Re: Andrew, could you replace your mailer and attach Subject?

[Masataka Ohta <mohta@xxxxxxxxxxxxxxxxxxxxxxxx>]

> >Though it has nothing to do with MIME, it is a good example of how absurd
> >10646 is.
> >
> >The above model has a fatal defect in interactive environment. You can't
> >determine where symbols are separeted unless you read the next non-combining
> >mark. Thus, you can't display a symbol unless you look ahead an extra
> >character.
> 
> Your reasoning is flawed.  Nobody said that a keyboard has to generate
> characters, it could generate text elements.

I can't understand why you mention a keyboard here.

Nobody said "keyboard".

Though I don't like your use of the terms "character" and "text element",
because their meaning is quite different from their usual one, let's
use it for a while.

The problem is that the unit of transmission in transport layer is, at most, a
character. Even if some entity (which might be a keyboard, but which might
be a program) produces a text elements as a unit, it arrives at the
destination as a byte stream. The problem is that while characters are
self delimiting, the text elements are not.

Thus, if you receive

	a base character 'A'

only, can you display 'A'? Or should you wait until you receive the possibly
following combining characters?

If you wait, when can you stop waiting? The next character might not arrive
forever.

If you display 'A' and then recieve

	a combining character "diaeresis"

what can you display?

Do you want to backspace and redraw 'A' with diaeresis again?

But, displaying 'A' when 'A' with diaeresis was required is the
incorrect behavior.

For example, if the other output stream outputs completely different
character sequences at the same time, how can be displayed?

What if, you are using a typewriter on papers? Though you can simply
overstrike diaeresis in this case, there are much fancier combination
of a base character and combining characters.

Or, do you want to modify TCP protocol to ensure the transmission
of possibly-infinitely-lengthy text element as a unit?

> PS. I am not trying to defend 10646, of which I know nothing.  I just
> have the feeling that your opposition is more emotional than rational.

Can you explain how my opposition can be emotional?

						Masataka Ohta