Re: comments from Clark and Haugen

[Prasad Yendluri <pyendluri@webmethods.com>]

Karsten,

Yes, RosettaNet transactions (Activities as they are called now) need not have any signals. Especially with the synchronous exchange scenarios, where the response comes back on the same communication protocol level connection as the request. PIP 2A9  is an example of this. Excerpt from 2A9 is listed below:

"The entity body of the HTTP response to the same connection that received the Query, shall be the EC technical information Response business action.  There shall be no intervening business signals between Query and Response. The HTTP status code shall be “200 OK” regardless of any PIATransaction status code.."

Regards, Prasad

--
Principal Architect, ATG; webMethods Inc.,
432 Lakeside Drive, Sunnyvale, CA 94088-3793, USA
Tel: (408) 962-5226 mailto: pyendluri@webmethods.com

-----Original Message-----
From: Karsten Riemer [mailto:Karsten.Riemer@east.sun.com]
Sent: Tuesday, January 02, 2001 11:16 AM
To: Karsten Riemer
Cc: ebxml-bp@lists.ebxml.org; ebxml-core@lists.ebxml.org
Subject: Re: comments from Clark and Haugen

Upon closer reading of Jim's and Bob's comments, I have a request for clarification. Jim states:
A transaction will have zero or one responding BusinessDocument, but always a
ReceiptAck and maybe also an AcceptanceAck

Is this really true? Must there always be a ReceiptAck? Why?
I was under the impression that a notification (which is also a business
transaction) would have no responding document and also no signals. And if a
transaction does have a responding document, then why require a ReceiptAck?

What is the RosettaNet implementation. Do they allow a transaction to not
have any signals?

-karsten

>Hi,
>for our meeting today, here is a set of comments received from Jim Clark and
>Bob Haugen. To view the comments, use M/S words menu option view->comments.
>I anticipate that we can agree to incorporate most of the comments and move on
>to submit the document to QR tomorrow as planned.
>Note that the document we will submit to QR will contain both the DTD sent out
>by Cory Casanave yesterday, as well as the set of interaction patterns
>contributed by Jim Clark. The version of the document I published yesterday
>had neither, only to keep the document short for transmission purposes.
>
>thanks,
>-karsten
>
>(Paul Levine will send out meeting notice, but it will be at 12 noon EST)