wsrp-pfb message

Subject: RE: [For ebXML] Re: [wsrp-pfb] [UDDI#3] use of UDDI's wsdl address tMod el

From: Andre Kramer <andre.kramer@eu.citrix.com>
To: 'Chiusano Joseph' <chiusano_joseph@bah.com>, wsrp-pfb@lists.oasis-open.org
Date: Fri, 23 Jan 2004 09:02:01 -0000

Title: RE: [For ebXML] Re: [wsrp-pfb] [UDDI#3] use of UDDI's wsdl address tMod el

For WSRP producers, our approach for UDDI has been to make the wsdl authoritative and to delay factoring out the ports/bindings contained in the wsdl until later. I think it is fair to say that we are waiting for the best practice to shake out a bit. The service wsdl always has a default SOAP http(s) binding that all producers must support but may contain other protocol bindings.

So publishing a service wsdl as the soap:address location (WSDL soap binding) seems the minimum required but you want to support at least the default SOAP binding directly and leave it open as to which is to be used for other (unspecified by us) binding types. I would say that the WSDL should always be published (even if the SOAP binding details are directly published).

For portlets, one needs a reference to a producer service (e.g. by a UDDI service key) and the handle. Both of these would be "accessURIs"?

regards,
Andre

-----Original Message-----
From: Chiusano Joseph [mailto:chiusano_joseph@bah.com]
Sent: 22 January 2004 23:48
To: Richard Jacob
Cc: Andre Kramer; wsrp-pfb@lists.oasis-open.org
Subject: [For ebXML] Re: [wsrp-pfb] [UDDI#3] use of UDDI's wsdl address
tMod el

In light of this discussion, I'd like to also clarify regarding ebXML
please:

Our Service Information Model optionally specifies the Web Services
access point in the "ServiceBinding" content model (see "accessURI"
attribute):

Should we defer to the WSDL SOAP binding to provide the access point in
all cases? (the "location" attribute of the "soap:address" element), and
therefore never specify the "accessURI" attribute? In cases where both
are specified, our TN on "Registering Web Services in an ebXML Registry"
leaves it as an implementation issue as to which address takes
precedence (accessURI or WSDL SOAP binding) when they are different.

Thanks,
Joe

Richard Jacob wrote:
>
> However, we may decide to not to support the v2 technote right from the
> start.
> But on the other hand it specifies exactly what we want - indirection and
> the wsdl being the authoritative source.
>
> Mit freundlichen Gruessen / best regards,
>
>         Richard Jacob
> ______________________________________________________
> IBM Lab Boeblingen, Germany
> Dept.8288, WebSphere Portal Server Development
> Phone: ++49 7031 16-3469 - Fax: ++49 7031 16-4888
> Email: mailto:richard.jacob@de.ibm.com
>
> |---------+---------------------------->
> |         |           Richard          |
> |         |           Jacob/Germany/IBM|
> |         |           @IBMDE           |
> |         |                            |
> |         |           01/19/2004 01:11 |
> |         |           PM               |
> |---------+---------------------------->
>   >--------------------------------------------------------------------------------------------------------------------------------------------------|

> | |

> | To: Andre Kramer <andre.kramer@eu.citrix.com> |

> | cc: wsrp-pfb@lists.oasis-open.org |

> | Subject: RE: [wsrp-pfb] [UDDI#3] use of UDDI's wsdl address tModel |

> >--------------------------------------------------------------------------------------------------------------------------------------------------|

