A native speaker wouldn't say this.

Not knowing the French I find it hard to suggest what it should be, the following are only guesses:

"What else do you want to do?" "What else are you working on?" "What else do you want to get started on?" ?

?

Oh yeah ? I just heard and read it in a US movie...

It's Bruce Dern, who got the Palm in Cannes, saying it in "Nebraska".

There can't be a more US movie than this one...

http://tatoeba.org/fre/sentences/show/2024265

Australian, huh ?

Thanks for the comments. One guess I didn't make "What else have you got going on?" which is probably the correct English for this sentence (?)

The only comment I'd make is that there is some pretty ungrammatical English used in films and songs. As the English sentences in Tatoeba (as I understand it) are to be used to teach people English, I question the usefulness of putting ungrammatical sentences onto the corpus.

One song that comes to mind is "Is you is or is you ain't my baby?" by Dinah Washington (and many others). https://www.youtube.com/watch?v=7kefYrJQ9tc

A great song and, as a native speaker, I know exactly what the title means, but I won't be putting it onto Tatoeba any time soon (even if it wasn't copyrighted).

> As the English sentences in Tatoeba (as I understand it) are to be used to teach people English, I question the usefulness of putting ungrammatical sentences onto the corpus.

We'd be throwing away a good portion of the corpus if we were just aiming to accommodate the ESL folks. Tatoeba isn't opinionated about how its data is used and mostly just aims to capture a living language, not become another phrasebook of half-useful stilted sentences. If someone tomorrow was interested in better representing African American slang as it's actually used or been used, it would actually enrich the corpus. The only condition set forth in the guidelines for this case is for it to be clearly tagged as a dialectal usage that isn't the current standard.

This sentence does sound a bit off to my ears actually, but I have no clue if more natives would have the same reaction as you. I wish we had some sort of rating system for things like this so we can get some quick metric to decide if it maybe should warrant a tag or some sort of reconsideration or research.

Neither of you two know how people speak English in Montana. So what is this debate ?

as for copyright, it applies only to large portions of creations, not to snippets.

Maybe it's "got" and not "get". Hard to tell, really, when you hear it.
Since I find ample examples of "What else you got going on" in Google Books https://www.google.fr/search?so...fe=off&tbm=bks

but none with "get", I'll suppose it has to be "got" so I change it. But I'd like a local opinion...

Actually I didn't comment to start a debate or anything, and I think you should keep it the way it was. I just wanted to answer patgfisher's comment about ungrammatical sentences. Though personally, I think tatoeba should have a classification scheme of some sort, where a linguistic corpus can be subdivided based on the certainty we have that something is actually representative of a language or dialect, and the highest levels would employ a guilty until proven innocent approach, so we don't have to argue over sentences like this to death, and could just wait till the user who added it, is actually a native from that area and is willing to start representing his dialect, or the non-native but competent user who added it can document this sort of usage and tag it appropriately, without anyone arguing with him. I'll write more about this in a proposal or something later on. Hopefully, the community can improve on this idea or offer a better alternative.

>I think tatoeba should have a classification scheme of some sort, where a linguistic corpus can be subdivided based on the certainty we have that something is actually representative of a language or dialect

We already had this debate many times here. So I repeat I'm against rating sentences and the reason is simple :
It's not because a sentence is not used (and even not known, which is so often the case in English !) by 99.99% of the speakers of a given language that it is not valid nor representative of the way people speak in a certain area, community, profession...

Tags and comments are good enough tools to provide information to users.
Besides, so-called "crowd-wisdom" is irrelevant when it comes to languages (as it is irrelevant in sciences), since a majority of people can be grammatically wrong and actually are, most of the time.

Of course, you will retort that if a majority of people say something in a given way, it has to be the correct one, but that works only with a few languages such as English, but it doesn't work with many others such as Arabic, French, Spanish, Mandarin, ...where consistency rules largely apply and scholars prevail.

