From: Asmus Freytag (asmusf@ix.netcom.com)
Date: Sat Jul 31 2010 - 18:34:19 CDT
On 7/29/2010 1:15 AM, Khaled Hosny wrote:
> I don't buy in Unicode idea of
> encoding different sets of decimal digits separately, they are all
> different graphical presentations of the same thing.
>
>
Two observations:
1) During rendering, everything turns into a graphical representation.
The decision one has to make is where to keep the information
about what graphical representation to use. The choices are
a) document content (character encoding)
b) documen format (CSS etc.)
c) local environment at display time (fonts available,
locale settings, etc).
Now you can always argue which (combination) of these factors
works best.
2) For better or for worse, Unicode has made a decision, and
picked a) for representing numbers.
This choice allows the text content to carry the information,
and documents will normally not change based on how they
are viewed and on whether style information is present.
For some tasks, this may not be the best choice, but given the
importance of numbers for the text content, it's definitely a
defensible choice.
The most important bit here is that the choice has been
made and given the goal of an immutable encoding, that's what
you are stuck with.
In some cases, you can make your own choice - the display of
programatically generated numbers is one such case, because it
may not actually involve interchange of character codes, you are free
to use your own solution.
A./
This archive was generated by hypermail 2.1.5 : Sat Jul 31 2010 - 18:39:56 CDT