I argued for the virtues of the existing algorithm, above, because it does work very well, and most users need not give it a second thought. The original question was whether or not it was working right. Yes, it is working as designed. BUT... (I hope no one minds my mentioning my algorithm ideas again) ...
On Feb 19, 12:19 pm, Francisco José Fiuza Lima Júnior <[email protected]> wrote: > I disagree! O.K. I'll show you I understand you perfectly. For a long time now I have been running modified algorithms (only for myself because I'm not a real programmer). Here is what I've been using for some months (simplified version- this is not real code): grade 2: new_interval = old_interval (<< There it is Francisco!) grade 3: new_interval = old_interval * 1.6 grade 4: new_interval = old_interval * 2.5 (this = grading it 3, twice (1.6 ** 2)) grade 5: new_interval = old_interval * 6 (this = grading it 4, twice (2.5 ** 2)) This is strong-arm control... I picture myself holding the hard ones "close to the vest", while flinging the easy ones into the future. I score 90% pretty consistently. That seems about right. Here are my "grades" statistics: Grade 2 : 73 (4.4 %) Grade 3 : 89 (5.4 %) Grade 4 : 1032 (62.2 %) Grade 5 : 463 (27.9 %) (this would be lower, but I recently added a lot of easy cards) I don't use the "easiness factor", because too often an item's difficulty will change instantly, when it "clicks" in the mind. I tried making grade 2 a decrement, like new_interval = old_interval / 2, but I really don't like this, myself. Just fail it, for the reasons I discussed. At this time I have a 100 day limit on the intervals. (I've grudgingly increased this from 30 originally, as my collection of cards has gotten bigger.) I increased the length of the statistics output to show the whole 100 days, out of curiosity, and to see that it is working right. I disabled the data upload because I didn't want to interfere with the mnemosyne-project. (But the author has said I'm not doing any harm.) > What is the point of increasing the interval if you barely remembered it > giving a grade 2? Increasing the interval, you will most likely forget it Until you want to edit the program, or run a v2.0 plugin, I suggest you simply fail the card, giving yourself another chance to master it! I'm eager to offer a variety of v2.0 algorithm plugins, if it isn't too hard. Mnemosyne remains ideal for me because it is easy to modify. Thanks. (This post is an improved version of one I posted previously.) --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
