Re: Multicode

Quoting John Cowan <cowan@...>:

> Danny Wier scripsit:
>
> > I'm looking for more information, but I'm having little luck.
>
> This scheme depends on a hopelessly broken encoding trick which requires
> one to read a text from the beginning (alternatively, to look very far
> back) before being able to even *start* figuring out what's going on.
> In UTF-16, you can do random access to any 16-bit codepoint and always
> know its meaning, and even in UTF-8 you have to look back a maximum of
> three bytes before you are in registry.
>
> Furthermore, it founders on two obviously bogus notions: that the boundaries
> between languages are hard and fast, and that there are only 255 of them.
This gave me the skeleton of a conculturing idea: In some holy, or otherwise
unquestionably authoritative, text it is stated that, say, 300 languages are
spoken in the world.

Centuries later, the bearers of the culture in question have developed a global
reach, and run into a thousand or more speech varieties that they'd normally
consider separate languages, but, because the authority of that text, they can't
accept the implication, and instead lump more and more different 'dialects' as
the same 'language' the farther away the speakers live, all the way to the
antipodes, where completely unrelated languages are considered "the same" to
squeeze under that limit of 300. Each time some trader or missionary runs into a
new language, he must classify it as a dialect of one of the 300 accepted languages.


I do not have any intention ATM of actually using this idea, but I thought
someone might like it, if not for incorporation in a coniverse, perhaps simply
as an amusing anecdote.

                                                          Andreas

Reply