Re: the "possible" need to timestamp EDI transactions

["J. David MacDonald" <davemac@xxxxxxxxxx>]

My first contribution to this group.

I think what we're talking about here a method to independently verify that
an EDI document was created on a certain date.

This would be well suited to a time-stamping service.  For instance, I use
Verisign's time-stamping service along with authenticode technology to sign
and date the executables that I distrubute.  What it really means is that
when I timestamp the executable and distribute it, I no longer have the
option to go back, change something, back-date my computer and sign it again
because the timestamp is supplied by Verisign, not by me. Changing anything
would invalidate the timestamp.  This keeps me honest!

The same thing applies to an EDI transactions.  We want an independent
source to verify that the document was in a certain state at a specific
time.

I can think of other situations as well.  If I'm late filing my B3
accounting documents to Customs, I will be fined even if they're
electronically lost or delayed by a 3rd party such as a VAN or ISP.  This
situation might change is I can prove that the document was prepared on a
certain date and timestamped by an independent 3rd party.

Of course, this brings up the subject of a reliable time-stamping service
capable of handling EDIINT type messages. None come to mind.

I guess the long and short of it is:  We need a place in the EDIINT
structure that can accomodate independent timestamps supplied by 3rd
parties.

If any of this make sense, let me know. I live by the addage:
"If you keep your mouth closed, they might think you're stupid. If you open
it, they're gonna know you're stupid".   I might be treading thin ice here.

Dave MacDonald
david.macdonald@xxxxxxxxxxxx


-----Original Message-----
From: Rik Drummond <drummond@xxxxxxxxxx>
To: 'Werner, James K' <James.Werner@xxxxxxxxxxxxxx>; 'ietf-ediint@xxxxxxx'
<ietf-ediint@xxxxxxx>
Date: Wednesday, January 21, 1998 5:35 PM
Subject: RE: EDIINT WG Minutes fo the IETF Dec 98 Meeting


>Then get involved and make it happen. How do you see this happening?  What
would the process be and who would be involved?
>
>Regards, Rik
>
>-----Original Message-----
>From: Werner, James K [SMTP:James.Werner@xxxxxxxxxxxxxx]
>Sent: Wednesday, January 21, 1998 1:25 PM
>To: 'ietf-ediint@xxxxxxx'
>Subject: RE: EDIINT WG Minutes fo the IETF Dec 98 Meeting
>
>imho, the "possible" need to timestamp EDI "transactions" is a complete
>misstatement: timestamping is an *absolute* repeat *ABSOLUTE* (volume
>added for emphasis) requirment - no waffling, no qualifications, no
>'divide and conquer' strategies - timestamping is a must!.
>
>
>James K. Werner
>The Boeing Company
>PO Box 3707, #MS 6C-FH
>Seattle, Wa. 98124-2207
>phone: (425)237-6274  fax: (425)237-8510
>email: james.werner@xxxxxxxxxxxxxx
>
>