Firefox Status Bar and Encodings

[roddy]

I'm not sure anyone will know this, but I've been amazed at what some people on here can explain previously, so . . .

This post

In Firefox, when I mouseover those two links, the Ebay one displays, with the Chinese characters shown, in the status bar. The Baidu one isn't shown. Ebay is UTF-8 encoded, Baidu is GB2312. This site meanwhile is in UTF-8.

In IE, both links show but in neither case are the chinese characters shown - you get the raw (ASCII? I have no idea.) version of the characters, %E6%8A%A4%E5%85%B7 on ebay, and %BB%A4%BE%DF on baidu.

Basically, what I don't understand is . . .

Why doesn't Firefox at least show the 'raw' version of the Baidu url, even if it can't manage the characters?

Why can't IE manage characters for either of the encodings?

Will encoding ever be easy to understand?

Roddy

[Jizzosh]

Perhaps you've garnered your answer already, but:

"Internet Explorer is not usable for all Unicode entities since in many cases it tries to combine an entity with a preceding character, which results in a wrong display."

Something I found on the limitation of IE, don't know if it's relevant, but if IE can display html incorrectly, it serves that the program itself carries the same limitation, forcing the actual code, where as FF can interpret the character.

As to why FF interprets the Ebay link and not the Baidu link, I think it's because FF as a program can interpret UTF characters, where as a more specific output like GB2312 does not get interpreted into the status bar. With my version of FF (1.5), I get chars for the Ebay link and code for the Baidu link. I think the links that use unicode will get characters and the links that don't get code. Perhaps the older version of FF tried to do somthing with the GB code, but it was lost in translation...

It could be how the programs were coded also. I imagine IE was coded in Visual Basic while FF might be in C++ or something and the coding of VB may not handle any encoding natively while C++ or whatever they use for FF sees and interprets UTF without a problem. (I think I do remember something about the love of UTF from my old linux loving friends and their disdain for Microsoft because they said MS had a habit of not following standards.)

I know this is mostly speculation, but perhaps it makes sense enough to answer your question. For more detailed answers, I would suggest asking on the Mozilla forums.

[smalltownfart]

In your FF browser, I think you will find that the chacter encoding being used is UTF-8 (I suppose this must be because it's the encoding for this site).

So if you change your Firefox display encoding by setting it to:

"View-Character encoding, GB2312", you will see the hanzi rendered correctly for Baidu as well.

As Roddy noted, the ebay link is UTF-8 encoded, while baidu one is GB2312.

I think Jizzosh is correct, here is my attempt to explain it another way:

In order to see UTF-8 characters, you do not need any more additional information, the char code tells you what the character should be. It is by design supposed to be language independent. You shd see the correct character no matter what language you are using.

To render GB2312 on the other hand, you need to know what encoding is being used, in addition

to the character code(s). This is because the older non-unified charsets need a mapping table to know what to display. This is why you see French/funny accented characters/garbage when you somehow get the mapping table wrong or mixed up when you are using GB2312 or Big5 etc.

In the absence of such information, I think Firefox just doesn't try to render it at all, which seems like a smart thing to do.

On the other hand, IE doesn't seem to render the hanzi no matter what encoding you use. So at least both browsers are consistent in how they treat the 2 URLs.