Tom Harminc wrote:
>Surely a Unicode-aware agent will never try to display UTF-8 directly.
Yes, for sure. But I don't expect to have a fully Unicode-aware operating
system, including Unicode text files and Unicode editors, before about the
year 2003. Some applications will come earlier (currently I have at least
partially Unicode aware web browsers and news readers at my hand). However,
due to the universal scope of Unicode (think of identifiers in programming
languages) there will be a longer time of overlap between old technology
and the new technology to come.
I see *all* proposed UTF schemes as migration schemes which help a smooth
upgrade untill the fully Unicode aware operating systems are on everybodies'
>I'm not quite getting it - what sort of errors would the presence of
>C1 characters cause ? Are you referring to the case where a
>non-Unicode-aware agent sends the UTF-8 directly to a screen or
>printer ? Why would display or printer devices be sensitive to the
>C1 range ?
The VT terminal protocol uses the C1 character set. It will cause several
terminal actions there. The VT terminals allow display of e. g. Latin-1.
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:35 EDT