Next in thread →
Next in month →
Re: [xml-dev] UBL
On Mar 17, 2006, at 3:55 PM, Chiusano Joseph wrote:
I know that the current effort for the US federal government's NDR is influenced by the UBL NDR, as well as others. However, it is certainly not "based up" the UBL NDR.
"influenced by" or "based upon" -- this might be poor word selection on my part. However, the documents I've read make it appear that the intent is to hew as closely to the UBL NDR as possible.
Not sure what you mean by "ebXML schema", as ebXML has never included an XML vocabulary (i.e. for defining a message payload) as part of its framework. Some of the various parts of the ebXML framework themselves (e.g. ebXML Registry) have XML schemas, but there is no reason that they should be UBL NDR compliant.
When I say ebXML schema, I mean schema that are defined in the framework (not the ebXML payloads.) I'm not suggesting that the framework ought to follow the UBL NDR. However, there is a belief (not my own) that if you are going to use ebXML technology, you should use the UBL NDR to define your payloads. This is one of the reasons given in our organization for adopting the UBL NDR.
On Mar 17, 2006, at 3:55 PM, Chiusano Joseph wrote:
A few clarifications:"The US Federal government, e.g., has come up with their own NDR based upon the UBL NDR."I know that the current effort for the US federal government's NDR is influenced by the UBL NDR, as well as others. However, it is certainly not "based up" the UBL NDR."since the ebXML schema are not themselves UBL NDR compliant"Not sure what you mean by "ebXML schema", as ebXML has never included an XML vocabulary (i.e. for defining a message payload) as part of its framework. Some of the various parts of the ebXML framework themselves (e.g. ebXML Registry) have XML schemas, but there is no reason that they should be UBL NDR compliant.JoeJoseph ChiusanoAssociateBooz Allen Hamilton700 13th St. NW, Suite 1100Washington, DC 20005O: 202-508-6514C: 202-251-0731Visit us online@ http://www.boozallen.com
From: Dave Scott [mailto:[email protected]]
Sent: Friday, March 17, 2006 4:48 PM
To: [email protected]
Subject: Re: [xml-dev] UBLOn Mar 17, 2006, at 8:56 AM, G. Ken Holman wrote:I hope this helps.This does help -- particularly the references to DSDL. If the schema our workgroup develops must strictly follow the rules laid out in the Naming and Design Rules document, those of us who feel we really need these attributes can draft a document detailing how to layer them on top of the official schema using the techniques described there.I'm not sure what you mean by a "custom" attribute.
I meant a "user defined attribute". Rule ATD1 in the spec saysUser defined attributes SHOULD NOT be used. When used, user defined attributes MUST only convey CCT:SupplementaryComponent information.This effectively prohibits any attribute definitions from an NDR compliant schema except for CCTS metadata.I'm confronted with this issue on this particular work group. In our organization, there is a follow the leader attitude regarding the UBL NDR. But our organization is not alone in taking the NDR as a guiding document. The US Federal government, e.g., has come up with their own NDR based upon the UBL NDR. As XML practitioners, I think we're likely to find the UBL NDR a very influential document outside of UBL itself. (As the paradigm implementation of CCTS, it's having a big influence in the ebXML world too -- which is ironic, since the ebXML schema are not themselves UBL NDR compliant.) I guess that is why I feel this is a question for the XML community at large not just the UBL community.Dave
On Mar 17, 2006, at 8:56 AM, G. Ken Holman wrote:
At 2006-03-16 21:46 -0600, Dave Scott wrote:I was curious about opinions and experiences people on the list mighthave regarding the UBL NDR:- Am I alone in finding it too be overly draconian?The "Garden of Eden" approach to UBL declarations have allowed me to mechanically analyze the schema expressions using XSLT that I think would have otherwise not been possible (thankfully, I didn't have to try to do it otherwise).- What practical limitations (particularly processing limitations)have you encountered when implementing the standard?How often is processing impacted by schema constraint expressions?Perhaps my biggest problem with the NDR is the prohibition againstcustom attributes.I'm not sure what you mean by a "custom" attribute. If you mean attributes from a foreign namespace, then a combination of ISO/IEC 19757-3 Schematron (to test the presence of the attribute) with ISO/IEC 19757-4 NVDL (to test the validity of the attribute and the validity of UBL without the attribute), perhaps even (optionally) in combination with ISO/IEC 19757-2 RELAX-NG (to express the constraints of the attribute) would seem to me to do the trick.All without needing any blessing or accommodation by the UBL committee since this is all totally orthogonal to the UBL schema expressions.I'm contributing to a workgroup to develop a datastandard which will be used to transmit potentially huge data sets.Understandably, we'd like the standard to be friendly to streamprocessing. The ability to add "type" and "role" attributes in keyplaces in the schema would greatly aid processing the data in a waythat would only require examining the current element stack duringstream processing for our most typical use cases.Sounds good.The UBL proponentson the workgroup propose requiring such items to be a first child(rather than attribute) of the node they qualify so as to act as aprocessing flag for its siblings and siblings' children.Yuck ... I would far prefer flagging such as attributes. Of course if the information you need to include is structured you would have to use elements.Also, I believe you shouldn't addordering constraints to your schema, burdening both the productionand validation of the data set, unless there is a semanticmotivation.Indeed.A couple of key individuals on the workgroup are adamant aboutadhering to the UBL NDR (although not very articulate when it comesto justifying either the individual requirements of the NDR or theblanket decision to follow it slavishly.)Fine ... you can (and I believe *should*) leave the UBL schema expressions created by the NBR untouched and sacrosanct. Layer your requirements on top of a read-only schema expression using ISO/IEC 19757 DSDL http://dsdl.org parts.Unfortunately, theliterature I've been able to turn up regarding the UBL NDR all readslike marketing literature rather than frank evaluations of the trade- offs the NDR involves. (My best source has been the UBL mail archivesbut it's fragmented and skimpy on this topic.) Can anyone point meto some more thoughtful evaluations of the NDR? Or, perhaps,implementations of CCTS that allow custom attributes? Or, betteryet, UBL endorsed mechanisms for integrating custom attributes withotherwise UBL NDR compliant schema?The propositions to use DSDL layered on top of UBL schema expressions are only now being considered. No decisions have been made and no recommendations are written down. You will find a discussion of supplementing UBL schema expressions in:As tools mature and experience using these evolving standards builds, I firmly believe we will be seeing wide-spread acceptance of these techniques.I hope this helps.. . . . . . . . . . . Ken--World-wide on-site corporate, govt. & user group XML/XSL training.G. Ken Holman mailto:[email protected]Crane Softwrights Ltd. http://www.CraneSoftwrights.com/x/Box 266, Kars, Ontario CANADA K0A-2E0 +1(613)489-0999 (F:-0995)Male Cancer Awareness Aug'05 http://www.CraneSoftwrights.com/x/bcLegal business disclaimers: http://www.CraneSoftwrights.com/legal-----------------------------------------------------------------The xml-dev list is sponsored by XML.org <http://www.xml.org>, aninitiative of OASIS <http://www.oasis-open.org>The list archives are at http://lists.xml.org/archives/xml-dev/To subscribe or unsubscribe from this list use the subscriptionmanager: <http://www.oasis-open.org/mlmanage/index.php>
Next in thread →
Next in month →