Federal CIO Council XML Working Group
Meeting Minutes, January 16, 2002
Logistics Management Institute, DAU 3/4
Please send all comments or corrections to these minutes to Jessica Glace.
Co-chair Mr. Owen Ambur opened the meeting. He announced Co-chair Mr. Marion Royal was at a speaking engagement in Florida. There are Developer’s Guides in hard copy in the back of the room and they were also distributed to the listserv.
Attendees introduced themselves.
Mr. Ben Dean from the Social Security Administration (SSA) thanked the Federal CIO Council XML Working Group for inviting SSA to speak and opened the presentation.
What are SSA’s challenges from now through 2010?
Mr. Dean planned to quickly review the business plan and concentrate on the more technical aspects of their project.
There has been an explosion of technology requiring new electronic interfaces and the SSA is working to use these new technologies. Some examples of the new technology are Web, wireless, and video.
The Public wants faster and more convenient access to information. SSA coverage is virtually everyone in the USA. New mission requirements have expanded SSA responsibilities. The baby boom retirements will dramatically expand SSA workloads.
SSA has a diverse staff dealing with people of different social and cultural backgrounds. SSA expects to lose up to 38,000 of its skilled, experienced staff via attrition (retirements) and will face a knowledge management challenge. Therefore, policy and procedural instruction systems will become even more important.
SSA uses private insurance companies as a performance benchmark. SSA has measurable performance goals and current high levels of performance. Our challenge is thus to take an excellent Agency and make it better. We feel SSA can achieve its' service delivery goals through established principles and enablers of service delivery driving well-executed, detailed, published plans with achievement of measurable goals throughout the journey.
SSA Service Delivery Principles are similar to those of other government service organizations. They are: consumer choice, first point of contact, privacy, one-stop government service, proactive service and stewardship.
SSA Service Delivery Enablers are: technology enhancements, access to electronic records, operational flexibility, external alliances, public communications and internal working relationships.
Around 1993, the policy process was taking too long to move from enacted legislation or court cases to policy implementations such as regulations and instructions. SSA has improved in this area, but is still working on further improvements.
The major body of information for SSA is the Program Operations Manual System (POMS) that all of our field units use. It guides us in making decisions on all retirement, disability and SSI claims. We have the largest group of administrative law judges and supporting staff attorneys in the world located in offices nationwide. They handle appeals of any of our claims decisions. With a clear communication of consistent, timely policies and procedures, we increase the quality of our decisions and lessen the need for appeals.
PolicyNet is the full life cycle of a policy document.
Some PolicyNet challenges are assuring the right policies and procedures are available to the right people, at the right time, in the right format, at the right cost, and in a secure manner.
Some initial social security retirement and post-entitlement transactions are now available via the Internet. We have 80+ screens of self-help for disability on the Web. This language needs to be kept current and consistent with the language in our instructional manual (POMS).
We are trying to make sure guidance is available immediately.
Also, SSA Front-line staff want policies and procedures available in a context-related manner (e.g., help screens related to the exact point in a transaction where they need help) and at a very granular level so that they don’t have to manually wade through many pages of policy background to find answers.
Islands of automation evolved from a wide range of users/employees putting systems in place throughout the country to take care of unmet needs from our legacy enterprise systems. This information now needs to be linked and integrated. Security is always an important part of this work.
SSA is moving to a new, Web-centric computing environment. Whatever SSA implements must be extensible. We want an XML-based architecture that is extensible so it can be flexible for new developments in W3C.
SSA will be driven to use XML wherever it fits. XML is the tool, but the bottom line is SSA wants to produce, and use effectively, good, timely and consistent policies and procedures. This now means writing for the public as well as internal SSA and State DDS staffs. The need for public self-help through filing claims via the Internet is primarily due to the increase in staff workload caused by the increasing needs of baby boomers.
Mr. Terry Hynes took over the presentation.
Mr. Hynes spent 18 years in field offices dealing with the Public on a daily basis. He has first hand experience with the workload as well as the support staff and public.
Being able to access SSA information in a timely manner is a self-confidence builder to the SSA staff and also to the public. Many interviews will be more smooth resulting in a higher confidence level by the public.
PolicyNet is the full life cycle of a policy document. PolicyNet is comprised of an Oracle database and authoring system and a reference area. The reference area is the source users access to find the policy instructions. The other piece of PolicyNet is, QuickPlace for collaboration and review of writing or rewriting instructions.
