I was originally going to post this in the smart.fm closing down thread, but I thought it warranted its own discussion, as most of the conversation there consisted of what people will be switching over to. I've been using Smart.fm for a while, and rather than logging into the old site, which will only be available for a couple of months, I thought I'd start over, from the beginning, with the iKnow version.

I noticed some huge changes with Smart.fm's transition to iKnow!

Just to see what had changed, I decided to start over, right from the beginning, doing the core 2k. I was on lesson 5 in the old lessons - but I started over, right from lesson 1.

changes I've noticed...first, the ones that seem like positive changes to me.

the core vocabulary lists aren't divided into "Core 2000" and "Core 6000" anymore - they're "Core 1000" "Core 2000" etc - breaking each set up into a smaller unit, of a thousand words. This seems like a positive change to me, because it increases the feeling of progression - it feels a lot easier to get through a thousand words at a time than 4000.

Similarly, each lesson is smaller as well - 100 words instead of 200. Again, seems like a positive change for similar reasons.

The ability to do 20 items at once, not just 5 or 10. A good idea, although I don't see why a 15 shouldn't have been in there too. With each lesson consisting of only 100 words,
you can get through a lesson in only 5 sessions by doing 20 at a time.

A new series of "Checkpoints" along your journey through a lesson, which are skippable, but seem like a good idea. They have you practicing your listening, typing what's dictated, and you can even speak into your microphone repeating the sentences afterwards, to see how your voice measures up against the voices in the lesson. Again, seems like good idea, nice that it's skippable if it's not something you want to do, probably good for overall Japanese skill but not really sure it adds to the long-term retention that Smart.fm/iKnow was known for developing.

A lot of the items are associated with more than one sentence now! GREAT idea!

The order of the items seems changed somewhat - not too much, but there are some items at the beginning that weren't there before. It seems logical, the way they've done this. Why learn "Watashi" in the first lesson but wait until several lessons later to learn "Watashitachi"?

There's no Yes/Maybe/No anymore - when it asks if you know, the answer is either yes or no. I never knew when to click maybe - ie do I click it if I think I know the answer but I'm not 100% sure? Do I click it if I can't quite remember but think I can get it if I see the available choices? this makes sense to me - you either know it or you don't, and if in doubt, just click no.

Finally, the scheduling system seems to be better - you can tell it how much you'd like to study in a week, it'll keep track of everything for you and let you know whether you're hitting your goals, and it will also suggest the next lesson for you once you're done with all of your available items for the day. Good changes here, at least to me.

Now, for the changes that I either don't like or am not sure about:

You have to use a mix of hiragana, katakana and kanji. no romaji for those who would like that option, and no kana-only option. Gotta learn the kanji - and yet sometimes, when you're learning the kanji, the sentences that are paired up with the item don't even bother to use them, they use kana instead. On the bright side, there's more practice for each kanji, it seems - the way that iKnow is set up now it seems more geared towards helping one to remember the kanji, not just the word and its meaning. Still, it'd be nice to give people the option of learning with romaji or kana, as this might scare off absolute beginners - although things are made a little easier because the new curriculum has a pretty good introduction to hiragana and katakana available for study before the core vocabulary. I tried it out of curiosity although I already know the kana, and it seemed to be well done.

Another huge change: when you're going over the items, you don't type them all at the end like the old system - the word entry part seems broken up over the course of the entire lesson. I don't like this as much, don't know how it will be long term, but I like having the end of the lesson make you remember the word and type it, as it shows whether you've really absorbed it or not for that session at least. As these words are items I've already done, I remember them well so I'm not sure how this would affect retention of the items.

THE CHANGE I DISLIKE THE MOST:

When I did my first reviews, the second day with iKnow, I was shocked. It doesn't quiz you to see if you remember the words you're reviewing - it just presents them again as if they were new items! I don't get it! I thought the whole point was to see if you remember or not first before you're reminded of everything again. I wonder how this will affect long term retention. Maybe it won't at all, or maybe it's actually better, but I can't tell because these are all words I've already done and know well. It seems to me, though, that if you're reminded at the beginning of each lesson, it isn't as good for long term retention because when you're quizzed on it during the lesson, you can't tell if you're remembering because you actually remember it from before or because you were just reminded of it a minute ago. I'm confused by this change, and depending on how it works out over the next month or so it just might make the difference between me getting a paid subscription or not - without this change, I'd have no real hesitation in paying for the service, as I find the learning method very effective for me.

These are just a few of my thoughts on the new iKnow, entirely apart from the whole decision to go pay, etc etc that others have been talking about - please discuss

I haven't played with iKnow much, and I don't plan on it, I only used the site for my initial learning and then as items were mastered threw them into anki anyways.

That last change you described seems pretty ridiculous, I would not want to be re-taught words instead of re-quizzing them, this would REALLY annoy me as well....

It sucks that there is no kana or romaji only versions, I always used the full kanji, but not everyone wants to, they're simply limiting their userbase making this change.

Re-sorting into 6 groups seems like a positive change, hopefully they've fixed a few other sorting issues as well, to easy the student past the first stages of the language.

It is nice that you posted these concerns here, because this is as close as anyone at iknow will come to caring about what you think. A year ago, they created a feedback page simply so they would not have to read angry emails. None of the issues raised on that page amounted to changes that they had not decided upon previously anyways. They were even so bold to say, "Thank you, but we know what we are doing better than you do". I am quoting from memory, but not paraphrasing. Their responses were exactly that arrogant and dismissive.

bodhisamaya Wrote:It is nice that you posted these concerns here, because this is as close as anyone at iknow will come to caring about what you think. A year ago, they created a feedback page simply so they would not have to read angry emails. None of the issues raised on that page amounted to changes that they had not decided upon previously anyways. They were even so bold to say, "Thank you, but we know what we are doing better than you do". I am quoting from memory, but not paraphrasing. Their responses were exactly that arrogant and dismissive.

