Meeting Report

National Committee for Information Technology Standards (NCITS)

L8, Data Representation, Special Interest Group

September 28 & 29, 1998 at NIST in Gaithersburg, MD

Attendees:

Bruce Bargmeyer Environmental Protection Agency

Jim Carpenter Bureau of Labor Statistics

Joe Christensen Australian Institute of Health and Welfare

Eliot Christian United States Geological Survey

Tom Culpepper 3M

Bruno Felluga Consiglio Nazionale Delle Riceriche, Italy

Larry Fitzwater Environmental Protection Agency

Sheila Frank Health Care Finance Administration

John Garrett NASA, Raytheon

Dan Gillman Bureau of the Census

Judy Griffin HAZMED

Beverly Hacker Environmental Protection Agency

Glenda Hayes MITRE

Henry Heffernan EDPNS

Pat Heinig Department of Energy

William H Kenworthey, Jr. Self

Judith Newton National Institute of Standards and Technology

Douglas Mann Battelle

Robert Mayes Health Care Finance Administration

Phong Ngo Science Applications International Corporation

Burton G. Parker Paladin Integration Engineering

Paolo Pelini Consiglio Nazionale Delle Riceriche, Italy

Jim Pipher Department of Defense/DISA

Marilyn Slater Department of Defense/Environmental Security

Dan Schneider Department of Justice

Genevieve Speier Health Care Finance Administration

Linda Spencer Environmental Protection Agency

Glenn M. Sperle Department of Defense/Health Affairs

Harry S. White, Jr. Retired

Dan Gillman and Judith Newton chaired the L8 SIG meetings.

We were pleased to welcome Harry White, who chaired X3L8 and SC 14 before Bill Kenworthey. Many of the new participants had not previously had a chance to meet Harry. We enjoyed meeting with him and gaining his insights on the past and present work of L8.

1. Discussion of New Work Items (NWIs)

All of the NWIs discussed below relate to metadata registries that are based on ISO/IEC 11179, Specification and Standardization of Data Elements. The NWIs extend the content of 11179; create connections to important technologies (such as Object technologies and XML); or relate to the operation and population of 11179 based metadata registries. The SIG agreed on leaders for the US work on each NWI. These leaders will get the US efforts underway for each NWI. We desire to engage the international members of WG 2 in the process of developing the NWIs and in selecting editors or co-editors for each NWI. We are looking for the best means to communicate between the US persons interested in each NWI and the international WG 2 members who are interested in each NWI. L8 members wish to engage international participation at the earliest possible stage in the exploration of the NWIs. There is still considerable flux in the content of each NWI with considerable regrouping of ideas--clumping and splitting topics to be included within the NWIs.

We discussed the NWI proposals, explored the potential content of each, identified next steps, and passed around a sign-up sheet for people to use to indicate their interest in participating on particular NWI groups. The results were added to the sign-ups from the last L8 SIG meeting and are shown below. There was general agreement that we are ready to engage the content of the NWIs, even while the discussions continue about what topics to include in each NWI. At the next L8 SIG meeting, each of the NWI groups will meet separately for one day (over a two-day period) and a final day will be used for inter-group coordination and discussion of general topics. Each NWI is described below, along with next steps. In general, the next steps involve development of a clear statement of what the NWI is about, recruitment of national and international participation, development of a draft of the proposed standard/technical report, and organization of a panel discussion for the WG 2 Open Forum on Metadata Registries to be held February 16-19, 1998 in Washington D.C. We also decided to set up informal email reflectors for each NWI group. We want to invite international WG 2 members and others to participate via the email reflectors.

