XML Community of Practice
Meeting Notes
August 17, 2005
The Department of Justice (DOJ) hosted this meeting. Following self-introductions by the participants, Owen Ambur thanked DOJ for hosting the meeting and introduced Jim Burch, Deputy Director of DOJ’s Bureau of Justice Assistance, for a few opening remarks.
Jim gave a brief overview of a number of DOJ initiatives, including privacy policy guidelines and “fusion center” guidelines on which they are working with the Department of Homeland Security (DHS). He also mentioned the Global XML Data Model (GJXDM), help desk support for the GJXDM, and collaboration on the National Information Exchange Model (NIEM), version 1.0 of which is scheduled for release in the fall.
He cited the National Sex Offender Public Registry as a real success story for XML, having been implemented under budget in 22 States in 58 days. Owen noted that although XML itself is a pretty simple concept, using angle brackets to demarcate data elements from each other, the experience of DOJ with the sex-offender registry is typical of the results that can be expected from applying XML in a well-disciplined manner across organizations.
With respect to the committee DOJ has formed under the Federal Advisory Committee Act (FACA) to support the Global Justice community, Jim noted that performance metrics have become an issue in OMB’s Performance Assessment Rating Tool (PART) process. While it is obvious that consultation with State and local stakeholders is important, FACA committee meetings are costly and it is difficult to measure the results and establish metrics demonstrating the outcomes in terms of a benefit/cost ratio. Jim invited anyone who has suggestions with respect to best practices in that regard to share them with the Global Justice community. Themes Jim cited that may prove to be pertinent include such things as the number of transactions facilitated and the preservation and effective sharing and use of well managed electronic records.
Finally, Jim noted that DOJ is always looking for partners in XML and information sharing efforts.
Following Jim’s remarks, Owen briefly displayed the most recent submissions to the ET.gov site at http://et.gov/component_search.aspx and called attention in particular to the following components:
● Adobe became the first major software vendor to register a component at ET.gov, having identified Portable Document Format (PDF) as a proposed specification for inclusion in the Federal Enterprise Architecture (FEA) Technical Reference Model (TRM). Although PDF is widely used by government agencies, it is not currently included in the TRM. Also, PDF-Archive (PDF-A) is nearing issuance as an international (ISO) standard for archival of records for long-term accessibility, and Steve Levenson of the U.S. Courts has expressed intent to register it as well.
● The Urban Institute is the first nonprofit organization to have registered an ET component – X-UDED, an XML vocabulary for grant budgeting and financial reporting by recipients of Federal, State and local government funds. The XML instance document is posted on the Web site hosted by the National Association of State Auditors, Comptrollers and Treasurers (NASACT).
[Editor’s note: One of the benefits of the XML-enabled architecture chosen for the ET.gov site/process is that submissions can be indexed by others as well. For example, X-UDED is also readily discoverable via Google at http://www.google.com/search?hl=en&q=x-uded. Citizen-centricity is a guiding principle of the President’s Management Agenda, and that means citizens should be able to use any tools they choose to interact with government.]
● At the request of the Department of the Interior’s Chief Architect, Owen has registered and will be aiming to build a community of practice (CoP) around an enterprise architecture (EA) metamodel containing the core elements that are common to the practice of EA among all agencies at all levels of government. http://www.doi.gov/ocio/architecture/eametamodel.xml Owen expressed the hope that such a metamodel will become a test use case for the FEA Data Reference Model (DRM). http://colab.cim3.net/file/work/drm/schema/Draft_FEA_DRM_Schema.htm
Owen also pointed out the potential for someone to identify Internet Protocol version 6 (IPv6) at ET.gov as a specification for inclusion in the FEA TRM and for vendors to use the site to identify IPv6-enabled components for discovery and potential use by government agencies and CoPs. Then Owen turned to Mark Crawford to brief the group on the status of the xmlCoP’s initiative to provide updated guidance on the naming and design of XML data elements and schema.
