What and how much do you ignore?
Hi, regarding the count of different forms of the verb the dicision to learn them or to ignore them lies in the various aims of the learner. Every learner has different priorities.
I for myself don't care about the number of known words. Actually I am on a B2+ level in French and Italian and a C1 level in English, but my "known words" say 31, 5300 and 5062 for these languages, despite the fact that I reach my daily (mini) goal for all three of them for almost a year now. That's because with every language I learn, I use LingQ differently.
When you start on LingQ with a language you already know quite well, you could for example "ignore" all the words you already know and just move the new words to known after you learned them. Then the "known word" list just shows the number of words you learned on LingQ. (That method does not work well when you compete in a 90-day hard core challenge. But that does not deter me from participating... )
I usually ignore words that I don't want to learn: Names that are the same in my native language, some words that are part of a dialect (for example in a few older britisch novels), unnecessary street names or names of (for me) unimportant places,...
As a learner I would never "ignore" compound words in German. Let's say we have the well known German word Autobahn. Even if you learned the two components Auto [=car] and Bahn [= train/ tram] before, how do you get the meaning of "highway" out of that? As someone who want's to repoduce the language you need to learn the compound words because even if the parts are a clue to meaning, sometimes the first part is in singular, sometimes in plural, sometimes there is a letter inbetween...
In other languages it is the same. You could just learn every word in itself, but often two words together have a special meaning. Just a little example in English: I see the new word "flowerpot" and I say >>oh I know the words "flower" and "pot"<< and I even have a mental image of both words in my head. So I ignore that compound word... Then a few weeks later I want to reproduce that word in a conversation and I say "flowers sauce pan"...
Also for conjugations, personally, I think that knowing every single form is necessary, when you want to be able to write properly. If you'd move them to "ignore" you would not have any way to track them. But as I said, if you just want to understand a text on the surface, or you just want to mark words that are important to you, you can ignore them for sure.
Also, I see the plural forms of nouns and all their forms of declension on LingQ as "full words", because I want to be able to understand and reproduce them.
As for the different methods you can use LingQ: While I read or listen only in two of my target languages, I use LingQ very untypicly with my A2 language, which I learn more intensely at the moment. I first read a sentence here. I look up all words I don't know yet, but I don't mark them. Instead, I write the translation of the sentence into my (real life) notebook. Then I try to translate the sentence back into the target language the next day. I repeat that over the next week until I get the sentence right three times. THEN I mark all new words as a lingq. And everytime I come across a lingqed word in another sentence or text and I am able to produce the translation without looking the word up (and the meaning fits with the sentence), the status of the words moves up. This is in fact a very slow process, but I was able to produce a well written text in my language course quite soon after using that intensive method. I guess, I will change to another method when I change my main focus to another skill, soon.
The word count in LingQ is a small ego boost but otherwise a meaningless number as nobody else in the world counts words this way. In Korean I think it's more out of control than other languages with all the conjugations. So ignoring words doesn't really matter much like you said. BUT - just to do things correctly, I ignore all names and anything that you can't look up in the dictionary or Google translate. You sometimes get things you just can't translate for whatever reason. Ignored.
I ignore almost anything I don't care about or is so obvious I could never forget it. Everyone else has listed these kinds of things in detail.
One I didn't see is compound words that are common in Germanic languages. I'm using LingQ for Norwegian where compounds can be 20 or more letters. But when you break them down, they are often all words I already know or have LingQ'ed. So I just ignore them. Sooner or later, all of the words in the compound will appear singly and I'll deal with them then. If they don't, obviously I was safe to just ignore them.
Ultimately, it doesn't matter. I read everything word for word. I don't just scan for words I don't know. Often, a word will be marked known, but I forgot it. So I remark it with a 1. Down the road it may be marked Known again. I don't use anything between 1 and Known. I don't care about degrees. I either know it or I don't.
I'm pretty strict.
All place names
Any word which is identical in English
All people names
All n', l', and d' words in French
Most plural words. E.g., if I see "homme" and "hommes" on the same page, I would ignore the plural
I always try to get as close to the root word as possible and ignore the rest.
That's quite convenient. I noticed with the verbs that after some time you'll probably have 20 different conuigations as "known words" but in fact you only know one word. Also with most nouns you'll have all 4 versions in Italian after a short period of time. But when it comes to verbs, I really lost track...