SSA’s main program policy and procedure document is the POMS, the Program Operations Manual System, consisting of 35,000 pages. It was moved to CDRom in 1991, but is still being produced in paper and is now moving to SSA’s intranet and to the Internet.
Mr. Hynes reviewed the workflow diagram.
1989/1990 SSA implemented a toll-free telephone line and developed a Teleservice Center Operating Guide for those employees who answer the phones. This has become a heavily used document on PolicyNet, supporting the use of the SSA intranet for quick access to program policy. SSA is anticipating heavier use of the Intranet within the Agency.
Now information is available to everyone without many barriers since the information is available on the Internet. For the first time the public has easy access to the agency’s program policy instructions.
Because of the amount of information the agency produces and how dispersed the information is in SSA; it is easy for information to become less standardized, or out of date. But now having one database where information is stored in a central location eases management and control of the information. Mr. Hynes feels this will be the case throughout the government. It will lead to everyone using the same information, which is most current.
The production process for CDRom has a lead-time of up to 6 weeks; whereas the Web publishing lead-time is 24 hours for incorporation into the full body of any document. POMS can be updated real time using the Web. We issue one e-mail message each day to users so they can directly link to all new information for the day. .
Internal users are gaining more confidence in the central repository of the agency’s program policy instructions than they had when they had to keep it updated via paper transmittals. This transfers to the public. Everybody in every office is using the same information. This is a huge improvement in the burdensome paper process, eliminating redundancy. The information is produced one time, in one place, and everybody can link to that information. Everyone is looking at the same thing.
SSA goes much further than Social Security. There is also SSI and a multitude of other programs. This makes it difficult to keep up with changes. PolicyNet is over coming that burden.
The public can also access this information on-line rather than coming into an SSA office, which reduces office or even phone interview time, freeing employees for case processing. With online instructions we have been able to build a FAQ area so internal staff doesn’t continue to research and answer the same questions. We hope to have XML link much of this information together to provide better, quicker processing of information making our data more accurate leading to more confidence in our agency by the public we serve.
Mr. Michael Leff from Lockheed Martin took over presentation.
SSA elected to use the “Unified Process”. The Unified Process is an iterative process that focuses on mitigating risk throughout the lifecycle. Mr. Leff assured Mr. Ambur that in light of previous email exchanges with Mr. Ambur concerning modeling efforts undertaken by the Bureau of Indian Affairs, the approach being taken was not an attempt to model the entire enterprise.
Mr. Ambur stated that the Bureau of Indian affairs has modeled their business processes. He also said that organizations should try to understand the entire process while also being practical because business needs to function as improvements are put in place.
Mr. Leff continued, saying they used the Universal Modeling Language (UML), which graphically depicts artifacts of a software-intensive system. They modeled the business process for understanding and to summarize system functionality resulting in Activity Diagrams and Use-Case Diagrams. The mapped SSA “as-is” process highlights the fact that there is data stored in many different places and is hard to link. XML was identified as a way to help move toward consolidated data and to improve the ability to link between disparate sources.
An XML Repository is part of the “to-be” structure. Because SSA has a content-centered model, they are looking at XLink, XPointer and XPath to provide linking capabilities between separate bodies of content.
The W3C XML Linking effort was started in 1996. It provides enhanced linking functionality utilizing an XML vocabulary (XLink) and an addressing scheme (XPointer). The objective of the XML Linking Working Group is to design advanced, scalable, and maintainable hyperlinking and addressing functionality for XML.
XLink W3C Recommendation (27 June 2001) defines the linking language which allows elements to be inserted into XML documents and to create and describe associations between documents.
XPointer W3C Candidate Recommendation (11 September 2001) is the definition of a language to be used to identify a resource based on XPath and can return any user selection in a document.
XPath W3C Recommendation (16 November 1999) and Version 2 Working Draft (20 December 2001) is used to address and select specific groups of nodes and attributes in an XML document.
The reason SSA is interested in XPointer is because XPointer provides a finer level of granularity, but the current XPointer protocol can retrieve only a small piece of the document. It also doesn’t guarantee that the retrieved XML is still well-formed and valid.
