At 20:27 +0100 2001/02/09, Frank Mittelbach wrote:
>your opinion is that natural language doesn't contain digits; those who have
>argued believe you are mistaken. Just like `a' might be semantically a
>variable belonging to a formula, even if the formula is just consisting of
>this variable, it might alternatively be the letter a or indeed the word
>a. Why do you accept the latter but claim that this can't be true for digits?
As you know, math formulas in the beginning were expressed solely using
words. Only gradually, in fact rather late in history, one introduced
If one carries out your argument, then there should be no math environment.
-- I merely carry out the logic of increased markup a bit further. As I see
there is no simple end to it -- one can continue the game as long as you
please and have effort.
>> (a) For people with GUI interface like Scientific Word, the software has
>> to deal with the ambiguity as best it can.
> and what I am saying is that they can't, really, deal with it. they have to
> provide a math markup and there is no way to identify something as "this
> should be math" automatically. so there is no real solution to the problem
> that people by mistake leave the needed markup out.
If one leaves numbers out in a text mode, then those that want to use
numbers will have to do some additional markup, be it "math" or "phone" or
-- Whenever there are some shortcomings of current markup, one eventually
will have to improve it, so that the stupid computers can get to know the
intentions of the human authors.