stroke order for 瓦

56 views
Skip to first unread message

Hubert

unread,
Jul 25, 2011, 6:27:07 AM7/25/11
to KanjiVG
There is an inconsistency between RTK and kanjivg for 瓦. Heisig has
the hook as the third stroke, then the drop in the middle and finally
the drop at the bottom. KVG has the drop at the bottom as the third
stroke, then the hook and the drop in the middle as the final stroke.

I have no idea which order is correct.

Regards,

Hubert.

Hubert

unread,
Jul 25, 2011, 6:30:02 AM7/25/11
to KanjiVG
FWIW, the kanjivg data for 瓶 has the Heisig order for the right part
of the kanji.

Ben Bullock

unread,
Jul 25, 2011, 8:35:54 AM7/25/11
to kan...@googlegroups.com

Hubert

unread,
Jul 25, 2011, 9:53:20 PM7/25/11
to KanjiVG
Ben,

thanks for the links. I looked up 瓶 on the first site you posted
(http://www.csse.monash.edu.au/~jwb/cgi-bin/wwwjdic.cgi?161344_%C9%D3)
and it has the same order as 瓦 in kanjivg for the right part, where
kanjivg has the Heisig order for the right part in 瓶. As I said, I
have no clue at all what is correct. But unless I was not quite awake
this morning when I looked up the stroke order on http://kanji.sljfaq.org,
it seems like the kanjivg data is inconsistent between 瓦 and 瓶. The
site you posted shows a consistent stroke order that is different from
the RTK book. Could the stroke order be different for 瓦 when it is
part of 瓶?

Thanks,

Hubert.

On Jul 25, 8:35 am, Ben Bullock <benkasminbull...@gmail.com> wrote:
> The wwwjdic stroke ordering is the same as the kanjivg one:
>
> http://www.csse.monash.edu.au/~jwb/cgi-bin/wwwjdic.cgi?163476_%B4%A4
>
> The kakijun web site has the same order:
>
> http://kakijun.main.jp/page/kawara200.html
>
> Yet another web site agrees:
>
> http://www.taishukan.co.jp/kanji/qa05.html
>
> http://www.taishukan.co.jp/kanji/images/020401.gif
>

Ben Bullock

unread,
Jul 26, 2011, 2:59:13 AM7/26/11
to kan...@googlegroups.com
On 26 July 2011 10:53, Hubert <hubert....@gmail.com> wrote:
> I looked up 瓶 on the first site you posted
> (http://www.csse.monash.edu.au/~jwb/cgi-bin/wwwjdic.cgi?161344_%C9%D3)
> and it has the same order as 瓦 in kanjivg for the right part, where
> kanjivg has the Heisig order for the right part in 瓶. As I said, I
> have no clue at all what is correct.

I don't have a definitive answer either, but my guess is that the
kanjivg data for 瓶 contains a mistake and the 瓦 is correct.

> But unless I was not quite awake
> this morning when I looked up the stroke order on http://kanji.sljfaq.org,
> it seems like the kanjivg data is inconsistent between 瓦 and 瓶.

I'm sorry if I misunderstood what you meant in the previous post.

Looking at the following list,

http://kanji.sljfaq.org/mr.html?b=127

of nineteen kanji with the kawara pattern, I only see the 瓶 kanji with
this pattern, so I think that 瓶 is just an error.

Incidentally I implemented a simplistic lookup system for the stroke
order diagrams to check them on-site, e.g.

http://kanji.sljfaq.org/minidic.cgi?k=%E7%94%8C

I specifically put this there to do these checks just now. I'll be
improving that page gradually.

> The
> site you posted shows a consistent stroke order that is different from
> the RTK book. Could the stroke order be different for 瓦 when it is
> part of 瓶?

I really doubt it, unless some specific reference says that it is so
then I would just assume it is an error.

> Thanks,
>
> Hubert.
>
> On Jul 25, 8:35 am, Ben Bullock <benkasminbull...@gmail.com> wrote:
>> The wwwjdic stroke ordering is the same as the kanjivg one:
>>
>> http://www.csse.monash.edu.au/~jwb/cgi-bin/wwwjdic.cgi?163476_%B4%A4
>>
>> The kakijun web site has the same order:
>>
>> http://kakijun.main.jp/page/kawara200.html
>>
>> Yet another web site agrees:
>>
>> http://www.taishukan.co.jp/kanji/qa05.html
>>
>> http://www.taishukan.co.jp/kanji/images/020401.gif
>>
>> On 25 July 2011 19:27, Hubert <hubert.schre...@gmail.com> wrote:
>>
>>
>>
>> > There is an inconsistency between RTK and kanjivg for 瓦. Heisig has
>> > the hook as the third stroke, then the drop in the middle and finally
>> > the drop at the bottom.
>

> --
> You received this message because you are subscribed to the "KanjiVG" group.
> For options and unsubscribing, visit this group at
> http://groups.google.com/group/kanjivg

Ulrich Apel

unread,
Jul 26, 2011, 9:02:35 AM7/26/11
to kan...@googlegroups.com
There are several "correct" versions for 瓦.

