@Pfirsichbaeumchen

Search: Nedda ɣer

https://tatoeba.org/eng/sentenc...rom=kab&to=und

@TatoebaAdmins
@Pfirsichbaeumchen
@CK
@AlanF_US


I just want to draw Pfirsichbaeumchen's attention to two facts:

1- The pattern is the same:
"Nedda ɣer" literally means "We went to".

You can find it here in many other example sentences:
https://tatoeba.org/eng/sentenc...rom=kab&to=und

The only thing that changes is the place name/destination. Tubiret, Bgayet, and Budwaw are all Algerian towns.

Now is this acceptable? Is this ethical in first place? Is this fairplay? And... what's the point of this and how far can it go?

Can I take the lists of all the 3000 counties of the United States + 1600 municipalities of Algeria + the 36,681 municipalities of France and contribute something like 40,000 sentences all starting with "Nedda ɣer" to expand the corpus of my language and make it gain 40,000 more sentences?

Looks like those who are doing this aren't against this idea.

Yet this website isn't supposed to become like that.

I know the world is changing. I know that some day, black will be white and white will be black. I know that the sky was once blue and now, although it's still blue in some parts of the world, they might be people who would call it "pink" or "purple" or even "black" although they too see it as blue as you and everyone else sees it. But I also think, in my very humble opinion, that Tatoeba should always continue to be a website primarily made for translation and learning, and not be used as a race track to dump the biggest possible number of sentences just to have a corpus bigger than anybody else's, and sentences that:

1- Hardly make any sense.

2- Often, untranslatable (I could suggest you to challenge some of the authors to even explain you some of those sentences if they can - I'm a native speaker, I know what I'm talking about).

3- Hardly serve any purpose.


2- This profile seems to have been created by some member just to do this type of dumping.

That's an issue with some of these contributors, and I agree with you in that aspect. Just be sure not to lump all of them in as one and blame them all, including those that have been doing a good job with the guidelines so far.

Let's focus on taking care of the ones that are going at it misguidedly

@DJ_Saidez

I have always enjoyed good, constructive, and transparent debate. Thanks for your suggestion.

Maybe we need to set up rules and even appoint people to watch the corpus and enforce those rules. Maybe we've got to determine how many sentences with a certain pattern is acceptable and how many is not...

I know it's a complex issue but no matter how we solve it, this nuclear explosion of insipid undigested material that these guys are flooding the website with is quite alarming and needs to stop.

Just take these two sentences for instance:

Ur nejbi ara ɣer Iɣil n Yefri.
https://tatoeba.org/eng/sentences/show/9353379

The verb "jbu" can hardly be translated by the author. I would say that the sentence means "We haven't been to Iɣil n Yifri". I don't even know where this "Iɣil n Yifri" is located and neither can you, I guess. I'd also translate it as: "We didn't go as far as Iɣil n Yifri". Anyways, this is not really a good example sentence. Maybe the author of this sentence could try and translate it to us.

Afeg ɣer At waggur.
https://tatoeba.org/eng/sentences/show/9353362

Literally: Fly to At waggur.
Or:
Fly to the sons of the moon.

As a native speaker of Tamazight (Kabyle dialect, etc.), I don't understand the meaning of such a sentence. Is "At waggur" the name of a village? Then in this case, it needs to be spelt "At Waggur". Is it "the sons of the moon" or "the tribe of the moon" or the "people of the moon" or "the aliens that live on the Moon"? Then in this case, the sentence doesn't make any sense. I don't see any way how a user of Tatoeba, be they a learner of Tamazight, a teacher of Tamazight, or simply a linguist, would use or need a sentence like this.

Tatoeba should primarily be a LEARNING TOOL IN THE SERVICE OF TRANSLATION AND LANGUAGE LEARNING. Although its primary goal is to give example sentences, it's also preferable that these sentences be translatable, MAKE SENSE, are pedagogical (teach the language), are diverse by teaching new vocabulary. However, with sentences like those I just showed you, they hardly teach anything.