XLink has two types:
· simple link elements, which are similar to HTML and
· extended link elements, which allow multiple links and have sub-elements:
o locator, which provides locations to participate in an extended link,
o arc, which defines navigable connections between locations resources,
o title, which provides a human readable sub-element, and
o resource, which defines participants in a link.
An XLink href supplies the Uniform Resource Identifiers (URI) to allow the XLink application to locate a remote resource.
XLink attributes can be split into three categories:
1. Semantic, which contains:
· role, which provides a URI reference that describes the intended property of the XLink,
· arcrole, which provides a URI reference that describes the intended property of the arc, and
· title, which describes the meaning of the link in human readable terms
2. Behavioral, which contains:
· show, which describes the desired presentation of the data and possible values are new, replace, embed, other, or none, and
· actuate, which describes the desired timing of the link and possible values are onLoad, onRequest, other, or none; and
3. Traversal, which contains:
· label, which provides an identifier that can be used to identify how different types of resources and locators are connected within a link,
· from, which identifies the origin, using a label value from one of the locator or resource elements of an extended link, and
· to, which identifies the destination using a label value from one of the locator or resource elements of an extended link.
HTML linking has some limitations such as:
XLink is extremely flexible:
XLink separates associating resources by using a link from traversal amongst resources by using arcs.
Mr. Tim Marr concluded the presentation with a case study.
Currently there are two disparate repositories in SSA. There is a policy repository that contains program instructions for processing claims maintained in XML and claim processing training material maintained in Microsoft Word (being converted to XML).
As policy was changed, it was hard to get training materials updated quickly. SSA has requested bi-directional linkages between the policy and the training material.
From the research conducted, Mozilla and Amaya were the only browsers that supported XLink. Internet Explorer (SSA's standard browser) didn’t support XLink, therefore, the idea of doing client-side linking within SSA was not an option.
In addition, if the training material is linked to policy using in-line links, the underlying documentation cannot be modified.
An example was presented that showed a section of:
unstyled training material in XML that used XPath, XPointer and XLink to link to policy data that was also in XML.
Then the training material was shown before the XLink was resolved and after the policy data was pulled into the training document. Finally, the training manual with the resolved XLink was shown with an applied Style Sheet.
It was warned that it is important as you create XLinks that your XSL stylesheet be aware of the structure of the data pointed to by the inserted links in order to display or transform the information appropriately.
Lockheed Martin wrote a limited XLink processor for practicality, but there are COTS products available that are beginning to implement the XLink specification. If you’re doing a major implementation, it would be prudent to look into one of the COTS products.
There are still many generalities in the specifications that need to be resolved. For example, the specification does not specify how to handle multiple links within a document that have the attributes of actuate=onRequest and show=replace.
Below is a chart identifying various XLink processors including the one written by Lockheed Martin and their compliance with the prescribed W3C specification features.
|
Name |
simple
links |
extended
links traversal |
third-party
links |
show= |
show= |
show= |
actuate= |
linkbases |
|
XLip
(Fujitsu) |
Yes |
yes |
yes |
yes |
yes |
yes |
incomplete |
yes
(with onLoad) |
|
X2X
(Empolis) |
N/A |
N/A |
transform |
N/A |
N/A |
transform |
N/A |
partial |
|
Linkit |
|
|
|
|
|
|
|
|
|
Mozilla
(and Netscape 6) |
Yes |
no |
no |
yes |
yes |
no |
incomplete |
no |
|
Amaya
(W3C) |
Yes |
no |
no |
no |
no |
no |
no |
no |
|
XLink2HTML |
Yes |
yes |
? |
? |
? |
? |
? |
no |
|
4XLink
(FourThoughtLLC/Python) |
Partial |
partial |
? |
? |
? |
yes |
yes |
no |
|
LM
XLink Processor |
Yes |
no |
no |
yes |
no |
yes |
yes |
no |
SSA is moving toward a J2EE platform.
Mr. Houser asked if SSA has looked at traceablity of requirements through legislation.
Mr. Marr said that is being accomplished through a related project that Lockheed Martin is supporting, but it’s not using XLink.
Mr. Henry Breton said we are looking at the entire domain throughout the policy planning, creation, and management processes.
Mr. Dean said with all the architecture work we are trying to do a good risk analysis so things don’t have to be redone. This is very complex; but we feel we will have a good XML-based architecture. We appreciate xml.gov and have tried to follow the standards set by the Federal CIO XML Working Group.
