It has become fashionable to design and develop business systems by starting in the middle, with the latest technological or e-commerce fads, which, supposedly, can solve most business problems.

Business intelligence (BI) software can prevent your company from suffering the next Enron-like meltdown. BI software can even improve homeland security -- at least this is what some vendors' marketing claims purport.

Is there a single action that could virtually guarantee improvement of software estimation? Well, you might say, if you knew of one, you would have implemented it already. But what if the answer were making software projects smaller and simpler? Undoubtedly, estimating a small project is easier than estimating a large one.

Assertion #60 from the Cutter Business Technology Council states:

The inherent tradeoff between agility and quality is an intuitive concept. With everything else kept the same, increasing agility typically decreases quality and vice versa. Agile decision-making, for example, is often exercised at the expense of exhaustive and time-consuming analyses.

There are many people who believe that software "engineering" is not an engineering discipline at all but rather a set of sometimes-followed best practices that, when all is said and done, aren't really the best.

This Update reports on data recently gathered by Cutter Consortium. We asked a variety of questions about recent trends and will be reporting the results over the course of the next three to four months. Overall, we had 235 responses from IT and business managers throughout the world. Of that total, 34% were companies engaged in the production and sale of software.

IT organizations continually spend vast sums to develop and maintain proprietary software, mostly for internal use. Some estimates place this figure at nearly half a trillion US dollars worldwide per year. This has created a huge and growing inventory of software assets, only a few of which provide real competitive advantage.

I was reminded of the tale of Saint George and the Dragon (see text box below) about three years ago in a conversation with the president of a Chicago-based company that seemed incapable of delivering any piece of software on time. He was looking for someone to come in and slay the dragon -- a sweep of the sword, a thrust of the spear, and the software overruns would disappear.

Successfully managing IT expenditure and ensuring successful delivery of IT projects and levels of service continue to be foremost in the minds of corporate executives, particularly in light of today's tight economy and struggles within the technology sector.

The subject of risk seems to come up often at Cutter Consortium, either voiced by our subscribers or discussed by our Senior Consultants. But one has the feeling that, as Mark Twain intoned about the weather, "Everyone talks about it, but no one does anything about it." So, risk endures as part of the messiness of life: not as part of the richness or the challenges -- just the muck.

Variation in your environment -- whether furniture, heating, air-conditioning, transportation, or your technology infrastructure -- is expensive. But while there are great savings embedded in standardized environments, the whole is fraught with emotion.

Everything you buy, you have to support -- or hire someone else to support. That's the problem with buying and deploying lots of stuff: it all needs attention. To make your support dollars effective, there are a number of things that you need to know.

The Object Management Group (OMG) began developing CORBA in 1989. The idea was that organizations would be moving to object and component technologies and that everyone would be better off if they had a common, open standard to use when they wanted to link together objects or components on different platforms.

It's no secret that insurance companies are big IT spenders. Larger insurers have been consistently spending hundreds of millions of dollars per year on software alone. For some time, insurers have been focusing the vast majority of their technology spending in two areas: computing infrastructure and business process automation.