Lfn-nepal
Mesajes
Introdui e viajores
Multe grasias per tua labora sur esta du pajes. Los apare aora en la presenta xef a elefen.org. — Simon 2021/03/11 09:35
Gramatica nepali es publicida
Lodas a tua completi rapida de la tradui nepali de la gramatica! Me ia ajunta tua pajes a la presenta xef. (Me ia usa codigos fea per la sinias devanagari en la adirijes de la pajes: cisa me va pote reali un plu bon sistem de trascrive roman, simil a lo cual nos usa en la gramatica rusce. Ma acel es un taxe per me a un otra dia.) — Simon 2021/03/18 21:45
- Alo Simon! I’m sorry for typing this in English, but I’m still learning elefen and cannot understand long sentences properly. However, I do have a PDF file and an EPUB file for the grammar of elefen in Nepali. Please let me know how I can upload those to the website, in English, if it’s not a trouble for you. Thank you! — LFN-Nepal 2021/03/19 04:42
- It’s even more impressive, then, that you have done all this work! The PDF and EPUB would be excellent to have. If each file is no larger than 2MB, I suggest you upload them to the wiki using the “Fixador” link at the top of this page – click on the “Carga” tab, then “Eleje fixes” to select the two files, and click the “Carga” button to upload them. I can then take them from there and copy them to the website. — Simon 2021/03/19 08:05
- To translate what I said above:
- “Nepali grammar is published. Congratulations on your rapid completion of the Nepali translation of the grammar! I have added your pages to the main presentation. (I have used ugly codes for the Devanagari characters in the addresses of the pages: perhaps I’ll be able to implement a better system of Romanisation, similar to the one we use in the Russian grammar. But that is a task for me on another day.)”
- Well, today is another day, and I have indeed implemented a system of Romanisation that is at least vaguely readable 🙂 Your page on spelling and pronunciation is now at https://elefen.org/gramatica/ne/haijajae_ra_uchachaarana.html, and so on. This system should also work well with the dictionary search engine, if you start to add Nepali entries there. — Simon 2021/03/19 11:35
- By the way, there are a few places where links are broken. For example, the link to la in the opening section on the determiners page tries to jump to a section called “artículos”. It might be worth you checking through all the links. — Simon 2021/03/19 11:42
- I have added the PDF file; however, for some reason, I am not able to upload the EPUB file. I will check the links, thank you. The romanisation has some errors, so it must be corrected if possible. The correct romanisations are: hijje_ra_uchchaaran, vaakya, naam, nirdhaarak, sarvanaam, visheshan, kriyaavisheshan, kriyaapad, sambandha_kaarak, samyojak, prashna, upavaakya, gananaakram, shabdarachanaa, sankshepak_chihna, lekhya_chihna. Thank you! — LFN-Nepal 2021/03/19 13:55
- I have copied the PDF file to the main area. I took the romanisation from code in this project. I can’t easily read Devanagari, although I have some understanding of how it works. In order to correct the romanisation, I would need a complete list of all the characters and what the romanisation of each one should be. — Simon 2021/03/19 22:15
- In fact, why don’t I provide you with what my romanisation code currently says, and you can simply correct it here? That’ll be easiest! Here we go: — Simon 2021/03/19 22:18
// nepali - transcrive prendeda de https://pypi.org/project/nepali-roman/ 'क्' => 'k', 'ख्' => 'kh', 'ग्' => 'g', 'घ्' => 'gh', 'ङ्' => 'n', 'च्' => 'ch', 'छ्' => 'chh', 'ज्' => 'j', 'झ्' => 'jh', 'ञ्' => 'n', 'ट्' => 't', 'ठ्' => 'th', 'ड्' => 'd', 'ढ्' => 'dh', 'ण्' => 'n', 'त्' => 't', 'थ्' => 'th', 'द्' => 'd', 'ध्' => 'dh', 'न्' => 'n', 'प्' => 'p', 'फ्' => 'ph', 'ब्' => 'b', 'भ्' => 'bh', 'म्' => 'm', 'य्' => 'y', 'र्' => 'r', 'ल्' => 'l', 'व्' => 'w', 'श्' => 's', 'ष्' => 's', 'स्' => 's', 'ह्' => 'h', 'ष' => 'sa', 'स' => 'sa', 'का' => 'kaa', 'को' => 'ko', 'कौ' => 'kau', 'कि' => 'ki', 'की' => 'kee', 'कु' => 'ku', 'कू' => 'koo', 'के' => 'ke', 'कै' => 'kai', 'कं' => 'kam', 'खा' => 'khaa', 'खो' => 'kho', 'खौ' => 'khau', 'खि' => 'khi', 'खी' => 'khee', 