If I take the example that I know the best : French in Paris and in Brussels, here is approximately the situation :
95% of sentences are common to the 2 groups of speakers.
3% make people of the other group smile, laugh or raise their eybrows, but they understand it.
1% is not understood at all by the other group.
1% is believed to be understood, but actually causes complete havoc because they mean entirely something else.

There are 65 million people in France, and only about 4.4 million French speakers in Belgium, so Belgian francophones are outnumbered almost 1 to 15 by the French, so their 1% of sentences that are not recognized by the French don't ever stand a chance in a "representativeness" classification scheme.
This is actually exactly the same phenomenon that is responsible for the disappearance of dialects, languages, cultures, animal species, plant species, etc...
You start classifying, and we all end up eating the same insipid green apples, and all the delicious yellow, orange or red ones are no longer on offer in shops.
Or people find asian turtles more beautiful, so they buy them and put them in their gardens and they end up destroying native european turtles (because the asian turtles eat them !)
Sentences are like these species in danger of extinction. The process of selection is actually one of eradication.

This one sentence might not be uttered in most of the English-speaking world, but it might nevertheless be representative of English in North Montana...

> so Belgian francophones are outnumbered almost 1 to 15 by the French, so their 1% of sentences that are not recognized by the French don't ever stand a chance in a "representativeness" classification scheme.

But in fact it would stand a chance in the scheme I'm proposing. If we had a belgian person that other french speakers see as having competent french in general, all it would take is his single stamp of approval for belgian sentences to be included into the 'high certainty' part of the corpus. It's a white listing scheme not a voting scheme where a majority dominates a minority.

> If we had a belgian person that other french speakers see as having competent french in general

But they just wouldn't ! That's the problem !

Why is it that there still has to be, in the understanding of many, THE correct way?

> But they just wouldn't ! That's the problem !

Now I'm curious. Why wouldn't they? Didn't you mention that french speakers would still agree with whatever belgian speakers would say 95% of the time? That's more than enough for them to get into the web of trust of the main french corpus even if we assumed that all of our contributors are from France. But this gets me thinking that maybe our use of iso codes to divide linguistic corpuses might not be that great to begin with...

> Why is it that there still has to be, in the understanding of many, THE correct way?

yeah why should we be making a corpus in the first place? let's just dump in anything that looks like a language, case-in-point look at the japanese corpus, it's not a pretty sight and needs years to become reliable in any sense or for any use case.

Why should you be using the unethically fallacious argument of reductio ad absurdum, which is fallacious because everything looks bad when reduced to absurdity?

What I mean is I don't like the idea of there being the one correct way which is the best because it is correct and is correct because it is the best. There are often several useful ways to express things. Not every way is useful though, of course.
(I capitalised 'the' for a reason, you did fine to miss it.)

> Why should you be using the unethically fallacious argument of reductio ad absurdum, which is fallacious because everything looks bad when reduced to absurdity?

Because it actually is a reality on tatoeba. I gave an example of an extreme. I'm still not sure what your contribution to this discussion was with that comment and assumed that you don't like the idea of offering some sort of reliability scheme for tatoeba's data to begin with.

> What I mean is I don't like the idea of there being the one correct way which is the best because it is correct and is correct because it is the best. There are often several useful ways to express things. Not every way is useful though, of course.

Then we actually are in agreement on this issue... I'm just not sure how it's relevant to the topic here.

>Didn't you mention that french speakers would still agree with whatever belgian speakers would say 95% of the time? That's more than enough for them to get into the web of trust of the main french corpus

That's where you're wrong.
If I consider a contributor is wrong 1% of the time. Then, to me, he/she's not reliable enough...
That's the way I think and that's the way contributors to Tatoeba think, in general, in my (relatively long) experience. And that is normal. Because that is the way we assess mastery of our own language by people in general.
In France (but I know it's even more severe in England, where any upper-class individual is able to place anybody's language in a level of 7 classes !), we judge the level of mastery of French along 2 or 3 basic criteria, and the slightest mistake, under these criteria, rules you out for all the rest.
Don't forget that language is, first and foremost, a discriminatory tool to differentiate social classes.

