RE: Question on OBI - pseudo EDI translator

[Rik Drummond <drummond@xxxxxxxxxx>]

I spent 1 hour scanning the document a week or so ago.... It has two focuses: 1) the process issues are very well described for open buying on internet, and 2) the technology.  I think the technology part needs a few of small adjustments: 1) they don't use MIME, they seem to define a new data structure, kind of object oriented, and they use pkcs7, but not s/mime.... to name a few.

They don't really seem to need a translator as Peter says. It is all Web based as far as I can tell.

I thought it was very well done by the way and is a major advancement on the process side. Again, I think it needs a little work on the technology side to align it with other internet efforts.


Later, Rik

-----Original Message-----
From:	Peter Rawlinson [SMTP:PeterR@xxxxxxxxxxxxx]
Sent:	Wednesday, July 02, 1997 12:20 PM
To:	'AJAY'; 'ietf-ediint@xxxxxxx'
Subject:	RE: Question on OBI - pseudo EDI translator

Ajay
>From my understanding, OBI is stating that a full-blown commercial EDI
translator is not required for the translation of Order Requests, and
that a simple mapping function would do, both at the supplier and the
buyer sides.

Pete

>-----Original Message-----
>From:	AJAY [SMTP:sanghi@xxxxxxxxxxxxxxxxxxxx]
>Sent:	Wednesday, July 02, 1997 1:23 PM
>To:	'ietf-ediint@xxxxxxx'
>Subject:	Question on OBI - pseudo EDI translator
>
>
>I read the OBI 1.0 specs (very professionally done, excellent reading
>material). On page 80 of Section 5.5 titled OBI Order Format Specification,
>it suggests that "implementation should not require the use of EDI
>translator" even though the order request and order data is in EDI format
>(X.12 850). It would be appropriate to provide an IC number to the way
>implementation or use of 850 is suggested (fully edi compliant IC).
>
>It seems to me that what is intended is a flat-file which immitates segment
>ids(BGM, REF etc) as record ids (instead of HDR or DTL records) which
>integrates with the business application directly. This in my opinion only
>adds to the complexity of the business app. In fact  the business app is now
>required to develop a mini edi translator. I agree that a "loaded" mapper is
>not required but it is probably wrong to say that edi translator is not
>required. In fact most implementations would prefer to have a mini-edi
>translator also in place.
>
>Please correct my understanding. 
>
>Thank you,
>
>Regards,
>
>@jay