slucido Bilingual Diglot Senior Member Spain https://goo.gl/126Yv Joined 6701 days ago 1296 posts - 1781 votes 4 sounds Speaks: Spanish*, Catalan* Studies: English
| Message 17 of 60 27 October 2007 at 10:54am | IP Logged |
gidler wrote:
I strongly recommend Anki. Having used SuperMemo, FullRecall, Mnemosyne and others, I think that Anki is superior to its competitors in all aspects. (Except if you care about memory usage, in which case I would recommend buying more memory. It's practically free these days.) And it's free, cross-platform, GPL, being actively maintained and developed, etc. :) I'm using it for Japanese, Swedish and occasional English words.
|
|
|
I have tried Mnemosyne, Jmemorize and Anki for a few days and Anki seems better. I have tab-separated text files with Assimil sentences, words and its translations to Spanish. It was impossible to import to mnemosyne and jmemorize, but I had no problems with Anki.
In regard the sound issue, if I use texts from Assimil or other audio-text programs I think it will be less problematic. On the other hand, I have a text-to-speech software and it useful to use on the go with the flashcards, without adding any sound to them.
I suppose all this sotware have the same spaced repetitions feature (Leitner...).I will try this... and the wordsgallore flashcards that works great.
1 person has voted this message useful
|
furrykef Senior Member United States furrykef.com/ Joined 6498 days ago 681 posts - 862 votes Speaks: English* Studies: Spanish, Japanese, Latin, Italian
| Message 18 of 60 27 October 2007 at 2:02pm | IP Logged |
Mnemosyne (and SuperMemo) don't use the Leitner system. They're much more advanced than that.
1 person has voted this message useful
|
gidler Senior Member Finland Joined 6649 days ago 109 posts - 118 votes 1 sounds Speaks: Finnish*
| Message 19 of 60 27 October 2007 at 4:39pm | IP Logged |
Speaking of the Leitner system and scheduling algorithms, I want to point out that complexity does not automatically imply better results. SuperMemo (or more specifically, its Windows version) uses a very complex model which tries to approximate the user's learning curve. However, after using SuperMemo for about five months, the curve the algorithm had created was quite malformed and the intervals it generated seemed to be getting worse. Either my memory doesn't work like it is supposed to, or the algorithm tripped on its own complexity (possibly implying that approximated forgetting curves is a flawed approach).
Mnemosyne uses an older, much less complex SuperMemo algorithm. While it works quite well, it can be argued that since it does not separate the initial memorization stage and long-term memory, the difficulty values of cards can get too conservative leading to too short intervals (although this fixes itself over time).
Anki uses a custom algorithm which does not compute card-specific difficulty values. A feature unique to it (as far as I know) is that new cards are kept in an initial memorization stage until the user indicates that they are ready for longer intervals. Anki also gives the user more power and responsibility in deciding how long the scheduled intervals are.
In the end, scheduling algorithms don't really matter that much. All of the above-mentioned apps (and the basic Leitner system) work relatively well and it is hard to show whether one is more optimal than another. I would recommend trying them all and using whatever one is most comfortable with. (Mnemosyne users: psst, Anki has more features including nice graphs, and it can import Mnemosyne decks. :)
Edited by gidler on 27 October 2007 at 4:42pm
1 person has voted this message useful
|
slucido Bilingual Diglot Senior Member Spain https://goo.gl/126Yv Joined 6701 days ago 1296 posts - 1781 votes 4 sounds Speaks: Spanish*, Catalan* Studies: English
| Message 20 of 60 27 October 2007 at 4:48pm | IP Logged |
furrykef wrote:
Mnemosyne (and SuperMemo) don't use the Leitner system. They're much more advanced than that.
|
|
|
I am new with this software and it seems I don't understand anything.
What's the difference between Leitner and SM?
I am using mnemosyne and Anki and it seems very similar, all about spaced repetition, but I am using this the last three days...
I was not aware than Anki can be used on line as well. If it works properly, it is great, because it's possible to work with different computers, for example on vacations.
1 person has voted this message useful
|
gidler Senior Member Finland Joined 6649 days ago 109 posts - 118 votes 1 sounds Speaks: Finnish*
| Message 21 of 60 27 October 2007 at 7:45pm | IP Logged |
slucido wrote:
I am new with this software and it seems I don't understand anything.
What's the difference between Leitner and SM? |
|
|
To put it bluntly, the Leitner system is a very simple variation of spaced repetition. SuperMemo, Mnemosyne, Anki etc. use more sophisticated methods for determining the repetition intervals.
slucido wrote:
I was not aware than Anki can be used on line as well. If it works properly, it is great, because it's possible to work with different computers, for example on vacations. |
|
|
Yep. You can use the "sync" feature of the standalone application to keep it synchronized with your online deck. The feature has been around for a while so I'd say it can be considered stable.
Edited by gidler on 27 October 2007 at 7:56pm
1 person has voted this message useful
|
furrykef Senior Member United States furrykef.com/ Joined 6498 days ago 681 posts - 862 votes Speaks: English* Studies: Spanish, Japanese, Latin, Italian
| Message 22 of 60 27 October 2007 at 9:09pm | IP Logged |
You're right tht complexity doesn't necessarily mean a better algorithm. But how well does the Leitner system handle extremely large numbers of cards, like 10,000 or more? The impression I got is that it doesn't scale as well as the Mnemosyne and SuperMemo algorithms do, especially for items that are inherently difficult to remember, like kanji.
- Kef
1 person has voted this message useful
|
gidler Senior Member Finland Joined 6649 days ago 109 posts - 118 votes 1 sounds Speaks: Finnish*
| Message 23 of 60 28 October 2007 at 7:17am | IP Logged |
furrykef wrote:
You're right tht complexity doesn't necessarily mean a better algorithm. But how well does the Leitner system handle extremely large numbers of cards, like 10,000 or more? The impression I got is that it doesn't scale as well as the Mnemosyne and SuperMemo algorithms do, especially for items that are inherently difficult to remember, like kanji. |
|
|
The difficulty of your cards only matters if the scheduled intervals get constantly too long and you end up forgetting them. This should not happen with SuperMemo and Anki because they let you adjust the scheduling. Mnemosyne has its card-specific difficulty values so it should work fine too, except that initial intervals might be too long for very difficult cards.
The classic Leitner system does not scale well to large numbers of cards, because the reviews pile up to big clumps instead of being distributed on a wider span of time. I don't know if some of the Leitner-based applications keep card-specific intervals to evade this, though.
1 person has voted this message useful
|
leosmith Senior Member United States Joined 6576 days ago 2365 posts - 3804 votes Speaks: English* Studies: Tagalog
| Message 24 of 60 28 October 2007 at 7:50am | IP Logged |
Do any of these allow one to skip a day or more? That's the thing I hate the most about supermemo.
1 person has voted this message useful
|