Last modified: 2011-03-13 18:04:45 UTC
Unicode has only approximations of some accented characters used in Lingala. An encoding such as AFRLIN-104-BPI_OCIL fit the needs of the lingala language perfectly. See http://www.progiciels-bpi.ca/tcao/apercu.html#h-an3 Unfortunately the AFRLIN-104-BPI_OCIL has very low support, only free recode currently supports it. Some people have mentionned Windows and Linux support for it but I haven't had access to it yet.
this appears to be a bug in Unicode rather than the website :-) has the Unicode standards body refused to integrated support for this language, or has it not been proposed, or...?
I don't know what is the status of those characters in Unicode, but I know for sure that they aren't included. I'll look into asking Unicode to be compatible with that encoding.
please notify unicode working groups about missing characters. they should be dealing with unicode covering all charsets.
Ok, I'll try to push this to Unicode. In the mean time ln.wikipedia.org is using an approximation of some accented characters.
Unicode will not add precomposed characters for several reasons. The first one being that they are supposed to be composed as we do know. It is Unicode's policy to only add characters that cannot be composed with existing characters. This means that fonts and applications should be able to display composed characters correctly, and applications should allow keyboard entry to be similar to that for precomposed characters. The other reason is simply that the encoding previously mentioned is not really in use, so it is not valid for inclusion as a legacy encoding. For such precomposed characters to be included, countries needing them should develop standardized encoding and push them in Unicode. I'm not in the position of doing so. Composing characters with Unicode is the best thing around, as long as the apps support it. Pango should support it as long as the font does it (it's in CVS) so any gtk+ based application should do that too. The new Qt is pretty decent with diacritics regardless of the font. Mac OS X handles it very well and Windows can too with Uniscribe. This can be left as a wontfix or even as closed since it's not a mediawiki issue, unless we want to support the AFRLIN-104-BPI_OCIL encoding. But I'd rather see applications and fonts handling Unicode properly.