1.1 Revise ISO/IEC 11179, Part 3 (five year revision) with amendments to other parts of 11179 to make all parts consistent [Joe Christensen editor]

This is an approved project, but is included with the NWI discussion, because the work is just starting. The project is to update ISO/IEC 11179, Specification and Standardization of Data Elements, Part 3, Basic Attributes. This work is the result of the 5-year review of Part 3. Related to this would be an effort to develop corrigenda or changes in the other parts of ISO/IEC 11179 to make the separately developed parts consistent between themselves and to make them consistent with the revised Part 3. The recently completed ANS X3.285 incorporates virtually all of 11179, Part 3, extends the list of attributes, and presents the information in the form of a model. It includes an informative annex with translations to several popular modeling notations. X3.285 is to serve as a basis for rapidly updating Part 3. Doug Mann will serve as the lead for the US work on this project. Joe Christensen, in Australia, is the WG 2 editor for this project.

Joe Christensen participated via teleconference from Australia. He is beginning the editing work. Joe will prepare a draft and circulate it informally to all SC 32/WG 2 participants for comment. When there is general agreement, he will forward the draft to the SC 32 Secretariat for circulation on a CD ballot. Joe is using ANSI X3.285 as a base document and is recasting it in ISO format. We discussed some changes that are desirable. We agreed that UML notation is preferable to extended Chen (Entity-Relation) notation for the normative expression of the metamodel. UML has greater acceptance in the target IT community. Therefore, Joe will move the UML model from the informative annex to the normative portion of the standard and the extended Chen (ER) model will be moved to an informative annex. We also agreed that we could postpone coordination with editors of other parts of 11179--to make all parts consistent—until the work on Part 3 nears completion. Dan Gillman has received some comments relevant to this work. He will forward the comments to Joe.

Next Steps include recruitment of national and international participation, development of a draft document, and organization of a panel for the Open forum.

1.2. Object extensions to ISO/IEC 11179, Specification and Standardization of Data Elements

Tom Culpepper described work underway on the new work item for Object Extensions to 11179. Tom is collecting "use cases" -- user statements about the types of queries desired, the types of information desired, and other requirements that describe the browser capabilities and run time requirements desired for an object interface. Several people suggested material that they can contribute. Everyone is invited to send suggestions for "use cases" to Tom. In addition to the metadata registry interface capabilities included in this effort, we discussed the need for extending the metamodel to cover object behaviors for the individual data elements and groups of data elements that are documented in 11179 metadata registries. We do not yet have a good handle on what is needed to accomplish this.

Tom also reported on the September 1998 OMG meeting held in Seattle. Tom discussed 11179 topics with the OMG members. L8 has requested Tom to act as liaison between L8 and OMG. To effect this, Tom picked up an OMG template to use as the basis for a formal document that will establish the liaison relationship. The next OMG meeting is scheduled for Burlingame, CA, starting Nov. 12, 1998.

Tom Culpepper is the US lead for this NWI. The next steps include continued development of the written description for this NWI, recruitment of national and international participation, development of a draft document, and organization of a panel for the Open forum.

XML extensions for ISO/IEC 11179

This involves development of a set of XML metadata tags based on XML-Data and/or XML/Resource Description Framework (RDF). This would make ISO/IEC 11179 useful in the new XML enabled web technologies. One topic in this effort will be to enable metadata interchange among metadata registries. Another topic is to enable access to metadata registries from XML based web pages, including pages generated from database queries. Another goal is to enable XML applications (such as XML-EDI) to directly access metadata registries. In order to accomplish this, requirements will be generated for XML syntax and schema work underway in the World Wide Web Consortium. We are working out arrangements for liaison with the relevant W3C workgroups. Frank Olken and John McCarthy are leading the US work on this NWI.

Next Steps include continued development of the written description for this NWI, recruitment of national and international participation, development of a draft document, and organization of a panel for the Open forum.

Extend ISO/IEC 11179 for Complex data

This NWI project examines the information technology characteristics of data that are derived, aggregated, composed, or grouped from other data or that have complex internal structure. Currently, 11179 and X3.285 do not adequately address these more complex types of data. However, much data is generated by means of calculation, aggregation, composition, grouping, or other derivation methods. In addition, new complex datatypes in multimedia and advanced scientific applications are increasingly used. This project will take an incremental step toward a developing the metamodel constructs needed to describe these more complex types of data.

Dan Gillman talked with Mike Zeug, Chair of NCITS L3, Coding of Audio, Picture, Multimedia and Hypermedia Information. L3 is looking at metadata for multimedia data. Mike is interested in liaison between L8 work and multimedia data. Dan also described the Data Documentation Initiative, led by the University of Michigan, ICPSR (Inter-university Consortium of Political and Social Research). ICPSR is coming up with recommendations for documentation of statistical data. Dan discussed our work on derived data with the ICPSR group and it sounds like the group is satisfied with our approach.

Pat Heinig described work done by DOE on the Comprehensive Epidemiological Data Resource (CEDR) system. It may be useful in helping us work out how to handle complex types of data. John McCarthy of LBL is a good contact for following up on CEDR. See www.cedr.gov. It has a template for recording information about data files. We also discussed the scope of this new work item. The types of data range from individual data elements to various groupings and aggregations of data elements. We discussed the need for "pedigree" or "provenance" information. (We might use the word provenance to describe certain descriptive information. Data provenance could be defined as "the fact of coming from some particular source, origin or derivation. The history or pedigree of data or datasets; a record of the ultimate derivation and passage of data through its various owners." –based on the Oxford English Dictionary definition for "provenance".)

Dan Gillman will write a document defining the issues to be included in this new work item, given the discussion at this meeting. Next Steps include continued development of the written description for this NWI, recruitment of national and international participation, development of a draft document, and organization of a panel for the Open forum.

1.5. ISO/IEC 11179 Metamodel issues Not Elsewhere Classified

Doug Mann described the remaining issues that he found in the list of open issues remaining from the work on X3.285. Most of the issues are already included in other proposed New Work Items. We discussed the issues and decided to group them into two areas. The first area covers levels of abstraction (generic, core, reference, abstract -- vs. application data elements). The second area is about units of measure and dimensionality. Judith Newton will prepare a document describing the issues to be addressed under levels of abstraction. A leader is needed for the measure and dimensionality work.

Technical Report on ISO/IEC 11179 Metadata Registry Content

The exchange of metadata between ISO/IEC 11179 metadata registries depends not only on registry software that conforms to the standard, but also on metadata contents that are compatible between registries. While the 11179 standard has general coverage for data element specification and registration, there are pragmatic issues pertaining to populating the registries with content. Based on the experiences of organizations that are implementing the standard, a technical report to explore content issues could help current and future users. This could also possibly be an implementation guide. Larry Fitzwater is the leader for the US work on this NWI.

1.7. Terminology extensions for ISO/IEC 11179

This work will focus on extending the 11179 to better accommodate semantic management of concepts and the terms by which concepts are referenced. An important foundation for semantics management is the ability to specify concepts and to associate the each concept with the linguistic expressions (term or unintelligent identifiers) used to name or reference the concept. Concepts may be expressed as definitions and each linguistic expression (term, identifier) for the each concept should be in a context. For example, the context for a term may be a specific language (French, English, …) or a specific scientific discipline (chemistry, materials engineering, …). Once concepts and terms are specified, they may be organized a variety of structured sets such as controlled vocabularies, keywords, data element values, data element components, thesauri, themes, topic trees, taxonomies, or ontologies. Each of these structured sets of concepts/terminology can be deployed in various technologies. For example, topic trees and thesauri can be deployed in search engines to facilitate discovery of relevant documents and data, ontologies can be deployed in intelligent information services (request brokers, query agents, resource agents, etc.), and data elements can be deployed in DBMS technology. Extensions to 11179 are required to establish a means for specifying concepts and terms and to establish a means for organizing concepts/terminology into structured sets. Reference implementations will be used to demonstrate the practical utility of the proposed standards and technology. Bruce Bargmeyer is currently leading the US work on this NWI.

Bruno Felluga and Paolo Pelini discussed the General European Environmental Thesaurus (GEMET) and the EUROVOC (multilingual terms in use by the EU Parliament), two multilingual terminology systems developed in Europe. The European efforts have addressed many of the terminological issues that we will encounter in developing terminology content. We expect to work closely with them on many topics.

1.8. NWI sign-up sheets

Sign-up sheets were passed around. These will be added to the sign-ups from the last meeting. Participants could sign up for more than one topic and could give a preference. Preferences are shown in parentheses ( ), below The results will be used for scheduling the NWI breakout sessions for the next L8 SIG meeting. The results are below:

1.8.1 Revise ISO/IEC 11179, Part 3 (five-year revision) with amendments to other parts of 11179 to make all parts consistent

Joe Christensen, editor

Douglas Mann

Burt Parker

1.8.2 Object extensions to ISO/IEC 11179, Specification and Standardization of Data Elements

Tom Culpepper (US lead)

Eliot Christian

Jim Carpenter (3)

Bruce Bargmeyer

Robert Mayes

Bill Kenworthey

Phong Ngo (1)

Sheila Frank

Marilyn Slater (1)

Glenda Hayes (3)

Jim Pipher

Genevieve Speier (2)

Glenn Sperle (1)

Elizabeth Fong

Tom Rhodes

1.8.3 XML extensions for ISO/IEC 11179

Frank Olken, John McCarthy (US co-lead)

Bruce Bargmeyer

Eliot Christian

Glenda Hayes (1)

Jim Pipher

Tom Culpepper

Genevieve Speier (3)

Glenn Sperle (2)

Robert Mayes

Elizabeth Fong

Tom Rhodes

Len Gallagher

1.8.4. Extend ISO/IEC 11179 for Complex data

Dan Gillman (US leader)

John Garrett

Bill Kenworthey

Larry Fitzwater

Burt Parker

Marilyn Slater (2)

Glenda Hayes (2)

Jim Carpenter

1.8.5. ISO/IEC 11179 Metamodel issues Not Elsewhere Classified

Burt Parker

Marilyn Slater (3)

Glenda Hayes (4)

Judith Newton (US lead on levels of abstraction)

Jim Carpenter (2)

Phong Ngo (3)

Glenn Sperle (3)

Bill Kenworthey

Sheila Frank

1.8.6. Technical Report on ISO/IEC 11179 Metadata Registry Content

Larry Fitzwater (US lead)

Dan Gillman (2)

Bob Mayes

Douglas Mann

Pat Heinig

Phong Ngo

Elizabeth Fong

1.8.7 Terminology extensions for ISO/IEC 11179

Bruce Bargmeyer (US lead)

Jim Pipher

Sheila Frank

Dan Gillman (3)

Tom Culpepper/Harold Solbrig

Beverly Hacker

Robert Mayes

Bill Kenworthey

Len Gallagher

L8/T2 merger and combined meeting

Doug Mann circulated a document describing how these committees might be combined. It was reviewed at the last SIG meeting. We had hoped to have a joint meeting in October, but the schedule works out better for November. It is now proposed to have a joint L8 – T2 meeting on November 18 – 20, 1998. A Plenary would be held on November 18, and SIG breakout meetings would be held on the 19^th and 20^th. A letter ballot will follow this meeting.

Status of draft standards

3.1 ISO/IEC 11179 Part 1

Dan Gillman was given the comments from the SC 32/WG 2 meeting in Brisbane. He has completed the revisions and is sending a copy to the SC 32 Secretariat for circulation as a Final CD.

3.2 ISO/IEC PDTR 15452 - Specification of data value domains

Judith Newton reported that she made the changes as agreed at the SC 32/WG 2 editing meeting. The new draft was sent to the SC 32 Secretariat for ballot.

Review of JTC 1 N 5491, JTC 1 N 5493, and JTC 1 N 5496

Phong Ngo distributed ballots on these documents via email. For those who had not responded, he brought paper copies for vote and signature.

4. Review of ISO/IEC 15046 Geographic Information – Part 15 Metadata

Doug Mann reported that he worked with the L8 review group formed in the August meeting and with FGDC members to formulate comments for the ballot on this CD. Doug helped them to align the standard with 11179 and suggested various improvements.

