Developing XML Standards.
The key benefit of XML is that it provides a standard, just like ANSI X12 EDI (electronic data interchange) standards did. And, just like X12, there must be widespread agreement among trading partners on exactly how that standard is manifested. Moreover, as with X12, standards are defined at two levels, according to Benoit Lheureux of the GartnerGroup, Stamford, Connecticut. The "foundation" level defines the basic nature of the language, such as how it handles document linking, security and so forth.
This level is currently being performed by the special interest group (SIG) of the World Wide Web Consortium ([W.sub.3]C), which recommended adoption of Version 1.0 for XML in February 1998. Members of [W.sub.3]C include all major vendors, government and academic organizations and large users. In addition, the Organization for the Advancement of Structured Information Standards (Oasis) is building an information repository where XML schemas can be posted (www.xml.org).
The second level, the so-called agreements level, includes the XML formats, vocabulary and definitions. They are developed by consortiums established by individual industries such as banking, insurance, automotive, etc. Very much like X12, each industry defines its standards--most commonly via DTDs (document type definitions). These establish the schema, the logical structure of the document. Schemas contain data element types, naming conventions, where the elements may occur in a document and the relationships among the elements.
All documents can be validated against the common DTDs, and an XML-structured document can be compliant with multiple DTDs. This means a document need only be created once and the data used for multiple purposes. In this fashion, a field such as "borrower name" in a loan record could be read by applications that produce everything from an approval letter in Microsoft(R) Word to a name on a title policy generated from a mainframe application to an invoice on a minicomputer-based accounting system.
Moreover, once the document has its XML tags in place, more or enhanced DTDs can be developed to read more and more data elements without the original document having to change formats. In other words, the standards can continue to develop without the document owner having to do conversions. This was a major drawback to EDI.
Thus, if all parties write their documents in conformance to the DTD, a broker could send a loan file electronically to any lender's LOS (loan origination system) for an underwriting decision; the lender in turn could send that file to any investor's analysis program for a purchase decision and eventually send it to any servicer's system. This can be done today by writing interfaces to individual systems; XML lets the same record be sent to any firm using XML. This gives industry players many more opportunities to shop loans or have multiple trading partners; it facilitates "bidding," and it enhances competitiveness.
The financial industry has announced FIXML (financial industry XML) for securities transactions and [F.sub.p]ML (financial products mark-up language) to facilitate derivatives dealing and information sharing on interest rate and foreign exchange products. The National Association of Realtors and others have worked to develop two standards to govern real estate listings: RETS (real estate transaction standard) and RELML, an XML-based standard for multiple listing services. All XML advocates agree that these industry consortiums are essential, but they also agree that there can be only one such effort per industry. Otherwise, there will be competing "standards" and no true uniformity.
The mortgage industry is currently facing such competition. On one side is the XML-MP (mortgage partners) group, comprising some vendors and large lenders. On the other side is the Mortgage Bankers Association of America's (MBA's) newly formed Mortgage Industry Data Standards Maintenance Organization (MISMO), which maintains that standards developed by a tiny subset of the industry, without broad input and support from all affected players, will not be adopted or sustained over the long haul. Both sides have released DTDs using XML.
The key issue is how to preserve the openness and inclusiveness of the process that worked with X12 while responding in "Internet time" to e-commerce requirements. The advantage of a small group is that a small number of players are likely to be able to hammer out a standard fairly rapidly. The advantage of MIS MO is that its broad representation ensures that all players' needs are met and the standards do not give one or another player an unfair head start by simply adopting their proprietary solutions.
In my view, a compromise approach might be to have XML-MP serve as the MBA's XML work group, but with its representation broadened to ensure equity. This would include lenders; representatives from both Freddie Mac and Fannie Mae; LOS and document management vendors; mortgage insurance, title, credit and appraisal industries; and some Web service providers. The work group would focus strictly on development, using as aggressive a timetable as it chose. There should be a common Web site to eliminate any image of competition.
The work group would be obliged to summarize the logic behind its decisions. This requirement would address one of the flaws of the X12 work groups. When we sought to integrate the efforts of the various groups to build a common data dictionary, we were unable to capture the rationale behind some of the choices. When reviewing any standard, it is extremely helpful to not only know the proposed answer, but to also know what the choices were and why a choice was selected over others.
It would be the job of MISMO to ensure the standards were posted, to receive and forward all comments and to eventually publish the standards and educate the industry. To speed up the process, MISMO could eliminate mailings, instead handling everything through Web pages and e-mail. After all, anyone interested in e-commerce should be completely comfortable with electronic communication. Timeframes for review could be substantially shortened--to weeks rather than months.
No standard is useful unless it is adopted. Once again, MISMO can overcome an X12 flaw and create a generic, online tutorial for XML implementation. Each DTD developed would simply be a chapter added on.
MIMSO and XML-Mortgage Partners recently announced that two key players in the online mortgage arena have stated they will develop a universal data dictionary for the mortgage industry's use in electronic commerce. The work group will eventually produce three other e-commerce tools: an XML-based representation of the mortgage data, a mortgage data model and repository tools.
David Williams, assistant director of Industry Technology for the MBA, points out that "MISMO's future is dependent on the success of merging multiple efforts to create a uniform standard for the real estate finance industry. We believe all members should familiarize themselves with the subject by accessing www.mismo.org. The Web site contains white papers, proposed standards and opportunities for comment and voting."
In sum, let's be true to the concept of e-commerce and eliminate the paper, be entrepreneurial and come up with one set of standards with broad applicability. Let's harness all the talent to perform a real service that will put the mortgage industry at the forefront of technology--for a change.
Leilani Allen, Ph.D., is a partner with Summer Point Consulting in Mundelein, Illinois. Summer Point provides strategic planning, process improvement and technology assessment services and has offices in the Chicago and Dallas metropolitan areas.
|Printer friendly Cite/link Email Feedback|
|Date:||Feb 1, 2000|
|Previous Article:||MORTGAGE.COM TO CREATE OPENCLOSE SUBSIDIARY.|
|Next Article:||Training as a Strategy and a Tactic.|