Home > Uncategorized > Rewriting the long-term study list, Part 4.

Rewriting the long-term study list, Part 4.

As I’m getting more things done in the long-term study list, I’m also facing more problems so the work has slowed down a bit. There are still quite a few things to be done before it can be released.

Statistics.
I tried to find a solution to save old statistics somehow and convert them to a new format, but it has become apparent that I can’t do that without giving up the freedom I need for creating the new test I imagined. I want to display the estimated time left, and in order to do that I need more detailed statistics. It’s not enough to take the average time spent on a single item and multiply that with the number of items, because for example the new items are usually repeated several times during the same test before the student is confident enough to accept them as learned. Instead of this, zkanji will make a rough estimation on the number of times an item from a given deck is repeated usually, and on how much time is spent on such item in average.

Better spacing of items.
In the current form of the long-term study test, the program uses a small list of the ratio of forgotten items in each deck. The values are used to calculate the number of days till an item is tested again based on the deck it is in. This system works fine without much tweaking, but does not avoid those cases when several items of the same word are scheduled for the same day. I will need to figure out the maximum deviation in scheduling that doesn’t ruin the system but will solve this problem.

Automatically adding words.
This seems to be the most difficult problem right now. As I mentioned previously, I would like to borrow a list of words that will most probably appear in future JLPT, but I still haven’t contacted anyone about this (I will do it once everything else is out of the way). Of course this in itself is a problem too, but not what I wanted to write about. 🙂 The real difficulty will be deciding the order of these words. In my experience kanji can only be studied efficiently, when at least 2-3 words are studied at the same time that contain the same reading of the kanji. When there are several readings, this number can be as high as 10-15 words for a single kanji.

If you think about this for a moment you will see, that studying for the JLPT from JLPT word lists can be an obstacle when the aim is not only passing the JLPT itself, but also being able to read Japanese newspapers/books/manga. I don’t want to tell anyone how to study, because each student’s aim is different, so I won’t add random or even frequent words that are not on JLPT word lists, but I will need to find a good algorithm to reorder the lists so words with the same kanji are grouped somehow.

User Interface changes.
I don’t mean how the windows look like. This is more about menu items, placement of buttons, labels showing how many items are due today etc. There is not much to write about this, except that it usually takes as much or sometimes even more time to design a good UI as writing the code itself. (This is what most authors of open-source or free programs usually skip…)

Advertisements
Categories: Uncategorized
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: