DBMS Interview - June 1995
Platinum Technology has been on the move lately. Founded in 1987 by Andrew J. (Flip) Filipowski as a systems management tools provider for DB2/MVS-based environments, Platinum has been quietly developing products, acquiring technology, and merging with/buying up companies with one main purpose: to be the number one systems and data administration tools vendor by the 21st century. Key to this plan is the Open Enterprise strategy -- an architecture for total systems management across an enterprise-wide networked environment. Filipowski has led the company is this effort, expanding the company's focus from DB2-on-the-mainframe to client/server environments. In 1991, Platinum was ranked number 11 in Inc. magazine's list of fastest growing companies, and number two in Business Week's list of best small growth companies in America.
Filipowski spent the initial years of his career in the commercial marketplace, working for Time Inc., Motorola, and AB Dick, where he ran the data-processing departments. In 1974, Filipowski moved to the vendor side of the business, taking a job with Cullinet Software Inc. (the producer of IDMS, Cullinet Software was acquired for $333 million by Computer Associates in 1989). In 1979, Filipowski left his job as executive vice president and chief operating officer at Cullinet Software to form DBMS Inc., a provider of additive tools for IDMS users. Eight years later, he saw the potential for DB2 and formed Platinum Technology to provide add-on tools for that environment.
But that's not all that's Platinum. Filipowski is also CEO of Platinum Rotisserie Corp. and Platinum Food Service Corp., which together operate 80 Boston Chicken food stores in North and South Carolina. He is chairperson of Platinum Development Corp., a developer of commercial, medical, and residential real estate, as well as chairperson of Platinum Entertainment Inc., an entertainment company with recording labels in the gospel music segment. Filipowski is also CEO of Platinum Construction Corp. and Platinum Venture Partners. He is also known for his commitment to protecting endangered species, and is chairperson of Panthera Productions Inc., a film and production company specializing in wildlife documentaries. He serves on the boards of several diverse companies, including Dynasty Technologies Inc.
As you might expect, Filipowski doesn't have much time for interviews. However, Associate Editor Theresa Rigney managed to track him down on the phone to ask him a few questions about Platinum's aggressive moves into the client/server market. An edited transcript of their conversation follows.
DBMS: What was your vision for Platinum Technology in 1987?
FILIPOWSKI: When I formed the company, I was trying to define a narrow enough segment of the systems management marketplace, in which I thought, as a start-up, [Platinum] could have a significant opportunity to become a leading player, especially because I had a database background, from my earliest days of doing pioneering work with IMS, IDMS, and other DBMSs. In 1987 it became pretty clear that RDBMSs were going to be a good opportunity -- although it wasn't clear that they would be as dominant and as fundamental a building block as they are today. At the time, it was pretty easy to spot the up-and-coming success that DB2 would enjoy, so we [at Platinum] latched onto that particular market and created the kind of tools we knew data administrators, operational people, and application development folks who had to deploy and exploit the capabilities of a RDBMS would need. We built up a business around that.
In that earliest time frame, from about 1987 to when we went public in 1991, we had to compete with a dozen or more companies, some of them start-ups that had the same vision we had, and some of them the larger-sized companies that existed at that time. Around the time we went public, we had emerged as kind of a duopoly in the DB2-based systems administration business with BMC, which kept up the pace. In fact, with the two of us keeping up the torrid pace of development and releasing of new products and functionality and capability, we basically eliminated most of the other competitors. The start-ups were basically bought up, and the larger companies couldn't keep up the pace and fell out of the race.
DBMS: When did you decide to change the focus of Platinum from DB2-based tools to client/server?
FILIPOWSKI: In 1991 or 1992, we started seeing a fundamental change occurring in the business, which was that people, instead of getting a larger computer when they started running out of computing resources, were going a different route -- they were getting another computer and hooking it up in a network. That alone, in its earliest stages, had us expand our view of the systems management marketplace from concentrating exclusively on DB2 to providing a suite of similar solutions across Oracle, Sybase, Informix, and DB2 in its various flavorings as it became available on the OS/2 and RS/6000 platforms.
