A (suspended) investigation of an obsolete form of glossika

[scivola]

I, too, have just recently started using Glossika for French. And I'm doing the GSR method. I've actually been doing 2 lessons per day, and I completed day 50 yesterday. There are 303 total lessons, so I've still got a long way to go. But I'm far enough in to be able to say that I like the course, and I think I have seen some improvements as a result of it.

Some of it is just on the physical level of being able to enunciate the sentences - some things that were real tongue twisters a few days ago are now pretty easy to say. But I'm also starting to get a sense of the sentences sort of building on each other. So if some earlier sentences focused on grammar point A and others on grammar point B, I can pretty easily see how they combine in a new sentence I hear today that incorporates both of them. There have definitely been times when I would hear the English prompt for a new sentence for the first time and knew how to say it in French when I definitely wouldn't have known a couple of weeks ago.

So, so far I'm off to an encouraging start.

:edit:

Oh, and to give a little more information about the structure of the course, it is divided into three parts - Fluency 1, 2, and 3. When you buy the course, you get all three of those, you don't buy them separately. There are three separate ebooks, but other than that I don't see any reason why it is divided up like that. Each of those modules contains 1000 sentences, so the course consists of 3000 sentences in total.

The ebook for the Fluency 1 module says this about the GSR method I'm following: "Our sound files include an algorithm that introduces 10 sentences every day, with review of 40 sentences, for a total of 1000 sentences in 104 days." A few pages later, it says "You will hear each sentence more than a dozen times over a 5 day period." So it seems that each module consists of 12,000+ repetitions, or something over 36,000 for the whole course.

That's kind of the low end of what's possible with the course. There is another method you can follow - GMS or Glossika Mass Sentences - that use a completely different set of audio files. The most hard-core version of that method involves doing around 97,000 reps. There is a study schedule document you receive that presents different options based on how much time per day you have available, how long each option takes to complete, etc to help you decide which method to follow.

[Ani]

I just picked up the FR-RU Glossika. It seems a lot of the sentences are repeated. Did you notice that when you were making your anki cards?

I'll be interested to follow your progress! I'm still trying to decide which study schedule to use.

[sfuqua]

There is a lot of repetition in the gsr files, but the book I made the cards from doesn't seem to have exact repetition, at least that I've seen yet. Many of the sentences are closely related, which produces something similar to audiolingual substitution and transformation drills.
Right now, I'm just counting new cards in Anki for reps. I read somewhere that the GSR cards have 180 reps on each one., except for the first few and the last few. This means that the way I am counting g things now, I am doing 200 reps per language per day. This would give me 30000 reps after 150 days. I will have finished only half of the gsr files, but I will have been exposed to all of the sentences in Anki. At that point, I'll try to test my production in each language. I think I'm going to record myself reading a page of each L2 aloud and then score it for hesitations and also speed. Perhaps I should just do a brief monologue for each language and count take stumbles and hesitations. I haven't done a "before" test yet, since I'm traveling and I didn't bring a Laptop, but I can probably do this before day 7.
If I like the results after 150 days, I may do the last half of the course.

I'm doing two languages each day, and - surprise, surprise- I seem to do much worse on whichever language I do second each day. Yesterday I did French first, and I felt like I had a bunch of interference from French when I did my Spanish. Today I did Spanish first and I wasn't really aware of interference from Spanish on my French; I just made a bunch of mistakes that made me mad.

I called this an investigation instead of an experiment, since I intend to do other things to study as well as glossika. In fact I'm going finish this post and read a chapter of_Leviathan Wakes_ first in Spanish and then in French.

[Ani]

There is a lot of repetition in the gsr files, but the book I made the cards from doesn't seem to have exact repetition, at least that I've seen yet. Many of the sentences are closely related, which produces something similar to audiolingual substitution and transformation drills.

Good to know, thanks! Maybe it is just this course. For example, sentences 23 and 49 out of the GMS booklet are both Quel âge as-tu?, and many sentences that are answers to the previous question are varried in the French, but just translated to Да or Нет in the Russian. Ah well.

[Brun Ugle]

Some of the sentences are repeated in the GMS files, but on the other hand, some sentences are actually two sentences since they give both question and answer. The quality varies from language to language. There are always a few mistakes, but the worst I've seen is the Mexican Spanish, which is bad enough that I think they should redo it. However, I do like Glossika in general and find it useful when I actually use it. I tend not to use it enough though.

[blaurebell]

Good to know, thanks! Maybe it is just this course. For example, sentences 23 and 49 out of the GMS booklet are both Quel âge as-tu?, and many sentences that are answers to the previous question are varried in the French, but just translated to Да or Нет in the Russian. Ah well.

This is rather realistic though, I remember being told over and over by my English teacher that in English you can't just answer "No" and be done with it. Monosyllabic replies aren't necessarily seen as awkward in other languages though, like German or Russian.

[Ani]

Good to know, thanks! Maybe it is just this course. For example, sentences 23 and 49 out of the GMS booklet are both Quel âge as-tu?, and many sentences that are answers to the previous question are varried in the French, but just translated to Да or Нет in the Russian. Ah well.

This is rather realistic though, I remember being told over and over by my English teacher that in English you can't just answer "No" and be done with it. Monosyllabic replies aren't necessarily seen as awkward in other languages though, like German or Russian.

