Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Re: 臺灣和香港Big5 HKSCS vs UAO分析和結論
126 views
Skip to first unread message
Kang-Hao (Kenny) Lu
Apr 22, 2012, 6:16:05 AM4/22/12
to "Mozilla 開發（mozilla.dev.general.zh）", Yuan Chao, Philip Jägenstedt, Hung-Te Lin
(12/04/22 17:05), Philip Jägenstedt wrote:
> On Sun, 22 Apr 2012 02:23:12 +0200, Yuan Chao <yuan...@gmail.com> wrote:
>> Emm... Here you think the current firefox solution is not perfect and
>> the needs in Taiwan is negligible so it's better to use big5-hkscs to
>> replace the big5 (seems to be CP950?)? I'm an experimental high energy
>> physicist. The best way to resolve a debating and validate a theory is
>> to do experiment and measure it. :) Maybe you can just implement it in
>> Opera and make a survey to see how both HK and Taiwan users appreciate
>> it?
>
> It will definitely be an improvement for Opera since HKSCS will start
> working and UAO has never worked, but if there's something even better
> we could do I'd really prefer that. A better test would be to see the
> reactions if Firefox changed, but that's not an experiment I can run :)
I am forwarding Firefox-related discussion to the relevant list, for
which I happen to help with spam screening. I am also looping in the
person who made big5-uao in Firefox happen.
哈囉 mozilla.dev.general.zh、piaip
public-html-ig-zh[1] 上這個月討論了不少 big5 到底要怎麼處理的問題。
Firefox 目前是用單向 Unicode 補完[2]。阿菲用蜘蛛研究的結果[3]是：big5-
uao 只在台灣的 big5 網站中的 0.13% 比 big5-hkscs 好，而且更多的網站會給
出奇怪的中文字，而在香港的 big5 網站中則 big5-hkscs 全勝。然後我發現如果
把一些 big5-2003 的映射弄進 big5-uao 的話，這些中文字應該就沒有了[4]。
這裡 Firefox big5 的走向兩個截然不同的方向（補一個對照組）：
一、不要再動了
二、改成預設用 big5-hkscs 解碼器
三、把 big5-2003 的一些映射弄進現在的 big5（big5-uao）解碼器
阿菲的意思可能是比較傾向方向二一點，我自己則是比較傾向方向三一點（不過老
實說，沒什麼太大的意見，不過方向三感覺比較麻煩一點 :( ）。我們有可能往二
的方向做一些實驗嗎？有沒有人知道要怎麼進行比較好，丟一個測試版本？
（實際來說，我感覺就算做了這個更動，使用者應該也不會有什麼反應。那這樣的
話，是不是就以全球有一個相同的 big5 的預設處理：使用 big5-hkscs，比較好
呢？）
[1] http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/thread
[2] https://bugzilla.mozilla.org/show_bug.cgi?id=310299
[3] http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/0046
[4] http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/0047
此致
Kenny
> On Sun, 22 Apr 2012 02:23:12 +0200, Yuan Chao <yuan...@gmail.com> wrote:
>> Emm... Here you think the current firefox solution is not perfect and
>> the needs in Taiwan is negligible so it's better to use big5-hkscs to
>> replace the big5 (seems to be CP950?)? I'm an experimental high energy
>> physicist. The best way to resolve a debating and validate a theory is
>> to do experiment and measure it. :) Maybe you can just implement it in
>> Opera and make a survey to see how both HK and Taiwan users appreciate
>> it?
>
> It will definitely be an improvement for Opera since HKSCS will start
> working and UAO has never worked, but if there's something even better
> we could do I'd really prefer that. A better test would be to see the
> reactions if Firefox changed, but that's not an experiment I can run :)
I am forwarding Firefox-related discussion to the relevant list, for
which I happen to help with spam screening. I am also looping in the
person who made big5-uao in Firefox happen.
