Bitcoin is in its infancy. It is an experiment that will be a lot harder to interpret if the 'protocol' is tweaked too much before even the first drop in block-reward happens. If miners start dropping like flies, there is a solution: pay higher transaction fees. If enough people pay enough transaction fees, mining will become profitable again.

The protocol taking time to adjust to changes in network performance is a good thing. It allows miners to respond to problems on human time scales. It guards against network take-overs during localized outages. The 10 minute block interval reduces the number of stale blocks processed, while still allowing transaction to happen within a reasonable ammount of time.

Put another way: you are thinking in the extreme short term. In my opinion any crypto-currency, no matter how good, will fail in the next 20-50 years. Modern computers are simply too insecure to handle a cryptocurrency. I predict that secure, proven correct on pain of liability, computers of equivalent complexity will take generations (or about 150 years) to develop. IMO, due to the 1996 WIPO 'Copyright' and 'Performances and Phonographs' treaties*, computer security has been regressing for the past 15 years.

* Articles 11 (Copyright)  and 18 (P&P) may be a symptom, rather than a cause of corruption. The fundamental issue is who controls the computer: the user or the manufacturer?
PS: I realize the 'state of the art' has advanced over the past 15 years; but people are still installing insecure sofware (like adobe flash player) as the administrative user. You can't install software making use of DRM as a limited user unless it plays nice in a virtual machine.