'खु' => 'khu', 'खू' => 'khoo', 'खे' => 'khe', 'खै' => 'khai', 'खं' => 'kham', 'गा' => 'gaa', 'गो' => 'go', 'गौ' => 'gau', 'गि' => 'gi', 'गी' => 'gee', 'गु' => 'gu', 'गू' => 'goo', 'गे' => 'ge', 'गै' => 'gai', 'गं' => 'gam', 'घा' => 'ghaa', 'घो' => 'gho', 'घौ' => 'ghau', 'घि' => 'ghi', 'घी' => 'ghi', 'घु' => 'ghu', 'घू' => 'ghoo', 'घे' => 'ghe', 'घै' => 'ghai', 'घं' => 'gham', 'ङा' => 'naa', 'ङो' => 'no', 'ङौ' => 'nau', 'ङि' => 'ni', 'ङी' => 'nee', 'ङु' => 'nu', 'ङू' => 'nii', 'ङे' => 'ne', 'ङै' => 'nai', 'ङं' => 'nam', 'चा' => 'chaa', 'चो' => 'cho', 'चौ' => 'chau', 'चि' => 'chi', 'ची' => 'chee', 'चु' => 'chu', 'चू' => 'choo', 'चे' => 'che', 'चै' => 'chai', 'चं' => 'cham', 'छा' => 'chhaa', 'छो' => 'chho', 'छौ' => 'chhau', 'छि' => 'chhi', 'छी' => 'chhee', 'छु' => 'chhu', 'छू' => 'chhoo', 'छे' => 'chhe', 'छै' => 'chhai', 'छं' => 'chham', 'जा' => 'jaa', 'जो' => 'jo', 'जौ' => 'jau', 'जि' => 'ji', 'जी' => 'jee', 'जु' => 'ju', 'जू' => 'joo', 'जे' => 'je', 'जै' => 'jai', 'जं' => 'jam', 'झा' => 'jhaa', 'झो' => 'jho', 'झौ' => 'jhau', 'झि' => 'jhi', 'झी' => 'jhee', 'झु' => 'jhu', 'झू' => 'jhoo', 'झे' => 'jhe', 'झै' => 'jhai', 'झं' => 'jaum', 'ञा' => 'naa', 'ञो' => 'no', 'ञौ' => 'nau', 'ञि' => 'ni', 'ञी' => 'nee', 'ञु' => 'nu', 'ञू' => 'noo', 'ञे' => 'ne', 'ञै' => 'nai', 'ञं' => 'nam', 'टा' => 'taa', 'टो' => 'to', 'टौ' => 'tau', 'टि' => 'ti', 'टी' => 'tee', 'टु' => 'tu', 'टू' => 'too', 'टे' => 'te', 'टै' => 'tai', 'टं' => 'tum', 'ठा' => 'thaa', 'ठो' => 'tho', 'ठौ' => 'thau', 'ठि' => 'thi', 'ठी' => 'thee', 'ठु' => 'thu', 'ठू' => 'thoo', 'ठे' => 'the', 'ठै' => 'thai', 'ठं' => 'thum', 'डा' => 'daa', 'डो' => 'do', 'डौ' => 'dau', 'डि' => 'di', 'डी' => 'dee', 'डु' => 'du', 'डू' => 'doo', 'डे' => 'de', 'डै' => 'dai', 'डं' => 'dam', 'ढा' => 'dhaa', 'ढो' => 'dho', 'ढौ' => 'dha', 'ढि' => 'dhi', 'ढी' => 'dhee', 'ढु' => 'dhu', 'ढू' => 'dhoo', 'ढे' => 'dhe', 'ढै' => 'dhai', 'ढं' => 'dham', 'ता' => 'taa', 'तो' => 'to', 'तौ' => 'tau', 'ति' => 'ti', 'ती' => 'tee', 'तु' => 'tu', 'तू' => 'too', 'ते' => 'te', 'तै' => 'tai', 'तं' => 'tam', 'था' => 'thaa', 'थो' => 'tho', 'थौ' => 'thau', 'थि' => 'thi', 'थी' => 'thee', 'थु' => 'thu', 'थू' => 'thoo', 'थे' => 'the', 'थै' => 'thai', 'थं' => 'tham', 'दा' => 'daa', 'दो' => 'do', 'दौ' => 'dau', 'दि' => 'di', 'दी' => 'dee', 'दु' => 'du', 'दू' => 'doo', 'दे' => 'de', 'दै' => 'dai', 'दं' => 'dam', 'धा' => 'dhaa', 'धो' => 'dho', 'धौ' => 'dhau', 'धि' => 'dhi', 'धी' => 'dhee', 'धु' => 'dhu', 'धू' => 'dhoo', 'धे' => 'dhe', 'धै' => 'dhai', 'धं' => 'dham', 'ना' => 'naa', 'नो' => 'no', 'नौ' => 'nau', 'नि' => 'ni', 'नी' => 'nee', 'नु' => 'nu', 'नू' => 'noo', 'ने' => 'ne', 'नै' => 'nai', 'नं' => 'nam', 'पा' => 'paa', 'पो' => 'po', 'पौ' => 'pau', 'पि' => 'pi', 'पी' => 'pee', 'पु' => 'pu', 'पू' => 'poo', 'पे' => 'pe', 'पै' => 'pai', 'पं' => 'pam', 'फा' => 'phaa', 'फो' => 'pho', 'फौ' => 'phau', 'फि' => 'phi', 'फी' => 'phee', 'फु' => 'phu', 'फू' => 'phoo', 'फे' => 'phe', 'फै' => 'phai', 'फं' => 'pham', 'बा' => 'baa', 'बो' => 'bo', 'बौ' => 'bau', 'बि' => 'bi', 'बी' => 'bee', 'बु' => 'bu', 'बू' => 'boo', 'बे' => 'be', 'बै' => 'bai', 'बं' => 'bam', 'भा' => 'bhaa', 'भो' => 'bho', 'भौ' => 'bhau', 'भि' => 'bhi', 'भी' => 'bhee', 'भु' => 'bhu', 'भू' => 'bhoo', 'भे' => 'bhe', 'भै' => 'bhai', 'भं' => 'bham', 'मा' => 'maa', 'मो' => 'mo', 'मौ' => 'mau', 'मि' => 'mi', 'मी' => 'mee', 'मु' => 'mu', 'मू' => 'moo', 'मे' => 'me', 'मै' => 'mai', 'मं' => 'mam', 'या' => 'yaa', 'यो' => 'yo', 'यौ' => 'yau', 'यि' => 'yi', 'यी' => 'yee', 'यु' => 'yu', 'यू' => 'yoo', 'ये' => 'ye', 'यै' => 'yai', 'यं' => 'yam', 'रा' => 'raa', 'रो' => 'ro', 'रौ' => 'rau', 'रि' => 'ri', 'री' => 'ree', 'रु' => 'ru', 'रू' => 'roo', 'रे' => 're', 'रै' => 'rai', 'रं' => 'ram', 'ला' => 'laa', 'लो' => 'lo', 'लौ' => 'lau', 'लि' => 'li', 'ली' => 'lee', 'लु' => 'lu', 'लू' => 'loo', 'ले' => 'le', 'लै' => 'lai', 'लं' => 'lam', 'वा' => 'vaa', 'वो' => 'vo', 'वौ' => 'vau', 'वि' => 'vi', 'वी' => 'vee', 'वु' => 'vu', 'वू' => 'voo', 'वे' => 've', 'वै' => 'vai', 'वं' => 'vam', 'शा' => 'shaa', 'शो' => 'sho', 'शौ' => 'shau', 'शि' => 'shi', 'शी' => 'shee', 'शु' => 'shu', 'शू' => 'shoo', 'शे' => 'she', 'शै' => 'shai', 'शं' => 'sham', 'हा' => 'haa', 'हो' => 'ho', 'हौ' => 'hau', 