@Amastan

We know very well what you think of the Kabyle language (a dialect that should not exist).

I’m going to disappoint you because the Ath Waggour village does exist. It’s obvious that you don’t know Kabylia and even less the Kabyle language.

https://www.depechedekabylie.co...ur-la-commune/

An official Algerian website.

But why do you come to the kabyle pages all the time to sabotage the kabyle language? Do kabyles come to your pages (to do the same thing)?

@Igider That was uncalled for. And this brings up the question of wildcards. Even if the sentence is correct and is describing an actual place (not saying it's not), if this logic is carried through there'll be 40,000 sentences that are exactly like each other, save for the place being referenced.
I don't know much about the conflict, but I'm not against a Kabyle language.
All that he did just now was suggest that we find someone fit to be a corpus maintainer for Kabyle, which I fully agree with.

@DJ_Saidez

Afeg ɣer At waggur.
https://tatoeba.org/eng/sentences/show/9353362

@Igider
I edited my comment because I didn't read your reply closely enough.

That's like me adding the following sentences:
I live in Escondido.
I live in Nakano.
I live in Fallbrook.
I live in Staten Island.
I live in Osaka.
I live in Tijuana.
I live in San Gabriel.
I live in Hermosillo.
I live in Oaxaca.
I live in Menifee.
I live in Marseille.
I live in Tucson.
I live in Denver.
I live in Kobe.
I live in Santa Monica.
I live in Nogales.
I live in Des Moines.
I live in Guasave.
I live in Veracruz.
Etcetera

Do you want to translate all these sentences?

@TatoebaAdmins
@Pfirsichbaeumchen
@CK
@AlanF_US
@DJ_Saidez

I'm publicly announcing that I won't post any direct reply to people from the Kabyle dialect community that attack me personally. It's completely pointless to argue with them. Some of them have a bitter personal grudge against me. This is completely irrelevant here.

I just propose that we discuss this topic in a very honest, transparent, peaceful, and constructive way, away from personal attacks and accusations of "political crimes."

Any other opinions would be welcome.

Attacking people personally,
Going off topic
Questioning people's attentions when they bring up a sensitive topic

All these are vain attempts meant to stop this sensitive topic from being addressed.

If a person claims that they're from an "ultra-fragile delicate ultra-sensitive paranoid persecuted insanely targeted tiny small political ethnic cultural language community" doesn't place them above the rules managing a website like Tatoeba.

Likewise, those people who claim they're from an "ultra-small microscopic delicate ultra-sensitive paranoid passive-aggressive insanely targeted UV-sensitive ultra-allergic tiny small nanometric political ethnic cultural religious language community" shouldn't allow themselves to attack and insult the admins of a website in case the latter choose to apply harsher rules to manage the website.

The status of "ultra-small microscopic delicate ultra-sensitive paranoid passive-aggressive insanely targeted UV-sensitive ultra-allergic tiny small nanometric political ethnic cultural religious language community" shouldn't be a justification to destroy Tatoeba as a website.

@Amastan Maybe you could turn down the wording a bit there, but we should definitely do our best to stay civil
If both sides never come to a resolution, that just creates a mess that no one wants.
That's easy for me to say because I'm a neutral party, and I'll admit that, but that doesn't take away from my point that discussion is key.

>Maybe you could turn down the wording a bit there
Yeah.....I think you really should. It's obvious who you're talking about.

This attitude of "if-you-touch-me-or-dare-tell-me-anything-I'm-gonna-cry-and-tell-Mama-and-report-you-to-the-UN-Security-Council-and-try-you-in-a-court-of-law-for-war-crimes" no longer pays off here on Tatoeba. If we want to discuss something about just anybody and anything, we should be free to do it.

@Amastan
>I'm publicly announcing that I won't post any direct reply to people from the Kabyle dialect community that attack me personally. It's completely pointless to argue with them.
You're just provoking them at this point

@DJ_Saidez

> Turn down the wording...

