I am keen to "protect my learning investment" performed using SM7.
Q/A export from Supermemo7 unfortunately contains no learning
history.
A key-term in Supermemo is "E-factor" (EF), ranging from 1.3 to 3.0,
1.3 indicating "very hard to learn" and 3 indicating "yo know this
item really well".
Proposed manual conversion procedure of a ten year old SM7-DB with 10
000 items: Since SM7 allows me to export ranges of E-factors
1) I create 3 (probably 10 to achieve more refinement) exported Q/A
textfiles of items
A. 2000 Easy items (E-factor 2.3 - 3.0)
B. 7000 Medium (E-factor 1.6 - 2.2)
C. 1000 Hard (E-factor 1.2 - 1.5)
2) I adjust the date of the computer 10 years back, import to
Mnemosyne A. "Easy items", then "learn" by pressing spacebar while
drinking coffe or watching a movie
3) Set date of the computer to correct date, repeat the "pressing-
spacebar-while-drinking-coffe-process" (Will the Mnemosyne algorithm
then schedule these for repetition "way into the future, as they were
in SM?)
4) Do the same for B. Medium starting 5 years back instead of 10
5) Combine Mnemosyne A. and B. and continue my learning process
6) When time and eagerness permits: import portions of the "Hard
ones" (C.), simultanously "cleaning them" (they often have a low E-
factor because they are not well formulated.
After lots of coffe and spacebar pressing I should be happily flying?
/ Henrik