No offense to those who put it up, but without word parsing it’s unusable.
Example:
Cà phê = coffee
Lingq:
Cà = egg plant, phê = criticism
I’m not a programmer but it seems like there’s some options on github.
2 Likes
Currently it is what it is due to being a beta language after all.
That’s a great idea to implement parsing abilities from GitHub code if it works on their system. At the moment, we can only read and lingq with caution to attempt to guess the right words from the wrong pairs but you’re not wrong too ahaha. It slows progress by a lot but it’s better than nothing I suppose
You’ll find that LingQ Support is well aware of this issue, see eg:
My example word was ironically Việt Nam 
Feedback I’ve had is that they are looking at how improvements could be made, but that was a few weeks ago and I’ve heard nothing further.
But I agree that this issue really is almost a deal breaker. You start off thinking that it only affects a word here and a word there, but honestly the cumulative effect when trying to learn a monosyllabic language like Vietnamese with LingQ becomes endlessly frustrating and time-wasting. It’s so demotivating, which is really sad when I’m otherwise very keen and willing to spend time on learning the language.
As an aside, I do see this comment that some languages like Vietnamese are still in the beta category, but I’ve never found out where this is documented anywhere. Is there a list somewhere of LingQ beta languages tucked away? And does it have any specific implication other than the usual generic meaning of beta? For example, I can see that the Ministories are complete (though I have criticisms of the translations), but there are no courses as yet. I wonder what else is missing?
Not sure if this helps, but I’m learning Chinese and at times encounter a similar issue. Lingq has an inbuilt dictionary function when adding words, you could use that instead of the automatic lingq recommended meaning. That’s what I do whenever I see that the word has been translated completely wrong. Obviously takes a couple more clicks to do and is tedious, but it might help.
Really not sure what this means. Yes LingQ is a software program (what other sort is there?) and GitHub is a massive repository of users’ own software creations. It’s like saying there’s a book in the library which explains how to perform some task, but completely omitting any further details. Which specific word-parsing algorithms or code libraries did you have in mind?
I have previously made a couple of suggestions for improving matters (see linked thread) such as suppressing LingQs for individual words when a multi-word LingQ supersedes it (not the best solution but much easier to implement I think and it would still be a big step forwards) or checking text against Vietnamese word frequency tables that explicitly include multi-words in the frequency hierarchy.
Maybe keep reading it over and over until you understand.
Here you go:
The online links used to work and you could dump text in and it would parse for you. I’m not a programmer but I don’t think it would be that hard to implement. But what do I know? (nothing)
IMO they threw the Vietnamese one up on linq to appease people (AFTER MANY YEARS) but don’t have the money or will to make it functional.
full disclosure, I haven’t studied Vietnamese in over a year, and only studied the mini stories. I’m not currently paying for a premium membership but I was a paying member when I was studying it and I thought the linking feature overcame this pretty well for me. Is that helpful for you at all?
@amemac You mean making LingQs manually, ie treating the Vietnamese ‘multi-words’ as phrases that you add as a LingQ manually.
Yes, this is the only option but, as explained in more detail elsewhere, has 2 big drawbacks. First it is very tedious and time-consuming to do, bearing in mind that 30-50% of Vietnamese text will consist of such multi-words. So progress through eg a mini-story becomes mind-numbingly slow.
Second, LingQ still also automatically creates LingQs for the individual words in a multi-word, so you end up with eg vocabulary lists that are 3 times longer than they should be, adding considerable complexity to the task of picking out the correct meaning for a word.
The combination of these factors plus others like Vietnamese being a pretty difficult language for Westerners to learn in the first place and (IMHO at least) the ministories not being optimally translated** plus the lack of a Northern Vietnamese voice and you end up with LingQ being very tough going for A1/A2 beginners in Vietnamese.
**Note I am not suggesting that the ministories are badly translated, but I rather suspect that they have been translated too accurately. What I mean is that the translators chose the most accurate words to use in Vietnamese rather than the simplest words. Ministories may have high frequency words in English, but that doesn’t automatically mean that high frequency words have been used in Vietnamese. I can’t see that it matters one jot whether the translation is 100% accurate or not, instead the priority should be that the translation into the target language uses only high frequency words and constructions.
BTW none of this is intended as major criticism of the LingQ team. I know they have only limited resources and are trying their best to cover many (too many???) languages. And I appreciate also that certain languages throw up unusual challenges like Vietnamese multiwords. But I do think that LingQ should stop trying to cover too many bases and resist releasing new languages until they are properly ready, Or at least have new languages beta tested and then issued only with a clear and prominent health warning if major issues remain. (An obscure beta designation somewhere non-obvious is insufficient - I still haven’t managed to find such a label for Vietnamese.)