I go where the challenge is. That keeps me motivated.
- Ron Hawkins
Ron Hawkins has an excellent background for a career in IS management: His father was in the watch repair and jewelry business, and his uncle was a professional hockey player. From his father, Ron inherited an ability to focus on details and work carefully until the job was done and done right. That talent for concentration has served him well in his rise from programmer to consultant to top-level executive.
At the same time, Ron admits that his uncle, Art Wessier, exerted an even stronger influence. A large, physically powerful man, Wessier was a defenseman for the Montreal Canadiens in the 1930s. From his uncle, Ron says, he gained an understanding about common sense, patience and the intelligent use of power. Even now in business, Ron finds himself asking how "Uncle Art would have handled this." He adds, "My uncle's lessons have really stayed with me. In a crisis, I seem to have this clarity of thought. It might seem confused around me, but I know what to do."
His first position was as a programmer at Leesona, Inc., a textile mill that also makes looms for winding yarn. He worked at Leesona for three years, fine-tuning financial and manufacturing applications. "That's when I gravitated from programming to the technical support side," he says. Ron eventually worked as a member of a technical support group that maintained an in-house teleprocessing monitor that controlled terminals and scheduling programs, something like what IBM's CICS transaction monitor does today.
In 1977, Ron left Leesona to work for Harris Corp. out of Melbourne, FL. He provided database administration tech support for one of Harris' manufacturing systems. He also helped implement the system at Westerly, RI, moving back to Rhode Island and continuing to work for Harris for two years.
His next move was prompted by the departure of a business manager at Harris who went to work for Sperry Univac. The manager was hired to support a Sperry Univac manufacturing system called Unis, and he started a team for pre- and post-sales support. Ron joined his team, helping customers during implementation. The job entailed a good deal of travel, with 80 percent of his time spent on the road. After two years of being a road warrior, Ron relocated to a Sperry Univac branch at Wellesley, MA, then moved again to Waters Chromatography at Milford, MA, a high-performance liquid chromatography and chemical separations business. Waters was in the process of moving to an IMAPS database management system (DBMS) from Cullinet (later absorbed by Computer Associates). Ron helped with the move and continued to expand and refine its DBMS.
At the time, Millipore's IS department was facing a major transition. In the early 1980s, the corporation had decided to adopt a centralized strategy using the proprietary HP 3000. It was using the HP 3000 worldwide and was trying to expand the HP 3000 environment in the United States. However, the HP 3000 program in North America had failed to meet expectations, and a new vice president of IS was called in to take a fresh look at the situation, especially since new upgrades were imminent.
A team was appointed to determine IS strategies for Millipore throughout the 1990s. The team came back with four principles that it felt were critical:
One of Ron's biggest tasks these days is moving Millipore into the world of data warehousing and expanded decision support. Ron points out that Millipore's personnel are highly educated, demanding users of information. Any decision support system will have to be equally sophisticated.
Currently, Millipore's corporate information system is divided into transaction processing and decision support platforms. Some data warehouse applications have been written in-house, but the company still requires major development efforts to reach its goals. At this point, it has completed building the infrastructure for data warehousing, running Oracle applications in all its subsidiaries. Millipore has been collecting data in the Oracle databases for several years, and now it is ready to start delivering this data to two primary audiences in the corporation.
One audience is senior management. Ron says that these executives will need an interface, what he calls a "dashboard," to help them navigate through the database. They will need to "drill down" to extract increasing amounts of data for questions about sales, growth, revenues and expenses. They will also need to be able to "spin the cube": to look at an issue from different perspectives until the relevant answers emerge. The other main audience consists of a team of financial analysts working in the office of the CIO. These analysts are fairly high-level PowerPC users who use modeling in Excel. They want a multidimensional application to do true financial analysis from warehoused data.
At this point, Ron and his department are facing the classic issues of setting up a data warehouse. First of all, do they go with relational or multidimensional architectures? Ron admits that multidimensional online analytical processing (OLAP) has a reputation for speed, but he also knows that relational OLAP (or ROLAP) vendors say speed is not an issue when the database is tuned properly. Ron is also wrestling with problems about metadata--the "map" that shows where and in what form data resides in a database. Finally, he has to develop interfaces that can handle ad hoc queries in an easy-to-understand fashion. None of these issues has simple answers, but Ron feels that "1996 is going to be the year of the data warehouse for Millipore."
Ron feels that more managers these days are trying harder to balance the two sides of technology and business. Even a few years ago, technology solutions were based on relatively mature technology, such as mainframes, and were more shrink-wrapped. "You had a fixed set of tools, and you could apply them to pretty much the same set of problems." Managers therefore concentrated more on the business case: how a particular purchase could reduce expenses, increase revenues or otherwise justify its investment and overhead costs. Today, with client/server and open system environments, plus groupware, collaborative computing and desktop systems, the technical issues are more complex, so buying decisions have to address a wider range of technical questions. Ron also notes that the general pace of implementation has increased. "I need to provide a flexible delivery cycle, because the delivery has to be done in a matter of months now."
Despite the complexities, Ron describes his job in the most general terms. "You have to put a computing infrastructure in place that's a utility. Not necessarily a Web site or accounts receivable or order entry. Rather, a basic utility of servers, databases and networks." Changing his metaphor a bit, Ron adds, "I'm like a grocer; I provide shelf space. I don't care what's on them, but I have to provide enough shelves and make sure people can get to them." This down-to-earth explanation sounds like one his uncle would appreciate.