Mr. Ambur agreed that it can be very complicated, but suggested that SSA's work has a great deal of application to the use of XML by others. For example, Congress has created a DTD for legislation and NARA and GPO have been considering the use of XML in the Federal Register, Code of Federal Regulation, and U.S. Code. XLinking would be very useful for establishing cross-references in legal documents such as these. IRS is another agency whose guidance, like SSA's, is very complex and could benefit greatly from the use of Xlinks.
Mr. Houser added that the Veterans Administration has been working with XML as well.
Mr. Dean said SSA hasn’t had contact in this arena.
Mr. Ambur said to let him know if the Federal CIO XML Work Group could help with making some contacts.
Mr. Houser asked for expansion on the Spewak EAP {Enterprise Architecture Planning} and the Zachman framework, specifically more about architecture and how that relates to XML.
Mr. Leff said he met Mr. Zachman, a key motivation behind enterprise architecture. We are trying to use the Zachman framework at a high level and understand the implications of what we’re modeling; what we aren’t modeling; and what the architectural implications are. In relation to the Spewak pyramid, we feel we’ve done a lot with the top of the pyramid and are now at the third level. What’s not clear is the role and impact that XML and it’s surrounding technologies will have on shaping the architecture. The question remains as to when the use of XML will begin to expand into the larger enterprise architecture.
Mr. Houser asked if XML expands across J2EE and whether they have looked at particular cell within the rows and columns in Zachman.
Mr. Leff said they are focusing on the first couple of rows, which he feels equates to the Spewak pyramid. We are using the Unified Process because we can incorporate the iterative process.
Mr. Houser asked what tools are you using.
Mr. Leff said we are not using an enterprise architecture tool; we are using Rational Tools.
Mr. Dean said that Mr. Ambur advised against enterprise modeling. Mr. Ambur responded that his concern relates to efforts to "model the world," as opposed to focusing on more practical changes that can actually be accomplished in day-to-day business processes. Mr. Ambur noted that if you try to do everything at once, you will surely fail.
Mr. Dean said it will take time in SSA to roll-out because the distribution is so large. We have to move fairly conservatively. We will follow the technology and be ready to implement.
Mr. Marr said that will be important.
Mr. Houser agreed that it is hard to integrate internally because of firewalls.
Mr. Dean said working with security was a challenge to be able to use the Extranet.
Mr. Marr said OMB is a tremendous driver. OMB is a driving force within the agency of how initiatives are going to work.
Mr. Houser said he is impressed with the architecture.
Mr. Ambur asked about user interface issues with respect to browsers and XML editors. He said he'd like to schedule a presentation on that topic at some point but that he has had trouble finding knowledgeable individuals who are willing and able to share expertise in that regard. Regarding the E-Gov Action Plan, Mr. Ambur asked, “Is your initiative being driven by the CIO office?”
Mr. Dean responded E-Vital is being driven by the CIO, but XML is not being driven by the CIO.
Mr. Hynes said that E-Vital is doing some XML. He believes they are creating an XML Schema.. SSA will pilot with up to 8 states For instance, instead of a paper birth certificate for verification, employees could look at POMS and have a URL that would allow them to link directly to vital records within the appropriate state for verification. They are talking about some XML as a base.
Mr. Ambur said homeland security will be a big driver. He noted that the system can and will be distributed, that the key is for widely distributed systems to interoperate more effectively together, and that XML is an enabler of interoperability.
Mr. Houser said my concern is I hope SSA will get involved with ebXML. It is frustrating that the government created their own rules rather than working with W3C for Section 508 standards.
Mr. Ambur asked if there is anything the XML Work Group can or should do to accelerate this process.
Mr. Crawford said the standards bodies are open. Some work groups are closed, but more than likely they will accept participants or at a minimum observers. Lockheed Martin should get involved in W3C. At last count nine federal agencies are members of W3C.
Mr. Houser said he would like feedback on the SSA implementation.
Mr. Ambur said if he were OMB he would like to know how many government employees and contractors are working on home-grown systems, as compared to the number participating in voluntary consensus standards organizations in order to accelerate the development of interoperability standards and avoid spending the taxpayers' money to reinvent the wheel.
Mr. Houser said if you look at NIST, all you see is a downward spiral as far as money is concerned.