Unfortunately, this is one of the problems of the Han unification <http://en.wikipedia.org/wiki/Han_unification>. We have to find ways how to deal with the fact that there is only one code point in Unicode for several kanji variants.

The Unicode consortium always has an explanation on its character lists:
"The shapes of the reference glyphs used in these code charts are not prescriptive. Considerable variation is to be expected in actual fonts."

This is topic, which is of course worth looking deeper into. It should be possible to describe several kanji variants, find ways of naming them -- probably using stroke types --, find ways for look-up of variants and suggest a main version of character. Perhaps I might have some ideas in the next few weeks.

瓦 is not part of the Joyo kanji, and for Japanese, there is no official character shape and stroke order.

Best wishes

Ulrich

Hubert

unread,
Jul 26, 2011, 1:02:11 PM7/26/11
to KanjiVG
> Incidentally I implemented a simplistic lookup system for the stroke
> order diagrams to check them on-site, e.g.
>
> http://kanji.sljfaq.org/minidic.cgi?k=%E7%94%8C
>
The plot thickens! It seems that 瓶 and 瓦 are the only two where 瓦 has
five strokes. The remaining kanji from your list with 瓦 in them only
use four strokes to draw 瓦! Now I am even more confused. According to
Ulrich, there is no official stroke order. But stroke count? Can the
stroke count for 瓦 change in different kanji?

Thanks,

Hubert.

Ulrich Apel

unread,
Jul 26, 2011, 2:33:36 PM7/26/11
to kan...@googlegroups.com
EMORI, Kenji (2003): Kaigyôsô – Hitsujun jitai jiten. Sanseidō, Tokyo.
gives two stroke orders for 瓦. The one in KanjiVG is the second. With this stroke order, it is clear, that there are five strokes.

With the other stroke order, it seems that the second and third stroke can be written without taking the brush from the paper so that they become one stroke. This how the character looks in the kyokasho font and in kaisho fonts.

The best would be to have data for all three variations and for example to suggest one as the official one.

Best wishes

Ulrich

Hubert

unread,
Jul 26, 2011, 11:10:29 PM7/26/11
to KanjiVG
Ulrich,

the stroke order in kanjivg for the kanji 瓦 is such that the second
and third stroke could be combined into a single stroke, but 瓦 has
five strokes in kanjivg. The kanji 瓶 in kanjivg is the only kanji
where the stroke order changes (ceiling, walking stick, fish hook, ice
in Heisig terms). This kanji also has five strokes for the 瓦 element
because the second and third stroke cannot be combined. I do not have
access to the reference you mention, but it seems like 瓶 is the only
one in kanjivg that uses the second stroke order from your reference.
The kanji 瓦 by itself appears to use the first stroke order, but
retains five strokes. All other characters except for 瓶 that contain 瓦
appear to use the first stroke order as well, but combine the second
and third stroke into one and only use four strokes.

I agree that it would be great to have all three variations in
kanjivg, but I don't know if that is feasible. From the perspective of
someone who is just trying to learn some kanji for the fun of it, I
think it would be best to have consistency in the stroke order and
stroke count. In the multi-radical search engines, 瓦 is listed as
having five strokes, so it might make things easier for someone like
me to use five strokes even if the first stroke order is used. Of
course, the quick fix to achieve a certain level of consistency would
be to change 瓶 and just remember that 瓦 loses one stroke when used as
an element in other kanji.

Viele Grüße,

Hubert.

Ulrich Apel

unread,
Jul 27, 2011, 9:53:49 AM7/27/11
to kan...@googlegroups.com
Hubert,

thanks a lot for your comments! I guess, I got the point. For learners, one version would be enough, and it should be consistent where ever a kanji appears as component. It is not alone a question of right or wrong.

Kanji are arbitrary sometimes, but choosing the most consistent versions should be possible. Developing a consistent systematization might be more work than making the three variations, but you are right, that this would be worthwhile.

I hope that I can make suggestions on this soon.

Viele Grüße

Ulrich

Reply all
Reply to author
Forward
0 new messages