Yeah, if I was starting again, I would ignore some of the verb conjugations, but not sure which ones. I also ignore all slang.
Honestly, I wouldn't worry too much about it. Even though there are 20 or 100 verb conjugations in some languages, it doesn't necessarily mean you have marked them all as known. For some of the rare verbs, you may have only encountered one or two of the conjugations. Because you can only mark words as known when you encounter them in reading, your word count actually underrepresents what you know.
Your word count does not represent either word families nor head words by default. I wouldn't even try to make it do so for several reasons: (1) you forget which words you have already marked as known so would end up double marking words, (2) it requires a lot of time and effort, (3) when do you decide that you 'know' a word family? Eg. You may incorrectly get the noun wrong all the time, but you get third person singular present tense of the verb correct 100% of the time but half the time misspell the subjunctive present third person plural. It's all pretty arbitrary.
If you really want to estimate how many word families or head words you can recognise/'know', find one of the various online tests to estimate it. If you actually want to do it with more accuracy, find a frequency list of word families or head words and go through it marking all words you know. Count them up after you go through the most frequent 5k or 15k most frequent words. This is easiest done in a spreadsheet.
When I reached 6,700 Known Words on LingQ, I went through the most frequent 4,998 head words in Italian. I successfully recognised 3,227 of them. I know I also knew, and had marked as Known on LingQ, many mid- and low-frequency words, which were not in this list. So my actual number of 'known' of head words was higher than this. The difference between my number of head words and my 'LingQ Known Words' was not too significant.
It's probably best to think about your Known Words on LingQ as simply 'my LingQ Known Word count'. It works best when you compare it to where you were last month or last year. It's great seeing your Known Words grow. The numbers that I personally prefer to focus on, because I can control them, are my number of words read and my number of listening hours.
jmuehlhans and Nobodysfuel, while this isn't the approach I've taken, your general approach is one that I have before wished I'd taken.
Moreover, I wish the LingQ programmers would have done this approach. I believe with access to dictionary files, they could have coded most of these concepts to handle much more smartly.
I have almost 30,000 known words in French and part of why is the word count inflation associated with not doing all of this manually.
Having done this for now a year and a half or so though, I don't think it really matters. I hardly care what the word count on LingQ says--I simply use this as a tool.
Now that I'm getting more and more into upper-level vocabulary of two academically closely related languages, much of the vocabulary that LingQ calls "new" is what everyone here has identified--person names, place names, foreign words, misspellings, etc.
All that said, if I were to do it all again, as of right now, what would I do? Probably the same thing, with the primary benefit of saving key strokes and the primary consequence of an overinflated word count.
The LingQ word count is not "overinflated." It does exactly what it was designed to do. That is, record every unique word form. You can only think it is "overinflated," if you think it represents word families or head words. But it was never designed to that and it simply does not do that.
Imagine how you would implement what you're wishing for. It would be an absolute headache, costing a lot of time and money, for LingQ to implement Known Words representing word families across all languages. I also disagree from a theoretical perspective that foreign language acquisition should be measured by word families. Honestly, I prefer the way LingQ currently does it.
If you really want to know how many word families you know, do an online estimate or manually count words in a frequency list. You can do an online estimate in the matter of several minutes or half an hour for a more detailed one. Or you can manually count all the word families you know in several hours.
nfera & gmeyer. I agree with nfera, in that the coding could become tricky for the LingQ coders. I suspect there are other features they spend their time with. nfera's estimate of about half LingQ known words as being "real" known words from a recognized list as a good estimate.
The LingQ "bells and whistles" are for motivation purposed (i.e., gamification) and not necessarily linguistically scientific. Still, their useful in a ballpark kind of way.
Honestly, that all makes complete sense.
For ease of programming and gamification, labeling as "words" makes much more sense than "unique whitespace and punctuation-delimited strings."
I do it similar as you. I started marking everything as known, but fortunately became aware soon enough.
As for places, I decide based on their popularity. For example, I added Firenze to my vocabulary which is "Florenz" in my native language German and thus, different. Same goes for Danube (Donau) + Reno (Rhein). But I am not adding names of places which are almost identical.
Also, I ignore words like "hmm", "uh...", etc.
