 ACORD ACORD’s Experiences using W3C Schemas Dan Vint Senior Architect

ACORD’s Experiences with W3C Schemas SLIDE 2 10/18/02 Background  ACORD  Not-for-profit Insurance industries standards body, we have standards for Life, P&C and RLC lines. Used in US, Canada, England/London Market, and Europe. Growing interest in Japan/China  Three standards started in 3 different ways:  P&C – based upon an earlier EDI standard. Originally designed for DTD use, but added schema support when it was available (currently using both)  Life – started as a MS initiative then brought into ACORD. DTD based originally, now only using schema  RLC – started in a European standards group, also based upon an EDI standard. Using DTDs currently trying to add schema support  Business Issue: How to bring the 3 standards into one design  For ACORD the best thing would be a common model, design and naming conventions  Can’t get there because we are on different cycles. Life and RLC cannot break backwards compatibility, P&C is currently redesigning  Each as a level of adoption and entrenchment.

ACORD’s Experiences with W3C Schemas SLIDE 3 10/18/02 Things that we like:  Having the schema format be an XML document has been very useful  We generate schemas and DTDs from a meta-format that is maintained in a database  We slice the schema into smaller subsets via XSLT  Support for basic data types was crucial for our needs.  Having a built-in documentation standard is very nice.  Support for namespaces and import/include/redefine is useful, but rough around the edges.  Can we get this sort of support reversed engineered into XML and DTDs?

ACORD’s Experiences with W3C Schemas SLIDE 4 10/18/02 Things we don’t like about W3C Schema  Took a simple 30 page standard and blew it out of proportion.  XML took a simplifying approach, schema reversed that and put in experimental features and complexity  Fallout:  You need tools to manage a schema  No two tools agree on their implementations  Incomplete implementations  Should have left out xsd:all, DTDs did and Schema implementation is not generally usable anyway.  I can teach DTD syntax in an hour or less; it takes a day for schema  Entities requiring the use of DTDs.  “Yuch! DTDs, that’s old technology and a kludge” Surely if the schema guys had wanted entities to be used they would have defined a new method (or they wouled have at least documented this use.

ACORD’s Experiences with W3C Schemas SLIDE 5 10/18/02 Things we don’t like about W3C Schema (cont)  No way to clearly specify the Schema  Hint attributes!  Confused world of support and no support for the hints, inconsistent on what should happen when they appear  No way to version schema that is supported by tools  xsd:version attribute has no use or support  Namespaces only way to force recognition of a change  Namespace designation of extension – want a different namespace rather than the original ACORD. (namespace pollution)  Extensions, did content really have to be added outside the original choice or sequence?  Extensibility of enumerated lists  Probably 1/3 of the elements in ACORD are based upon code lists, and we can’t extended them.  Need a way to identify the extended code values. We have our tpe based upon QName, but this doesn’t work for external lists.

ACORD’s Experiences with W3C Schemas SLIDE 6 10/18/02 Things we don’t like about W3C Schema (cont)  No easy way to both validate or ignore extensions  We need extensions, but not all trading partners will know what the extensions are, so they need to at least be able to validate the basic ACORD information and structure  We have two designs within our standard  One uses global elements and groups with restriction and extension  Another uses an element of that is of type xsd:any and is added at the end of our aggregates  How about a processContents="strict“ or processContents=“lax"  We could use co-constraints and/or “business rules”  We publish a huge schema – 700 messages, 29,000 lines need help in modularizing to make the smallest schema possible  Business level validation – just because a date is the wrong format, should I reject a $50 million transaction?  Don’t like substituionGroups because it is not easy to determine the final content model  I like the old SGML notion of “no surprises”

ACORD’s Experiences with W3C Schemas SLIDE 7 10/18/02 Where we need help  Need an integrated approach that supports all tools  Code generators my biggest headache currently – don’t support all features, don’t expect to and have decided what subset of XML they think is appropriate  Groups  Redefine  Attributes on data level elements  Too many levels of inheritance in classes – create complex code  ‘abstract’  Possible solution – profiles  XML DTD level support plus data types (import, include, namespaces)  DTD+ with base types (abstracts, redefine)  Schema extreme (key, unique, substitution groups)  Should the W3C consider specifying a language that uses XML as a data type? Scripting or compiled  XML should become a native data type  Consider the adoption of a catalog approach to schema location, general processing, and versioning  One standard has a wide content model for messages and uses a type code to designate the type of message being sent  Everything is optional in the core design of elements  We need a very rich method for managing co-constraints based upon several hundred type codes  Would really like to somehow get auto-magic schema validation of these real message requirements  We want to standardize on the resulting data stream when it comes to extensions  Need a way to prevent the use of features we don’t want to support.

ACORD’s Experiences with W3C Schemas SLIDE 8 10/18/02 Where we need help (cont)  Reuse of other schemas  Good intentions, but impossible to do unless designed at the same time  Design principals and modeling techniques  Based upon different datatypes – with/without id attributes  I want to share part C of an outside schema, outside schema has a definition for Party and it is used in C, I also have a Party definition and I need to substitute it in part C to make it useful  No easy way to take a part of the other schema unless it is in a subset file  All I want is codelist A  All I want is the Party definition  Need to promote a standard way to define codelists and encourage standards producers to publish them in XML based upon that standard. See previous bullet.  Want extensions to be identified against the extender, not the originator  Propose a new attribute "hideBase" to manage “namespace pollution”

ACORD’s Experiences with W3C Schemas SLIDE 9 10/18/02 Why ask why?  Why do xml special attributes require definition in schema, but not DTD?  xml:space requires and include file, but xmlns is automagically there!  How will the xml:id attribute be implemented?  Why is a simpleType not always simple?  Why does adding an attribute make a simpleType become complex with a simple content?