Mr. Ambur said associating metadata with standards will be very useful for other agencies that don't have many resources to conduct research to identify standards that should be incorporated into their systems. [Editor's note: ANSI has initiated consideration of a core set of metadata for standards documentation. See http://www.ansi.org/rooms/room_5/public/xml_forum_agenda.html]
Mr. Dean said we were zero funded until lately. It was difficult to prove that we needed this.
Mr. Ambur asked whether XML was being used to format SSA GPRA plans or reports. He suggested that using XML in a reasonably standardized manner to render all agencies' GPRA plans and reports on their websites would be very useful not only for oversight agencies such as OMB, GAO, and Congress but also to other stakeholders, including citizens.
Mr. Dean said I think that is all coming. We were in SGML already. It will come once we’ve shown the value of this, especially what we’re doing with the Repository. We need to establish positive results.
Mr. Houser expressed an interest in their security model. He asked if it appeals to federal courts and how does SSA ensure the integrity of reports.
Mr. Dean said we have all the issues everyone has with going to electronic media.
Mr. Hynes said we have had meetings with NARA regarding what we have to keep. Something could come up internally that happened a long time ago so we need long storage times on data. Now we can do that with electronic archiving. We are working with NARA to define the operating procedures that we should follow. We have had preliminary meetings. There is lots of incidental data, but we now have versioning control.
Mr. Dean said there are data integrity controls. People have held onto CDROMs because they are stamped. There are certainly many issues that we have been dealing with.
Mr. Hynes said we had an old proprietary database that was the only one I counted on. I wouldn’t trust paper because it got out-dated too quickly.
Mr. Houser asked if they are using PKI for document signatures.
Mr. Dean said these are all valid points. We understand the issues, but we think there is as much, if not more integrity in this data.
Mr. Leff said how do you know what’s being link and presented is a whole other issue- were these the right links, etc.
Mr. Houser said Oracle databases have a series of caching servers that have images of the policy. Does that cause any issues?
Mr. Dean said any decision we make can be appealed, so this can always be reviewed.
Mr. Ambur said link management is already a big problem with HTML and will be an even more important requirement when XLink is widely implemented.
Mr. Dean said in Chicago one of SSA's applications has 14 thousand granular links to the POMS . These need to be linked and integrated.
Mr. Ambur noted that NARA is leading an E-records effort under the Administration's E-Gov Action Plan, and he suggested that the E-records project should be identified as a cross cutting initiative since all information technology systems create records that should be managed appropriately.
Mr. Mark Crawford of LMI presented the document.
There are still many opportunities to enhance the document. Mr. Marion Royal is chairing this effort. The goal is to get the working group charged to work on moving the guide forward to be something that we can take to the CIO and then to OMB.
Michael Jacobs and Brian Hopkins issued a Department of Navy (DON) Developer’s Guide in October 2001. At the November/December Strategy Session it was decided that a Federal developer’s guide would be created using the DON Guide as a blueprint.
EPA is having a first draft of an XML Developer’s guide created. Steve Vineski (chair of the EPA’s Technology Assessment Group—TAG) plans to issue that document to xml.gov.
Mr. Royal tasked Mr. Crawford to federalize the DON Guide.
The DON needed time to deliberate on the XML technologies that need to be sifted and analyzed. But DON’s Taskforce Web was to move 50 applications to the Internet in a short period. They were told to do it using XML. While they wanted time to develop a good guide, they had an impending need for it. So, they issued a general guide with guidance that wasn’t overly strict.
Background influences of the Federal Developer’s Guide were the DON, Schema Best Practices, EPA, ebXML, Oasis Registry Technical Committee, UML and RUP, LMI Federal Tag Standards for XML, etc.
The Guide layout uses the RFC 2119 terminology to express requirements. These terms are MUST (NOT), SHALL (NOT), MAY, and REQUIRED.
There are seven chapters in the document:
1. Background
2. Recommended XML Specifications
3. XML Component Conventions
4. Schema Design Conventions
5. Document Annotation Conventions
6. Attribute/Element Conventions
7. Federal XML Registry
8. Appendices
Mr. Crawford reminded the Working Group that this is a draft boiler plate for Marion’s group to continue to develop.
The recommended XML specifications have strong wording that both DON and EPA have adopted. That is that any production level system will only use W3C recommended standards.