Agree with you, buddy.

Who calls for hatred and accuses sovereignists with words as separatists, dangerous, criminals.... with words full of hatred and insults?

https://tatoeba.org/fra/sentenc...rom=und&to=und

https://tatoeba.org/fra/sentenc...&sort_reverse=


What can we do with those sentences in Tatoeba ? What can we do with person like that? Why he behave like an administrator ? Why @TatoebaAdmins let him do that ?

@DJ_Saidez

What do you suggest to solve this issue?

@Amastan Not have this discussion in the comments of a random sentence. And keep pathos out of it.

I have overdue essays to write and sleep debt to pay off, so I need to head out for today. But definitely keep admins in the loop to decide how to best moderate this.

I saw what Igider linked, and saw how they could use that as basis for feeling attacked. So if you stop it with that kind of behavior and set an example, they won't have any more relevant evidence to attack you with, and we can go from there.

For example, instead of using the word in a derogatory manner, change the sentences to more neutral sentences, so that the word stays in the corpus.

@DJ_Saidez

Do you suggest we go to the Wall?

I would, but maybe double-check with the admins.

@DJ_Saidez

They're dead silent for now.

Thanks for your advice :-) Glad that everything is clear to everyone ;-)

There are two major issues here: how to relate to other community members, and what type of sentences to post. For years now, there's been a pattern of mutual provocation and retaliation between people who identify as Kabyle and Berber speakers for years now. I'm not interested in discussing that right now other than to ask both sides, yet again, to stop it.

As for the bulk production of near-duplicate sentences, it's clearly a means of trying to generate as many sentences as possible in order to win some kind of individual or group record for number of sentences. It should be just as obvious that collections of such sentences are nearly useless compared to more varied sets of sentences, and that writing such sentences requires no skill.

Both kinds of behavior have the tendency to destroy the usefulness of the collections for individual languages, or of the site as a whole. If people are not actually interested in making their sentence collections useful, and if they're not embarrassed to be seen acting like children, there are limits to what can be accomplished. Admins can suspend accounts or delete sentences, and we have, but that should always be a last resort.

Regarding sentences of the form "Nedda ɣer X", there are only 32 of them, and they're written by a number of different people, so this is hardly the worst example of near-duplicate sentences from a single pattern. Still, my pride and my desire to make my contributions useful would lead me to try to create other types of sentences.

@yiwenkan

Perhaps you could add a Kabyle translation to this one, so we get a Kabyle translation for this, plus a Kabyle indirect translation to all the other languages in this series.

[#2739060] We went to Boston. (CK) *audio*

@AlanF_US

You said:
If people are not actually interested in making their sentence collections useful, and if they're not embarrassed to be seen acting like children, there are limits to what can be accomplished. Admins can suspend accounts or delete sentences, and we have, but that should always be a last resort.

Yes, but how do you propose we monitor the corpus to try and stop this behavior?

I have a few questions:

1- How much is too much?

2- How to monitor the website?

3- Who should monitor it?

4- Who to report this to so that they take the appropriate actions?

I just want to bring up that it's important that we have both sides involved in this, so someone from the Kabyle side who the Kabyle speakers will listen to, and someone from the Berber side who the Berber speakers will listen to, and both of these should be willing to cooperate with each other and the admins to come up with a legitimate agreement.
So if just you (a Berber supporter) and the admins came up with an agreement, Kabyle speakers can use this as grounds to retaliate, seeing as the agreement is likely to be seen as against them. But if someone from their side gave them a voice, they're more likely to adhere to any new regulations.

> so this is hardly the worst example of near-duplicate sentences from a single pattern

hardly ;)
https://tatoeba.org/eng/sentenc...&=Any+language
37905

@DJ_Saidez

You said:
so someone from the Kabyle side who the Kabyle speakers will listen to, and someone from the Berber side who the Berber speakers will listen to, and both of these should be willing to cooperate with each other and the admins to come up with a legitimate agreement.

