After 61 hours of recorded PSP development work and untold hours of writing about that work, I've learned a great deal about personal software processes. The application of defined processes and engineering tools (linear regression, defined prediction processes, defect analysis with feedback into the development process, etc.) has had a profound effect on my own development practices and has taught me a great deal about myself and how I produce software.
Will I continue using it? Perhaps not all parts, but yes, I will. The focus on early detection and elimination of defects made quite an impression on me, and the use of tools to log and track development efforts is very revealing. In fact, most of the PSP is good enough that I'm inclined to implement some automation to simplify elements such as the data-centric postmortem, etc.
Is the PSP worthwhile? To me, a developer with little formal education on software process, the answer is a resounding yes. I have spent too many years in hazy code-and-fix development practices, and while I've done reasonably well, that is no way to continue. Unfortunately, most of the software world is mired in old practices, particularly the entertainment industry (which seems to survive on young, idealistic developers fueled entirely by caffeine).
The idea of software engineering as a profession is nothing new; it's been around for some time. But books such as [McConnell99] show a resurgence of interest in the idea of a defined software engineering profession. While it's unclear at this time what will constitute the bulk of a software engineer's training, I do feel that defined personal processes such as the PSP could well form the foundation of a professional's habits. The time spent has been worthwhile, and I look forward to using these tools in the future.