Re: 臺灣和香港Big5 HKSCS vs UAO分析和結論

On Sat, 21 Apr 2012 03:49:42 +0200, Yuan Chao <yuanchao@gmail.com> wrote:

> 2012/4/21 Philip Jägenstedt <philipj@opera.com>:
>>> (12/04/19 1:01), Philip Jägenstedt wrote:
>
>>>>  • 298 pages with mixed/broken encodings
>>>>   • 190 pages that would yield U+FFFD with HKSCS, but instead produces
>>>> bogus Chinese characters using UAO, some of them user-visible:
> Phillip, which OS are you using? To me, they are all visible as
> squares with code
> ID in it with HKSCS under Ubuntu!

Ubuntu. To test user-visibility I'm using Firefox's Big5 as an  
approximation of UAO and Opera's Big5-HKSCS as an approximation of HKSCS.  
It's using Firefox that you get the bogus Chinese characters.

>>>> Using Big5-UAO for Taiwanese sites would give mixed results. Correctly
>>>> encoded Big5-UAO is very rare, so the tested mapping (Firefox)
>>>> introduces almost as many user-visible misencodings as it fixes and
>>>> masks many others.
>>>
>>>
>>> 我不知道該說什麼才好了，感覺為 Big5-UAO 把 big5-2003 的東西加回去一些可
>>> 以解決很大部份，另外，上面這些字都不是日文漢字，所以也不影響我對 Big5-
>>> UAO 的要求 :p，有人知道這部份的編碼對應是在可以動手術的範圍還是不行？
>>
>>
>> 按照上面的，用Big5-2003並不是很完美的。MozTW的映射好像不是完全可靠，所以我不知道該根據什麼去定義Big5-UAO。
>>
>> 問題的範圍畢竟是0.043%的臺灣網頁的幾個字符。現代的瀏覽器只有Firefox能顯示，而且他們的映射還造成別的問題……
>>
>> 在這種情況下，我覺得嘗試跟受影響的網站聯繫還是有希望。反正這是唯一的辦法能夠讓香港和國際的用戶也看得到。
> I don't know... to me the original thought of big5-hkscs doesn't seem
> to dominate, and looks
> like big5-uao is not dominate either according to "bing". (I just
> realize that our
> "frequent-visit-sites" with big5-uao are not under "*.tw". Some of my
> treasure sites can
> only be found in internet-archive now) To my surprise is that there
> are quite a lot of cases can
> be explained with big5-2003 PAU though. (probably to Kenny too) At
> least HK friends can
> live with a hack in firefox to force big5-hkscs=big5; ie is ok if the
> official patch installed (the
> font with extended glyphs is needed for up to win xp). I'm curious
> about the browser share
> in HK?

I don't know about Hong Kong browser market shares, but I assume that IE  
has been dominant for a long time and probably still is.

If there are popular/important sites that depend on UAO, then I'd really  
recommend contacting them to see if they can either escape the conflicting  
code points as &#1234; or even use UTF-8. That would fix the problem for  
all users immediately, instead of only Taiwan-locale users in a few years.

Perhaps there does exist a non-HKSCS mapping that would work better than  
Firefox's Big5 for Taiwan sites, but I'm really not sure how to define it.  
One would probably need to figure out which OS and fonts were used to  
produce the content and base the mapping on that. Still, Big5 should  
certainly be a synonym for Big5-HKSCS by default, so that other mapping  
would have be either locale-dependent (i.e. not work for me) or depend on  
sniffing.

Futher suggestions welcome.

-- 
Philip Jägenstedt
Core Developer
Opera Software

Received on Saturday, 21 April 2012 11:56:32 UTC