It's arguable if it's worth the time, but I prefer to have the word count as accurate as possible. :)
Thanks for all your answers, you do have a point with not holding me back too much counting words. However going through some older lessions is also a good repetition :-) Even though I like things clean, I also see that this numbers and counters don't really matter or tell me much about my progress.
Still good to see that many of you also ignore roughly the same things.
My general rule of thumb is any word I would find in the dictionary stays. Including newly added foreign words (mostly from English) in what I imagine would be in the equivalent of Urban Dictionary. This even includes the shocking but rare words, where even the English plural is also used (i.e. ending with an 's')!
Things I ignore:
- People's names
- Street names
- Jibberish (like the random letters of a website link)
- Foreign words not actually used by an Italian
- Names of most cities (the exceptions being some large important cities, which are different than the English word, eg. Londra, Parigi, Francoforte, Pechino)
In the end, it doesn't mean much, especially at the higher levels.
Don't waste your time by going back to corect them. Maybe if you encounter the word in future lessons, you could change it's status, but it's not worth spending hours trying to correct it. Unless, as you mentioned, it's fun. Then go for it. ;) As long as you are rereading and relistening to the lessons at the same time.
I would keep country names and language names, as they sometimes aren't cognates (i.e. in Russian, Georgia is pronounced "Gruzia" and Hungary is "Vengria;" Germany is "Germania," but German is "Nemetskyi"). Just like "telefono" is a cognate that you would count, country names are the same in my opinion.
Agree with town names and peoples' names - ignore them. Also ignore when someone says something in English.
That said, honestly, don't worry about going back through your list to cull out the ones you added that you wouldn't now. Just keep pressing forward. The known word count is an estimate anyway. If it's a few hundred more than it should be when you're at 15k words, it's not a big deal.
Yes, I ignore all proper nouns and also dialect. I come across a lot if Napolitano in what I’ve been reading lately and don’t save any of it. I’ve got over 50,000 known words now and I’m sure there’s lots of words there that I didn’t ignore at the very beginning, but I don’t worry about them at all. If it’s fun for you, however, you’re probably learning so no need to stop.
I ignore all English words (mostly from book imports) that are clearly English as it might be involved in an English sentence. Loan words or common words to languages I'll tend not to ignore, particularly if there isn't a likely word in my target language. I don't think too hard on this and am probably pretty inconsitent on this...to dgbeecher's point, I don't want to reflect too much on it and once I've made that decision I don't worry about it.
I ignore names of people and most names of places. Exceptions to this might be if I want to save a bit of information about that person or place (I more often do this with places). For example I probably have "Alexanderplatz" saved off (a public square in Berlin). Mostly because I figured I might come across it again at the time and might want to just quickly see what it was again without needing to look it up. Same with SOME names.
For cities, or country names, I've tended not to ignore many of these. In German they often have much different spellings or "words" for them than "English". If it's one I know very well I might ignore it. i.e. "Köln vs Cologne".
That's about it. I don't think too much about it and my rules I probably break all the time. To dgbeecher's point, to some degree I can't be bothered.
I certainly wouldn't go back through my list trying to find words to ignore. Many you probably will never come across again, and if you do, you can always ignore it then.
The fewer decisions I have to make when reading, the more attention I can pay to the content. Hence, I can't be bothered to pay any attention to the Ignore button. I never use it. I just mark all such words as Known.
Going through my Known words and moving them to Ignored does not seem like it would be a good use of the limited time I can dedicated to language learning. ...That said, neither does posting on these forums, but I like our community here so I do it anyway :)
Another thing I pay minimal attention to is the Learning level. Hence, I pretty much use level 1 for everything. Occasionally if I have to move something from Known back to Learning, I'll choose a different level. But there's no real logic to it ... or, more importantly, there is not much thought put into it. These things are not worth the thought effort (in my view).
I have not experienced (or at least not noticed) any negative consequences from having an inflated Known count, because it's the progress that matters, not the absolute number. I also don't care about having a less-than-accurate Learning level for my LingQs, possibly because I don't use the flashcard review feature.
I'd say, do what's fun and keeps you engaged with the content.
Yes. I really only move words to known; not 2,3,or 4 or ignore. It means I can read more.
WANT TO LEARN A NEW LANGUAGE?
Learn from content you love!Sign Up Free