TO: Mark
My brief answer - at the header level, these are not duplicates. At the
document content level (855s) they might be. You may have to check what the
business and trading partner agreement has to say about this. Chances are,
it will say "it is blowing in the wind". Either way, this will have to be
resolved, else you have potential "ticking time bomb" in your hands.
Your question spawns other age old questions. In the practice, what drives
the definition of duplicates? Is it the header information (ISA/IEA, GS/GE,
ST/SE, TA1, UNA, UNB/UNZ, UNG/UNE, UNH/UNT, ICE/ICS, etc.) which are
artificially generated and checked by translators, parsers and mappers for
tracking, reconciling and matching documents/ messages. OR, is it the
application which could care less about control numbers, sender/receiver
ids, date and time stamps - and more often than not, oblivious to the
existence of a front-end/back-end translator/parser/mapper? OR, is it the
user, who can override both the translator and the application software?
CONSIDER THE FOLLOWING:
Assume the following baseline: I have commercially a available translator
brand A operational in my system; I want to send four (4) 855s (purchase
order acknowledgment numbers BAK03=AB001, AB002, AB003 and AB004,
respectively) to trading partner XYC; the trading partner profile for XYC
has the following baseline control number: ISA13/IEA02=5, GS06/GE02=5,
ST02/SE02=20; Version 004010 at the GS level.
1. I process the first 2 855s (AB001 and AB002) through the EDI translator
and transmit them to trading partner XYC. Assuming that nothing else is
using the translator, the header control numbers are now set to: ISA/IEA=6,
GS06/GE02-6 and ST02/SE02=22.
2. After step 1 is done, I went back to EDI translator, assuming I have the
privileges of course, I reset the header control numbers to
ISA13/IEA02=5, GS06/GE02=5, ST02/SE02=20. In reality, this is not
necessarily intentional - for instance, let us say, the system crashed and I
have to reinstall or recover from files prior to the 855 translation - which
is the equivalent of resetting the control numbers of the headers.
3. Then I process the next 2 855s (AB003 and AB004) through the EDI
translator and transmit them to trading partner XYC.
4. Guess, what happens? Do we have a duplicate or not? Somebody is not
going to be happy about this, right?
Lessons learned:
1. Duplicate determination should not be limited to header level
information.
2. TA1, date and time stamps are significant.
This is probably not the scenario that you have, Mark. But I hope it
illustrates some possibilities.
Thanks,
David Paraiso
http://www.edi-info-center.com
http://www.technology-info-center.com
> -----Original Message-----
> From: Mark Jasper [SMTP:[EMAIL PROTECTED]]
> Sent: Thursday, March 02, 2000 1:07 PM
> To: [EMAIL PROTECTED]
> Subject: Duplicate Transaction Identification
>
> List members, a question:
> If a trading partner sent you these two 997's, would you consider them to
> be
> duplicates?
> Note:
> The ISA segments are identical
> The GS segments have different receiver identifiers
> All Control Numbers are the same
>
> ISA*00* *00* *14*AAAAAAAAAAAAA *ZZ*BBBBBBBB
> *000218*0931*U*00302*000000001*0*P*:
> GS*FA*111111111111*BBBBBBBB*000218*0931*00001*X*003020
> ST*997*32376
> AK1*PR*59
> AK2*855*0114
> AK5*A
> AK2*855*0115
> AK5*A
> AK9*A*2*2*2
> SE*8*32376
> GE*1*00001
> IEA*1*000000001
> ------------------------------------------------------------------
> ISA*00* *00* *14*AAAAAAAAAAAAA *ZZ*BBBBBBBB
> *000218*0931*U*00302*000000001*0*P*:
> GS*FA*111111111111*XXXXXXXX*000218*0931*00001*X*003020
> ST*997*32376
> AK1*PR*59
> AK2*855*0114
> AK5*A
> AK2*855*0115
> AK5*A
> AK9*A*2*2*2
> SE*8*32376
> GE*1*00001
> IEA*1*000000001
>
> Thanks!
> Mark Jasper
> EC Programmer Analyst
> Memec, Inc.
> 9980 Huennekens Street
> San Diego, CA 92121
> Tel: 858-450-8919
> E-Mail:[EMAIL PROTECTED]
>
> =======================================================================
> To signoff the EDI-L list, mailto:[EMAIL PROTECTED]
> To subscribe, mailto:[EMAIL PROTECTED]
> To contact the list owner: mailto:[EMAIL PROTECTED]
> Archives at http://www.mail-archive.com/edi-l%40listserv.ucop.edu/
=======================================================================
To signoff the EDI-L list, mailto:[EMAIL PROTECTED]
To subscribe, mailto:[EMAIL PROTECTED]
To contact the list owner: mailto:[EMAIL PROTECTED]
Archives at http://www.mail-archive.com/edi-l%40listserv.ucop.edu/