>But this gets me thinking that maybe our use of iso codes to divide linguistic corpuses might not be that great to begin with...

That is indeed an issue. But if 2 dialects share 90 to 95% of their corpus with each other (which is the case between Québec French and France's, or probably between UK English and Australian's...), creating different corporae would be a waste of energy...

Again, I think tags are ideal and suffice.

I had actually asked that question in a sort of rhetorical way, but (a stupid idea, perhaps) half-hoping Celtic would reply to it out of the blue because we know he likes going off on tangents to explain his ideological position (which I see as sometimes great and sometimes distracting).
---
I myself support a very simple idea: if something feels natural to a person, they should be completely free to add it, while we'll trust they're not a madman whose perception is uniquely modified far beyond reason or a troll who is playing with the corpus on purpose (and take action if they turn out to be such), and I think we can generally trust those who would be willing to take part here on that.

> If I consider a contributor is wrong 1% of the time. Then, to me, he/she's not reliable enough...

That's actually pretty extreme imvho. There are sentences that amastan writes that I find sort of weird. They're definitely more than 1%. I still wouldn't doubt he's a native and wouldn't hesitate to vote him into a web of trust for the standard arabic corpus. Same goes for Samer or Eldad. Eldad isn't even a native but he's pretty good, a lot better than some of the natives I've had to read their writing.

I wouldn't of course say that this isn't possible. But I'd like you to propose a way to fix this, since you actually seem to have more experience with this sort of thing. Maybe we can separate the web of trusts related to a particular corpus by regions and have them created by admins for new regions if we get contributors that we can confirm somehow that they actually are from there, for that variety of the language to get a foothold?

> Again, I think tags are ideal and suffice.

Tags are a hidden tax whenever they're used for something that should be part of a streamlined process. In this case at least, the kind of controversy generated by sentences and people who want to tag them with whatever because they think it isn't representative enough of the mainstream and isn't good for general consumption without ample warning, ends up being a monumental waste of time and energy and a constant source of unnecessary conflict and quit rates of major contributors. This isn't a pleasant situation for anyone. Everyone loses including tatoeba.

> "Is you is or is you ain't my baby?"

What does that mean?

>Maybe we can separate the web of trusts related to a particular corpus by regions and have them created by admins for new regions

But that's where it doesn't work. Because linguistic "regions" don't exist.
From Paris to Brussels (currently I'm trying to inventory different accents and expressions between Charleroi and Brussels, which are just 60 kilometers away...) and from Casablanca to Cairo and Baghdad, there are INFINITE variations !

Take Rovo and me (2 main contributors of the French corpus here) : he's from Eastern France border with Luxembourg and Germany, and I was born in the West, studied and worked in Paris and now live in Brussel and work in Paris most of the time.
His French sounds somewhat "German" to me (even though I am very much used to German myself). And mine probably sounds also bizarre to him.
That is actually why we hate each other so much, because when he popped up here, I dared tell him that he was writing a few germanisms. He hated me as soon as I hinted that, and has never ceased since (consult the history of our exchanges when in doubt)

People just hate when they're spotted ! Everybody wants to be the center of the universe...

> Because linguistic "regions" don't exist. From Paris to Brussels (currently I'm trying to inventory different accents and expressions between Charleroi and Brussels, which are just 60 kilometers away...) and from Casablanca to Cairo and Baghdad, there are INFINITE variations!

Are you saying people from belgium would have trouble recognizing each other's status as native speakers of french?

> He hated me as soon as I hinted that, and has never ceased since

This is precisely why we need an implementation of policy or feature that puts an end to this. You at least agree with me that this is a problem and needs a solution right?

> "Is you is or is you ain't my baby?"

What does that mean?

@ alexander
It is extremely bad grammar. The meaning is "Are you or aren't you my baby?" "baby" here meaning "lover" "sweetheart" "boyfriend" etc.
The point being that the singer laments the poor behaviour and attitude of her lover/boyfriend etc. which leads her to question whether he is, in fact, still her lover/boyfriend etc.

