On Tue, 2011-10-18 at 11:27 +0900, Masahiro Sekiguchi wrote
> ... the difference of the style you are talking is something like
> traditional vs modern preferences. It appears that, for whatever
> reason, the designer of VL Gothic chose the traditional style for
> this particular one. However, I don't think it is a bug; it is a
> matter of preferences.
I never said it was a bug (I think someone else used that word). I
just said it was an odd feature. Thanks to the Ministry of Education
page that you showed me, I have now learned that these are examples of
kanji where the printed form can differ from the accepted handwritten
> > My teacher told me that if any student wrote these kanji in the
> > VL-Gothic way, they would be corrected.
> I'm not sure what your teacher actually said, but I believe you
> misunderstood her explanation.
What she said is consistent with what several Japanese people have
told me. The VL-Gothic forms of soto/GAI, uchi/KA, etc, would be
regarded as wrong if a student hand-wrote them. Teachers would correct
them to the customary handwritten form. However, it seems that there
is more freedom in printed fonts than in handwriting. There are some
fonts that use the handwritten forms. But VL Gothic is not one of them.
> I'm not sure what you want, but just as an example, you can download
> a PDF copy of the latest (revised last year) Joyo kanji hyo (List of
> regular use Kanji)... it is a part of definitive document of modern
> Japanese orthography for legal/official uses, issued by the Prime
This is a very useful resource, thank you. Indeed, my Japanese is
nowhere near as good as your English, and I had to get help in reading
it. At the top of p7 it talks about kanji where there is a difference
between the handwritten form and the printed form (at least in some
fonts). The list there includes the kanji I asked about (soto/GAI and
uchi/KA) as well as some other well-known cases where printing can
differ from handwriting, such as hito/JIN, i(ru)/JYU, hachi/YA.
Actually a vaguely analogous situation exists in English orthography
too. The handwritten form of lowercase "g" and "a" is the
"single-storey" version, but many printed fonts use the "double-storey"
However, it would be very unusual to use the double-storey versions of
these letters in handwriting.
Many thanks for your help.
> All the native speakers of Japanese that I have consulted tell me that
> these kanji look odd, definitely non-standard. They have never seen
> "soto" or "uchi/KA" written this way in handwriting, and hardly ever in
I guess you consulted (relatively) young Japanese speakers only. My
understanding is that the difference of the style you are talking is
something like traditional vs modern preferences. It appears that, for
whatever reason, the designer of VL Gothic chose the traditional style
for this particular one. If you don't like his/her decision, that's
fine. You can have your own opinion. However, I don't think it is a
bug; it is a matter of preferences.
> My teacher told me that if any student wrote these kanji in the
> VL-Gothic way, they would be corrected.
I'm not sure what your teacher actually said, but I believe you
misunderstood her explanation.
> Does anyone know of a computer kanji font that uses the standard form of
> these characters, the form that is used in printing books and is taught
> to students in Japan?
I'm not sure what you want, but just as an example, you can download a
PDF copy of the latest (revised last year) Joyo kanji hyo (List of
regular use Kanji) from a web site of Bunka Cho (Agency for Cultural
Affairs, a Special Body in Ministry of Education.) Although it is
called "List", it is not just a list; it is a part of definitive
document of modern Japanese orthography for legal/official uses, issued
by the Prime Minister.
The PDF is available on the following page:
# The page is entirely in Japanese, and the "English" link on the upper
# right doesn't lead you to a corresponding English page...
On 24th page of the above PDF, you can see the character in question.
You will find the document shows the shape you called a bug as the
reference shape. Note that this PDF has all required fonts embedded,
and the shape of characters you see is what authorities saw when they
approved this publication.
On 9th page, you can also see the same character under the item (6)
(jsut above 3.) It is an example of the explanation beginning on the
7th page. I'm not sure you are able to read the explanation, but your
teacher should. It essentially says the difference of the shapes you
pointed out are both acceptable.
In summary, the shape you called a bug is not a bug. It is perfectly
acceptable. On the other hand, the shape you prefer is also
acceptable, too. it is just a matter of preference.
I myself has no strong preferences over any of these shapes for this
particular character. I will not object if you get a majority support
from Fedora community to modify the shape of this character. However,
I don't agree to call it a bug.
Masahiro Sekiguchi <seki(a)jp.fujitsu.com>
> > But in the vl-gothic that my Fedora 13 has, it looks different: the
> > diagonal stroke in the right half of the character goes through the
> > vertical bar, when it should only go to the right of the bar.
> > How does it look in your vl-gothic font?
> Aah, I see what you mean. In both characters the last stroke is placed
> unusually. I'm not sure if it's a bug or feature though, the IPA fonts
> have it also this way (both Gothic and Mincho style). Might be worth
> asking some actual Japanese whether it's wrong or acceptable
It's a design thing. Both shapes are perfectly fine.
As long as the rasterized image reflects the designer's intention, it
is not a bug.
Masahiro Sekiguchi <seki(a)jp.fujitsu.com>