🚀 go-pugleaf

Vadim Antonov (avg@rodan.UU.NET) writes:
>Now, i think everybody agrees that the "ultimate encoding" is
>the one which provides the complete information about which
>language is used -- it sovles all the problems.

No, if it doesn't tell which font is used I cannot use it to typeset
the text properly.

Seriously, I disagrees with the statement above, and I think most
other people do. Putting the language into the coding gives more
problems than it solves. How deal with umpteen instances of the
same character because it is treated different language?

My modem - or rather the stinking telephone line - prevented me from
read all of Vadim Antonov's text and hang up when I had gone through
18%. The far I got the article seemed to be based on the same erroneous
presumptions that he has presented before. This horse has been beaten 
dead all over again, so I will be brief.

Vadim, your KOI-8 solution might have been the only possible solution
at the time of its creation, but it was still not able to even support
the former USSR entirely. (Assuming that you could not squeeze in Armenian
and Georgian scripts and the extra letters need for Estonian, Latvian,
and Lithuanian. Even less would have been possible if you want to embed
sort rules with the character code, since for Estonian and English you
would have to duplicate at least parts of the alphabet.)

But Unicode is not a partial solution for the Soviet Union in the 1980s. 
Unicode is a tool for complete solutions for the world in the 21st century.
-- 
Erland Sommarskog - ENEA Data, Stockholm - sommar@enea.se