An important note, make sure that the inscription (SPQR) of the Latin flag is legible. I'm not sure if it was the contrast of colors or just the size of the letters, but I could not make them out as clearly as in the flag we already have for Latin. That being said, I DO like the colors of flag #2 much better than the colors of the flag we already have; I would just request to maybe increase the boldness of the letters a little bit.

I'm in favor of #1 for Latin and #2 for Ancient Greek.

They have already requested the language properly and Kabyle is all set to be added on the next update. Belkacem77, we just need to be a patient a LITTLE bit longer for the site to be updated.

https://www.wasabi-jpn.com/japa...mar/quotation/

Though we cannot add Taiwan Mandarin as a separate language, we have a tag number 2636: "Taiwan Mandarin". If you contribute a sentence (or see another sentence on this site) in Taiwan Mandarin, you can write "Taiwan Mandarin" as a comment for the sentence and somebody can come and tag that sentence.

Hi, those flags were suggested by native speakers of those languages, despite the concerns you mentioned above. Also, other languages are spoken in Samar as well, so indeed, I am in favor of changing these languages' "flags" to a Philippine flag with the ISO 639-3 code, in all cases but Cebuano, which uses the flag of Cebu. Shall I change the Waray and Chavacano flags? And how about Hiligaynon (Ilonggo)?

We're very happy to have you here; it's been a long time since we've had a Gujarati user. If you have any questions, please feel free to ask on this wall or as a private message to any of the members who regularly write on the wall!

>> Many of these short sentences are near-duplicates, like "Tom is tall" / "John is tall" or "The car is green" / "The car is yellow". These duplicates are really tedious to translate and aren't helpful.

Near duplicates may be near duplicates in some languages and not in others. If Tatoeba will become as large as we hope it will someday, it will incorporate all types of languages with all types of noun classes/declensions/etc. Some languages differentiate between 'gender', animate/inanimate, whether the noun is possessed or its relative use to humans, whether it is part of a greater whole, etc. Even 'yellow' could be interpreted differently in different languages. And let's not even mention verbs (though I will, very briefly): verbs like 'go' can vary if you are on foot/in vehicle or going uphill/downhill in other languages.

Welcome aboard! We all make mistakes trying to figure out the workings of the site, and some of us do get frustrated with all the rules here, but once we get to understand the reasons for them, we realize just how neat the site is and what a great tool it can be :) We thank you in advance for your contributions!

- Cueyāyōtl, one of the site's Corpus Maintainers

I agree with odexed that censorship here is quite vague, but it is more important to have more of a variety of sentences rather than many sentences of that format: "X is a **** head!" just by replacing 'X' with one of many other names. In the end, it may just be best to use names like Tom or Mary.
I think the main instance where such sentences were deleted was when using one of our users' names.

Thank you for your input; I'm sorry if my words were confusing, but I meant to say that we have "failed to obtain" or, in other words, that we have "lost opportunities to expand our corpus".

Some of us come from places where people have been or are routinely ostracized or even killed for speaking minority languages. Organizations like the Summer Institute of Linguistics come to our villages, document and study our languages, and then bogart the information in volumes and other materials which require payment to be accessed. They in the end rarely help the people whose languages they study; I have personally experienced cases where the SIL uploaded a dictionary onto their site and no one from the villages where they speak the language knew about it. Tatoeba gives us the opportunity for us to share our cultures and languages freely, to create learning materials and recordings free of charge so that anyone who wants to learn a language, can. Realistically, we cannot add the thousands of "languages" that have ISO 639-3 codes, but we must do all in our power to include everybody.

>> Are we just adding these so certain members can show off their (likely less than perfect) language ability, or so that the Tatoeba Project can brag that it supports a large number of languages?

I have no idea how one can come up with such petty reasons... Maybe someone who likes playing numbers games?

+1

This is definitely a great idea, and I hope it can help us minimize the number of languages with few contributions.

We've lost a huge number of possible contributors due to us not having their languages on our site. Those who have been on this site long enough know that it is possible to contribute in languages not yet added to the site, but unfortunately new users have no way of knowing this. I've added either languages of previous users who have left the site or languages with many native speakers in order to increase our chances of having a native speaker (in fact I was able to catch a native Minangkabau speaker who created an account and wasn't adding sentences: hopefully we will see much more of her here), don't worry though, I always consult native speakers before making contributions :) As has been seen with languages like Marathi, Berber and Macedonian, all it takes is a single, dedicated user to create a large corpus. Now that we can add languages to our profiles and filter out other languages, there is really no reason why we couldn't add the over 7000 "languages" with ISO 639-3 codes; it is purely for scholarly purposes and exploiting the potential of those who browse the site, "bragging" that Tatoeba supports many languages has nothing to do with it.

I've linked the indirect English translations that I felt matched.

First of all, we are sorry you feel this way, however there simply have to be certain conventions by which we must abide in order for this project to be successful.

1,3) I will reply to these two issues together. Copyright laws are stronger in some countries than they are in others. Even YouTube follows copyright laws and takes down millions of videos that break them. If you like poetry, then great! So do we :) Just make sure that the actual poetry cannot have any copyright protection applicable. If the artist died more than 50-100 years ago, you can be sure that their works can be uploaded here on Tatoeba.