哈囉 mozilla.dev.general.zh、piaip
public-html-ig-zh[1] 上這個月討論了不少 big5 到底要怎麼處理的問題。
Firefox 目前是用單向 Unicode 補完[2]。阿菲用蜘蛛研究的結果[3]是：big5-
uao 只在台灣的 big5 網站中的 0.13% 比 big5-hkscs 好，而且更多的網站會給
出奇怪的中文字，而在香港的 big5 網站中則 big5-hkscs 全勝。然後我發現如果
把一些 big5-2003 的映射弄進 big5-uao 的話，這些中文字應該就沒有了[4]。
這裡 Firefox big5 的走向兩個截然不同的方向（補一個對照組）：
一、不要再動了
二、改成預設用 big5-hkscs 解碼器
三、把 big5-2003 的一些映射弄進現在的 big5（big5-uao）解碼器
阿菲的意思可能是比較傾向方向二一點，我自己則是比較傾向方向三一點（不過老
實說，沒什麼太大的意見，不過方向三感覺比較麻煩一點 :( ）。我們有可能往二
的方向做一些實驗嗎？有沒有人知道要怎麼進行比較好，丟一個測試版本？
（實際來說，我感覺就算做了這個更動，使用者應該也不會有什麼反應。那這樣的
話，是不是就以全球有一個相同的 big5 的預設處理：使用 big5-hkscs，比較好
呢？）
[1] http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/thread
[2] https://bugzilla.mozilla.org/show_bug.cgi?id=310299
[3] http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/0046
[4] http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/0047
此致
Kenny
Kang-Hao (Kenny) Lu
Apr 23, 2012, 10:44:31 PM4/23/12
to "Mozilla 開發（mozilla.dev.general.zh）"
很遺憾 Google Groups 的存檔不知道搞什麼出毛病了，重新寄一次。這裡的原文
討論串在：
http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/thread#msg46
(12/04/22 17:05), Philip Jägenstedt wrote:
> On Sun, 22 Apr 2012 02:23:12 +0200, Yuan Chao <yuan...@gmail.com> wrote:
>> Emm... Here you think the current firefox solution is not perfect and
>> the needs in Taiwan is negligible so it's better to use big5-hkscs to
>> replace the big5 (seems to be CP950?)? I'm an experimental high energy
>> physicist. The best way to resolve a debating and validate a theory is
>> to do experiment and measure it. :) Maybe you can just implement it in
>> Opera and make a survey to see how both HK and Taiwan users appreciate
>> it?
>
> It will definitely be an improvement for Opera since HKSCS will start
> working and UAO has never worked, but if there's something even better
> we could do I'd really prefer that. A better test would be to see the
> reactions if Firefox changed, but that's not an experiment I can run :)
I am forwarding Firefox-related discussion to the relevant list, for
which I happen to help with spam screening. I am also looping in the
person who made big5-uao in Firefox happen.
哈囉 mozilla.dev.general.zh
討論串在：
http://lists.w3.org/Archives/Public/public-html-ig-zh/2012Apr/thread#msg46
(12/04/22 17:05), Philip Jägenstedt wrote:
> On Sun, 22 Apr 2012 02:23:12 +0200, Yuan Chao <yuan...@gmail.com> wrote:
>> Emm... Here you think the current firefox solution is not perfect and
>> the needs in Taiwan is negligible so it's better to use big5-hkscs to
>> replace the big5 (seems to be CP950?)? I'm an experimental high energy
>> physicist. The best way to resolve a debating and validate a theory is
>> to do experiment and measure it. :) Maybe you can just implement it in
>> Opera and make a survey to see how both HK and Taiwan users appreciate
>> it?
>
> It will definitely be an improvement for Opera since HKSCS will start
> working and UAO has never worked, but if there's something even better
> we could do I'd really prefer that. A better test would be to see the
> reactions if Firefox changed, but that's not an experiment I can run :)
I am forwarding Firefox-related discussion to the relevant list, for
which I happen to help with spam screening. I am also looping in the
person who made big5-uao in Firefox happen.
哈囉 mozilla.dev.general.zh
0 new messages