'हि' => 'hi', 'ही' => 'hee', 'हु' => 'hu', 'हू' => 'hoo', 'हे' => 'he', 'है' => 'hai', 'हं' => 'ham', 'क' => 'ka', 'ख' => 'kha', 'ग' => 'ga', 'घ' => 'gha', 'ङ' => 'na', 'च' => 'cha', 'छ' => 'chha', 'ज' => 'ja', 'झ' => 'jha', 'ञ' => 'na', 'ट' => 'ta', 'ठ' => 'tha', 'ड' => 'da', 'ढ' => 'dha', 'ण' => 'na', 'त' => 'ta', 'थ' => 'tha', 'द' => 'da', 'ध' => 'dha', 'न' => 'na', 'प' => 'pa', 'फ' => 'pha', 'ब' => 'ba', 'भ' => 'bha', 'म' => 'ma', 'य' => 'ya', 'र' => 'ra', 'ल' => 'la', 'व' => 'wa', 'श' => 'sha', 'ष' => 'sha', 'स' => 'sa', 'ह' => 'ha', 'ँ' => 'n', 'ं' => 'm', 'ः' => 'h', 'अ' => 'a', 'आ' => 'a', 'इ' => 'i', 'ई' => 'i', 'उ' => 'u', 'ऊ' => 'u', 'ऋ' => 'ri', 'ए' => 'e', 'ऐ' => 'ai', 'ओ' => 'o', 'औ' => 'au', 'ा' => 'a', 'ि' => 'i', 'ी' => 'ee', 'ु' => 'u', 'ू' => 'oo', 'ृ' => 'ri', 'े' => 'e', 'ै' => 'ai', 'ो' => 'o', 'ौ' => 'au', // virāma, danda (estas no ia es en la codigo de nepali-roman) '्' => '', '।' => '',
- I have made some corrections to the romanisation scheme. However, it shows only the main letters that are used in Nepali. Devanagari is an abugida, so different letters can be joined together to form different conjunct consonants. So while the scheme properly displays the romanisation of letters used in the language, it does not show the romanisation of conjunct consonants. Furthermore, the letter व is pronounced as ‘w’ or ‘v’ depending upon the word, but I have put ‘v’ according to the IAST romanisation. — LFN-Nepal 2021/03/20 04:08
- Ah yes, I had entirely forgotten about the conjuncts. As a result, I’m generating far too many instances of the default vowel a. Do I just need to process the virāma ् better? I think the problem is that, in a word like प्रश्न – which is actually held as the sequence [प ् र श ् न] – I’m romanising each of those seven characters separately. प्र is in the list above, but प gets converted first (to pa), and the virāma is then converted to nothing. (The fact that प्रश्न looks like it’s only two characters, two of which are conjunct, is the result of clever font trickery, so I don’t think I need to worry about that.) — Simon 2021/03/20 08:07
- However, I think our dictionary engine will require a blunter approach. Its searches take place in romanisations – actually pure ASCII: not even an É or an Ü is allowed! – and it then reconstructs the correctly accented form based on what it has found. This is highly convenient in most European languages. But the virāma is different. For the dictionary, I’ll probably have to represent it with an otherwise unused letter, such as x, so that प्रश्न would get treated as paxrashaxna behind the scenes. That shouldn’t be a problem, because the romanisations are never shown to the dictionary’s users. When the user types प्र, the software will convert this to paxra and thus find paxrashaxna, which it will convert back to प्रश्न for display. — Simon 2021/03/20 08:07
- That sounds like a great idea! Nepali phonology is a tricky one, as the final schwa, and in some cases the middle schwa, is deleted in some words but it kept as it is on other words. प्रश्न (question) and क्रियापद (verb) are perfect examples of such words, as the schwa is retained in प्रश्न and is pronounced as “prashna” where as the schwa is deleted in क्रियापद and is pronounced as “kriyaapad”. This random deletion of the schwa creates a problem when it comes to romanisation, as the correct romanisation based on native pronunciation becomes “kriyaapad” whereas the romanisation based on orthography becomes “kriyaapada”. Please let me know what we can do about this. — LFN-Nepal 2021/03/20 09:36
- Indeed. I was just reading up about this, on noticing that you said उच्चारण should be uchchaaran, not uchchaarana. I don’t think we can do much about this, nor do I think there’s any reason to do anything about it. Devanagari orthography is based on Sanskrit, where all implicit vowels are pronounced, and speakers of modern languages like Nepali are presumably very accustomed to suppressing this vowel in the appropriate places when reading. So the same should be true in the romanisation. Besides, I’m not suggesting that anyone reads pages of text in romanisation – it’s merely a way to generate convenient URLs. — Simon 2021/03/20 11:10