My reply:
Thanks for your suggestion.

To move forward, I suggest @Ubezwi1 to be the trusted person for the Kabyle dialect corpus and, since I'm the most active member in the Berber corpus, I might be the person representing this corpus and its community. Ubezwi1 is a personal friend and I guess he's trusted by everyone in both corpora.

What's next?

>What's next?
I'd say asking other Kabyle users if they're willing to trust Ubezwi1 as their representative. Better safe than sorry.

@brauchinet Oh wow

@DJ_Saidez

Yes. Good idea. Maybe one of the admins could do that.

Seems we've made some progress here, @Pfirsichbaeumchen. Could you please do that for us?

Hello

the Kabyle corpus has already admins to manage it, we do not need to point another admin to validate sentences, BEF contributors can make suggestions to help us having better sentences and we will improve our sentences if needed, but definitely they can't decide for anything regarding Kabyle corpus since they are managing BER one.

Regards

@imalaqvayli We were asking if you're OK with Ubezwi1 as your representative to discuss how to move forward

@Dj_Saidez Are you un Admin?

It seems like a decisions were made behind closed doors...

You and @Amastan are fellows...it's ok but here we are not in club, we are in scientific project.

Regards.

Dears,
If something is wrong on the Kabyle corpus, it is up to Tateba's admins to point it out and ask the author to do the suitable corrections.
We didn't attack anyone and didn't blame anyone. We're just working seriously and assiduously.
Any suggestion/recommendation that did not come from the Tatoeba admins has no value.
If something needs to be corrected for one reason or another, we are ready to do so. But to bind us or compare us or force us to work with another team, or someone from another team suggest who would be our representative.... I don't see any justification for that.

KR

@Pfirsichbaeumchen
@AlanF_US

What do you think about DJ_Saidez and my suggestion to monitor the Kabyle and Berber corpus and report the cosmic-scale dumping of near-duplicates?

I'll clarify something before I step out of this conflict for now

My suggestion is that Kabyle speakers agree on a representative to speak for them. Then they can get together with the admins and the Berber representative (presumably Amastan), argue their points about certain topics (such as near-duplicates, or inflammatory entries), and hopefully come up with a resolution
I'm not trying to solve the entire conflict between these sides. I just want to find a way to keep things civil.

Also, I know it looks like I already teamed up with Amastan, but let me clarify my position in this. I'm not against a Kabyle language, and I feel like Amastan could do a better job of listening to them instead of calling them names. But this is not the place to fight about what's happening in Algeria and elsewhere, or to go against common practices to make a point. This is a language corpus. Focus on developing your respective corpus, and following the guidelines that everyone should follow in the first place. My hope was that I'd be able to moderate some sort of truce, but now I'll leave that to the admins, who are much more qualified to deal with this than me.

Bien dicho, DJ_Saidez.

We're still waiting for the admins :-)

@Pfirsichbaeumchen
@AlanF_US

Thank you @Dj_Saidez,

It’s very kind of you to try to get the wolf into the fold.
This is really nice from you.

Regards.

;-)

https://tatoeba.org/eng/sentenc...rom=eng&to=und

more than 32

@DJ_Saidez
We already have a representative from the beginning who is Amazigh_Bedar.

KR

@TatoebaAdmins

I'm a computer engineer.

I'm working on Voice Models to generate a Kabyle voice assistant model for GPS on OSMAnd, an opensource GPS.

Kabyle has no presence on such tools, we can't afford GMAPS or other proprietary solutions for Kabyle. So, we have to use some proper nouns. These sentences will be recorded later and we will train our model based on these sentences recorded.

We will also translate these sentences to English and French for another purpose : MT (Machine translation)

So, I asked people in my community, first to translate sentences, and second to produce sentences with kab Named Entities (towns, rivers, roads, villages, people...) and translate them into English and French.

