I am learning Finnish. When I started learning Swedish as well, I noticed that different words have a different number of coins to it. At the same time all Finnish words seem to have 1 coin attached to them.
I was wondering what effects this could have? Also, does this make progress in Lingqs be slower than for other languages?
I had this same problem. It really made the 90 day challenges frustrating!
It turns out that all beta languages assign one coin per word. The non-beta languages assign a number of coins based on their frequency.
Well … that is annoying. I guess if the mass of words is insufficient this reasoning is valid, but the frequency of words in Finnish seems to be … well … sufficient. If really does make stuff more frustrating.
Do you know if the number of linqs is affected? I would think not, but you never know. I mean, if a word has 5 coins, while another has 1 coin, does that affect number of linqs at all? Or is it only for gold-coin reward?
Beta languages do not have word frequency yet. Hopefully we will be able to implement it for all languages soon.
Can you inquire if there is a planned date for this? or a planned period? or any indication?
I’m also learning Finnish and it would be nice if LingQ parser could just run things through wiktionary or something. It’s somewhat tiresome to see the same word as new over and over again simply because the form is different. I understand that a truly perfect parser is difficult even for morphologically simple languages, yet alone Finnish but some effort in this direction would be greatly appreciated.
After we launch LingQ 5.0 we’ll see if we can add work frequency to beta languages too.
Thank you very much. I understand these things don’t come without a lot of effort. I do appreciate it!
You are actually pointing out a different issue. If you wish to pursue it, then create a new thread. Having said that, I tend to disagree.
In my experience, I benefit from the fact that lingq does not recognize all conjugations of a verb or declensions of a noun or adjective as the same. They are not the same and often have a different meaning.
Should lingq count them as the same, I would have a big problem in terms of meaning. If I want to understand sentences, I need to understand meaning of the different declensions and conjugations. As it is now, I can make notes for each declension of a noun or adjective. That is priceless.
If you understand a word, it is pretty easy to make the status 4 or k (known). Then it will get out of your way in all documents.
The reason I mentioned it here is that it contributes greatly to the coin inflation and would make it near-on impossible to realistically reward LingQs based on word frequency.
Finnish declension is fairly regular once one knows the rules, so it could be argued that there is little practical value for treating all forms differently. Moreover, while translations may be form-specific, it can also lead to serious misunderstandings since the declension of a word does not necessarily impact its semantic meaning, rather it may just be required by the syntax of the sentence or the preceding verb.
Finally, just practically speaking, Finnish morphology permits MANY more forms than other languages. Even a simple word such as ‘koira’ (dog) can appear in 50+ forms, albeit with varying frequencies. This can make it very cumbersome to use LingQ and essentially makes any estimate for the number of known words invalid.
I understand your reasoning. Whether I agree or not is irrelevant. The current title and thread is about the lack of frequency analysis in Finnish (and apparantly in all beta languages). In general, people select on title while browsing the forum. Please don’t hijack the thread. Create a thread with a proper title so it can be discussed properly.
How could you possibly implement frequency analysis with Finnish in its current state. Almost every word would have an inordinately low frequency. That’s the point.
I don’t know. We will see. Zoran said they are working on it.
Yep, Zoran has already cleared this up and explained it but I noticed the very same thing when I read Norwegian here and figured it was because it was still a Beta language. I have this autistic like quirk in having to finish and max out everything, so this meant I obsessively read copious amounts of Norwegian to be able to get all the background items for the avatar, where I didn´t need nearly as many “known words” in French to do that.
I get your frustration. Remember what you really want is to learn a language a good as possible and in the end the avatar is just a means to that end. But still, hopefully sooner rather than later.
Yep. I know this is more like a flaw in the way I think than anything else and it should not matter to me whether some abstract picture of some little creature has certain items in it or not. Now I somehow feel I need to finish at 60K words in Fench, even though I´ve gotten fairly sick of reading and would probably be better off trying to get people to talk to in French. It´s also hard for me to not finish books I start reading, even if I don´t like them.