- I also have a few questions regarding the EPUB file. The PDF file uses a Nepali non-Unicode based font named Preeti, which shows different characters based on different input. For example, “k|Zg” in Preeti displays the word “प्रश्न”. However, while converting the font from PDF to EPUB, the formats got changed, consequently displaying “k|Zg” where it should have been “प्रश्न”. Please let me know if I can convert the PDF file to EPUB without such problem or if I should use a separate Unicode-based font for the EPUB. Also, the “Fixador” said that “.epub” was a forbidden extension and cannot be uploaded. Please let me know about that as well. — LFN-Nepal 2021/03/20 09:36
- Interesting. It would probably be better to use a Unicode font, because Unicode is so widely supported nowadays. With a non-Unicode font, any attempt to copy and paste from the PDF would probably fail. I have an automated script that I use for generating the grammar PDFs in the other languages. If you like, I could easily modify this to process the Nepali grammar, too. Then it would look consistent with the others, which would be nice 🙂 — Simon 2021/03/20 11:10
- I’m not sure why the wiki is forbidding the “.epub” extension, but you can doubtless circumvent the problem by changing the extension to “.zip” before you try to upload it. I can then switch it back to “.epub” on the server. — Simon 2021/03/20 11:10
- I forgot to comment on the problem you’re having with the font in the EPUB. Our other EPUB grammars have been produced by Michel Gaillard: I’ll ask him if he’d like to try the Nepali one. — Simon 2021/03/20 12:05
- I have also added Nepali translations to the LFN course in Wikibooks, and have created a Memrise course for Nepali speakers who want to learn Elefen at https://app.memrise.com/course/5935379/elephen-aadhaarbhuut-shbdaavlii/. Please add them to the main presentation if you think it should be. Thank you once again! — LFN-Nepal 2021/03/20 09:36
- Perhaps you could add the Memrise link to the main page of the Nepali grammar, where it’s more likely to be found by Nepali speakers. — Simon 2021/03/20 11:10
Here is my latest romanisation scheme. The published grammar pages now use this in their addresses. I’m using @ to represent the implicit vowel, as well as anything that cancels this vowel, i.e. the virāma or any of the combining vowels like ा. Having romanised a word or sequence of words, I then delete any occurrences of @@ and replace any remaining instances of @ with a. Following the names of the consonants in the Unicode standard, I’m using ng, ny, tt and so on to represent what IAST writes as ṅ, ñ, ṭ, etc. And following IAST, I’m using c for च and ch for छ. It seems ugly to use ch for the former and then chh for the latter, but if you feel strongly about it, it’s simple to change. Note that I’ve retained ee for ी and oo for ू even though Unicode more logically calls these ii and uu. (I’m guessing the very Englishy ee and oo originated in colonial days, perhaps related to the lack of a need for long E and long O, these historical sounds having shifted to ai and au.) — Simon 2021/03/20 11:10
// consonantes + "a" (la "a" desapare si la consonante es direta segueda par un virama ्)
// velal <k kh g gh ṅ h>
'क' => 'k@', 'ख' => 'kh@', 'ग' => 'g@', 'घ' => 'gh@', 'ङ' => 'ng@', 'ह' => 'h@',
// palatal <c ch j jh ñ y ś>
'च' => 'c@', 'छ' => 'ch@', 'ज' => 'j@', 'झ' => 'jh@', 'ञ' => 'ny@', 'य' => 'y@', 'श' => 'sh@',
// retroflexe <ṭ ṭh ḍ ḍh ṇ r ṣ>
'ट' => 'tt@', 'ठ' => 'tth@', 'ड' => 'dd@', 'ढ' => 'ddh@', 'ण' => 'nn@', 'र' => 'r@', 'ष' => 'ss@',
// dental <t th d dh n l s>
'त' => 't@', 'थ' => 'th@', 'द' => 'd@', 'ध' => 'dh@', 'न' => 'n@', 'ल' => 'l@', 'स' => 's@',
// labial <p ph b bh m v>
'प' => 'p@', 'फ' => 'ph@', 'ब' => 'b@', 'भ' => 'bh@', 'म' => 'm@', 'व' => 'v@',
// vocales
'अ' => 'a', 'आ' => 'aa', // <a ā>
'इ' => 'i', 'ई' => 'ee', // <i ī>
'उ' => 'u', 'ऊ' => 'oo', // <u ū>
'ऋ' => 'ri', 'ॠ' => 'rii', // <ṛ ṝ>
'ऌ' => 'li', 'ॡ' => 'lii', // <ḷ ḹ>
'ए' => 'e', 'ऐ' => 'ai', // <e ai>
'ओ' => 'o', 'औ' => 'au', // <o au>
// vocales combinante con la consonante presedente
'ा' => '@aa', // <ā>
'ि' => '@i', 'ी' => '@ee', // <i ī>
'ु' => '@u', 'ू' => '@oo', // <u ū>
'ृ' => '@ri', 'ॄ' => '@rii', // <ṛ ṝ>
'े' => '@e', 'ै' => '@ai', // <e ai>
'ो' => '@o', 'ौ' => '@au', // <o au>
// sinias spesial
'्' => '@', // virama, sutrae la vocal presedente
'ं' => 'm', // anusvara <ṃ>, marca la vocal presedente como nasal
'ँ' => 'm', // candrabindu <~>, marca la vocal presedente como nasal
'ः' => 'h', // visarga <ḥ> /h/
'ऽ' => '', // avagraha <'>, marca la elidi de la vocal inisial seguente
// dijitos
'०' => '0', '१' => '1', '२' => '2', '३' => '3', '४' => '4',
'५' => '5', '६' => '6', '७' => '7', '८' => '8', '९' => '9',
- The romanisation scheme might have some more problems. ॠ, ऌ and ॡ are Sanskrit vowels, which are not used in Nepali. However, if you want to add them to romanise Devanagari, it’s good to have them. However, if the scheme aims to romanise Devanagari as a whole, there are many more modified characters. The other thing is that in Nepali, the anusvar (‘ं’) can have two pronunciations, depending on the word, and is represented by ‘m’ or ‘n’. For example, the word संसार (world) is transliterated as ‘sansaar’ whereas the word संयोजक (conjunction) is transliterated as ‘samyojak’. Similary, the candrabindu (‘ँ’) represents a nasal sound, which is quite often represented by ‘n’. An example of this is the word चाँद (moon), which is almost always translitered as ‘chaand’. Furthermore, repeating the letters for retroflexes might cause problems for the readers. For example, ‘matt’ can be interpreted as ‘मट’ (pot) or ‘मत्त’ (tipsy). To avoid this, we can do what is done in Romi Konkani. In Romi Konkani, which is Konkani written in the Roman script, the retroflexes are represented by repeated letters, and repeated dentals are separated by an apostrophe. For example, the word कन्नड (Kannada, the language) is written as “kon’nodd”. Simply writing “konnodd” could mean be interpreted as “कणड”, “कन्नड”, “कन्नद्द” or “कणद्द”. However, I think it’s best to use ‘t’ to represent both dental and retroflex sounds. — LFN-Nepal 2021/03/20 12:19
- The nasal consonants ‘ङ’ and ‘ञ’ are also somewhat tricky, because they have different pronunciations as standalone characters and as conjuncts. ‘ङ’ is pronounced similarly to the ‘ng’ in ‘bang’. However in ‘ङ्क’, which is a conjunct of ‘ङ’ and ‘क’, it is pronounced like the ‘n’ in the elefen word ‘ance’. As a standalone character, ‘ञ’ is pronounced like the Spanish letter ‘ñ’. However, in ‘ञ्ज’, which is a conjunct of ‘ञ’ and ‘ज’, it’s pronunciation is similar to the ‘n’ in ‘range’. So, unless these letters are the first or the last letters of the word, it is advised to simply use ‘n’ instead of ‘ng’ and ‘ny’. These phonetic changes can be differentiated properly with diacritics; nonetheless, in ASCII, it does create problems. — LFN-Nepal 2021/03/20 13:03
- Another quirk of the language is the pronunciation of ‘व’, which is somewhere between ‘v’ and ‘w’. In words like वकिल (lawyer, pronounced ‘wakil’) and वहाँ (there, pronounced ‘wahaan’), ‘व’ is pronounced as ‘w’. However, in words like वन (forest, pronounced ‘van’) and वाक्य (sentence, pronounced ‘vaakya’, the letter is pronounced as ‘v’. Majority of Nepali speakers also pronounce ‘व’ as ‘b’, as in व्याकरण (grammar, pronounced ‘byaakaran’). This can cause problems in romanising as well. Please let me know what can be done about it. — LFN-Nepal 2021/03/20 13:03
- Very interesting information, but I don’t think it affects the romanisation, which is only used for generating URLs. It’s not a pronunciation guide. If the doubling of retroflexes produces an offensive word, then it would be worth adjusting, but we’re looking at fewer than twenty URLs in total. And for the dictionary, it will be crucial that each character be romanised in exactly the same way each time it appears. — Simon 2021/03/20 13:43
- By the same token, though, I suppose we ought to simply the retroflexes and nasals as you suggest 🙂 This I have now done. — Simon 2021/03/20 14:07
- I have generated a PDF, in Unicode from the wiki pages: https://elefen.org/pdf/gramatica_ne.pdf. I hope I haven’t unwittingly filled it with mistakes! — Simon 2021/03/20 13:43
