On 24 Nov 2013, at 15:39, Steve Newcomb <> wrote:
> As a practical matter, is there *any* difference between text and binary
> data *other* than the necessity of worrying about record-end handling in
> the former, but not in the latter?
>
Reminds me of the problems we used to have in ICL because our mainframe binary files (in particular those used to hold object code) had significant record boundaries: they weren't just a sequence of bytes, but a sequence of records each holding a sequence of bytes. It became very difficult to transfer these to and from other (Unix and Windows based) systems that didn't recognise such a concept.
Michael Kay
Saxonica
