Software is simply a layer of abstraction on top of this. For instance, when the USAF wants software for the F-22, you better believe there are requirements and processes just as stringent as Intel's process for designing/developing a CPU. And the people producing that code, imho, are every bit as much engineers as the EE's.
But there is plenty of software out there (maybe most software) like ERP/CRM, desktop/web applications, games, etc ... , where the technical parts are not that important, or aren't really the main problem. The main problem is to try to create the most clear representations of real life items (customers, business processes, human resources, etc.), and make them clear to other programmers in the source code, and to the customers (who most of the time aren't technical).
It's probably true that my theory doesn't make much sense for lower level/close to the hardware software development. But for high level things I still think I'm absolutely right ;).
If you want to develop software without strict practices, that's just dandy, and you don't have to refer to yourself as a software engineer. It does not follow that software engineering is a joke: http://news.ycombinator.com/item?id=132640.
On the other hand, if you do want to be a software engineer, these practices can be applied to all projects, no matter how small or high-level. I think this may be what your professor was trying to say.
Unfortunately my main thought always seems to get lost. Maybe I should try to make it more clear in the article. In my opinion the core focus of software development, and therefore also the most difficult part, is to create an image of the real world (both in source code and the program). An engineer's main focus is completely different.