the braille unification of the 19th century is invalid? Because that's all this is. — kwami (talk) 18:28, 29 July 2012 (UTC)

BTW, Unesco here calls the first section the "International Braille" method, the second the "Concurrent Sequences" method (formerly used in the US with w; also in old Greek, 4 of the Arabic brailles, and in th Uniform Indian Braille of 1944), and the third "Scientific Rearrangement of the Signs [...]" (American, old German, Hebrew < 1938, Sinhalese < 1940, plus four in India, also China, Korea, Japan). Thus, if there are still objections to the wording "French-ordered", we have a RS here to call them "International Braille". — kwami (talk) 20:47, 23 August 2012 (UTC)

Need up-to-date confirmation
We need modern confirmation for all braille articles based on the Unesco doc. Current Greek braille, for example, has several diffs, not counting the typos that we've found. Just a heads-up that we probly shouldn't consider the articles reliably sourced until we can confirm the Unesco description. — kwami (talk) 08:19, 24 July 2012 (UTC)

I cannot confirm Cantonese or Armenian Braille. Or the Nigerian alphabets or Motu either, for that matter, but they are so little different from English Braille that I'm not too worried. Of the other langs on WP-fr, we don't have Galician (didn't bother, it's so similar to Spanish and Portuguese) or Faroese (a rather odd alphabet). It would be nice to get Georgian, and we need to confirm how Urdu braille is used in Pakistan. — kwami (talk) 21:52, 21 August 2012 (UTC)

Turns out that part of the problem of the Unesco docs is that they are only recommendations. In at least some cases the alphabets shown were never implemented. Maltese, for example. — kwami (talk) 21:54, 23 August 2012 (UTC)

3rd edition out
UNESCO (2013) World Braille Usage, 3rd edition. (thanks, VanIsaac)


 * All entries are covered. "1990" in the list below means that Unesco (2013) merely republishes Unesco (1990) without verification. (Unesco (1990) is an extremely poor source.) Some of these have been verified from other sources, but most require confirmation – Armenian and Georgian, for example; and in other cases there may be questions where Unesco is incomplete, as with Burmese, or apparently has the wrong alphabet, as with Dutch and Maltese. — kwami (talk) 03:26, 10 November 2013 (UTC)

Country entries: Armenian 1990, Azeri 1990, Bangla (cf. Indian Bangla), Belarus 1990, Bhutan: Dzongkha; Bosnian 1990, Bulgarian, Cambodian, Croatian, Czech, Den: Faroese, Greenlandic; Estonian, Eth: Amharic; Finnish, Georgian 1990, Ghana: Twi, Ga, Dagaare; Icelandic, India: Hindi, Tamil, Marathi, Gujarati, Bengali, Kannada, Punjabi, Assamese, Malayalam, Nepali, Oriya, Telugu, Urdu; Indonesian, Iran: Farsi; Israel: Hebrew; Kazakh, Kyrgyz, Lao, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy 1990, Maltese [error], Mongol, Myanmar: Burmese; Nepali (cf. India: Nepali), NZ: Maori; Nig: Hausa, Yoruba, Igbo; Norwegian, North Sami; Pakistan: Urdu (cf. India: Urdu); PNG 1990: Dobuan, Huli, Kuanua, Motu, Pidgin; Par: Guarani; Phil: Tagalog, Ilocano, Cebuano, Hiligaynon, Bicol; Polish, Samoan 1990, Serbian, Slovak, Slovene, SA: Afrikaans, Zulu, Xhosa, Sotho, Tswana, Venda, Swati; SK: Korean; Spain: Galician, Basque; Sri: Sinhala, Tamil (cf. India: Tamil); Tajik 1990, Thai 1990, Tibetan, Togo 1990: Bassa, Kabiye, Konkomba, Moba, Tem; Turkmen 1990, Ug: Ganda; Ukranian 1990, UK: Welsh; US: Hawaiian, Iñupiaq; Uzbek 1990, Viet, Zambia: Lozi, Bemba, Nyanja, Kaonde, Lunda, Vale, Tonga; Zimb: Shona

Albanian, Arabic, Cantonese, Catalan, Danish, Dutch, English, Esperanto, Ewe, French, German, Greek 1990, Greco-, Hungarian, Irish, Italian, Japanese, Kiswahili, Malay, Mandarin, Ndebele, Portuguese, Romanian 1990, Russian 1990, Spanish, Swedish, Turkish 1990


 * All confirmed alphabets with more than the letters of the basic braille alphabet have articles, though sometimes they are grouped together by country. All others are redirects.  Punctuation, where given, is covered by our articles, though for most redirects it's just a mention that it's the same as that of English Braille.
 * Missing punctuation: Maltese, Catalan, most Mongol, many unconfirmed alphabets. (A lot of punctuation is variable or inconsistent between sources.)

3rd edition still requires independent verification
It seems that the 2013 edition sometimes retains data from the 1990 edition without acknowledgement. It's more reliable than 1990, but contradicts RS's in several cases, such as Maltese and Dutch, and so still needs to be confirmed with additional sources. — kwami (talk) 07:23, 16 November 2013 (UTC)

⠃ ⠗ ⠇ for "braille"
Just curious: the titlebar now has ⠃ ⠗ ⠇ for "braille". Is this the more common writing? The Braille infobox shows the seven letters transposed, and for me (a layman) this looks better as an illustration (more to the point, so to say). -DePiep (talk) 17:46, 13 February 2014 (UTC)


 * Depends on your audience. ⠃⠗⠇ is normal English Braille.   is Grade-1 braille, which is more accessible cross-linguistically.  — kwami (talk) 22:37, 13 February 2014 (UTC)

Braigo
I added this template to the article for Braigo, but wasn't sure if I should add that link here. I did add it as a "see also" article at Braille embosser, which is in the technology section of this template. B7T (talk) 21:11, 4 May 2015 (UTC)

Cyrillic Braille subfamily (and Other non-Roman mediated Brailles)
I think it's fairly safe to propose a 'Cyrillic subfamily' within the French lineage of Braille systems. Almost every nations using Cyrillic as their national script, namely Bulgaria, Soviet Union (and its member SSRs) and Mongolia, modeled their Braille system on the imperial Russian Braille - except Yugoslavia, which preferred the interchangeability between Roman and Cyrillic writing systems just like in their printed books, but the usage of Cyrillic in Yugoslavia is relatively limited than the usage of Roman (used only in North Macedonia and (partly) Serbia).

In contrast, I doubt such subfamily could be proposed for, e.g. Arabic script - Persian, Arabic, and (old) Urdu Braille are developed separately, not elaborating the 'intra-scriptal uniformity' but rather following the phonological and pragmatic principles. Same goes for the Tibetan and Dzonkha - even though they use the same Tibetan script in printed books, which in turn a part of Indic script family, their Braille are developed separately and are shockingly different.

Despite that, I believe a section named 'other non-Roman mediated Brailles' within the French-based Braille lineage can be helpful (even though a bit misleading) for purely navigation convenience.

HighVoltage 06:15, 24 February 2020 (UTC)

Edit: The above description about Perso-Arabic related Braille systems was based on a misunderstanding. Hence, I propose another subfamily, namely Perso-Arabic. I located it between the Russian-based Cyrillic systems and the Bharati Braille, according to the chronological order.

In conclusion, I propose the below:

HighVoltage 07:18, 24 February 2020 (UTC)