Have you heard about a company called Timeline and the patents it holds? I suspect not. However, this company has the potential to significantly affect every database professional in the world. I'll give you some background, but keep in mind that this story is the subject of many pending lawsuits and thousands of pages of legal documents. Calling this week's commentary an incomplete summary would be an incredible understatement.
Timeline holds three U.S. patents, for its Data Warehouse Generator and Data Mart Generator technology, which provides connectivity to account systems and automates the generation of data warehouses and distributed data marts. The patented technology lets Timeline's customers automate the building and maintenance of financial data warehouses in SQL Server for analytic applications, synchronize the warehouse and the general ledger, automatically update the warehouse and related Excel-based reports, and deliver data and reports to user desktops. Timeline asserts that its patents cover such common database functions as using a driver to interrogate a source (e.g., transaction systems), use the information from the source to determine its structure, and in turn use the structural information in a middleware layer to design a target data mart or data warehouse; reporting from a data mart built using this method; write back to a target database; and pull data from multiple sources into one target structure.
According to Timeline's Web site, other patent violations include scheduling the building of or refresh of target data marts and publishing of reporting databases by area of responsibility, natural language queries, caching query results, loading data from a source, using a user interface to select from various drivers, creating new tables in target databases; generating tables, and optimizing reporting marts (e.g., adding summarization, rollups, or new relationships through an interface).
Timeline's home page (http://www.timeline.com) provides information about the patents involved. The full text of each patent is available from the United States Patent and Trademark Office Web site at www.uspto.gov . I encourage you to read the patents to form your own opinion. My research revealed that Timeline asserts that anyone who uses a software tool that automates the process of building a reporting environment that integrates and summarizes data from multiple upstream transaction systems is potentially in violation of the company's patents. This blanket assertion probably includes almost every data-warehousing product on the market. I'm not a lawyer (and I didn't stay at a Holiday Inn Express last night) so that opinion might not be legally correct, but it's based on thoughtful diligent research for this article. Timeline's original patent claims in 1998 were limited to narrow use within a financial services context. Those claims have been expanded since 1998 to apply to any generic data transformation architecture.
Think about the implications of those expanded patent claims. You probably have a reporting system that integrates data into a summarized format in a database that's separate from the original transactional systems in which it was generated.
I was initially stunned to learn that Microsoft and many other major database and extraction, transformation, and loading (ETL) vendors paid for licenses to use the patented technology rather than fight the patents. Public records show that Oracle paid around $1 million to license the patents. Microsoft originally licensed the patents with the understanding that it would be able to sublicense the patents to their customers and to third party software developers who use Microsoft software and tools. Microsoft intended to provide this sublicense to its customers for free to ensure that the patent claims didn't directly affect customers. Microsoft sources told me that for this privilege, the company paid substantially more than other vendors for its license, although the exact figure isn't public. Microsoft filed suit against Timeline shortly after signing the license agreement in June 1999 because Timeline claimed that Microsoft didn't have the sublicensing rights. See the Microsoft PressPass article at http://www.microsoft.com/presspass/press/1999/jul99/timelinepr.asp for additional information about the suit Microsoft filed against Timeline. In December 2002, the Seattle Supreme Court ruled in favor of Timeline on this matter.
Timeline is now suing Cognos over related patent-infringement claims. Cognos has refused to pay a fee to license the patents, claiming the patents are invalid. Last week, Cognos counter-sued Timeline, claiming "unlawful assertion of said patents." Cognos hasn't issued a press release about the countersuit but on January 8, a firm representing Timeline issued a response, which you can read at http://www.newsandearnings.com/ViewFile.asp?ID1=8994&ID2=%5B:FileTracking:%5D&ssid=3&directory=204&filename=TMLN_CGNS_reply_final.pdf . The argument that Cognos makes is interesting because it questions the validity of the patents. A win for Cognos could mean significant problems for Timeline, which has made a substantial sum from suing other companies for patent infringement. Timeline posted a company profile this summer that states, "Patent licensing enforcement efforts have grossed more than $12 million to date and have the potential to contribute millions more in the future."
I wonder why few people in the industry seem to know or care about this pending litigation. It seems to me that these patents could affect every IT shop that uses third-party software to build a reporting database to summarize and integrate data from multiple databases. Deciding the future of all computer-generated reports across every industry that exists is big news to me.
I'd like to point out that Timeline has never sued an end-user company for building a solution using Microsoft and other third-party products. Timeline suits have been against OEMs and Independent Software Vendors (ISVs). This doesn't mean that Timeline's legal strategy couldn't change in the future. However, you don't have to immediately assume that you have to stop that really cool data warehousing project you're working on. At least not yet.