I don't think there is something wrong. Kabyle is a suffering language and we need Tatoeba to gather free corpora and work on NLP models (POSTAG, Voice recognition, Voice synthesis, MT, ....). All of this work and more is shared for free on Github.

Thanks to Totaeba, the kab community is doing nice job.

@TatoebaAdmins

If there is someone here on Tatoeba working on computer linguistics as I'm doing, he will understand why we need Named Entities in Kabyle. So, even if the pattern is the same, the use of different NE is necessary for the job we are planning to do.

@TatoebaAdmins

We are planning to ask to record all Sentences with Named Entities for the next weeks. That's why we asked by the past to use CC0 licence. We will use them to work on kab voice models and share our work with other communities, in particular, suffering languages like Kabyle.

In my view, this particular conversation raises some interesting questions, but can and should be wound down.

I would hope that most people agree that contributing sentences that vary in structure and content is more valuable than writing a large number that vary only slightly. There were at least three examples mentioned in the comments:

(1) Nedda ɣer + <Algerian town>. (Translation: "We went to <Algerian town>.") - 32 examples
(2) Vado a costruire {edifici|ponti|ospedali, etc.} in <country>. (Translation: "{I am|You are, etc.} going to construct buildings|bridges|hospitals, etc. in <country>.") - 37,905 examples (search pattern: "a =costruire in")
(3) The separatists are {fascists|cowards|hypocrites, etc.}. - 138 examples

Pattern 2 is arguably the most valuable because it (a) shows how the names of countries are written in Italian ("Germania", "Inghilterra", etc.), (b) introduces a little bit of vocabulary, and (c) demonstrates a little bit of grammar ("Vado", "Vai", "Va", etc.). Having said that, a much smaller set of handwritten examples that didn't undergo multiplicative explosion (X verb forms x Y nouns x Z countries) would of course be far better.

Pattern 1 is less valuable because there's not much in it other than the names of Algerian towns, which are unlikely to have a variety of translated names across a large number of languages. I would encourage people not to repeat this pattern anymore. I don't agree that there is NO value in sentences that don't have much translatable content, but I do think that these are the least valuable from Tatoeba's point of view (particularly when they don't contain much of interest in terms of vocabulary or grammar).

Pattern 3, which incidentally was written by the person who complained in the first place, does introduce some interesting vocabulary (despite the lack of grammatical variation), but is of course insulting to other people. Naturally, this is against the rules and also a bad idea. I noticed these sentences a while ago, and asked the author to stop writing them, which he did.

Admins are very busy both at Tatoeba and with everything else we need to do in our lives. We try as best we can to keep an eye on genuine abuses, but there's a limit to how fast we can react and how much we can do. Please don't make our lives more difficult by either perpetuating the abuses in the first place or making spurious complaints. No one here has the time to sit on a monitoring board, so please don't contribute the kind of sentences that would make one necessary, and please try to deescalate conflicts rather than breathe new life into them.

@Igider @Yazid_Bouhamam

Please do address your comments to admins of other communities except Berber. And please keep commenting about quality and purpose of the Kabyle team as we are discussing on our space on FB. We need to grow our corpora and you know about our NLP projects.

I wrote my comment at the same time as @belkacem77. If you are trying to produce sentences for a language model, then I can see why you want to expand the number of named entities (like Algerian towns). Is it possible, however, to add more variety to these sentences? And is it possible to also add names of larger geographical entities that are more likely to have different translations in other languages?

@Igider please use other verb aspects Nedda, Ddan, Ddant, Ddan, Ad ddun, tteddun, tteddunt, Ttedduɣ, Tetteddum, tetteddumt.... etc. It's the best solution because we need to keep Named Entities and record them all.

@DJ_Saidez

Iɣil n Yifri, At waggur, ... are named entities. The user named Amastan does not live in Kabylia, so he does not know the name of villages, places, roads, ... in Kabylia.

In Kabylia where I'm living and working, not only big cities and villages have names, also small parts of places where we live have names. We are working on another project (OpenStreeMap), where we are putting these names. These names will be recorded in sentences that we will use later.