>
> exactly this would mean a 3rd reference here, sorry I miscounted the
> tModels :-)
>
> The Bindings tModel inherits wsdl-ness (I like that :-) ) but this only
> expresses that the technical fingerprint is described in wsdl.
> However usually the access point is the real endpoint.
> While in our case the access point is the wsdl holding the endpoints.
> This indirection is expressed by the WSDL Address tModel in the UDDI
> technote v2.
> Could you also verify this
> (
> http://www.oasis-open.org/committees/uddi-spec/doc/tn/uddi-spec-tc-tn-wsdl-v200-20031104.pdf
> )
>
> ?
>
> Mit freundlichen Gruessen / best regards,
>
>         Richard Jacob
> ______________________________________________________
> IBM Lab Boeblingen, Germany
> Dept.8288, WebSphere Portal Server Development
> Phone: ++49 7031 16-3469 - Fax: ++49 7031 16-4888
> Email: mailto:richard.jacob@de.ibm.com
>
> |---------+---------------------------->
> |         |           Andre Kramer     |
> |         |           <andre.kramer@eu.|
> |         |           citrix.com>      |
> |         |                            |
> |         |           01/19/2004 11:31 |
> |         |           AM               |
> |---------+---------------------------->
>   >
> --------------------------------------------------------------------------------------------------------------------------------------------------|

>
>   |
> |
>   |       To:       wsrp-pfb@lists.oasis-open.org
> |
>   |       cc:
> |
>   |       Subject: RE: [wsrp-pfb] [UDDI#3] use of UDDI's wsdl address
> tModel
> |
>   >
> --------------------------------------------------------------------------------------------------------------------------------------------------|

>
> I had forgotten to ask about this. I noted that the wsrp_v1_Bindings tModel
> does "inherit wsdl-ness" by use of the keyedReference "uddi-og:types" /
> "wsdlSpec" (section 5.2.3).
>
> But do we not already have 2 tModels: WSRP_Producer and WSRP_v1_Bindings,
> so you are adding a 3rd WSDL Address tModel?
>
> regards,
> Andre
>
> -----Original Message-----
> From: Richard Jacob [mailto:richard.jacob@de.ibm.com]
> Sent: 19 January 2004 09:51
> To: wsrp-pfb@lists.oasis-open.org
> Subject: [wsrp-pfb] [UDDI#3] use of UDDI's wsdl address tModel
>
> The WSDL in UDDI V2 technote contains a section discussing WSDL as the
> authoritative source for port access points in Appendix A.
> It mandates to reference a tModel indicating that the access point is the
> WSDL URL and that port endpoints must be retrieved from the WSDL.
> Here is an excerpt of Appendix A:
>
> "If a WSDL Implementation Document is being used then the bindingTemplate
> MUST contain
> sufficient information to identify the port address in the WSDL
> Implementation Document.
> The mapping described here MUST be used instead of the mapping defined in
> section 2.4.5.
> In all cases where a WSDL Implementation Document is used, the URLType
> attribute of the
> accessPoint corresponding to each port MUST be "other", and the value of
> the accessPoint
> MUST be the URL of the WSDL Implementation Document.
> The bindingTemplate MUST contain a tModelInstanceInfo element with a
> tModelKey of the
> WSDL Address tModel. This tModelInstanceInfo element, in combination with
> the protocol
> and transport information from the binding tModel, provides the necessary
> information to
> locate and interpret the endpoint address."
>
> This would mean that our Producer's bindingTemplate (currently we have only
>
> one) would reference that tModel in addition.
>
> Mit freundlichen Gruessen / best regards,
>
> Richard Jacob
> ______________________________________________________
> IBM Lab Boeblingen, Germany
> Dept.8288, WebSphere Portal Server Development
> Phone: ++49 7031 16-3469 - Fax: ++49 7031 16-4888
> Email: mailto:richard.jacob@de.ibm.com
>
> To unsubscribe from this mailing list (and be removed from the roster of
> the OASIS TC), go to
> http://www.oasis-open.org/apps/org/workgroup/wsrp-pfb/members/leave_workgroup.php
>
> .
>
> To unsubscribe from this mailing list (and be removed from the roster of
> the OASIS TC), go to
> http://www.oasis-open.org/apps/org/workgroup/wsrp-pfb/members/leave_workgroup.php
> .
>
> To unsubscribe from this mailing list (and be removed from the roster of the OASIS TC), go to http://www.oasis-open.org/apps/org/workgroup/wsrp-pfb/members/leave_workgroup.php.