To the question
> > 3. What was the original intent in allowing two -
> > potentially conflicting - ways to define the length of a field?
Gary Smith writes:
> We can't know what the writers of the original specification were thinking,
We *can* know it. I have it from one of the veterans in the field, who
used to work at LC in software development, that the reason was (as I stated
yesterday) that the assembly language they used at the time had a convenient
command "skip to the end-of-field mark" (code 30). That's why they put that
code in, redundant though it is and anxious though one was at that time
of avoiding reduncancy.
There was the command "skip to the end-of-record mark" as well, that's why
they put in that code 29 at the end of MARC records too. This one is useful,
of course, for if the first five bytes indicating the record length are
disrupted, you can still find the beginning of the next record.
Universitaetsbibliothek, Postf. 3329,
D-38023 Braunschweig, Germany
Tel. +49 531 391-5026 , -5011 , FAX -5836
e-mail [log in to unmask]