As we matured as a company, we were looking for ways of expanding that slice of the systems management pie into a broader front. We actually entered a phase where we had to define our aspirations of becoming a major software enterprise, whether we would become a part of some other company or create that opportunity ourselves. Because it was very difficult for larger companies to justify the purchase of Platinum Technology, we never got beyond some investigatory stages.
So with that fundamental change of people going from larger host-based systems to creating a network -- which became their system -- the phenomenon that occurred was that all the systems software that had been created for host-based systems became for the most part obsolete in the context of the network. What Platinum defined for itself is a market space called the Open Enterprise, which is principally populated by Fortune 5000 companies that are hooking up networks that include at least MVS, Unix, and Windows in all of their various flavorings. There's rarely a situation where those components [MVS, Unix, and Windows] are the only things found in a network, but that was the target market -- a superset of the open systems marketplace (because "open systems" is really just a euphemism for Unix). We thought this was the space where we could define a broad range of systems-management products, rather than just dealing with a slice of that opportunity. We had $120 million in the bank, a very substantial distribution capability in place, and 40-plus subsidiaries around the world, so we felt we were ready for a major challenge. That challenge became a picture of the kind of company we wanted to be by the year 2000 -- what kinds of markets we wanted to address, what skill sets we wanted to possess -- and we came up with a four-division approach to systems management.
One of these divisions is POEMS [Platinum Open Enterprise Management System], which consists of eight solutions for what we perceive to be the eight largest problems in the open environment and networked system. So we composed that architecture, picture, and vision, and set out in a rapid pace to complete it and put together all of the component parts to provide our customers with a total solution in that networked environment.
We also found that we had a core competency in the RDBMS market that served us well: the warehouse. We found that the warehouse was a phenomenon that, over the last few years, had accelerated amazingly, and we had a series of point-product solutions that were available for customers to cobble together with other folks' point products. The Platinum Pipeline product lets users define warehouse structures, create metadata, and manipulate the various data stores that are loaded into the warehouse. But when we examined that marketplace, the shift we were seeing was that third-party data warehousing product integration was becoming imperative. So we acquired Trinzic, and that became the vision for the second division of the company. And with InfoPump, users can get data into the warehouse.
The third division was the one that was already there really: data administration. This is such a significant part of our business, we retained the data administration business as a separate division. Finally, the fourth division focuses on the challenge surrounding application development technologies, so we built the division around that.
Between those four product areas, combined with consulting and education, we felt we had defined a framework for a multibillion dollar business by the end of the decade. Doggedly we have pursued getting [the framework] done, so we can get on with execution.
DBMS: How do you view the transition to client/server, or Open Enterprise, in terms of its benefits and risks?
FILIPOWSKI: I really don't look at [the migration to client/server] from that perspective. I have a pretty ambivalent opinion of it -- I could care less why the clients are doing it, as long as I can figure out what they're doing [in the process]. It's up to them to figure out why they're making the move. Otherwise, Platinum starts getting into a situation where we start preaching, and that's not our job.
DBMS: Where do you see the database industry ultimately going?
FILIPOWSKI: We certainly see RDBMSs continuing to be a fundamental component of the infrastructure of information technology. The relational database management system will continue to be a very potent component through the end of this decade. We do see some of the ODBMSs taking a sliver of the marketplace, and perhaps emerging at the other end of the century as significant players in the operational area, as well as perhaps serving a much broader function. And we see the specialized data stores contributing to the OLAP end of the business, whereas RDBMSs may specialize in OLTP. So there's a pretty interesting period in which the emergence of a true OO data store and specialized query and warehouse data stores, along with the maturing of the OLTP data stores -- the relational database management systems -- will help all of them, for the next five to 10 years, enjoy a wonderful environment in which there will be one whale of a contest out there in terms of who gets what dollars -- perhaps to the chagrin of the user community. Platinum has recently made a series of acquisitions and alliances -- most notably, the March merger with Trinzic.
