Our data is released under various Creative Commons licenses.More information
If you love this content, please consider a donation.
Wall (7,174 threads)
Tips
Before asking a question, make sure to read the FAQ.
We aim to maintain a healthy atmosphere for civilized discussions. Please read our rules against bad behavior.
Thanuir
an hour ago
LeviHighway
11 hours ago
frpzzd
11 hours ago
LeviHighway
12 hours ago
LeviHighway
13 hours ago
AlanF_US
19 hours ago
LeviHighway
yesterday
PaulP
yesterday
Rok
3 days ago
Feniks
4 days ago
I am writing to report a significant issue regarding the language labeling of imported data from Translators Without Borders (TWB) on Tatoeba.
Specifically, a substantial amount of data imported from TWB has been incorrectly labeled as "Tigre" (tig) when it is, in fact, "Tigrinya" (ti). These are two distinct languages, and this mislabeling creates serious inaccuracies within the database.
As a native Tigre speaker, I, along with my colleagues, have been eager to contribute to Tatoeba. However, this existing error is preventing us from accurately adding Tigre language data.
We understand from the TWB user profile (https://tatoeba.org/en/user/profile/TWB) that "As the TWB translators will not be on Tatoeba to correct their sentences." This means that the responsibility for correcting this critical error falls to the Tatoeba administrators.
This is not a minor issue; it fundamentally compromises the integrity of the language data and obstructs the contributions of native Tigre speakers. The correct language codes are:
Tigrinya: "ti"
Tigre: "tig"
We urge you to take immediate action to rectify this mislabeling. We are prepared to provide specific examples of incorrectly labeled sentences if needed.
Hello Beshir,
Here in Tatoeba, we use three-letter ISO codes instead of two, so Tigrinya should be "tir".
Thanks for understanding :)
Thank you for your response. I understand you've confirmed the ISO 639-3 codes: "tir" for Tigrinya and "tig" for Tigre.
However, there is a critical error: sentences labeled as Tigre, specifically those contributed by user "TWB", are actually Tigrinya sentences.
As a native speaker of both Tigrinya and Tigre, I can confidently identify this mislabeling.
Here's a breakdown of the issue:
Incorrect Labeling: User "TWB" has submitted Tigrinya sentences that are currently labeled as Tigre.
Confirmation: This error is also noted in the comments of Sentence #10898663, where user "Pathawi" correctly identifies the sentence as Tigrinya, not Tigre.
Impact: Currently, there are 3,800+ sentences labeled as Tigre, contributed by users "dagnaq" and "TWB". The sentences from "TWB" are the problem.
Resolution: All sentences submitted by user "TWB" that are currently labeled as Tigre must be re-labeled as Tigrinya.
This mislabeling is negatively impacting both language communities:
It discourages Tigrinya speakers from using Tatoeba due to the incorrect categorization.
It discourages Tigre speakers from using Tatoeba due to the inaccurate data.
Correcting the labels will significantly increase the Tigrinya sentence count.
Correcting the labels will encourage Tigre speakers to contribute to Tatoeba.
Please take immediate action to correct the labeling of user "TWB's" contributions.
Hello Beshir,
I'm the one who created the TWB account to batch-add the Gamayun kits sentences using scripts.
Indeed, it seems that the sentences in Tigre are actually in Tigrinya. The error comes from the name of the source file, which uses the language identifier 'tig' instead of 'tir'.
https://github.com/translatorsw...ng-tig_TIG.txt
I apologize for this mistake. I'll try to write a script that changes the language identifier of these sentences. Sadly, this will take some time.
Is there's anything I can help with to speed up the correction process.
Does it help to temporarily remove the sentences from the Tigre category? Then, when you have time, re-import them into the correct Tigrinia category?
Can you provide estimated time
Thank you
My Tigre sentences have been moved to the Tigrinya corpus. Please feel free to adopt and improve them. This will increase their visibility for language learners.
Thank you very much for your prompt response.
I am very new here and still finding my way around. I joined because I heard we can translate sentences from one language to another and wanted to do the same for my native language Ao from English. I do not know how to get Ao added to the languages supported here. Can anyone please help me?
Ao uses just English alphabet and is written from left to right just like English. It has got 3 mutually unintelligible dialects (Mongsen Ao, Tsüngli/Jungli Ao and Changki Ao) so I do not know whether a single language Ao should be added and the dialects should be indicated in the translated sentences or 3 new seperate language entries should be made for each dialect (Mongsen Ao, Changki Ao etc.). I don't know what is the policy regarding such situations but I am okay both ways. For info, my native tongue is Mongsen Ao, but I am well versed in Tsüngli/Jungli Ao through formal education. Regarding Changki Ao, I can add only a few common expressions at most and therefore, best left to someone who speaks it natively.
Thank you for reading this wall post.
Hi and welcome. These are the steps to request a new language: https://en.wiki.tatoeba.org/art...nguage-request
Thank you so much. Will try and apply even if adding new languages is paused for now...according to the page.