Mr. John Esoman stated so RELAX won’t be used. He asked, “What about RIDDLE?”
Mr. Crawford said RIDDLE is not listed. We needed a single authority. Although, SAX is a special case exception.
The use of proprietary extensions is only for local program area implementations.
Chapter 3 XML Component Conventions addresses naming case convention, use of acronyms and abbreviations and component creation.
Chapter 4 Schema Design Conventions addresses schema languages, development methodologies, and capturing metadata.
UN/CFACT modeling methodology should be reviewed by the Federal CIO XML WG. It’s a new and exciting way of representing UML. It’s called UMM.
Chapter 5 Document Annotation Conventions, for versioning Part 10 of the Federal Implementation Guidelines for EDI should be reviewed. We need something comparable to Federal Part 10 for XML within the Federal government.
Mr. Houser asked for clarification on Part 10.
Mr. Crawford said Part 10 is supposed to be used by all federal agencies for enveloping and exchange in EDI. The latest version is a year and a half old. The FESMCC group is responsible for this.
Theresea Yee said it’s been updated and it is currently out for review.
Mr. Crawford said it can’t be used in its current form for XML, but it has cross agency agreement and can be used as a model.
Chapter 6 Attributes and Elements says elements should be used for all parsed data and attributes should only be used for metadata needed by the Server.
Chapter 7 Federal XML Registry addresses the Registry and reuse of XML artifacts.
This is a call for volunteers and to begin deliberations on this document.
Mr. Ambur said he is glad the draft has been distributed on the listserv because a draft is better than nothing, but that he would like to post it prominently on the xml.gov site in HTML and/or XML rather than PDF. Since a Registry is not yet available at xml.gov but will be piloted soon, Mr. Ambur suggested the timing of the guide is good. He expressed appreciation that the Navy and EPA have invested in drafting it. He suggested the Work Group should provide a tool that developers can use, namely the Registry, before we have any right to tell them what we think they should be doing. However, to the degree that developers find it in their own interest to use the guide, it is good to make it available to them, and when the Registry is available, the guide and the Registry will be mutually supportive.
Mr. Crawford said the DON Guide is focused on Schema. EPA is differentiating between DTD and Schema and between data centric and document centric.
Mr. Ambur said I think there is high value in that.
Mr. Dean said we see ourselves moving from document centric to data centric because of our users. If data centric and document centric guidance are different, are we at risk?
Ms. Jessica Glace responded the data centric guidance is more rigid to provide greater data integrity whereas the document centric guidance is more loose to allow greater flexibility. If you lean toward the data centric guidance you’ll be safe.
The question was posed as to why the ebXML model is being used rather than the DoD model.
Mr. Crawford said the DoD model is static—based on registration of artifacts. ebXML is based on a more dynamic run time. I think we’ll end-up with a combination of the two.
Toni Wire said we are considering a run time version where we would have something like that available.
Mr. Crawford said I hope they are looking at the OASIS registry specification. There is an opportunity for a marriage of convenience. EPA has looked at what is happening with DOD and is working with NIST to help develop a more dynamic registry.
A member from DEFAS asked what is your impression of the DON Guide.
Mr. Crawford said considering the tight time constraints, it’s a good start. They are also continuing to expand the document. I assume that other Agencies may be working on this type of document as well. Does SSA have any guidelines?
Mr. Dean responded that they do not.
Mr. Leff said it would be helpful to know how a developer's guide fits with an architectural guide.
Mr. Ambur said he had already had conversations with leaders of the CIO Council's Architecture Working Group and that they had expressed interest in referencing XML appropriately in their guidance."
Mr. Crawford said we need to make sure that there is good cross-agency representation. This will eventually go to OMB.
Mr. Houser said it really needs to get there fast.
Mr. Ambur announced that the CIO Council's Architecture and Infrastructure Committee would be briefed regarding the Liberty Alliance and indicated that portion of the meeting would be open to anyone who may be interested in attending.
Mr. Houser said the Arts and & Architecture Common Business Model will be presented. The DON Guide is more mature than the Architecture Model because DON has more resources.
Mr. Crawford said yes DON hired a contractor to do some of the work. The contractor, Logicon, drew from the Federal CIO XML Work Group.
Mr. Houser said we’re going to need to register namespaces.
