Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo

Close

OpenCards and others similar programs

ferossan
2008-04-24
2013-04-24
  • ferossan
    ferossan
    2008-04-24

    I read the wonderful article in Wired (see OpenCards Blog) and it was very interesting (quite long anyway). In the comments appears a mention to another similar project, Mnemosyne (http://www.mnemosyne-proj.org/index.php) which I tried and looked quite fine despite it offer just "long-term learning". Mnemosyne use an algorithm close to SM2 used in the past by SuperMemo. The algorithm deal (I guess) with how often the program must ask again for a card.
    My question is regarding to the algorithm. OpenCards use a SuperMemo based algorithm? Considering the ancient age of SuperMemo how better are the new algorithms? How better could be OpenCards against Mnemosyne or SuperMemo?
    Thanks

     
    • Holger Brandl
      Holger Brandl
      2008-04-25

      Hi,

      OpenCards implements a slightly modified version of SM2. Additionally it implements a ultra-insession-reschedulder which reschedules items within a learning session.

      > The algorithm deal (I guess) with how often the program must ask again for a card.
      The big question is rather "when" (although this includes "how often").

      > How better could be OpenCards against Mnemosyne or SuperMemo?
      AFAIK nobody knows. I'm not aware of any statistically significant studies in this direction.

      The main difference between all these apps is usability. Here (you've already guessed it ;-) ) I think OC to be superior in most cases. Only if you're using multi-sided cards (e.g. 3-sided for learning asian languages) I would maybe suggest to use anki ( http://ichi2.net/anki/ )

      Although Supermemo now uses SM11, I'm a bit skeptical that the huge complexity of the newer SM algorithms provides for a statistically relevant benefit. But because I'm using OC everyday, I wouldn't hesitate one second to implement another scheduling-approach if new scientific evidence should favor another learning algorithm.

      If you're interested in the implementation itself, just have a look at http://opencards.svn.sourceforge.net/viewvc/opencards/trunk/src/info/opencards/learnstrats/ltm/LTMItem.java?view=markup . This file contains almost all the SM2-logic. A quite good description of SM2 is available at http://www.supermemo.com/english/ol/sm2.htm .

      Best regards,
      Holger