5. Comments on IEEE P1489

Doug Mann reported that, as requested at the last L8 SIG meeting, he developed comments on IEEE P1489, Draft Standard for Data Dictionaries for Intelligent Transportation Systems. He sent the comments to the appropriate IEEE contact.

6. Open Forum on Metadata Registries

Bruce Bargmeyer reported that he sent out a draft call for participation in late August. He has received some comments. There was general agreement on the content of the Call. It was agreed that day 3 and 4 should be switched, with the NWI presentations on day 3 and the related presentations on day 4. Both of these days can have concurrent sessions, if needed to accommodate the number of panels and speakers. We are trying to complete the list of presentations and speakers. International input is needed. We have received some international input, but will work to get broader involvement. Bruce requested people leading panels or giving presentations to send him brief descriptions of their topics. Also, panel leaders are requested to determine whether their panel will last 2 or 4 hours. These are due October 6, 1998. Bruce will then revise the call for participation and send it out for another round of comments. The WG 2 informal reflector is not yet working and this is urgently needed to facilitate communication on this and other topics. We very much want to engage the international community on planning, organizing, leading and participating in the Open Forum and the New Work Items. WG 2 members, including L8 members, are invited to send nominations for speakers to panel leaders and/or to Bruce Bargmeyer and Eliot Christian.

We expect that all speakers for the Forum will be invited (no public call for presentations will be issued), but that written presentations can be accepted from any source and made available (or linked to) the Open Forum web site. Eliot Christian said that he would organize the review of documents sent by contributors. Papers not relevant to the Open Forum will be rejected.

7. Implementation reports

7.1. Environmental Data Registry

A new version is about to be released with capability to register EDI message sets, download individual data element value domains, register multiple definitions (as needed for the Global Information Locator Service data elements), and with other enhancements. Development work on the new version is complete, but deployment will take until late October or mid-November. Levels of abstraction are becoming important to the content of the EDR. Most of the EDR content is abstract (core, generic) data elements. Now more "application" data elements are being registered, including EDI data elements. Therefore EPA has considerable interest in how to relate the application data elements to the abstract data elements? There is also a question about how many layers of abstraction there should be?

A major effort is now underway to work on terminology. EPA is building a prototype terminology reference system as a module of the EDR.

7.2. Census

A number of coordinated activities going on. Have a prototype behind the firewall. Have encountered some restrictions on the ability to publish metadata for public accessibility. Are registering application data elements; haven't worked with abstract data elements. This is because the focus is on sharing existing application-level data. Are working on getting the metadata from the sources. Some documentation is word-of-mouth and may be separated in various organizational components. Entering into a contract to have a contractor build an architecture and a production system, using the structure that Dan has developed. The IT area will manage the architecture and implementation. Also are organizing a group that will help to manage the content.

7.3. 11179 Metadata Registry Consortium

A meeting was held in August to begin discussions between various organizations involved in implementing metadata registries based on 11179. There was a striking commonality in the problem space described by the participants. A common approach based on 11179 appears to be very useful. This can help to demonstrate the potential market to vendors, who could make commercial implementations of 11179. This consortium group can interact with L8 to identify and work on the technical issues that need to be established as standards. The next meeting will be held at NIST on October 8, 1998. All interested are invited.

7.4. Intelligent Transportation Systems Data Registry

ITS 1489 Part 1 standard was out for public comment. The balloted standard is for the ITS functional-ITS data dictionaries not the ITS Data Registry. An ITS Data Registry Configuration Control Committee and a Board of Directors committee have been formed and have met. The comment period is closed. DOT has funded the development of a registry design, which will be completed shortly. A development effort is expected to soon follow.

7.5. Health Care

This is a joint effort by agencies involved in implementing the Health Care Insurance Portability and Accountability Act (HIPAA). A prototype registry has been implemented and is being loaded with Health care EDI messages. It is also being loaded with data from National Committee on Vital and Health Statistics. Use cases have been developed. Demonstrations will begin soon.

7.6. Health Care cont.

The ANSI Health Informatics Standards Board has a proposal before it to implement and administer a 11179-based registry for the health care sector. This proposal will be voted on at the ANSI HISB December meeting.