Thank you for the great explanation.
By the way, "Are you or aren't you my baby?" would be in German something like
"Bist du denn nun mein Liebster oder nicht"?

And what I always wanted to ask, is
"that don't impress(in) me much"
also an such example of bad grammar? Could it be slang, dialect, an poetic licence ...?
---> https://www.youtube.com/watch?v=jQZTK6xJ1Bw

(In my abysmal naiveté I suppose that should be "That doesn't impress me much". ☺)

An interesting thread on this topic: http://dict.leo.org/forum/viewG...idThread=10808

>Are you saying people from belgium would have trouble recognizing each other's status as native speakers of french?

Belgium is quite interesting at that, indeed, because it's a fairly recent state. At the linguistic scale of time, French and Flemish have no unity because different regions were part of very different Empires. The city of Liège, which is francophone, and that of Antwerpen, which was mixed, we're part of the Holy Roman Empire, while other parts of Belgium were constantly disputed between Burgundy, France, Spain, Austria...
This ended up in no unified education system and the use of a very complex continuum of dialects. In a region as small as modern day Flanders, they speak no less than 5 different dialects which are not entirely understandable to each other.
Similarly, North France has been the theatre of border disputes and fierce wars with England, Burgundy, the Holy Empire, Spain and Austria, so the border was very variable over time. This ends up in a continuum of dialects between Northern France and Belgium, there are also - few - Flemish speakers in France. So some Belgians speak a French that is closer to that of the rest of France than to any other part of Belgium. So all Belgians think they speak the standard "Belgian" speak, but that doesn't exist at all.
It happens to me all the time that I try to apply expressions I just freshly gleaned in one part of Belgium, in another one and it falls flat because they don't understand it there and would all swear it's not Belgian...

>You at least agree with me that this is a problem and needs a solution right?

But I think there is none. It is in human nature to be very touchy on identity issues, and languages, as well as religion ( these two factors being often intertwined) are at the heart of identity.
And there is a real identity and a fantasised one. People often don't really know their real identity because it's so complex, being the result of thousands of years of history. Not everybody knows history and languages.
For example, if I make a remark to a belgian francophone, who knows Dutch, that he has just said a sentence that is a grammatical nederlandism, they will have to agree, because it's obvious. But if I make the same remark to another Belgian who doesn't speak Dutch, he will protest vehemently, and will consider it a casus belli because who am I, a foreigner, to tell him what is his own language ?
It's a bit like when you tell a chiite Iranian that his ancestors were sunite until the 16th century and that their further ancestors weren't Muslims but Mazdeists. If they don't know history, they will express shock and anger and your life might be at risk to utter such blasphemes...
We are all 100% certain of our identity and what defines it, although most of us just ignore what it is actually made of.

> But I think there is none.

I'm really sorry that you feel that way. While this discussion was interesting, I was mostly just interested in improving the corpus's reliability and not in linguistic relativity or philosophical arguments about identity. I really do think that in practice there can be a very simple solution that would work fine without being too restrictive. We'll have another round I guess on the wall, but by then there probably would be a better governing structure for this project and the community can decide on what ideas to experiment with in the near future so we can move forward not backwards on issues like these. Simply saying that this isn't an issue isn't gonna help certain corpuses on tatoeba to stop being such a joke. I guess I'll go back to talking less and coding more ^^"

I hear this sentence used a lot by my native-speaking classmates, I also use it myself sometimes. All it'd need is a "colloquial" tag maybe.

The punctuation is another argued-to-death matter.

Edit: The right word (to me) is "got", not "get". It's probable that in the movie, the accent made it sound like "get".

Could the space be removed from before the question mark now?

Done, I have adopted the sentence. Regarding whether this is a good English sentence - I hear native speakers say this sentence all the time, and I am one of them. (But indeed, it is colloquial.)