There is a feature of the MCQ window that is worth knowing, but is poorly documented, and until I put some context-sensitive tips within the program, I thought I'd spell it out here...
The MCQ window pauses after showing you the answer, so you can study the answer an commit it to memory. It goes away eventually, but it is quite slow.
If you want to get rid of it immediately, just click the answer again.
Sunday, May 15, 2011
Sunday, May 8, 2011
Version 2.68
Version 2.68 is available for testing, but might be a bit rough. When it is stable and bug-free, I will release it to the public but I would prefer it if a few users tested it out first, if anyone is willing.
Please email me for a download link.
Amongst other improvements, it has two new graphing modes to show you your progress and the state of your memory model. Just click on the main graph to cycle through the other graphs.
If you find any bugs, please just post them here as a comment.
Please email me for a download link.
Amongst other improvements, it has two new graphing modes to show you your progress and the state of your memory model. Just click on the main graph to cycle through the other graphs.
If you find any bugs, please just post them here as a comment.
Wednesday, April 27, 2011
Progress Report
Hi everyone,
There has not been a new release in a little while but that merely reflects the fact that I'm currently going through the Henshall list of 1,945 kanji, making sure that the kanji are in the correct sequence and checking the definitions for duplicates. This work should be finished in a couple of days.
Meanwhile, I have kanji lists for 'Reviewing the Kanji' (RTK) and also the KanjiDamage website, which will be made available for people studying from those sources.
A new feature of the upcoming version is that the RTK keyword will be displayed alongside the current Henshall-supplied or user-supplied definitions, to help disambiguating duplicates. Another new feature is a duplicate-cleanup tool that will flag (and remove if necessary) duplicated kanji and allow modification of duplicate keywords. This is very helpful for cases where the original Henshall definitions have more than one kanji for, say, 'child' or 'village'. One great advantage of learning the RTK keywords is that they are guaranteed to be unique, with exactly one kanji per keyword and vice versa.
Finally, I have added another new feature that makes life a bit easier when you are attempting complex kanji. If you get five strokes in a row correct, then you get an error credit, which shows up as a blue square below the drawing page. Your first error cancels out one credit, and so on. This means you won't be penalised if, in a 19-stroke kanji, your hand slips and you draw one line too short. It does mean that an occasional kanji is marked as known where you didn't quite know all the strokes, but you do still have to redraw wrong strokes so your knowledge will move forward. Your time is better spent learning new kanji than getting every stroke 100% perfect.
I found this last feature very useful when I was trying to catch up from a position that was 80+ kanji behind my target of 50 kanji per week. I'm now at 350+ kanji, ten ahead of target, despite taking a couple of weeks of to program.
There are also some improvements to the beginner-friendly IME, which will be the subject of another post.
Please continue to report any bugs, preferably here but by private email if you prefer.
There has not been a new release in a little while but that merely reflects the fact that I'm currently going through the Henshall list of 1,945 kanji, making sure that the kanji are in the correct sequence and checking the definitions for duplicates. This work should be finished in a couple of days.
Meanwhile, I have kanji lists for 'Reviewing the Kanji' (RTK) and also the KanjiDamage website, which will be made available for people studying from those sources.
A new feature of the upcoming version is that the RTK keyword will be displayed alongside the current Henshall-supplied or user-supplied definitions, to help disambiguating duplicates. Another new feature is a duplicate-cleanup tool that will flag (and remove if necessary) duplicated kanji and allow modification of duplicate keywords. This is very helpful for cases where the original Henshall definitions have more than one kanji for, say, 'child' or 'village'. One great advantage of learning the RTK keywords is that they are guaranteed to be unique, with exactly one kanji per keyword and vice versa.
Finally, I have added another new feature that makes life a bit easier when you are attempting complex kanji. If you get five strokes in a row correct, then you get an error credit, which shows up as a blue square below the drawing page. Your first error cancels out one credit, and so on. This means you won't be penalised if, in a 19-stroke kanji, your hand slips and you draw one line too short. It does mean that an occasional kanji is marked as known where you didn't quite know all the strokes, but you do still have to redraw wrong strokes so your knowledge will move forward. Your time is better spent learning new kanji than getting every stroke 100% perfect.
I found this last feature very useful when I was trying to catch up from a position that was 80+ kanji behind my target of 50 kanji per week. I'm now at 350+ kanji, ten ahead of target, despite taking a couple of weeks of to program.
There are also some improvements to the beginner-friendly IME, which will be the subject of another post.
Please continue to report any bugs, preferably here but by private email if you prefer.
Saturday, April 16, 2011
Kanji Vectors That Need Fixing
The following kanji have issues with unwanted or awkward intersections:
聞 : hear
原 : plain ; field ; origin
弟 : younger brother
They can still be drawn but the second stroke of the 'ear' radical and the second stroke of 'field' have to be started right on the line before; you'll almost certainly pick up an error dot or two. The last stroke of younger brother starts on the line.
These will be fixed in the next version.
Please report any other oddities in the vector data.
PS.原 : plain ; field ; origin and 弟 : younger brother are actually okay - they were affected by a rounding error, rather than a faulty vector set. The problem with 聞 is that the second last stroke must cross the vertical, not just start near it. This will be fixed in the new vector data set.
聞 : hear
原 : plain ; field ; origin
弟 : younger brother
They can still be drawn but the second stroke of the 'ear' radical and the second stroke of 'field' have to be started right on the line before; you'll almost certainly pick up an error dot or two. The last stroke of younger brother starts on the line.
These will be fixed in the next version.
Please report any other oddities in the vector data.
PS.原 : plain ; field ; origin and 弟 : younger brother are actually okay - they were affected by a rounding error, rather than a faulty vector set. The problem with 聞 is that the second last stroke must cross the vertical, not just start near it. This will be fixed in the new vector data set.
Thursday, April 14, 2011
Text Editor
The Kanji Sketch Pad now has a built in Text Editor and although it still needs a couple of refinements it is a handy way to write kanji before you know how they are pronounced.
Version 2.62 allows you to specify the font for kanji, kana and English - this is accessible through the Usage Preferences, because it primarily affects the appearance of the Usage pop-ups. The same fonts are used in the Text Editor. The kanji are coloured to make them easier to find when scanning though a long list.
Tuesday, April 12, 2011
Gotta agree with this
Stumbled across this page, and agree with the sentiment. You should own Japanese books before you can read them. Even though the Sketch Pad is built around memorising kanji in isolation, it should be followed up as soon as possible with actual attempts to decode Japanese text.
Subscribe to:
Posts (Atom)