- If the romanisation is just for generating URLs, then the romanisation wouldn’t be a problem. It’s also impressive that you could generate a PDF in Unicode. From what I can see, the PDF is correct for the most part. In the first page, the title at the first page should be “लिङ्ग्वा फ्राङ्का नोभाको व्याकरण” and the correct word for “table of contents” would be “विषयसूची”. In “प्रश्न”, there is an untranslated phase, i.e., it shows “No, thanks” instead of “नाइँ, धन्यवाद ।” In “गणनाक्रम”, the powers of numbers have not been represented as superscripts. The formatting can be changed; for example, the titles can be bolded and italicised as in the English version, and line spacing can be added as the font is kind of conjusted. — LFN-Nepal 2021/03/20 15:00
- Excellent feedback, thanks! I’ve corrected those points. The superscripts were missing because in the other languages, we use dedicated Unicode superscript characters (¹, ², ³ etc) which require no special handling. I have italicised the subheadings within each chapter, but I don’t seem to be able to generate bold Devanagari characters in the PDF, even though they are bold in the LibreOffice Writer document from which I am exporting the PDF. (I use a script that downloads the wikitext of the grammar, makes a few simple changes to it, then passes that through Pandoc, which spits out an ODT document to which I then make final adjustments, such as adding the table of contents, in LibreOffice Writer.) — Simon 2021/03/20 16:03
- I wanted the PDF file to be in non-Unicode based Preeti font, as Nepali uses a special form of Devanagari, called Balbodh, which is not used in most other languages that use the Devanagari script. For example, the letter “झ” is written in Nepali with a small tail after “भ”, similar to how “फ” is written with a tail after “प”. This feature is found only in Nepali and is recognised as the official way to write the letter by the Nepal Academy. However, this way of writing is not supported online. However, if it helps in creating an EPUB file, please feel free to use the Unicode version. Thank you! — LFN-Nepal 2021/03/20 15:00
- Very interesting. And how frustrating for Nepali speakers that Unicode, the world standard, doesn’t fully support their official orthography. If you add Nepali entries to the dictionary, they will have to be Unicode, too. (Are you planning to, by the way? It’s a long old task. I added the Esperanto entries in about three months last year, but I had rather more spare time available then than I do now… Michel took 7 or 8 months to add the French translations.) — Simon 2021/03/20 16:03
- It is kind of frustrating, because I have only ever seen the official orthography be used in Nepali Wikipedia in Android. However, in other websites, it’s almost non-existent. Also, I would love to add Nepali entries to the dictionary, if you could teach me how to. I also have to balance the entries with my classes, so I might take about 3 months or more, depending on how many entries have to be translated. — LFN-Nepal 2021/03/20 17:07
- Also, for the PDF file, the page numbers displayed on the table of contents do not match the actual page in which the topics start. Please fix that as well, thank you! 🙂 — LFN-Nepal 2021/03/20 17:13
- The other thing I have noticed is that in “शब्दरचना”, under “क्रियापद बनाउने प्रत्यय”, the word “र्याल” has been displayed as “र्याल”. Nepali and Marathi have two conjuncts for “र” and “य”, which include “र्य” for Sanskrit words and “र्य” for words of their own origin. While “र्य” is present in Unicode, the PDF document does not show it, probably because of the font used. Please change it if it is possible. The formatting of quotation marks and apostrophes for uniformity would be appreciated. Thank you once again! — LFN-Nepal 2021/03/20 17:16
- When I paste र्याल into LibreOffice Writer, it displays र्याल, unfortunately, regardless of what font I choose. Please do bear in mind that I can’t read Devanagari easily, so it took me quite a while to even locate the word you were talking about (“saliva”). I certainly can’t manually correct the quotation marks and apostrophes, because I can’t read the text that surrounds them! I could give you a copy of the ODT file and let you adjust it, if you think that would work. — Simon 2021/03/20 17:27
