On 24 March 2010 06:55, Peter Bienstman <[email protected]> wrote:
> This being said, the advice I can give you is not to overanalyse things. The
> algorithm is not clairvoyant, It's not exact science, but a heuristic. Just
> give it some time, and it will work more or less OK in the end.

Yes, it will still work fine - there's just a slight lag between
viewing a card for the first time and being able to say with some
confidence that you'll remember it. But after, say, 5-7 days, you'll
have gotten over that hurdle (without the alternative of seeing all
cards after one day for the first interval and perhaps spending longer
doing the same job).

I guess you could debate whether this time trade-off is productive or
destructive, whether it starts the journey to long-term memory faster
or slower, and whether it's appropriate for all card types.
The only way to find out really, would be to patch Mnemosyne and set
the first interval to ~1 day (or less, as people have requested
previously) and gather some statistics about learning rate over time
(could we have a scheme whereby when we patch Mnemosyne to change its
fundamental behaviour, we add/change a tag in the learning data it
exports so that that data can still be analysed but not merged with
the normal data? i.e. a "short-early-intervals" tag).

Remember that the point is not to learn a lot of facts that you
remember tomorrow, it's to learn a lot of facts that you remember over
a month from now.
Whether you drill more strictly in the first hours/days to avoid
forgetting new cards may not affect that rate of getting knowledge
into your long-term memory in an obvious way.

Oisín

-- 
You received this message because you are subscribed to the Google Groups 
"mnemosyne-proj-users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/mnemosyne-proj-users?hl=en.

Reply via email to