I'll reply on my own log so we don't take over here

[Axon]

I hope I don't come off too strongly in this forum as a glossika evangelist, but I do believe it's one of the most efficient ways to improve speaking. Also I'm a sfuqua fanboy so I couldn't help but post

Something fundamental that I think gets brushed over in most discussions of the system is that it's not about the words. Why do they use tennis as an example a dozen times when soccer is a much more international sport? What's the deal with all the names? How come it seems like half the questions are about watching TV? Why don't we learn how to say "Do you speak English?" until like five hundred sentences in?

It's not about the words - it's about the patterns. As the hundreds of repetitions tick by, you're building up a certain automaticity that will burn in key patterns like "she usually does X" and "he is [occupation]" and where in the sentence to put "yesterday." Over the course of three thousand sentences, you really will start to feel a native-like intuition for sentence structure, which prepositions go with which verbs, how to give a quick negative answer to a question without thinking about the verb, and more.

For that reason, it's not crucial for you to memorize each of these sentences through Anki. Some of them of course are high-frequency enough that it might be worth it, but as you get into the higher levels the sentences get more and more specific. Every sentence that I've written in this post is likely to be unique on the internet. But each of them follows a pattern, or is made up of a few patterns strung together. With the mass exposure to sentences, you internalize these patterns so that you can use them in situations relevant to you. That Anki time can be better spent on something more connected to your learning, like cloze drills or sentences mined from things you like.

The program and company aren't without their faults, for sure. After I finished the Chinese Business Intro I sent them a long list of problems with the course, and I've found a few even more serious problems with the courses I'm using now. But they are constantly improving, and apparently there's going to be some massive updates at the end of the summer.

[sfuqua]

GLOSSIKA
ES 30000: 1200 / 30000 reps
FR 30000: 1200 / 30000 reps
Since this is a process, it really makes more sense to put the bars in message instead of in the signature; if someone reads this a year from now they will know where I was in the course. when I made the comments.

First of all, everything Axon said.
It is trivial to come up with "better" sentences for a given language; no course is ever going to come up with optimal topics and vocabulary for a given learner.
I admire the brilliant "crudity/simplicity" of the course. They have created hundreds of hours of courses for multiple languages. If they had tried for an optimal course for each language, they would never have completed anything. I hope that they don't "improve" the courses too much. Fix problems, yes. Change the basic system, please no!

You see, after a week, I am pretty certain that the GSR files work pretty much as advertised at least for me.

The benefits are similar to the benefits one gets from FSI or shadowing Assimil, but there are some intriguing advantages to the GSR method. With shadowing Assimil, it is pretty easy to get to the point where you are "singing along" with Assimil without paying attention to meaning. I think this might have some use for pronunciation, but it probably isn't a very useful practice. At one point, when I was trying to wring every last drop out of Assimil Spanish with Ease, I found myself reading a book in English while I was blind shadowing a familiar lesson from ASWE.

FSI has it's own problems. I think one of the things that is wrong with the whole audio-lingual approach, at least the courses we have that were designed for reviewing material from the classroom, is that it is great at practicing things, but that it isn't brilliant at introducing things. It challenges and it tests, but that it doesn't teach very well. With shadowing, even if you foul up a sentence, there is no time to suffer, you are trying to keep up with the next sentence. I bet back in the old days at FSI, they did nice introductory exercises in class.

With GSR, you hear the L1, followed briskly by the L2, which I've been shadowing along with. After a few exposures, the L2 sentence pops into your head. Glossika suggests that you mumble along if you can't spit out the L2 sentence at the speed you should. I like this, because it gives you a way to succeed even if you are not up to speed. The L1 sentence keeps the meaning in your mind and the L2 keeps the form in your head also. I think this would not be useful for a highly advanced learner, but for someone trying to get over being tongue tied in a language, this is useful.

I agree about anki and glossika. I find that it is useful, but mostly because it helps me to figure out what the written form of some of the French sentences that are all slurred and warped together, the way that spoken French loves to do. I've actually changed the way I'm doing anki; I've put a bunch of sentences from García-Márquez and Proust in the deck. I originally was going to use them after I had done all of the GSR sentences, but now I've got them mixed in with the GSR sentences. I bet this will help my reading some.

I'm very encouraged so far that this is a useful intermediate level course, but I'm still very early in the process.

[sfuqua]

When I was in grad school in what they call the Second Language Studies course at University of Hawaii (it changed names a couple of times), one of the courses was great fun. Working as small teams of investigators, we were supposed to use learn as much as we could about a given language from a native speaker of an L2. The rules were simple, our language informant would do two things:
1. translate any sentence we gave in English into the L2, doing a very literal translation.
2. tell us whether an L2 sentence we tried to produce was actually correct or not.

The informant could say nothing in English, and looking things up about the L2 was considered cheating.

We studied Mandarin.

By picking our sentences to be translated and to try to produce carefully, we learned quite a bit about Chinese, and we also got to learn some of the strategies for learning a language without any resources, just based on what we know about languages.

A written form of this might make a fun thread here on the site sometime.

The course had a special meaning for me in that the language informant for my team was my girlfriend at the time. She was from Taiwan and was a fellow student in the grad program. After class, when we would go off to do couple things, she would crack up about how fun it was to watch us suffer trying to figure out her native language. She was an instructor in the UofH Chinese course, so she found it very frustrating sometimes not to be able to help us. She took her job as a language informant seriously and never gave me a clue about where to take our investigations.
I think she also got a certain sadistic satisfaction about watching her boyfriend and fellow graduate students suffer

Anyway, you can learn a lot about a language just based on having a native speaker translate.

Making a course based on this is a good idea. Of course this doesn't tell you everything you need to learn about a language, but it sure beats the heck out of nothing. As long as glossika keeps working on quality control, this may be a game changing set of courses for high beginner/low intermediate students.