Mojibake again

[Glenn]

I thought that the problem had been solved, because I have posted several times without any mojibake and without doing anything special to manipulate the situation (such as selecting Western European ISO encoding). However, in this message I ended up with some, without doing anything any differently than I had been before. Is there anything that I can do to prevent this in the future?

 

[PaulTB]

I thought that the problem had been solved because I have posted several times without any mojibake and without doing anything special to manipulate the situation (such as selecting Western European ISO encoding). However, in this message I ended up with some, without doing anything any differently than I had been before. Is there anything that I can do to prevent this in the future?

That's a known problem that I demonstrated in one of my posts in the test thread.

Basically each Japanese character takes two bytes, and every 50 bytes the forum inserts a space. If this space is half-way through a character you will get mojibake. This will only be the case if you mix western and Japanese characters.

The best way of avoiding it is to only type 25 Japanese characters in a row.

[VIEW THIS TEST IN SHIFT_JIS]
No spaces typed.
１２３４５６７８９０１２３４５６７８９０１２３４５６７８９０１２３４５６７８９０
One space typed.
１２３４５ ６７８９０１２３４５６７８９０１２３４５６７８９０１２３４５６７８９０
Western character in text. (first number 6)
１２３４５6７８９０１２３４５６７８９０１２３４５６７８９０１２３４５６７８９０

 

[Glenn]

Does that mean that, as an example, "a" would be 10011010, but あ would be 0111001001111001? So Unicode would solve that problem, right? For instance AA38 would be "a" and 29BD would be あ. Am I understanding this correctly? Please bear with me, I know hardly anything about computers, especially technical things like codes and languages.

 

[PaulTB]

Does that mean that, as an example, "a" would be 10011010, but あ would be 0111001001111001? So Unicode would solve that problem, right?

In theory. However I think there may still be computer / browser systems that have problems with Unicode.

 

[Glenn]

Here is a slightly different problem that I just experienced with encoding. I tried to enter "ﾃｴ" into a post ("o" with a circumflex, in case it does it again), and it came up as html code -- "ô" I suppose that it's significant that it was in the "Learning Japanese" forum. So, why did it do that, and is there anything that I can do to prevent it?

 

[PaulTB]

Here is a slightly different problem that I just experienced with encoding. I tried to enter "ﾃｴ" into a post ("o" with a circumflex, in case it does it again), and it came up as html code -- "ô" I suppose that it's significant that it was in the "Learning Japanese" forum. So, why did it do that

Because ﾃｴ is not represented in the Japanese encodings.

and is there anything that I can do to prevent it?

Try, before editing your message, right clicking and selecting a different encoding. Unicode should work, and (paradoxically) Western European might well also work.

 

[Glenn]

Try, before editing your message, right clicking and selecting a different encoding. Unicode should work, and (paradoxically) Western European might well also work.

I tried Japanese (EUC), Unicode (UTF-8), and Western European (ISO) and (Windows), and none of them worked. I guess this is a lost cause....

 

[PaulTB]

I tried Japanese (EUC), Unicode (UTF-8), and Western European (ISO) and (Windows), and none of them worked. I guess this is a lost cause....

Actually it does work - but you can only see it if you select Western European (or whatever) when viewing the message as well.

Shiritori

 

[Frungi]

I tried Japanese (EUC), Unicode (UTF-8), and Western European (ISO) and (Windows), and none of them worked. I guess this is a lost cause....

Did you try Japanese (Shift-JIS)? I think it's what the site uses, and it seems to work for me at least. I'm in Shift-JIS right now, so I'll try typing a circumflex-o.
窶ｦﾅ?
If you see the "^o" above, then use Shift-JIS. If not, then disregard this entire post.
I guess you could also keep trying different encodings and use the Preview Post button that I just scrolled down and saw (this is my first post)...

 

[Frungi]

窶ｦﾅ?

w00t, it works! In your face, Glenn! Ha-ha!
Ehh... I'll shut up now. ﾅ津､窶禿??堙遺?堋ｳ窶堋｢ ^^;

 

[Glenn]

Err, I see a single dot. It apparently worked for you, though. Anyway, I've given up on the idea for now. Maybe when I get a new Mac with Firefox it'll work fine; IE seems to really suck.

[Edit] I should've mentioned that I'm on a PC right now.

 

[Frungi]

ô
ô
ôôôôÔ
This was in Firefox (Mac). My previous post (which apparently only I can see it) was in OmniWeb. I tried it in IE, but I think it is a lost cause. So, can anyone see those o's up there?
[edit] Never mind, Firefox converted them to &# HTML code.

 

[Glenn]

Well, I can see them now, on a PC, with IE. Don't know what that means, though.

 

[Frungi]

IE seems to really suck.

[Edit] I should've mentioned that I'm on a PC right now.

Can't you get Firefox on that PC?

 

[Glenn]

Well, we had it but it was then removed by someone who knows more about computers than I, so I didn't want to go against that. Also, it's not actually my computer, so I don't want to mess with it too much.

 

[Elizabeth]

I thought that the problem had been solved, because I have posted several times without any mojibake and without doing anything special to manipulate the situation (such as selecting Western European ISO encoding). However, in this message I ended up with some, without doing anything any differently than I had been before. Is there anything that I can do to prevent this in the future?

It shows up seemingly randomly on occasion for me as well when writing in Shift-JIS and without inserting any Western characters. As it apparently isn't a structural encoding problem of the board, I'm not inclined to bother with it further beyond change the encoding for editing and see if that helps.