Tom Mochal made it quite clear in his article when he wrote "the vast majority of research and information on the subject shows that major work initiatives should be defined, planned, and organized as a project." If you don't invest the time and energy up front in defining your project how can you know if you are hitting the mark? This is where 90% of projects fail. Project definition is useful as a selling and negotiating tool which allows the target customer to opt out or make modifications before serious money has been spent.
Project management skills should be applied far more widely than they are - they work in any environment. Everything is either within a project or process environment or a hybrid of both. There are a lot ofsimilarities in the way they should be planned and managed.
I have taught project management programs to many kinds of people including civil engineers and IT specialists, many of whom were ready to leap onto a piece of software without any real idea of what has to occur before project implementation. The nice thing about PM is that it can really heap to build a team culture. Although everytime I run a monument building exercise in a workshop I have never had the same result twice!

A common difficulty is the tendency to look to project management for large or unusual projects that people have had little or no experience with. It's no wonder they often start to feel a bit edgey about the prospects of success. Neverthelss, one of my favourite sayings about estimating resource requirements is that plus or minus 50% is better than no idea at all.