Requests for new languages/Wikipedia Sindarin 2

From Meta, a Wikimedia project coordination wiki
This is an archived version of this page, as edited by Calad-ne-dúath (talk | contribs) at 16:31, 18 October 2021 (→‎Proposal). It may differ significantly from the current version.
submitted verification final decision
Discuss the creation of this language project on this page. Votes will be ignored when judging the proposal. Please provide arguments or reasons and be prepared to defend them (see the Language proposal policy).

The language committee needs to verify the language is eligible to be approved.

  • Check that the project does not already exist (see list).
  • Obtain an ISO 639 code
  • Ensure the requested language is sufficiently unique that it could not exist on a more general wiki.
  • Ensure that there are a sufficient number of native editors of that language to merit an edition in that language.
  • The community needs to develop an active test project; it must remain active until approval (automated statistics, recent changes). It is generally considered active if the analysis lists at least three active, not-grayed-out editors listed in the sections for the previous few months.
  • The community needs to complete required MediaWiki interface translations in that language (about localization, translatewiki, check completion).
  • The community needs to discuss and complete the settings table below:
What Value Example / Explanation
Proposal
Language code sjn (SILGlottolog) A valid ISO 639-1 or 639-3 language code, like "fr", "de", "nso", ...
Language name Sindarin Language name in English
Language name Eglathrin Language name in your language. This will appear in the language list on Special:Preferences, in the interwiki sidebar on other wikis, ...
Language Wikidata item Q56437 - item has currently the following values: Item about the language at Wikidata. It would normally include the Wikimedia language code, name of the language, etc. Please complete at Wikidata if needed.
Directionality LTR Is the language written from left to right (LTR) or from right to left (RTL)?
Links Links to previous requests, or references to external websites or documents.

Settings
Project name Limoseliad "Wikipedia" in your language
Project namespace Limoseliad - Oseliad Lain usually the same as the project name
Project talk namespace Pedad en Limoseliad "Wikipedia talk" (the discussion namespace of the project namespace)
Enable uploads no Default is "no". Preferably, files should be uploaded to Commons.
If you want, you can enable local file uploading, either by any user ("yes") or by administrators only ("admin").
Notes: (1) This setting can be changed afterwards. The setting can only be "yes" or "admin" at approval if the test creates an Exemption Doctrine Policy (EDP) first. (2) Files on Commons can be used on all Wikis. (3) Uploading fair-use images is not allowed on Commons (more info). (4) Localisation to your language may be insufficient on Commons.
Optional settings
Project logo File:Limoseliad.png This needs to be an SVG image (instructions for logo creation).
Default project timezone Continent/City "Continent/City", e.g. "Europe/Brussels" or "America/Mexico City" (see list of valid timezones)
Additional namespaces Parfod, Pedad en Pharfod, Antaen, Pedad en Antaen, Maeron, Pedad en Vaeron For example, a Wikisource would need "Page", "Page talk", "Index", "Index talk", "Author", "Author talk".
Additional settings Anything else that should be set
submit Phabricator task. It will include everything automatically, except additional namespaces/settings. After creating the task, add a link to the comment.

Proposal

  • Sindarin has a large vocabulary and allows you to easily construct new words using its word-formation mechanism. The same mechanism allows us to express in Sindarin information about things not only of the modern and familiar world, but also about things that we do not even know. Run on an ordinary user's PC, the program, the purpose of which is to create new combinations of roots, prefixes and suffixes, allows you to get a vocabulary of over a hundred million new words. At the very least, that's enough for a dialog.
  • The experiment I started in Wp/Sjn shows how flexible this language can be already on the example of an article about the language itself. I'm sure there are no insurmountable difficulties, and the goal of translating as many articles as possible into Tolkien's wonderful Sindarin language will be successfully achieved. At present, this project has not encountered problems with the rules, which are perfectly clear and strictly defined in the Sindarin language textbooks written by many excellent authors, to whom I am personally very grateful for their work and the tremendous work they have done to make its study possible.
  • During the translation of articles in Sindarin, I have gained some experience in this field and can call the problems of rules and vocabulary in Sindarin far-fetched, and referring to my experience, I can say that I personally have not had such problems. And I don't see any objective reason why anyone else should have them, especially people learning Sindarin.
  • The well-known and generally accepted rules of Sindarin do not cause too much dissonance when communicating between different speakers.

Someone would oppose me just because the difficult of developing the IME on typing Sindarin script, so I'd love to continue contributing via Latin transliterations, then make a gadget to show em. --120.6.9.47 23:02, 23 July 2021 (UTC)[reply]

