Re: [Dragon-char-datasets] New CVS character commits, small change to Editor

[Dr. H. A. <har...@pl...>]

ad...@yo... wrote:
> This is turinig into a real community project.  Pretty cool.

Just added another set of characters (both sets are 110+ in toto). I 
reanimated my old and trusty Wacom tablet -- but I should hack the 
character editor so that it makes using a pen even more straightforward; 
at the moment I've configured soft keys on the tablet, but that still 
needs too much hand movement. Also, maybe some automatic for drawing 
strictly horizontal and vertical lines could be helpful...

Now, if we could make a java version which even works with tablets. :) 
Hmm, on Windows we could write a JNI wrapper for the WinTab interface. 
For X there is some X extension for tablets available, as far as I 
remember...

> I had noticed the stroke order problem but had just factored it in to my
> brain.  Ah laziness.  There are similar problems with, for example dou1
> and na4 where the final  should only have two strokes as opposed to three.
>  The curved stroke is all one stroke, and I think whoever did it go it
> confused with The child character (as in final of hao3).

Okay, I will look into these two in the next days. But someone should 
skim through the others, really...

> Cool.  I wonder if we should make a stroke count game, as that was
> something my instructor regularly put on quizes.

Well, don't ask me on this one -- remembering stroke counts is something 
I never had an interest in; for instance, with the electronic Palm 
dictionary I don't need to care about this at all; I just need to sketch 
the character, et voila!

> I have been talking with Thomas offline about another game I have written
> for Dragon, which gives you the character and requires you to provide the
> english.  It worked in the pre hi-res font version, but now that we are
> doing the tones over the vowels, I have a couple problems to solve.  I
> have heard that the most common input method when typing is to type in
> pinyin, and ad the number afterward, as we have been doing, and that will
> put the ton over the word.  In order to delete, however, you merely
> backspace until you remove the voewl, and then renter.  The tone does not
> hide as the last character (this is defferent from how the vowels are done
> in Hebrew).  Do these rules make sense and  should I implement this way? 
> Also, will the new fonts take in to account the umlaut?

Hmm, I would just accept that the user can input a tone number, whenever 
she or he wishes to do so and that number is then attributed to the 
current syllable. So, no need to backspace to the vowel, just enter 
another tone number to change a possible existing one.

Oh, while you are at it: could you please be so kind as to make the 
program suppressing "5"? Currently it shows up in the display, albeit it 
should better be suppressed.

Hey, what about a Dragon Light, only with the drawing part? I have to 
admit that for learning terms I rather use Eric Youngdale's Chinese 
Flashcards. It's the traditional old-fashioned flash card principle, but 
I personally like it more than multiple choice. But that's really just 
my personal opinion.

-- Harald

-- 
Dr. Harald Albrecht
Chair of Process Control Engineering
RWTH Aachen University of Technology
Turmstrasse 46, D-52064 Aachen, Germany
Tel.: +49 241 80-97703, Fax:  +49 241 80-92238