I'll ask the kab team to use other verb aspect instead of using the same aspect (Scheme).

@Igider, please would you like use other verb aspects. Thank you.

@belkacem77

Absolutely, although we never go to the Berber pages. But, it’s taken into consideration, we remain focused on our project, and I’ve already begun to enrich these sentences.

@TatoebaAdmins

Look at this pattern and who wrote them.

The separatists are

https://tatoeba.org/kab/sentenc...rom=eng&to=und

By the way, The targeted separatists are we (Kabyle community on Tatoeba) working on the Kabyle language corpus. But that's okay, the user is free to think as he wants and we're not asking him to change his sentences except denigration.

I'm just asking kab contributors to keep calm and not respond to provocations. Our goal is to go far in the digital integration of the Kabyle language since we are suffring from no representation on digital tools and voice assistants.

Thanks, @belkacem77 and @Igider.

Thank you so much @AlanF_US for explanations and time spent.

Thanks @AlanF_US for the complete and valuable answer, we (kab contributors) are always focused on how to improve and make our languages better and wider, we do not want any conflict, we just want to contribute to raise our Kabylian language As you may know, we have also our personal life, and such as conflits are time consuming for everyone. Be sure that we always try de-escalate conflicts, even when we are attacked.

Thank you so much for your very thoughtful, constructive, and helpful answer on this matter, Alan, and thank you, everyone else, for taking these points into consideration. All that Alan has put so well here and elsewhere is precisely my view as well.

While I see that machine translation for Kabyle would help tremendously and would be a very desirable tool to have, one which we sorely lack, and that one would need for it a large pool of sentences featuring all conjugated forms and all kinds of regional names, I'm also a bit concerned about Tatoeba becoming a database for machine translation that is in large parts as though a script had generated it. It is without question that we see this kind of material a lot all over Tatoeba, but the one that Tatoeba is for in the first place is the human learner of course. I think it should be possible to enrich these sentences in ways that would both make them interesting, varied, and fun enough for our members and still allow them to serve their apparent purpose as material for other projects, but you already promised you would do that.

Using different names in sentences that are themselves varied and reflect as much of Kabyle or other cultures as possible is good. The shorter and more standard the sentences are, the more likely those near-duplicates become, of course.

People often seem to think that a Tatoeba contribution has to be a single sentence and has to be short, but that is not so. I often find that one way to make a contribution more interesting is to make it several short sentences, a long sentence, or a dialogue, like this:

• We went to <place name>. We wanted to show our friend <name1> the famous <name2> there.
• "He went to <place name>." "Why would he go there?"

Maybe it would be worth considering developing Tatoeba further to have a "related sentence" feature where "I went to <place name>", "You went to <place name>" all count as one contribution that have a "related sentence" kind of link between them so this flooding of Tatoeba with similar sentences that people perceive will not be so apparent, but right now we don't have that, so trying to find a compromise and to not overdo anything will probably be best.

Lastly, I agree that these conflicts really have to stop.

I'm confident that this discussion we've had together has demonstrated everyone's point of view, and that all will be a little bit better for everyone now. :)

@Pfirsichbaeumchen
Vielen Dank für Ihre Antwort. Das weiß ich sehr zu schätzen.
Es lebe Deutschland, mein Land des Herzens!

Ich sage auch danke schön! 😊

@Igider
Et c'est notre ingénieur chez Volkswagen en Allemagne qui nous a proposé de travailler sur le modèle vocal automobile produit de Deep Speech!! Toujours l'Allemagne!!

On aura besoin de phrases en relation avec l'automobile, conduite, code de la route... C'est aussi utile pour le GPS.

@belkacem77
Très intéressant ce que tu proposes. Le vocabulaire de l'automobile et tout ce qui s'y rattache est très utile. J'y penserai sérieusement. Justement je pense aussi au système GPS, à la robotique, au deep learning et e-learning, chat bot... merci pour tes idées ingėnieuses.