- If it can’t be corrected, then it’s best to leave it as it is. Quotation marks and apostrophes were brought to Devanagari from the Latin script. Therefore, there are no special rules related to those punctuations. However, I have noticed that the fonts for apostrophes and quotation marks seem different in different pages. Apart from that, I recently found out that in “नाम”, I had forgetten to translate “duke, duchess” and in “शब्दरचना”, I wrote “मिथ्रावाद” (mitraisme) instead of “मिथ्रावादी” (mitraiste), which I have changed in the website. I would be glad if they could be changed on the PDF file as well. — LFN-Nepal 2021/03/20 17:34
- No problem: “duxe, duxesa” and “mitraiste” now stand corrected. It may be possible for someone who can read Nepali to correct the quotation marks and apostrophes, but not possible for me. The fonts ought to be the same throughout: they are applied by a central stylesheet, which I have not changed at all. LibreOffice Writer substitutes another font (Lohit-Devanagari) whenever the standard font CharisSIL does not contain a character; possibly it has a bug there.
- I did look a bit further into the problem of र्याल versus र्याल. It seems that the former contains a “zero-width joiner” character between the र् and the य, while the latter doesn’t. I’ve experimented in various editors, and most of them seem to ignore the zero-width joiner altogether, hence the problem. In one editor, I found I could paste र्याल successfully only if it wasn’t the first word on a line! — Simon 2021/03/20 17:59
- I suspect we may have to go back to your original PDF, with which there was nothing really wrong, other than that it didn’t use Unicode – I was being technically picky about that, but a typical reader of your PDF wouldn’t notice, and would doubtless be glad to have the orthography correct. — Simon 2021/03/20 17:50
Michel has produced an EPUB, here, if you’d like to check it. I don’t know whether it includes your latest corrections. — Simon 2021/03/20 18:11
- For some reason, I am unable to open the EPUB file. And for the PDF file, it would be nice if I could edit the quotation marks and apostrophes for uniformity of font. How can I do so? Please help me out, thank you! — LFN-Nepal 2021/03/21 04:01
- Apologies – some irrelevant characters somehow got into the URL. It’s inexplicable that this didn’t simply give an error on clicking the link; instead it downloaded an empty file! I’ve corrected the link just above here, and the file certainly seems to contain the Nepali grammar. I don’t have a device capable of reading EPUBs, so I can’t check further. — Simon 2021/03/21 07:52
- Regarding the quotation marks and apostrophes in the PDF, I’ve placed a copy of the ODT file I generated here. You can download it, correct it and send it back via the wiki’s “Fixador” (file manager). But I actually think your original PDF was much clearer than anything I can generate. In particular, the use of a monospaced font for the Elefen words made the text look as though it would be much easier to read. So I suggest you adjust your PDF with the corrections you listed here yesterday, and then re-upload that via the Fixador. — Simon 2021/03/21 08:01
- Me ia restora tua PDF orijinal a la adirije xef, per la presente. — Simon 2021/03/21 19:10
How to edit the dictionary
Start here. Preferably read the explanation of the format, which is written in Elefen but hopefully not hard to follow. Then you simply use the navigation links at the top to edit the pages – it’s divided into 166 pages, partly for technical reasons and partly so that we seldom tread on each others’ toes. Insert Nepali glosses on lines with an ne: prefix, at the correct alphabetical position in the list of languages. You can do as much or as little as you like in a session – but if you do a lot of work in one go, I strongly recommend that you copy the content of the editor before you save it, because (very occasionally) the server will decide that it’s too busy to save your changes right now, and will simply throw them away with an error message. 🙁 Having made changes to the dictionary, you should click “Publici” in the navigational panel at the top, then click “Publici Xercan” in the “Disionario” section on the subsequent page – this will run a script to check the syntax and ordering of the entire dictionary, and if all is well, the results will be published here. I will have to add some code to the search engine to make Nepali available as a searchable language, but I won’t do that until you’ve got a reasonable number of translations in place. — Simon 2021/03/20 18:16
