Continue in 2 seconds


  • August 01 2002, 1:00am EDT

What factors do the items in the following alphabetized list have in common?

AD cycle (application development cycle), artificial intelligence, business intelligence, CASE (computer-assisted software engineering), client/server technology, COBOL (common business-oriented language), CRM (customer relationship management), data marts, data mining, data warehouses, e-business/e-commerce, EIS (executive information systems), formal language theory, GUIs (graphical user interfaces), Internet, object-oriented technology, OLAP (online analytical processing), OLTP (online transaction processing), personal computer, relational database, SAA (system application architecture), spreadsheets, structured analysis and design, and 4GL (fourth generation language) technology.

The short list is a description of most of the fads that have occurred in technology since the inception of the computer. In one way or another, there have been advocates of these fads that with enthusiasm have brought these fads to life and have made them a part of the technology landscape today.

With the perspective that there have been fads in the short history of computer technology, what are some of the characteristics of the fads?

  • Advocates of fads always overestimate what their fads are worth or the ultimate size of the fads. The advocates have hammers, and suddenly everything looks like a nail. Ultimately, all fads fall short of the future that the early advocates predicted.
  • Vendors twist the fad into the existing technology. Once the fad becomes popular, there is a cry from the vendor community. They say, "We have that. That is what our product has been doing all along."
  • Fads always leave a mark ­ some a large mark, others a small mark ­ on the world as they start to mature.

In most cases, the advocates of the fad approach the world of technology as revolutionaries. Their message is that the fad is the true and proper way and everything that has preceded the fad is incorrect, old and bad.
In a few cases, the fads were approached in an evolutionary fashion. In these few cases, the advocates of the fads recognized that it was important to build on the infrastructure of technology that already existed rather than try to overthrow all existing technology.

It is interesting to note the difference between evolutionary and revolutionary fads. In every case, the evolutionary fads have had a much greater impact than the revolutionary fads. In some cases, the revolutionary fads made their mark and faded fast.

The closer a fad or technology fit with architecture, the longer lasting and the more impact the technology had (or still has). For example, EIS proponents adamantly insisted that EISs were independent of operational or data warehouse technology. In truth, both OLTP and data warehouse technology provided necessary infrastructure for EISs. However, the fact that EISs insisted on being independent only contributed the notion that executives needed a different kind of data. The actual EIS technology does not survive to any great extent today.

To some extent, the same is true of CRM today. In many circles, CRM advocates insist that CRM is a discipline unto itself and that there is no need for an infrastructure for CRM. These shortsighted people are simply being ignored by the more sophisticated people that understand that CRM needs an infrastructure of data warehousing. Unlike the technology of EIS, CRM technology will survive, but the most successful implementations of CRM will be those that are closely supported by a data warehouse.

In the same category as CRM is data mining. There are those that will tell you that you don't need a foundation in order to conduct data mining. They maintain that to start a data mining project only requires that you find your data, integrate your data, cleanse your data, stage your data, periodically refresh your data, and so forth.

With this approach, by the time you get around to data mining, you most likely have forgotten the question. Instead, if you accept the fact that data warehousing is an ideal source for data mining, you have a ready source of data and may commence analytical activities immediately. The price to pay for not accepting the data warehouse as an infrastructure for data mining is that you have to build a data warehouse anyway before data mining can commence.

To a great extent, the Internet in its commercial application is subject to the same need for infrastructure. The dot-com companies, with a great deal of arrogance, proclaimed their independence of "old" technology. The notion the dot-com companies promoted was that they were new and superior technology ­ an entire new economy; however, the reality was that in order to succeed, dot-com companies needed the infrastructure of transaction processing systems and the integrated and historical data found in the data warehouse. There are successful e-commerce applications today. However, those e-commerce applications are nothing like what the Internet visionaries had in mind. The successful e-commerce implementations are tied to infrastructures that have already been built.

Data marts had the same experience. When data marts proclaimed their independence and said that they wanted nothing to do with the IT infrastructure, thereby creating what was known as the "independent data mart," the result was chaos ­ there was a proliferation of stovepipe DSS applications. Once it was recognized that data marts were properly a part of a larger architecture, they prospered; but as long as data mart advocates insisted that they be free and independent of the existing environment, data marts produced more misinformation than information.

The test of the long- term viability of a fad is whether there is a core definition and theory behind a fad. Some fads have had very strong core definitions and defenders of that definition. Relational technology with Ted Codd and data warehousing (my contribution) come to mind. Structured programming and analysis with Ed Yourdon and Tom DeMarco also comes to mind. These fads (if that is what you can call them) have stood the test of time and are likely to continue to stand the test of time. Other fads that have not had a core definition and defender have long ago diminished in terms of influence and support.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access