I want to thank the Kabyle side for explaining more thoroughly what their mission is here, and I want to also thank @AlanF_US and @Pfirsichbaeumchen for their amazing insights and all that the admins have done to keep things running smoothly, you guys are awesome ^_^

@Pfirsichbaeumchen @AlanF_US

Thank you for your comments. I'd just like you to delete the personal attacks that some individuals have taken the advantage to insert in their comments. Of course, I'm not going to answer them.

Thank you for the kind words, @Pfirsichbaeumchen, @AlanF_US, and @DJ_Saidez.

@Amastan wrote:

> I'd just like you to delete the personal attacks that some individuals have taken the advantage to insert in their comments.

While this may seem to you like a small request, there's little "just" about it, in the sense of "merely" or "fair".

Deleting/hiding comments or sentences is a tedious and thankless process. Aside from the mechanics of hunting down comments/sentences, pressing a button, and waiting for the action to process, it takes time to decide what should be hidden and what should be not, based on complicated criteria, and rarely will anyone be fully satisfied with the results. Furthermore, deleting comments in a discussion disrupts the flow, removing the context that explains later statements.

I suspect that you are not interested in having your own sentences or comments deleted, despite the fact that there are hundreds that deserve it, and despite the role that they play in your being exposed to sentences or comments that you don't like. So please respect our time and let's end this discussion.

One thought to consider is that If instead of creating series of sentences that only vary in place names, hoping that someday these will eventually be translated into other languages, you could translate existing English or French sentences with the city name "Boston," many of which already have translations into multiple languages.

This would mean that, not only would your Kabyle sentences likely be linked to native-speaker-owned sentences in those languages, but if you get volunteers to also record audio for those sentences, you would more likely get bilingual audio pairs.


Try these searches to find sentences to translate

English Sentences with "Boston" and audio, not yet translated into Kabyle.
http://tatoeba.org/sentences/se...&has_audio=yes
10,014 results

Proofread English Sentences limited to List 907 with "Boston" and no audio, not yet translated into Kabyle.
https://tatoeba.org/eng/sentenc...b&has_audio=no
3,681 results

French sentences, limited to ones owned by native French speakers, with the word "Boston" that are not yet translated into Kabyle.
https://tatoeba.org/eng/sentenc...sort=relevance
508 results



Find more useful links on these pages.

http://study.aitech.ac.jp/tatoe...hp?f=eng&t=kab
English to Kabyle

http://study.aitech.ac.jp/tatoe...hp?f=fra&t=kab
French to Kabyle

@CK I don't know whether you've considered that, but if the goal is to aid in the creation of a voice assistant for GPS that can be used to navigate Kabylia, a collection of sentences only about Boston is obviously not fit for the task. What's useful for you isn't necessarily useful for everyone else.

That said, if you sort "French sentences, limited to ones owned by native French speakers, with the word "Boston" that are not yet translated into Kabyle." so that long sentences appear first

https://tatoeba.org/eng/sentenc...rt_reverse=yes

there are quite a few examples of more creative sentences than just "I went to X."

In my ideal dream world, there would be a sentence for every city, town and hamlet telling you something unique about that place. "The oldest building of X is ... constructed in ...", "Olive oil grown in the Y region is sold all over the country.", "Trains from Z station leave every ... minutes, making it the busiest station in the region.", "The fastest route between A and B is via C." (Imagine the last example with A = B = C = Boston.) From a GPS perspective, such sentences have the advantage that they don't just include city names, but also points of interest (POIs) within cities that people will likely want to navigate to. In my neighborhood in Berlin, OpenStreetMap has markers for a theater, a dentist's, a restaurant in a former school gym, a church, a charger for electric cars, a fountain in a park, even individual park benches... I don't know how well Kabylia is mapped, but there should be plenty of things to write about.

Thank you, Yorwba. That is a great comment with points that one would hope will be considered. 🙂

Many thanks @Yorwba. And for all @admins.