@Sammer
Akka i yettaru ***. Dayen tura gziɣ. Tanemmirt-ik.

I'll beleive you if Amastan said that he will transfer his sentences to the Kabyle corpus.

Most of them are invited to add sentences just to oppose to the kab language. A campaign was launched on Facebook to do that and most of them wrote few sentences.

Their goal is to stop the kabyle nothing else.

Anyway, if Tatoeba is based in the USAs, some contributors based in the USAs will reach you to discuss about the issue.

@Trang,

Are there any contributors on the Kabyle Corpus complained about the Kabyle Flag?

Do you think if a Palestinian complains about the Israeil Flag, you will also remove it?

Do you know that there is a berber flag shared by all berbers (a socio reality) that should represent the berber language rather than the one put there because of political beleifs of its Admin?

I can't understand why people working on a language want to change things about another language and you think that it's a normal behavior.

You must know that we have struggled for decades to access the teaching of the Kabyle language. Our elders have been jailed, Kabyle language activists have been murdered by the Algerian government, and I see that these bans are also following us on free online projects.

I'm sure that if you remove the kabyle flag, the kab corpus will die, unless that it is also your goal.

@Trang

Please, I beg you to ask for more information about people opposed to the Kabyle Corpus and Flag. I'm sure they were recruited to do that job or the same person having several accounts.

@Sammer

At least if we know who you are.
All contributors working on the kabyle corpus have identities and we know who they are. Why are you hidding your identity? Because you are not the one you are pretending to be.
And I did not see any contribution from you on the Kabyle corpus. You are working on the Berber corpus.
Here, we talk about linguistics, not politics.
Could we know from where are you?
___________

Lemmer tessneḍ xarsum ad taruḍ s teqbaylit.
Imdanen yellan dagi ttwassnen. Kečč d uffir neɣ ala?
Nebɣa ad nẓeṛ anwa i k-ilan acku ur telliḍ ara d win nɣil.
Rnu ɣur-s ulac ula d yiwet n tefyirt s teqbaylit. Aql-ak tqeddceḍ deg tbarbarit neɣ ala?
Dagi ad nettmeslay ɣef tesnilest, mačči ɣef tsertit.
Ini-d wi k-ilan.

From the begining, there was a member (only one memeber) who was oppposed to the Kabyle corpus when I asked Tatoeba to add it. This member was working on the Berber corpus which is a set of more 22 (or more) different berber languages written with 3 scripts: Tifinagh, Arabic and Latin and there is a consensus about it.

You know, there was a real debate about linguistics between academicians to deal separately with Berber languages because of phonetics, morphology and syntax. Now, most of them are convincd that there could never be one language that can be shared between north africans, except if we create Newspeak.

You can see some translations in the berber Corpora. Most of them are Kabyle. If the other natives (Mozabit, Chawi, Rifain, Chleuh, Tergi, Zennati, Djerbi, Nefousi, Siwa...) come and translate, you will get more than 22 different translations. 22 phonetics systems, Morphoply systems, syntax and grammar systems And it's a nonsense!!!

Believe me, kabyles do not understand rifain language or Mozabit language or any other berber languagesexcept Kabyle. there is no standard for all north africans. The conception of the unique berber language is like if you gather French, Spanish, Italian, Portugeese, romanian and other languages derived from Latin and you put them in one corpora and call it Latin!

When we suggested the Kabyle flag, because it's a sociocultural reality. The Kabyle flag is not related to politics. The Kabyle flag is just a regional flag that identifies a region. Most of the contributors (if not all) as I am working on the Kabyle corpus are not involed in politics. But they are aware about linguistics issues and the identity of their region. They know that there is no a language called Berber shared by north african people. People arging for the concept of the unity of the berber language inherited such a concept from totalitarian ideologies such as Arab nationalism or political Islam and try to reproduce the same pattern on us. And we refuse. We have our own language and it's own visual identity.

I can't understant that person who was working on the Berber language, why in the begining opposed to the kabyle language and he knew that he could'nt remove it, he want to see it's identity disapear?

How can we imagine French Nationalists asking Tatoeba to change the Breton, Corsican or Bretagne Flag related to their regions/languages? And Spanish nationalists asking to change Catalan and Basque flags?

By the way, there are also a flags identifying: Shawi, Mozabit, Tamahaq a,d Rifain language.

I think the best thing we can do, is to split the Berber language to several real berber languages that correspond to reality rather than this useless corpora. None, could take advantage from it.

If people are opposed to the Kabyle flag, there is no a reason except politics, and we are not on Tatoeba to make politics.

Suggestion:

As Tatoeba community is growing arround the world, many activities are probably planned or done behind the project such as training, data processing... I'd like suggest a workplace where we can share information about these activities and publish planned actions.

As a contributor, I planned some actions like training sessions about Tatoeba itself but also evangelization about the importance of free linguistics corpora, develop some scripts that would help with data processing for everyone., contributions on github..

Language detetcion to analyse your corpus:

With a corpus of thousand of sentences, il would be hard to analyse them manually.

I shared two scripts on Github that deal with language detection using b-grams.
If you would like analyse a specific corpus, you can generate a model for your language from a corpus that you are sure contain only sentences of your language and then use the model to analyse all Tatoeba sentences and detect sentences that are not in your language.

The script will not work for macrolanguages (that contain several variants such berber) because of morphology.
You can find also a script to extract sentences tagged with your language code from the sentences file dowloaded from Tatoeba.

https://github.com/mozillakab/lang-detection

PS:
Theses scripts are written in Python. Models for Catalan, Spanish, Italian, French and English are also included. Later I'll add models for Hebrew, Euskara, Arabic, Persian, Russian, Portugeese. IF some one would a model for his language, please ask me.

Enjoy!

For those who are interested in dowloading the audio recordings for a specific language and extract only the sentences with audio for that language, I shared a script on Github.

The script is written in Python and tested on Windows and Linux. You have to install wget python library.

Don't forget to change the code of your language.

Enjoy!

https://github.com/mozillakab/K...-recordings.py

I will download again and try to fin out why I got less sentences within the file.

yes, I will download again and try to find out why.

The website shows 83190 kab sentences and when I downlowded sentences.tar.bz2 file, I found only 152000 kab sentences. Anything wrong?

There is an erro when trying to make a search:
https://tatoeba.org/eng/sentenc...rom=eng&to=kab

Welcome to Tamahaqt.

Tamahaqt Tahagart (Tatergit n uheggar) has just been added by Taoteba. Thanks the the tech team and especiallay @Ricardo and Aissa Mahfoudh who is working on Tamahaqt language.

We will do the best to share in info with the Tamahaqt speakers to create bridges between Berber (Amazigh) languages but also with the whole world languages.

For information, Tamahaqt is a Berber language. About 200 000 people in the South of Algeria (but also Mali, Niger, Libya, Mauritania, Burkina Faso, Senegal, and Tchad) are still speaking it.

Again, Welcome to Tamahaqt.

Done and thanks

How can I do to make all my contributions by default CC0?

Thanks for all job you did.
We are testing while browsing the site.
Thanks for everyone.

There is an update announced on Github

For Kabyle, I suggest @Amazigh_Bedar and he can suggest more names to help.
Bedar is a linguist (Kabyle and other berber languages)

Welcome