by Bill Trippe

XML Answers

The world of XML development surely doesn''t suffer from a lack of ideas. Name an industry, and I can name at least one XML initiative that is designed to solve its problems. These range from the broad (RosettaNet for trading networks) to the highly particular (the Bioinformatics Sequence Markup Language) to the whimsical (Facial Animation Markup Language, and, no, I am not kidding).

Sometimes I wonder if the XML community should not be so prolific and should instead focus on mastering a few key initiatives. For example, in my last column I wrote about XBRL, the eXtensible Business Reporting Language. The FDIC has directed the banks it oversees to use XBRL to report their performance and results. XBRL-encoded financial reports are both human and machine readable and can undergo efficient computer analyses that can''t be matched by conventional manual approaches.

What did it take to make XBRL work for the FDIC? A government mandate didn''t hurt, but the initiative also required key elements to be in place. Any XML initiative requires the core schemas or document type definitions be written and reviewed by people who know both XML and the data to be described. The initiative also requires tools that can support the XML version of the data. XBRL has had a recent boost with the announcement that two-thirds of accounting software vendors have either already modified their software to support XBRL or will do so by 2004. Impressively, this includes market leaders such as Oracle, Microsoft, SAP and PeopleSoft.

But more than anything else, XML initiatives need a critical mass to succeed. The FDIC''s adoption of XBRL is significant because it means all banks under the FDIC umbrella will be providing reports in XBRL form; this is a great example of critical mass. Contrast the FDIC''s approach with that of the Securities and Exchange Commission, which has made XBRL optional. So far, a number of companies have voluntarily filed XBRL reports with the SEC, but the vast majority haven''t. As a result, there is no critical mass among SEC-regulated companies.

Enter Concord, MA-based OneSource. OneSource is in the business of providing business content that supports sales, marketing and customer service functions. Like Standard & Poor''s, OneSource is able to charge a premium subscription because of the high value and broad coverage of its content databases.

Historically, financial reports from publicly traded companies have been a key component of the OneSource offering, and OneSource saw a natural fit between XBRL and the needs of its customers. But with so few companies providing financial results in XBRL form, OneSource had a chicken-and-egg problem. If XBRL is going to be successful, yet it lacks a critical mass of support, what to do? OneSource decided to step into the breach and provide the data itself.

OneSource''s XBRL financial report database was first published in March, and it debuted with data on 28,000 publicly traded U.S. companies. OneSource has since expanded the database to include all listed companies in the United Kingdom and all multinational companies. As a result, more than 775,000 companies are now listed, with XBRL-tagged fields such as number of employees, revenues and cost of goods sold.

The value of XML, and a particular application such as XBRL, is to provide data that is unambiguously encoded and can be easily isolated and manipulated by other programs and processes. In the case of XBRL, financial models and other tools can use the encoded data to automate vital analyses that formerly required much time and effort.

OneSource customers are using XBRL data for applications such as sales and marketing. One company, for example, automatically collects certain facts for prospective customers and feeds them into an ROI analysis they can provide for certain offerings. An insurance company uses another set of fields from both the income statement and the balance sheet to automatically produce proposals. OneSource provides a software developer''s kit and an Application Programming Interface that lets users access, manipulate and integrate the XBRL data with applications ranging from Microsoft Excel to high-end analysis tools such as those from Cognos and Hyperion.

OneSource''s approach is not unique. Standard & Poor''s Compustat division provides very similar data, as both ASCII data feeds and as XML-tagged content. In Compustat''s case, though, the XML data is not in XBRL, but in a unique XML tag set. Customers can develop similar applications using the Compustat data, but those applications end up being proprietary to that format.

OneSource''s efforts have provided some real value, but hopefully this will only be the tip of the XBRL iceberg. Ideally, the source data itself — that is, those reports submitted directly to the SEC — will soon be available as XBRL-tagged data. With Microsoft releasing an XBRL toolkit with the next version of Office (due out in October), this XML standard will begin to reach a much larger audience.

In the long run, as public data sources rely more heavily on XBRL, the financial community will be able to produce a vast array of useful data and tools. The result will be a much better informed financial and investor community, aided by a critical mass of XBRL-encoded data.

Bill Trippe (btrippe@nmpub.com ) is president of New Millennium Publishing (www.nmpub.com ), Boston, a consultancy specializing in electronic publishing, content management, SGML and XML.