W3C home > Mailing lists > Public > public-rif-comments@w3.org > December 2009

RIF PRD implementation report from Oracle

From: Gary Hallmark <gary.hallmark@oracle.com>
Date: Sun, 20 Dec 2009 13:11:31 -0800
Message-ID: <4B2E9303.9080501@oracle.com>
To: public-rif-comments@w3.org
# Your name, affiliation, and (optionally) the names of other people who 
helped with the implementation.

    Gary Hallmark, Oracle

# The name of your system, a URL for its website (if any), and a 
one-sentence description

    Oracle Business Rules
    http://www.oracle.com/technology/products/ias/business_rules/index.html
    Oracle Business Rules is a high performance lightweight business
    rules product that addresses the requirements for agility, business
    control, and transparency. It is part of the Fusion Middleware stack
    and integrates seamlessly across the entire Oracle SOA Suite and BPM
    Suite stack. It is also a core component for present and future
    Oracle Fusion Middleware and Fusion Applications products.


# Which dialects your software is designed to support (eg Core, BLD, PRD, 
or non-standard extension dialects). We would appreciate some brief 
commentary about why you chose these dialects, and what sorts of 
implementation techniques (eg algorithms) are being used.

    PRD (and of course Core). Business Rules products have traditionally
    been based on production rules systems (e.g. Blaze, ILog)

# Do you believe your system currently conforms to the RIF Candidate 
Recommendations? Does it pass all the test cases for your dialect(s)? If 
not, which features does it lack and/or which test cases does it not yet 
pass? Do you have plans to make it conformant, and make it pass all the 
test cases?

    PRD refraction is not implemented per the spec. The spec differs
    from several popular implementations, including Jess and CLIPS. The
    spec will probably need to be changed.

    Over half of DTB has been implemented, including numeric, list, and
    string builtins. Work continues on the remaining builtins. Only
    fully bound binding patterns have been implemented.

    Only Const terms are allowed as frame slot keys.

    General note: a number of unimplemented features (e.g. unbound
    variables as frame slot keys and as argument to list-contains) have
    no test cases.

# Does it implement any parts of RIF RDF and OWL Compatibility 
<http://www.w3.org/TR/rif-rdf-owl/>? Any issues?

    Not implemented.

# Did you implement the "at risk" features 
</2005/rules/wiki/Features_At_Risk>? If not, do you intend to, or do you 
think we should remove them from RIF?

    Indexing from 0 is implemented and is convenient because it allows
    mapping list builtins to java.util.List methods, which are also 0 based.

# We'd appreciate your evaluation of whether the RIF Candidate 
Recommendation is ready to proceed along the standards track toward 
being a W3C Recommendation. If not, please be sure to tell us what 
problems you think we need to address.

    PRD refraction is a problem because it seems to require rule engine
    changes and cannot be implemented in the translation layer of software.

# Which datatypes & builtins <http://www.w3.org/TR/rif-dtb> do you support?

    Over half of DTB has been implemented, including numeric, list, and
    string builtins. Work continues on the remaining builtins. Only
    fully bound binding patterns have been implemented.
Received on Sunday, 20 December 2009 21:12:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 20 December 2009 21:12:26 GMT