Mark noted that SAP has adopted ISO 15000 for implementation in NetWeaver, including the data types identified by UN/CEFACT. He also mentioned that the Department of the Navy (DON) naming and design rules identify 20 reusable data types. Mark’s presentation is available at http://xml.gov/presentations/lmi4/ndrg.htm
The key point is that the draft will include both guidance that agencies should apply as well as proposed rules that must be observed in order for a data element or schema to be approved as a Governmentwide component. Accordingly, the draft is now entitled XML Naming and Design Rules and Guidelines (NDRG). When the xmlCoP has achieved substantial consensus on the proposed rules and guidelines, Owen will forward the draft to the Governance Subcommittee of the CIO Council’s Architecture and Infrastructure Committee (AIC) for further consideration and perhaps for distribution and more formal review by agencies. Based upon recent progress in the NDRG working group, Owen hopes to be able to forward the draft to the Governance Subcommittee at the AIC’s October 20 meeting.
Another key point is that the NDRG working group has reached consensus that agencies should use URNs for namespace management but that they may use URLs instead, so long as they consistently conform to the specified structure. In that regard, Brand Niemann asked whether the proposed namespace guidance will address individual devices on the network, as provided by IPv6. Thus far, there has been no discussion in the work group in that regard, but if it makes sense to include such guidance and there is substantial consensus to do so, it could be included now or in a subsequent revision of the NDRG.
John Dodd asked about ownership and governance, and Owen noted that the role of the xmlCoP is to identify best practices and make recommendations but that the AIC Governance Subcommittee could engage in a more formal governance process and/or recommend that OMB issue policy that is binding on Federal agencies.
Following the mid-morning break, Paul Embley briefed the group on the status of the GJXDM XML naming and design rules (NDR). He noted they are not suggesting their NDR are applicable in any other CoPs but merely to foster understanding of the GJXDM. Since they were two days behind schedule, Paul was unable to get into the specifics of their NDR. They are following the UBL format but their NDRs are not a projection of or comparison with UBL. Nor is it a methodology for building exchange packages. Their draft will be posted for public review, including by the xmlCoP, and October 31 is the projected date for formal release of the GJXDM. Currently, the draft addresses 22 principles and they may add a few more. They are still debating the issue of versioning.
KC Morris asked how the GJXDM differ from the draft Governmentwide NDRG. Paul said the differences are not great and that they will highlight them.
Brand said ISO 11179 has been shown to lack sufficient semantics for machine processability and that related efforts are evolving toward RDF and OWL. He noted there are tradeoffs between XML and RDF. XML is easier to process but RDF is more scalable. He suggested scheduling a joint xmlCoP/SICoP meeting to consider the respective use cases for XML and RDF/XML.
Bob Greeves commented on the need and difficulty in keeping code tables synchronized, noting that the GJXDM contains nearly 100 code tables on such things as the automobile makers and models. He noted the need for authoritative sources and, thus, a policy on how to identify such sources by namespace.
Paul’s presentation is available at http://xml.gov/presentations/doj2/gjxdmndr.htm
Alice Marshall’s notes from this meeting are available at http://technoflak.blogspot.com/2005/09/federal-xml-work-group-august-17.html
Among those in physical attendance were:
Owen Ambur, xmlCoP
Patrick Ashamalla, OSD Studios
Tim Bornholtz, ED
[Deleted per attendee's request], BearingPoint
John Dodd, CSC & IAC
Paul Embley, PRG
Puja Goyal, NIST
Bob Green, DON C10
Robert Greeves, DOJ/OJP
Amin Hassam, i411
Holly Hyland, FSA
Steve Horn, FSA
Joab Jackson, GCN
Kim Koran, CSC
Chris Kupcyk, LMI
Lin Meng, DOT
Alice Marshall, Presto Vivace
Tom Merkle, NIJ
Michael Miri-Kavuma, VA
Roy Morgan, NIST
KC Morris, NIST
Vicky Niblett, SAIC/NASA
Brand Niemann, EPA
Chris Traver, DOJ/OJP
Allyson Ugarte, XBRL Spain & US
Among those who identified themselves as participating via teleconference were:
Rex Brooks, Starbourne Communications
Brenda Duvall, CSC
Tim Grapes, EDXL
Mark Kindl, GTRI
Marion Royal, GSA
Please convey any additions or corrections to Owen_Ambur@ios.doi.gov