Thanks for your helpful reply.

As a fairly new employee here, I'm (carefully) lobbying for a change to our
utility for converting records from MARC to our proprietary format.
Because it relies on the field terminator alone, the program produced a
record in which all the fields following the errant FT were assigned the
wrong tags.  It seems we have been very lucky in not having run into this
kind of problem previously (or lucky that no one noticed if we did).
Nevertheless I seem to be meeting some resistance in my effort and need all
the support I can get.


At 01:27 PM 4/27/99 -0400, you wrote:
>Virginia M. Iandiorio said:
>> My questions:
>>         1. Is there a rule or best practice tenet that
>> declares which of the field
>> length or field terminator takes precedence in interpreting a
>> USMARC record?
>Any such discrepancy should be regarded as a severe error.  That said, our
>software gives precedence to the length of field and would diagnose the
>premature terminator as an invalid character.
>>         2. Shouldn't a processing program look at both field
>> definition indicators and report a discrepancy?
>>         3. What was the original intent in allowing two -
>> potentially conflicting - ways to define the length of a field?
>We can't know what the writers of the original specification were thinking,
>but one of the virtue of this record structure is its redundancy, which
>makes it possible to detect errors.  This was vital in the days of delicate
>and unreliable magnetic tapes and has regained its importance with the
>advent of electronic transfers, which not infrequently do strange things to
>the data being transmitted.
>Gary L. Smith
>Senior Consulting Systems Analyst
>Database & Offline Products Development
>[log in to unmask]
Virginia M. Iandiorio
Inforonics, Inc.
[log in to unmask]