DBMS: How are these acquisitions fitting into Platinum's overall Open Enterprise strategy?
FILIPOWSKI: POEMS is Platinum's architecture and vision for constructing the set of solutions for what we perceive to be the eight most significant problem areas confronting IT organizations in the open enterprise environment.
DBMS: What do you define as the eight problem areas?
FILIPOWSKI: One is data administration -- the entire process of operating and administrating a database environment. We feel that this area has been and continues to be a significant core competency of Platinum Technology. I should add that we have not stopped, in any way, shape, or form either developing internally or acquiring the necessary components to ensure that all aspects of data administration are done to the requirements of the Open Enterprise. We acquired Datura in 1994, which gave us a desktop Windows entry to the data administration function. We then acquired Dimeric, which gave us a very interesting capability within the Unix environment, and we started constructing the Consolidated product line from that. We've done very well in providing data administration and operational tools not only across the platforms but also across the various DBMSs -- DB2 to Informix.
Moreover, job scheduling came out as a very serious priority, confirmed by not only our intuition and focus groups, but also by Meta and Gartner Group. So to that end, we acquired a company that is now our laboratory in Boulder, Colorado -- the AutoSys Labs. We feel that we now have the premier, best-of-breed job scheduler in the market, which addresses problem number two.
Problem number three is performance management. To that end, we made an acquisition last year of a company by the name of Aston Brooke in Philadelphia. We acquired that company in 1994, and have, without question, the best-of-breed product in that category. Performance management is a critical issue in the network, where there are so many component parts and so many things that need to be monitored and measured. So, with the UniVision product line of DBVision and ServerVision, we have a very complete solution.
We identified the fourth problem area as that of distributing applications, programs, and licenses to the various nodes of the network. To that end, we bought the Viatech Xfer product. This product happens to also be marketed by Open Vision -- just as AutoSys is marketed by Tivoli. We have very strong alliances with these companies, as well as with IBM, Oracle, Sybase, Informix, and others.
Another component part is the Help Desk, which is a very critical area when trying to solve problems that occur in the open systems environment. The acquisition of Answer Systems, a San Jose-based company, was announced in early April. That will become our Answer Lab, and Answer's help desk software really stands head-and-shoulders above other products in this area in terms of problem resolution.
Sixth is security administration. To that end, we acquired a product line from Memco, an Israeli firm, in mid-April. This technology provides security administration that we feel is by far the best-of-breed solution. We will develop a product line from this technology in the United States.
The seventh box is that of backup and recovery, and hierarchical storage management. We have a pending acquisition there that will be announced in the second quarter of this year.
The eighth box is event management. Although some of the other components, such as program management, preclude asset management and other component pieces that will be announced later, the remaining area is event management, which is being developed in our Philadelphia labs and will be announced as a product shortly. So event and configuration management will be a home-grown set of solutions from within our own laboratory structure. Inside this architecture, we felt it necessary to have an enterprise counsel that is part of the events management. We looked at Trinzic, and we needed the inference engines in order to define the business rules, and so they were being incorporated into that event management component. And we felt that repositories were going to play a vital role in the Open Enterprise, so we acquired both BrownStone Solutions and Reltech, which was announced at the end of March.
DBMS: Will all of the POEMS components be integrated into the central repository?
FILIPOWSKI: We pretty much have a strangle-hold on the relational-based repositories, and we are enabling all of them to function with the POEMS parts. The repositories also play a critical role in application development, as well as in the warehouse -- perhaps their most vital function will be in the warehouse. But the enablement of all of these components of POEMS allows users to define issues related to their enterprise just one time, and all of the products then contribute as well as extract information from the central objectory in order to carry out their functions. So, as these products encompass the MVS, Unix, and Windows environments on a vertical level as well as integrate on a horizontal level, we feel that we will be a major player within that space.
