I am also of the mind that neither of these incidences were Princess' fault. Naturally the Star was not, they are only guilty of trying to provide the nicest decking possible - teak. Passengers are warned every cruise not to throw cigarettes off the ship, and while we don't know for certain, that is the likley cause there.
Teak decks actually were not the issue aboard MV Star Princess. According to the report by the British authorities (which was linked in a thread on that incident), the problem was a plastic material used used for the partitions on the balconies themselves.
BTW, the reason for teak, rather than other materials, is its ability to withstand weather.
In this incident, though "computer errors" are rare, it is not out of the realm of possibility that it is a computer error. Princess didn't build or program the computer, those are companies subcontracted by the shipbuilder. I highly, highly doubt this incident was caused by operator error because a turn going the wrong direction at top speed just isn't somthing you choose do, either by programming it or manually.
That's basically true, but a computer program is only as good as the software engineers who develop it. Most worthwhile computer programs are complex enough so it is absolutely impossible to test the response to every possible combination of inputs and internal values of program variables, so the development team must identify situations that require particular attention by other means. As a software developer, I can say from experience that this is not exactly trivial and it is very easy to miss some special case that, no matter how improbable, can cause a disaster when it occurs.
New ships often have problems - that has been shown countless times.
Yes. Of course, sea trials that stress the ship's propulsion and steering systems should identify and resolve these problems.