From mario at epcc.ed.ac.uk Thu Apr 16 06:54:36 2009 From: mario at epcc.ed.ac.uk (Mario Antonioletti) Date: Thu, 16 Apr 2009 12:54:36 +0100 (BST) Subject: [BYTEIO-WG] ByteIO interop Message-ID: Hi, We have been thinking about (finally) doing interop in the DAIS WG and I was looking through the ByteIO interop document and near the beginning there is a statement: ByteIO implementations that wish to participate in interoperability tests MUST use the following binding: o The Port Type MUST be bound to SOAP 1.1 o The binding MUST use the "document/literal" style. In DAIS we have already hit the document/literal vs rpc/literal SOAP binding. I was wondering whether there was a rationale for having adopted document/literal rather than rpc/literal for the binding? or was this a quick and easy decision to get the interop testing underway? Please let me/us know. Thanks in advance, Mario +-----------------------------------------------------------------------+ |Mario Antonioletti:EPCC,JCMB,The King's Buildings,Edinburgh EH9 3JZ. | |Tel:0131 650 5141|mario at epcc.ed.ac.uk|http://www.epcc.ed.ac.uk/~mario/ | +-----------------------------------------------------------------------+ -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From mmm2a at virginia.edu Thu Apr 16 07:40:14 2009 From: mmm2a at virginia.edu (Mark Morgan) Date: Thu, 16 Apr 2009 08:40:14 -0400 Subject: [BYTEIO-WG] ByteIO interop In-Reply-To: References: Message-ID: <1239885614.6583.27.camel@irtahk> My understanding, which may be flawed, or out of date, or both, is that this was the defacto standard adopted by OGF, and if not by OGF, then certainly by the OGSA WSRF-BP; the rendering used by the Byte IO interop. -Mark On Thu, 2009-04-16 at 12:54 +0100, Mario Antonioletti wrote: > Hi, > We have been thinking about (finally) doing interop in the DAIS WG > and I was looking through the ByteIO interop document and near the > beginning there is a statement: > > ByteIO implementations that wish to participate in interoperability > tests MUST use the following binding: > o The Port Type MUST be bound to SOAP 1.1 > o The binding MUST use the "document/literal" style. > > In DAIS we have already hit the document/literal vs rpc/literal SOAP > binding. I was wondering whether there was a rationale for having > adopted document/literal rather than rpc/literal for the binding? or > was this a quick and easy decision to get the interop testing > underway? > > Please let me/us know. Thanks in advance, > > Mario > > +-----------------------------------------------------------------------+ > |Mario Antonioletti:EPCC,JCMB,The King's Buildings,Edinburgh EH9 3JZ. | > |Tel:0131 650 5141|mario at epcc.ed.ac.uk|http://www.epcc.ed.ac.uk/~mario/ | > +-----------------------------------------------------------------------+ From mario at epcc.ed.ac.uk Thu Apr 16 08:52:43 2009 From: mario at epcc.ed.ac.uk (Mario Antonioletti) Date: Thu, 16 Apr 2009 14:52:43 +0100 (BST) Subject: [BYTEIO-WG] ByteIO interop In-Reply-To: <1239885614.6583.27.camel@irtahk> References: <1239885614.6583.27.camel@irtahk> Message-ID: Hi, > My understanding, which may be flawed, or out of date, or both, is that > this was the defacto standard adopted by OGF, and if not by OGF, then > certainly by the OGSA WSRF-BP; the rendering used by the Byte IO > interop. I had a brief scan through the WSRF BP 1.0: http://www.ogf.org/documents/GFD.72.pdf but could not find any preference specified for the document/literal or rpc/literal styles. If OGF is recommending one style over the other it would be good to know what the reference for this is. Thanks Mark, Mario +-----------------------------------------------------------------------+ |Mario Antonioletti:EPCC,JCMB,The King's Buildings,Edinburgh EH9 3JZ. | |Tel:0131 650 5141|mario at epcc.ed.ac.uk|http://www.epcc.ed.ac.uk/~mario/ | +-----------------------------------------------------------------------+ -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From mmm2a at virginia.edu Thu Apr 16 09:42:34 2009 From: mmm2a at virginia.edu (Mark Morgan) Date: Thu, 16 Apr 2009 10:42:34 -0400 Subject: [BYTEIO-WG] ByteIO interop In-Reply-To: References: <1239885614.6583.27.camel@irtahk> Message-ID: <1239892954.6583.45.camel@irtahk> I can no longer find any such text either. Maybe I'm simply remembering incorrectly. Without such a document, I can't say for sure why that requirement was put into the interop document. Could be simply to simplify the interop process -- after all, the point wasn't to test the interoperability of SOAP, but only ByteIO which is an orthogonal concern. -Mark On Thu, 2009-04-16 at 14:52 +0100, Mario Antonioletti wrote: > Hi, > > > My understanding, which may be flawed, or out of date, or both, is that > > this was the defacto standard adopted by OGF, and if not by OGF, then > > certainly by the OGSA WSRF-BP; the rendering used by the Byte IO > > interop. > > I had a brief scan through the WSRF BP 1.0: > > http://www.ogf.org/documents/GFD.72.pdf > > but could not find any preference specified for the document/literal > or rpc/literal styles. If OGF is recommending one style over the other > it would be good to know what the reference for this is. > > Thanks Mark, > > Mario > > +-----------------------------------------------------------------------+ > |Mario Antonioletti:EPCC,JCMB,The King's Buildings,Edinburgh EH9 3JZ. | > |Tel:0131 650 5141|mario at epcc.ed.ac.uk|http://www.epcc.ed.ac.uk/~mario/ | > +-----------------------------------------------------------------------+ From mario at epcc.ed.ac.uk Thu Apr 16 10:57:55 2009 From: mario at epcc.ed.ac.uk (Mario Antonioletti) Date: Thu, 16 Apr 2009 16:57:55 +0100 (BST) Subject: [BYTEIO-WG] [DAIS-WG] ByteIO interop In-Reply-To: <49E74A04.9090300@fz-juelich.de> References: <1239885614.6583.27.camel@irtahk> <49E74A04.9090300@fz-juelich.de> Message-ID: Hi Bernd, > the WSRF spec does not mention the type of binding at > all, except for the (non normative) "Application notes" document: > > which recommends document/literal (section 8.1). Thank you. This is useful. > Most WSRF implementations use doc/lit anyway (an exception > being the Perl WSRF::lite, but this is very outdated) > I'm surprised that the OGF WSRF BP does not mention this, as it > is crucial for interoperability. I assume they just took doc/lit > for granted. Best to be explicit about these things else if one starts assuming you could end up in lots of places.... Mario +-----------------------------------------------------------------------+ |Mario Antonioletti:EPCC,JCMB,The King's Buildings,Edinburgh EH9 3JZ. | |Tel:0131 650 5141|mario at epcc.ed.ac.uk|http://www.epcc.ed.ac.uk/~mario/ | +-----------------------------------------------------------------------+ -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.