7.7. US Government CIO Council

Pat Heinig raised the possibility of including 11179/X3.285 into the architecture work underway for the CIO council and for work mandated by the Clinger-Cohen act. Are there compliance levels that could be established for 11179 and X3.285? Currently 11179 has two areas where compliance is specified: Mandatory/optional attributes and standardization level. The first relates to the compliance of the implementation and the second refers to the level of compliance of the content. Additional levels might include import/export, service interfaces, etc. It would be useful to work on additional areas where compliance can be specified. This discussion led to general agreement that we should include compliance criteria in the relevant new work items, however, the criteria may be at a level of granularity such that implementers can select particular capabilities to implement.

8. TC 215 liaison report

Bob Mayes reported on TC 215. Bob made a presentation at a preliminary meeting to introduce ISO 11179. There was interest in the Terminology Workgroup in perhaps incorporating metadata activity into their work.

9. HL7

Sheila Frank gave a presentation on the DoD/HCFA metadata registry project at the HL7 quarterly meting in San Diego. There was a great deal of interest and the National Council of Prescription Drug Programs (retail pharmacy EDI standards group) requested a presentation on metadata registries at their national technical conference in November.

10. Future meeting schedule

October:

10/21/1998 9 AM - 5 PM NIST North Rooms 145, 152 and 501 – NWI topic meetings (meet first in 152)

10/22/1998 9 AM – 5 PM NIST North Room 145, 152 and 501– NWI topic meetings

10/23/1998 9 AM - 5 PM NIST North Room 152 – L8 full group SIG meeting (Room 145 available for breakout)

10/21 - 23/1998 10/23/1998 Dining Room - A

A notice will be sent out in advance of the meeting to specify which NWI topic will meet in which room during day one and day two.

November:

11/18/1998 9 AM – 5 PM NIST Administration Building Lecture Room D – Joint L8 Plenary with T2

11/19/1998 9 AM – 5 PM NIST Administration Building Lecture Room C, B111, B113 & Employee Lounge– NWI & other topic meetings

11/20/1998 9 AM – 5 PM NIST Administration Building Lecture Room C & D, B111, & B113 – NWI & other topic meetings

11/19/1998 11/20/1998 Dining Room A & C

11/20/1998 11/20/1998 Dining Room A

December:

12/10/1998 9 AM - 5 PM NIST Administration Building Lecture Room E, Rooms B111 & B113 (LR E is located in the

basement of Administration Building) – SIG meetings. Meet first in Lecture Room E.

12/11/1998 9 AM - 5 PM NIST Administration Building Lecture Room E, Rooms B111 & B113 12/10/1998

12/10 - 11/1998 Dining Room C

Directions to NIST Administration Building: From Interstate 270, driving North, take Exit 10. The exit takes you under the freeway. Take a left into the NIST campus at the first stoplight. The Administration Building is the largest building on the campus. Visitor parking is in the large parking lot in front of the Admin Building. Other parking lots nearby are reserved for permit holders.

Directions to NIST North: NIST North is Building 820. The address of this building is 820 West Diamond Avenue. Note that the building is located across the street from the main campus of NIST. Instead of turning onto Bureau Drive toward the main Gate of NIST, turn the opposite direction (North) onto Bureau Drive and take the second possible right turn to get to the NIST North building, The building has the name "NIST" on it in large letters. (From Interstate 270, driving North, take Exit 10, which loops under the freeway, and then turn RIGHT at the first stop light--this is the intersection of West Diamond Avenue and Bureau drive.)

-------------------

X3L8 documents, including this meeting report, are available via Anonymous FTP from sdct-sunsrv1.ncsl.nist.gov. Change directory to X3L8 (X3L8 in lowercase letters) and check the "readme" file for details. Many of the documents are also listed on the X3L8 Home Page. The URL is http://www.lbl.gov/~olken/X3L8". If you have questions or comments on this document send email to bargmeyer.bruce@epamail.epa.gov.