Mr. Crawford said that’s why Marion’s group needs to get started. They need to decide how the Federal CIO XML WG is going to use standards. We also need a registration process (operating procedures).
Mr. Joel Reeves asked how to get more information on the DOD registry.
Mr. Crawford said go to http://diides.ncr.disa.mil/xmlreg/user/index.cfm. EPA requirements for an XML registry document is another good resource. This is something that Lisa Carnahan’s group will be exploring.
Mr. Eliot Christian said can’t we just get a group started.
Mr. Ambur said Ms. Carnahan is the Federal CIO XML WG expert.
Mr. Houser said Ms. Carnahan sent me e-mail that she got an e-mail for DLIS and plans to have prototype work for the end of February. Her focus has been on implementation.
Mr. Crawford said we need to review the Work Group’s action items.
Mr. Houser said I will ask Ms. Carnahan to arrange a meeting.
Mr. Crawford asked if Mr. Ambur or Mr. Royal have officially asked Ms. Carnahan to be chair.
Mr. Ambur said I haven’t contacted her.
Mr. Christian said the minutes show that Ms. Carnahan was asked.
Mr. Ambur said that this is voluntary and this group can’t direct anybody.
Mr. Houser said Mr. Royal should contact her.
Mr. Ambur said I have to give Mr. Royal and Ms. Carnahan the benefit of the doubt that they are concentrating on the right areas. I hear Mr. Houser volunteering to contact Mr. Royal / Ms. Carnahan to try to make a meeting happen.
Mr. Houser agreed to do that.
Mr. Ambur said Jon Bosak is scheduled to speak at next month's meeting. Responding to a question about why NIST chose to use registry software from DLIS rather than DISA, Mr. Ambur noted that both are under the auspices of DoD and that DoD leadership should work with us to harmonize the necessary standards for interoperability of XML registries.
Mr. Crawford said there is no formal policy. There is draft documentation from DII COE. DLIS has been on logistics information with a business orientation. If you look at the focus of DII COE, they are different. In fact, DLIS was the creator of EDI that was subsequently adopted by the business world. It is a technically sound solution that is well thought-out.
Mr. Ambur said we want to build on the good stuff everyone has done.
Mr. Crawford added and which is best suited for our needs.
Mr. Lex Poot said some combination of the two is probably a good approach.
Mr. Crawford said we need a namespace approach. I have some reservation about the DOD namespace approach.
Tina said each community of interest is in charge of their own namespace. They have to go through the COE to get a namespace.
Mr. Crawford said at a high level that isn’t happening. It’s not in the documentation. There is no mandate to create harmonized efforts.
Tina said once they can collaborate you’ll start seeing the enterprise namespaces.
Mr. Crawford said you have each area creating disparate data, there needs to be a way to create federal standards.
Mr. Houser said we need to agree on the Sandbox.
Mr. Christian said the interoperability is always in a context. There are differences between health care and logistics.
Mr. Crawford said the issue is that there are so many agencies and they interact with each other. That is why we should be working within the standards bodies. We need to articulate our needs into those standards bodies.
Mr. Mike Lubash said this has to be customer centric. The approach that we need to lay out needs to be based on common elements.
Mr. Ambur said our current charter only extends through the end of this fiscal year. Our resources are limited and thus far our budget allocation has not been confirmed
Mr. Houser said the Washington DC XML/SGML group is meeting and the topic is Topic Mapping.
The meeting adjourned at noon.
|
Last Name |
First Name |
Organization |
|
Ambur |
Owen |
Interior-FWS |
|
Breton |
Henry |
Lockheed Martin |
|
Christian |
Eliot |
USGS |
|
Crawford |
Mark |
LMI |
|
Duplain |
James |
SSA |
|
Esoman |
Jon |
|
|
Gellar |
Larry |
Vitria |
|
Glace |
Jessica |
LMI |
|
Houser |
Walt |
VA |
|
Joel |
|
Intel |
|
Leff |
Michael |
Lockheed Martin |
|
Lubash |
Mike |
DFAS |
|
Marr |
Timothy |
Lockheed Martin |
|
McKeever |
David |
i4i |
|
Napoli |
Frank |
LMI |
|
Poot |
Lex |
DTS |
|
Roberts |
Davis |
SAIC |
|
Weiland |
John |
NMIMC |
|
Weir |
Toni |
DISA |