DBMS: At a meeting with DBMS last July, you previewed the Integrator, an EDA/SQL competitor. How is the middleware market playing out?
FILIPOWSKI: The Integrator product is doing extremely well. It was renamed after the acquisition of Trinzic to be InfoSessions. We're prefacing all of our products in the warehouse space with the "Info" moniker, just because InfoHub and InfoPump [acquired with the Trinzic merger] have such good mindshare. We've announced our intention to expand our Pipeline product to InfoRefining, InfoReplicator, and InfoPropagator. The DataTransport product will become InfoExpress. This set of solutions in the warehouse space represents probably the only front-to-back solution from a single vendor, in terms of data extraction and refinement, or "scrubbing" -- whatever the popular term is for really working on the data accuracy and summarization (so that it can be put into a set of warehouses for interrogation and mining).
DBMS: At our meeting in July, you said there would be no application development environments for Platinum -- they're a "Trojan horse solution." Can you elaborate on that statement?
FILIPOWSKI: We have an entire group of products that allow application development to occur -- they're enabling technologies. We feel that we're going to carve out a very substantial part of the OO business as it emerges. In that arena, we're concentrating on making sure that we have a core competency developing in object repositories, databases, and development tools. That's what that ADT [Application Development Tools] division is all about.
DBMS: As a venture capitalist and entrepreneur, what vendors or kinds of vendors do you think are going to make it or not make it into the next decade? For example, you have money in Dynasty; how do you view a company like Forte?
FILIPOWSKI: I think that the application development business for the high-end environment is a viable business. The question is the size of the market. The current batch of application development toolsets are analogous to the Visicalcs of the PC revolution -- they were not built with the full knowledge or expertise of scaling to heavy-duty environments. So this is where there's going to be room for high-end solutions like those from Dynasty and Forte.
As to which vendors are going to "make it" is hard to say -- the software business is always changing. Software is a very dynamic and creative business, and there are new innovations all the time, so the most innovative, energetic, and aggressive companies -- the ones with the best, most competitive technologyý--ýwill probably be the ones with staying power.
DBMS: As Microsoft continues to drive down the prices of tools, will Microsoft play a role in the type of computing that Platinum is doing?
FILIPOWSKI: Platinum is focusing on the enterprise. And Microsoft will definitely play a role in the enterprise -- it dominates at the desktop and workgroup level, so its presence in the enterprise is secure. Microsoft has traditionally had a consumer focus, and the purchase of Intuit and what Bill Gates seems to be saying in his presentations is that the volume markets are in the consumer end of the business. But the question is where do you draw the line; do you need to vivisect or separate the consumer market from the business sector as in the past?
There's no question that Microsoft will be a potent force in the enterprise. But that doesn't mean that there's no room for other vendors to compete and thrive -- Oracle is an example that you can compete against Microsoft and still build a multibillion dollar business. And IBM is proof-positive -- it has a larger software business than Microsoft. And there's Computer Associates and others. And there's us.
Of course, Platinum will work with Microsoft on some levels and compete on other levels. Because Microsoft has had a desktop-centric history, I believe Platinum has an edge in the enterprise-wide systems management space.
DBMS: You are known for your help with endangered wildlife -- what is Platinum Technology's role in that effort?
FILIPOWSKI: As a company, we at Platinum try to counter-balance our technology and product aggressiveness with a benevolent attitude toward endangered species. We spend a lot of time, effort, and money to educate about wildlife and protect endangered species.
From its inception, Platinum has taken pride in being environmentally sensitive. We set up the Platinum Wildlife Foundation, which has several funds including the Black Rhino Rescue Fund dedicated to education and funding the protection of the Black Rhino.
In addition, each of our products is associated with an endangered species; and in our manuals and product literature, we try to dedicate some space for education and encourage folks to do whatever they can (in a subtle way). Getting the message out there is the important thing.