I agree and would also suggest to add specifically Mongsen Ao, like you suggest, and not "Ao" in general. Great to have this language here in the future!
I have translated an English book into German. The English is copyrighted, of course. But I do not think the German, taken sentence by sentence randomly, for educational purposes, would be considered a copyright infringement, would it?
Also, I would not want to enter the sentences here one at a time. Is there a method of loading the sentences in batches?
I believe derivatives of copyrighted work still fall under copyright, although if the selected sentences are likely to be used elsewhere in real life conversation it probably won't matter much.
Also, to my knowledge there aren't ways, at least for non-admins, to upload batches of sentences.
> But I do not think the German, taken sentence by sentence randomly, for educational purposes, would be considered a copyright infringement, would it?
It would. Please don't do it.
✹✹ Stats & Graphs ✹✹
Tatoeba Stats, Graphs & Charts have been updated:
https://tatoeba.j-langtools.com/allstats/
The content of this message goes against our rules and was therefore hidden. It is displayed only to admins and to the author of the message.
I want someone to record audio from my sentences!
You're not in my next batch of sentences to record, but I'll try to keep you in mind for the one after that.
For your native language, you can record your own sentences.
See this page in the wiki.
https://en.wiki.tatoeba.org/art...ntribute-audio
I'm using a CHROMEBOOK!
I mean, check your e-mail.
I would like to ask for an update for Kabyle and every other Amazigh languages in general because I wanted to write a sentence in Tifinagh, but I was told that Kabyle could only be written in Latin and that I had to put the phrase in the “Berber” language category. Tifinagh shouldn’t be just limited to Amazigh, it’s part of every Amazigh languages and dialects such as Kabyle and it hurts me to be forced to write my language in an other script than its own. So I would just like to ask to allow the simultaneous use of the Tifinagh and Latin script for Kabyle.
It is not the work of a week, a month or a year that you want to call into question, but the work of more than a century.
There is no book, no software, no academic work by a Kabyle in tifinagh.
I understand that Latin is the basic script for Kabyle, but Tifinagh is also part of it. For example, my cousin in Algeria learns Kabyle in Tifinagh at school and I think that if we continue writing Kabyle only in Latin, Tifinagh might go extinct soon. Furthermore, I thinks it’s unfair that people are allowed to use Tifinagh in Standard Moroccan Amazigh, but not in Kabyle. If you don’t agree with my statement, can I at least put the Tifinagh script of my sentences in parenthesis or as a bonus?
"For example, my cousin in Algeria learns Kabyle in Tifinagh"
No school in Kabylia teaches in Tifinagh. The only ones who do are the Tuareg in southern Algeria and the Moroccans.
I think you should respect this choice and add your sentences in Latin when it's under the Kabyle flag.
You can also add sentences in tifinagh under the Berber flag.
The teachers of that school voluntarily teach Tifinagh, but that’s not important. What I mean is that languages like Mongolian or Kazakh have the same situation as Kabyle. While Cyrillic is the official script of both languages, there are a lot of Kazakh people who write their language in Latin and there are a lot of Mongolians who write their language in the Mongolian script (ᠮᠣᠩᠭᠣᠯ ᠪᠢᠴᠢᠭ). It’s the same situation in Kabylia even though there’s not as much people who write in Tifinagh. I’m asking for at least an authorization to use Tifinagh as a parenthesis or as a bonus while still writing in Latin. If I put my Kabyle sentences under the Berber flag, it takes away the fact that it’s a dialect of Berber and not just Berber.
Hi, Amastan ! :-)
Name: Svan
ISO code: SVA
List: https://tatoeba.org/en/sentences_lists/show/173540
Flag: https://commons.wikimedia.org/w...mo_Svaneti.svg
Tatoeba has been updated today. What’s new?
※ New "sentence is original" search criterion. You can use it to search for sentences that were initially not added as a translation of some other sentence.
※ Two new languages have been added: Interslavic and West-Central Oromo (respectively requested by rul and abiniz). Tatoeba now supports 426 languages.
※ The language Old Tupi was renamed to Tupinambá and its ISO code was changed from tpw to tpn. This change follows the latest updates of the ISO 693-3 standard.
※ The Silesian language icon was updated to better represent Silesian: https://tatoeba.org/sentences/show_all_in/szl/none thanks to vxern!
※ For the more tech-savvy among you: the new API at https://api.tatoeba.org/ now allows searching for sentences by combining a plethora of criteria. Follow the link to check the documentation.
Many thanks to DJ_Saidez for helping with the language requests!
Hello sir,
Thanks for bringing these updates and new languages into Tatoeba. I have a question. when will you guys update a flag for these tongues?
The flags for Promo, Interslavic and Tupinambá are all up, but it may take some time for them to show up. You can refresh your browser cache or simply wait.
Thanks for ur prompt answer.
✹✹ Stats & Graphs ✹✹
Tatoeba Stats, Graphs & Charts have been updated:
https://tatoeba.j-langtools.com/allstats/