You see it as abuse, and I understand why, but from the perspective of a multi-billion-dollar business, this is not dissimilar from the people who think all non-GPL software is also abuse.
I am currently at a college and our PC Techs tend to get some interesting problems with PC's brought in by people who tried to repair them. It is kinda ugly.
I would not consider this cheating, or an evil way to prevent people from hacking on the items they legally own. If anything, it protects consumers from getting into things that they most certainly do not know how to fix, and even more, they don't know when they actually break the item further. Hackers that can own up to their responsibility of opening an item and accidentally damaging it will always open up and play with the internals. No one cares about that, they care about the guy that thinks he can replace parts he doesn't even know exists.
I've only "made" one watch from start to finish of which I still own. There really isn't much of a market for custom made watches. Plus all my tools are traditional hand tools so it can take a couple months to craft a single movement.
How can you "violate" a warranty. As most I think you can void it.
http://en.wikipedia.org/wiki/Magnuson%E2%80%93Moss_Warranty_...
I don't mean that like it's a bad thing. If you've ever had to reach up behind an instrument panel and attempt to guide a #00 philips screw in to its destination, you'll quickly come to appreciate the ingenuity of these odd-ball fastener types.
Another common cause is automated assembly. Fasteners like torx screws work better with automated machinery because they don't suffer from cam-out.
What I'm getting at is that it was probably an engineering decision, not a device security decision.
At these tiny sizes torx bits end up with some very thin material at the tips of the star pattern. The pentalobe has one fewer tips and the curves are inverse, resulting in more meterial in the bit and less in the fastener head. That makes for a more durable tool at the expense of the fastener durability. These screws weren't intended for frequent removal though, so it's a good trade-off from a tooling perspective.
If they were a company that sold equipment to R&D labs only, they would have used common fasterners.