If you add sentences in other languages, they may contain mistakes and be overlooked by native speakers (remaining uncorrected). This will unfortunately result in lowering the quality of the corpus, which we have to minimize at all costs. Even if your sentences are corrected, they may still sound unnatural because the sentence structure you suggested may not be the best, so that the sentence gets corrected AROUND the structure that you suggested, rather than reworded to sound more natural.
If you want to practice English, Lang-8 is an excellent site to do so! I often use that site myself :)

2) sil.org is full of great studies supporting their decisions for separating languages; definitely check them out. Your case is similar to mine; my native language is Nahuatl and we too consider it a single language. However, in all honesty, some linguistic varieties are just TOO distinct to be considered to be the same "linguistic unit"; it is even worse for Zapotec (which has been spoken in the area for thousands of years). We simply cannot have all varieties of Nahuatl added into ONE category; we must divide them into "linguistic units" which we call here "languages". Doing so is no easy task, as no matter what you do, somebody will disagree. But, it must be done through research and field studies, and careful analyses of linguistic variety, especially when dealing with a linguistic continuum.
You must recognize, that a rose by any other name... is still a rose. What we call the Arabic macrolanguage, you call the Arabic language, and what we call the Gulf Arabic language, you call the Gulf Dialect of Arabic. My Nahuatl language is divided into 30 "languages" through the ISO 639-3 classification, and to me they are "linguistic varieties", but in the end, I know what each "linguistic unit" means, and so does everyone else... so now we can work together.

We hope that you understand, and we do apologize if we hurt your feelings, as that is most definitely not our intention. Please do let us know if you have any other concerns that need to be address, we will be here for you :)

I agree with you 100%, and we ARE going to stick with ISO 639-3 classifications, no matter what anyone else says. As I said before, I STRONGLY disapprove of using the word "dialect" in the sense of "inferior/undesirable language", or as you put it in a less demeaning manner: "language without (standardized) grammar".

Gulf Arabic is one of the Arabic languages spoken by MANY of the Arabic users who come and go through Tatoeba. As most Arabic speakers who speak Gulf Arabic natively appear to not consider it a separate language, nobody has either requested it or contributed sentences in it. So we have lost many opportunities to develop a corpus of Gulf Arabic Sentences.

We will definitely add it (under the code AFB), but we must now consider a name by which to call it. My opinions on the subject seem to be identical to yours, so I have no issue with just calling it "Gulf Arabic" or even "Arabic (Gulf)", but I wonder what name we should use in order to attract the most native Gulf Arabic speakers and offend (and repel) the least amount of them.

Several Arabic speakers left their complaints here:
https://github.com/Tatoeba/tatoeba2/issues/1084

They mostly condemn the separation of "Arabic" into multiple languages rather than being dialects, and urge Tatoeba not to be "political" about its choices. I mainly disagree with everything they say, but please take a look for yourself.

If we want native speakers to contribute, we have to try to offend them as little as possible, so we have to do more unorthodox things (like naming languages in a more unconventional way). This is why I am asking other people's opinions on the matter.

My reply:

Right, Tatoeba is definitely not in the business of making political decisions, rather we follow ISO 639-3 where the SIL (Summer Institute of Linguistics) have made those decisions for us.

Being from Mexico, I must say how much I despise the word "dialect" as it is used in our country to mean inferior or undesirable language and usually is used to refer to Amerindian languages exclusively. The distinction between what constitutes a language and a dialect varies from source to source, which is why we have decided to use the ISO 639-3 classification.

Now, as for what we shall then call afb , I am EXTREMELY against using the actual word "dialect" in the actual name of a language, but I fear it may be our best choice in this case. "Arabic Dialect (Gulf)" may work for afb , however we cannot simply change all Arabic languages to this format, as attitudes differ in different regions that speak Arabic languages. In my experience, Egyptians call their language "Egyptian", and not even "Egyptian Arabic". I wonder how much renaming arz to "Arabic Dialect (Egyptian)" could hurt our chances of having native speakers; could it be that they hate the word "dialect" as much as I do when referring to their language? We should keep the name as either "Egyptian Arabic" or "Arabic (Egyptian)" The renaming of Arabic languages has to be on a language by language basis.

I am bringing a discussion on GitHub about the naming of Arabic languages (or "dialects" as some would prefer to call them).

loolmeh's original post:
While I'm still disillusioned by the discussion on (https://github.com/Tatoeba/tatoeba2/issues/1084) which amounts to a projection of a disgraceful and malicious victimhood mentality on none other than a technical issue tracker, I think something can be done to more explicitly indicate that Tatoeba isn't in the business of making political decisions.

This pull request has a few sql update statements to relabel ara and relabels arabic corpora with the pattern Arabic Dialect (...). Please check if I missed a table that should be updated and if active contributors on these corpora object to this change. Also I'm not sure what the impact on urls is gonna be from these changes for seo purposes.

@trang I think you should reconsider the current wave of migration of nontechnical discussions to github instead of confining them to the wall.