OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ubl-dev message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [ubl-dev] UUID


My feeling on this is that
1) as a technical feature a UUID concept suggests uniqueness even
between different copies of a document
BUT
2) as a business semantics feature it is required that the same UUID
be used for separate copies of the same document
As such the UUID in UBL is a semantic feature and is there for
business and legal purposes in the legal documents such as the Invoice

However I'm a little concerned that the suggested use for ID/IDREF
overlaps a little with the requirement for a technical feature so if
there is a temptation to place technical (1) requirements over semantics
(2) then this temptation should be resisted and something designed for
the technical added instead (e.g. actual XML ID/IDREF).

As an example to help demonstrate the importance of preserving the
semantic business function consider an invoice number: One wouldn't
expect a copy invoice to have a different invoice number to the original.

As an example of the separate nature (separate 'layer' nature) of the
technical requirements consider that a copy in the case of UBL is
different from the original anyway by virtue of its having CopyIndicator
set to true. Thus it is tempting to give such a document a different
UUID if only the technical matter is considered and if the UUID BBIE
is used for that there is obviously a breach of the semantic requirements.

All the best

Stephen Green




Quoting Martin Forsberg <martin.forsberg@amnis.se>:

> Hi Juha,
>
> When the xml-file is generated the UUID is created, but the   
> transmission and re-transmissions might happen in a later step. It   
> is still the same xml-file instance that is re-transmitted. The UUID  
>  is changed first when the document issuer is re-creating the xml   
> instance. This means that it might be up to the implementor and the   
> issuing system if the UUID is the same or changed when transmitting.
>
> In a scenario where conversions (between different document   
> standards) are involved the UUID might be lost and re created on the  
>  path to the receiver. This is also something to consider when using  
>  UUID on a larger scale. (i.e. UBL -> EDIFACT -> UBL).
>
> One other thing to consider is the definition of a document copy   
> (using the copy indicator). If the UUID is changed when the issuer   
> issues a copy, is the copy then really a copy? (the UUID is not   
> copied)
>
> Regards,
> Martin Forsberg
> Single Face To Industry, Sweden
>
>
> -----Original Message-----
> From: David RR Webber (XML) [mailto:david@drrw.info]
> Sent: den 2 september 2006 18:14
> To: Juha Ik&auml,valko
> Cc: UBL-Dev
> Subject: RE: [ubl-dev] UUID
>
> Juha,
>
> If you are using ebXML messaging and reliable messaging then the   
> delivery is controlled to the ebXML messaging specification (ebMS).
>
> Retrying will occur to the level of tries specified in the CPA   
> between partners.
>
> Failure recovery is then determined by the BPSS work flow.
>
> If a subsequent re-send is permitted - then yes - it will be an   
> entirely new exchange - but the business information of course can   
> be the same.
>
> It is the job and purpose of the ebMS to avoid duplicate messages -   
> and again that is also part of the specification.
>
> Tools like Hermes V1 from http://www.freebxml.org implement a   
> significant portion of this reliable interchange system.
>
> UUID algorithm is an open public specification and computes a unique  
>  32 byte string based on date/time and IP location.
>
> DW
>
>
>  -------- Original Message --------
> Subject: [ubl-dev] UUID
> From: Juha Ikävalko <juha.ikavalko@tieke.fi>
> Date: Tue, August 22, 2006 8:29 am
> To: "UBL-Dev" <ubl-dev@lists.oasis-open.org>
>
> Hi Developers,
>
>
>
> I got a question what is the purpose of UUID in UBL documents?
> Definition says "A computer-generated universally unique identifier
> (UUID) for the Order instance". My conclusion: ID identifies the   
> order on the business level and UUID identifies an order message   
> instance on the technical level. Thus, if a seller is not able to   
> receive the order message at the first attempt, a buyer sends a new   
> order message with same ID but different UUID. Is this the right   
> conclusion?
>
>
>
> Regards,
>
>
>
> Juha Ikävalko
>
> TIEKE Tietoyhteiskunnan kehittämiskeskus ry TIEKE Finnish   
> Information Society Development Centre Salomonkatu 17 A, 10th floor   
> FI-00100 Helsinki Tel +358 9 4763 0410 Fax +358 9 4763 0399   
> juha.ikavalko@tieke.fi http://www.tieke.fi <http://www.tieke.fi>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: ubl-dev-unsubscribe@lists.oasis-open.org
> For additional commands, e-mail: ubl-dev-help@lists.oasis-open.org
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: ubl-dev-unsubscribe@lists.oasis-open.org
> For additional commands, e-mail: ubl-dev-help@lists.oasis-open.org
>
>




[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]