  • I assume that the Sindarin Wikipedia project, with its automation technique, can open the door to many new language projects, the development of which will turn out to be accelerated the more the more new articles are written for them.--Calad-ne-dúath (talk) 11:52, 26 September 2021 (UTC)[reply]
  • I now use the OpenNMT neural network to translate into Sindarin. Since the neural network sometimes makes translation mistakes, I correct the translation manually. The same technique is very useful for a number of natural languages, including languages of Africa and various small peoples of the world.
  • Many interesting and relevant language projects can go the same way because of the relatively small population of peoples and/or the peculiarities of their languages. But this does not negate the need for them to convey information to their speakers, because their number is not zero.
  • The need to reconstruct individual words and speech patterns, such as those required for translation into Sindarin, may be useful for languages spoken today by linguistics specialists or small peoples of the world. These languages, and the Wikipedia projects based on them, may be of considerable value to their speakers, and to humanity as a whole.
  • Sindarin speakers. There is an ambiguous definition of a native speaker. When applied to an artificial language, a native speaker can be a person who has learned the language at a conscious age and has contributed to the development of the language. At the same time, there are cases where people have learned artificial languages from birth, which allows us to apply the term native speakers to these people, which is close in meaning to the term for native speakers of a natural language.
  • Learning Sindarin. Today, information about the Sindarin language is widely available. Various textbooks have been written, instructional videos have been recorded, publicly available e-courses have been created, and magazines publishing Tolkien's notes have been published. The speed of learning Sindarin language for beginners is 2 weeks.
  • Language development. People who speak Sindarin often invent something new: words, phrases, turns of speech. All together, this enriches the text base of the language, while demonstrating its capabilities. Software modeling shows the possibility of obtaining billions of new words in Sindarin, a number of which may have no semantic analogues in natural languages, but can be understood on an intuitive level even by a beginner. At the same time, professional linguists are engaged in the development of the language. Thanks to common efforts and the possibility of worldwide communication, there are no significant discrepancies in the rules of the language today. The irrelevant rules are found in the list of ancient ones. Therefore, if a person says Ered, his interlocutor will not get upset and will not explain to him about the correctness of Eryd, but both will know that it was about mountains.
  • Sindarin is a synthetic language created by J.R.R. Tolkien. There are two well-known facts about it:
  1. Tolkien created the language first, and then the story he placed that language in the world.
  2. Today, Sindarin has gone far beyond the fictional world, and it is used by people for a variety of needs. There are print editions, communities on the Internet, people write and communicate in it.
  • The two points above do not allow us to characterize Sindarin as a language created only for the fictional world, and which has no place in the real world. Tolkien is known to have had a passion for constructing languages from a young age, and the ease with which he spoke them, although this is a natural characteristic for a language creator. But people who had nothing to do with the creation of Sindarin have long shown that it allows the construction of verbal and semantic constructions that are relevant to the modern world. Practical research on the language and the communities that speak it has shown its great flexibility and the possibility of using it beyond the fictional world. This circumstance allows us to look at the Sindarin language from a different angle, viewing it as a phenomenon separate from the works of fiction in which it was used by its author.
  • I rely heavily on automation in the course of my work. Reconstructing the necessary words and turns of speech can be very time-consuming if done manually, but knowing the rules of the language, I write programs that function according to these rules, and they reconstruct the words and turns of speech for me. All the programs I have written allow me to automatically, with a minimum of manual edits, form a text corpus and train a neural network that helps me translate arbitrary texts into Sindarin.
  • It is a well-known fact that all language is abstract. No word is firmly attached to a subject, and this fact makes words replaceable and texts translatable. By following the grammar rules of a particular language, words and phrases can be formed and reconstructed. The grammatical rules of the language can be written into the program so that it deals with these tasks automatically. Using a similar algorithm, it becomes possible to reconstruct arbitrary languages without loss of meaning for their native speakers. Therefore, I assume that the example of a section written in Sindarin will inspire someone to translate Wikipedia into the languages of small peoples or ancient states.

Discussion

@Calad-ne-dúath: Here here. --120.6.9.47 23:04, 23 July 2021 (UTC)[reply]

I think it is not difficult to write a special program that makes it easy to write the Sindarin texts using the Tengwar runic alphabet developed by Tolkien for the elven languages. Moreover, modern development tools make it possible to perform such a development in the shortest possible time, and later to refine it as needed.

About the same applies directly to the process of translating articles in Sindarin: thanks to the proliferation of advanced machine translation tools, including neural networks, it is now possible to develop software for automatic translation of texts in Sindarin, which will help the translators involved in the project to do their job (although I personally prefer manual translation and creativity, but I consider automatic translation inevitable and in general do not avoid progress).

In general, I consider the addition of wikipedia sections written entirely in synthetic languages to be a natural progression of the project, and by now we have sufficient tools and information to translate the articles into Sindarin.

As the project develops, I am sure Sindarin Wikipedia will attract a wide variety of people working in different fields, and among them there will certainly be those willing to contribute to the development of the project. But, to do this, it is important to keep the project alive, not to cut it off at such an early stage.--Calad-ne-dúath (talk) 09:15, 24 July 2021 (UTC)[reply]

I have prepared two versions of the logo with captions in Latin and Sindarin Tengwar. But there's a little problem - I can't upload them in any way.--Calad-ne-dúath (talk) 22:40, 26 July 2021 (UTC)[reply]

Ned eraid vidui 'olthant tu'wi. Neviaid vinui gert lim a vae. Neviad gonodolui ger. For the last few days I have been training a neural network. The first results are impressive in terms of speed and quality. Automated translation in Sindarin is already here.--Calad-ne-dúath (talk) 21:16, 24 September 2021 (UTC)[reply]

          • If you are asking just me, then I would say A little support, the previous rejection seems like Unicode-related, but the test project is adopted somewhat a good Latin transliteration. --Liuxinyu970226 (talk) 02:05, 27 September 2021 (UTC)[reply]
            • Thank you very much for your response. I asked technical support and they told me that they fixed the bug that allowed adding translations for arbitrary language codes. I was told that they would add the language code sjn and I could continue translating in the future.--Calad-ne-dúath (talk) 08:24, 28 September 2021 (UTC)[reply]

Hello. I encountered a problem today with the inability to create new pages in the Wp/sjn project. Thank you all for your help, but it seems to be the end of the road. I think I will continue to stand for keeping this project on wikipedia, and want to dissuade those who don't see its future prospects. I will take the difficulties that arise and show my vision in action. It still takes some time, though.--Calad-ne-dúath (talk) 10:23, 10 October 2021 (UTC)[reply]