Haha, I remember them saying that.

The thing is, it's right. The average language learner does -not- know what's best for them yet. And that's their customer base. People generally want to make things easier, and that's not always the right direction when doing something as hard as learning a language.

Having said that, I don't think -anyone- knows what's best yet. Language learning, despite thousands of years of practice, still isn't a science. There are many, many different theories on how and why it works, and even more on how to learn. Probably none of them are 100% correct.

iKnow started out as an amazing product, if not yet mature. Then they started messing with the process and interface and generally made it worse at each iteration. Switching to Smart.fm continued that process. It doesn't surprise me that the process has continued further while they switch back to iKnow again.

entirely apart from whether the company is any good/accountable/listens to feedback/ripping people off/etc etc I'd really like to know what people think of the specific changes I've mentioned here. There's an entire other thread talking about the business side of things and the switch to pay - I'm more concerned with whether the changes I've listed help/hurt/don't affect long term recall of the items.

[ファブリス]

Thanks for the productive and informative post, LunaSlave.

I've barely used iKnow myself, just checked it out a while ago to see how the flashcards worked, so I can't comment on that.

Hi lunaslave, thanks for the detailed post. As noted in the other thread, I heavily use iKnow for learning vocab, and find it a great tool for initial learning and reviewing. I haven't looked at the new site, so your comments are interesting.

Multiple example sentences is good; do the new ones have audio? Sounds like they've put some money into the new core list, which is a positive.

Dictation throughout the lessons sounds good, though I hope the new ordering has sorted out the biggest problem with dictation mode; early sentences contain large amounts of vocab you don't know yet, making most beginner sentences totally incomprehensible.
On a side note, this issue also occurs with 2001KO, and I guess this is because both use example sentences from a 3rd party source, but for a learning resource it's inexcusable. Only Tae Kim grammar guide gets it right, as each example sentence builds on the previous ones.

Anyway I'm currently on lessons 8 of the original core 2000, and fully intent to finish way before the switchoff at the end of March. I'm not sure core 6000 is worth it (nukemarine doesn't think so, and i usually agree with him) , maybe better to move all core 2000 sentences to anki for retention and just start on native material.

I thought I would put my two cents in for the Chinese program.

Good)
I really like the way iKnow is presented. It's easy to navigate because of the tutorials and the calendar and schedule is top notch.

The audio and examples are nice quality, though a bit quieter than the sound effects.

Classes are seperated by 繁體字 (traditional) and 简体字 (simplified) hanzi. It's good for those of both character sets though it doesn't explain the differences between "普通话""Standard Chinese" and "國語" "Taiwanese Dialect of Chinese".

Chinese Media class is nice and most words have 2 examples! (Of course all with audio!)

Bad)
Writing "拼音" "Pinyin" is messed up with some sentences. I couldn't finish a lesson because of this. (I THINK they will fix thix soon though.)

They have too high expectations for dictation so early. After 20 hanzi (in their hanzi class) you have to do dictation -- which would be fine other than you have to do long dictation for words you don't know yet.

There isn't a Step Core like in Japanese.

If they continue to create content then I would reccomend this to a Chinese student.

lunaslave Wrote:...I'm more concerned with whether the changes I've listed help/hurt/don't affect long term recall of the items.

You asked about:
* dictation spread throughout lesson, and
* exposure to learned words before quizzing.

I see dictation as just a way to practice/test one way of knowing a word (sound to writing). I guess I don't see why you'd expect dictation to be the final test of the other ways of knowing a word.

As for exposure before testing, I doubt anyone would be able to answer that based on personal use. Since Cerego (iKnow!) made this change and advertises itself as the patented scientific method, presumably it's based on some research. So perhaps this is a question best directed to them?

There are a couple threads about the benefits of active recall over passive exposure (try searching for "testing"). IIRC, some articles/papers were referenced. Maybe you can try to leapfrog to a paper comparing the long term effects of exposure+testing vs testing alone.

Not sure what you mean by "long term", but one month might be on the short side. ;-)

LunaSlave Wrote:The order of the items seems changed somewhat - not too much, but there are some items at the beginning that weren't there before.

The order is actually changed a great deal in some places, especially once you hit the new core 1k steps 5 and 6 where less than half of the words are the same as the old step 3. They've also seem to have reworded a number of the definitions for better clarity/accuracy.

Quote:When I did my first reviews, the second day with iKnow, I was shocked. It doesn't quiz you to see if you remember the words you're reviewing - it just presents them again as if they were new items! I don't get it!

It only does this the second day you see the new word or possibly only if your knowledge of the word is below 25%. I'm not sure which yet. After that it just does normal reviews. It does seem a bit of an odd decision, though in general I'm finding the new lessons to be a pretty big improvement for retention.

Quote:You have to use a mix of hiragana, katakana and kanji. no romaji for those who would like that option, and no kana-only option.

They have a separate romaji series for beginners now and they've said that a kana-only version of the core series is coming soon for those that want it.

The biggest improvement for me has been the greater emphasis on learning kanji, but I can see why it would be frustrating for some users. I would have found it difficult in the beginning too.

I also like that they give you lots of cloze sentences now where you hear the audio and then type in the clozed word. Also clozed sentences with an English translation but no audio too to test reading comprehension, often requiring you to conjugate verbs. It does make me wish for a greater number of sentences though.

LunaSlave Wrote:When I did my first reviews, the second day with iKnow, I was shocked. It doesn't quiz you to see if you remember the words you're reviewing - it just presents them again as if they were new items!

They appear to have changed/fixed this now, it's back to the old way.