- It’s great to see you’ve made a start on the Nepali entries for the letter Z. Excellent! Do please note, however, that you mustn’t change the number of blank lines between entries. When I clicked “Publici” just now, it complained about more than 30 places where you had reduced 3 blank lines to 1 or 2. It’s easy to avoid this if you always get into the habit of adding a line for Nepali, rather than adopting what appears to be a pre-existing blank line. — Simon 2021/03/21 08:07
- Sure! However, I am a school student and have been informed recently that I will be having my exams soon. So I might take longer to complete the translations. I hope that won’t be a problem, as I’m translating little by little. — LFN-Nepal 2021/03/21 15:33
- No problem at all. I’m delighted to see the dictionary expanding into another language. It’ll take as long as it takes. More than that, no one can say 🙂 — Simon 2021/03/21 17:24
- = Tota no problem. Me es deletada en vide la disionario estendente a un plu lingua. Lo va dura tra la dura. Plu ca esta, nun pote dise 🙂 — Simon 2021/03/21 17:26
You’re making great progress already! 2,7% of the dictionary entries now have a Nepali translation. But please note that your edits have occasionally damaged existing entries – for example, I’ve just reinstated a Hebrew line and a French line that you had overwritten. I suggest that after each editing session you have a quick look at the display of the differences between the latest version and the previous one. This is easy to access if you click the “cambias” link for the appropriate edit on the recent changes page. — Simon 2021/04/04 17:28
Use of English
Your English is excellent – are you a native speaker? But it would be nice to reach a point where I could converse with you in Elefen, as I do with the other people on the wiki. You mentioned that long sentences are currently difficult for you to understand. I heartily recommend a simple exercise of reading my news blog, Aora Oji, where I post a short article each day on something major or minor that’s happened in the world. If you spend five minutes on that each day for a month or so, you should rapidly achieve reading fluency. — Simon 2021/03/21 08:10
- Comensa simple usa lo es vera la metodo la plu bon per aprende un lingua! E con la disionario e la spelador, comensa scrive e discute en elefen no es tan difisil. E si tu fa eras (como cadun), Simon o otras asi va coreti delicata tu. — Cris 2021/03/21 14:43
- Multe grasias per la coraji! Me es un parlor nativa de nepali e me es un abitor de nepal. Ma, en Nepal, los abitores aprende engles como un lingua du (as a second language, I couldn’t paraphrase it properly). En cuando comensa la pandemica, me ia crea articles en Vicipedia tra cual me ia descovre elefen. Aora me aprende espanyol e elefen, e me vole tradui la pajes en esta loca ueb afin plu persones, spesial parlores de nepali, pote aprende la lingua. Per favore coreta me si me ia crea eras. Gracias a un plu ves! — LFN-Nepal 2021/03/21 15:16
- Eselente! “Aprende engles como un lingua du” es coreta. Tua sola eras es peti: “los abitores” → “la abitores”; “comensa la pandemica” → “la pandemica comensa” (la sujeto presede sempre la verbo en elefen); “espanyol” → “espaniol”; “coreta me” → “coreti me” (“coreta” es un ajetivo); “gracias” → “grasias”. — Simon 2021/03/21 17:20
- Ah, grasias! Me lo confusa la speles e strutur de frases con espaniol. — LFN-Nepal 2021/03/22 14:00
nepali en la xercador
Tu progresa briliante: 11,5% de la entradas en la disionario ave aora traduis nepali. Multe lodas! Aora, me ia ajunta nepali a la lista de linguas disponable en la xercador. — Simon 2021/05/30 06:50