Re: Interesting web site about kanji (in Japanese).


<jwb@xxxxxxxxxxxxxxxxxx> wrote in message news:Q%Lpe.10975$F7.1948@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Danny Wilde <fuzakenbo@xxxxxxxxxxx> dixit:
Hmm. I'm using a Japanese set-up, and I very often see web pages in English
which come out as strange kanji, because the page doesn't bother to properly
indicate the charset. 

Things like "I'm" and "he's" seem to get muddled, for some reason. The
apostrophe and the next character get turned into a kanji.


That's because the the page has been written using software from
a shonky outfit in Seattle, which didn't feel obliged to use the
standard code for an apostrophe and invented its own (8-bit) code.


I get `'

If
your browser is locked onto Shit_JIS, it'll treat any pair of bytes
with the first having the MSB on as a double-byte character.

Because my operating system is in Japanese, the browser assumes that the page is in Japanese, but that's a different matter from "locked onto" shift JIS. It's just an unfortunate assumption. The problem is a mirror of Louise's problem with pages in Japanese which her browser doesn't detect properly.

The web site about kanji which I referred to is actually in shift JIS, but there is no information in the html source about that.

I found the following outrageous claim at

http://www.csse.monash.edu.au/~jwb/coding_inf.html:

"IMNSHO, it is very important that WWW pages state their coding in a way that browsers can detect. The traditional way to head up a page with Japanese in it is to have a "META HTTP" directive at the front. Here is what I have as default on the WWWJDIC pages:

<META http-equiv="Content-Type" content="text/html; charset=euc-jp"> "
Surely it is equally important for a page in English or Albanian to indicate itself?

Well, my version of Internet Explorer usually works perfectly well with any
Japanese coding system, although there did seem to be some problem with
Jim's UTF-8 pages.

Examples?

I was talking about the recent discussion of the radicals page you made. I thought those were in UTF-8, no? Were they in some kind of EUC coding?

.

Quantcast