Tomohiro KUBOTA: > BTW, usage of non-ASCII characters are not recommended even if > the character is expressed using entity.
Using entities is the correct way of using non-ASCII characters, and it should be promoted. Most browsers today are multilingual, and should be able to display them properly. > This is because text- based browser cannot display it. Then they should be fixed. > For example, read "This page is available in the following languages" > in http://www.debian.org/index.en.html using text-based browsers like > lynx and w3m. In Lynx, it looks like this (with a iso-8859-1 display): This page is also available in the following languages: català dansk Deutsch Ellynika' (Ellinika) español Esperanto français 한국어 (Hangul) hrvatski Indonesia Italiano magyar Nederlands 日本語 (Nihongo) norsk (bokmål) polski Português româna Russkij (Russkij) suomi svenska Türkçe 中文 (CN) 中文 (HK) 中文 (TW) Lynx automatically transliterates many alphabets (I can sort-of read the Russian page in Lynx, the page title is "Universal'naya Operacionnaya Sistema". The only things not showing properly are the far-east characters. But if I set my display to use UTF-8, I will see them as well. Just because some implementations cannot display the characters is not a reason for not using them (especially since we also include transliterations anyway). It's like deciding that we won't use PNG images just because some browsers can't display them, or to remove the content negotiations because it's not available in some browsers, and so on. The best solution would probably be to transcode *all* our pages to use UTF-8 and automatically remove the entities that can be expressed properly. -- \\// peter - http://www.softwolves.pp.se/ I do not read or respond to mail with HTML attachments. Statement concerning unsolicited e-mail according to Swedish law: http://www.softwolves.pp.se/peter/reklampost.html
