Buck Golemon wrote:
Latin1 explicitly gives no semantics to several byte values (for
example 0x81), but acknowleges that other standards will define their
semantics.
Unicode provides code-points with equally-undefined semantics so that
these bytes can pass through without change.
This allows a byte-level system using control codes in those ranges to
interact with a unicode-aware system, without loss of information.
Does that summarize well?
That should be good enough. It would be a poor process indeed that would
not convert the control characters 1-to-1, so that CR and LF would
become replacement characters or nulls or something.
--
Doug Ewell | Thornton, Colorado, USA
http://www.ewellic.org | @DougEwell
