Gattonero wrote: ..... The pads must be adjusted, this is not a "failure" of the design.
I don't entirely disagree with you, but I do think that many of these designs are extremely poor. Dangerous, even. Plenty of people have been caught out by the various issues that arise, and there have been product recalls. Type b) behaviour (cropping up part way through the pad life in many cases) is pretty much unforgiveable, and is a sudden failure mode. Type a) behaviour can be sudden too, depending on how stiff the cable housings are and how fast the pads wear; you can have good brakes at the top of a hill and no brakes by the bottom.
At one time I had BB5 calipers on both an MTB and a Road bike. Just at the point where the BB5(MTN) caliper works best, the BB5(Road) caliper arm is about to stop dead in its tracks, leaving you with no brakes! There is no indication on the outside of the caliper that this is the case.... this is just plain daft!
there are lots of things you could do, for example
1) you could make is obvious where the arm stops by incorporating a stop in the caliper body, perhaps with a rubber buffer or something. [This would make it comparable to a V-brake, where if it is set badly or the pads wear, the yoke hits the pinch bolt, which is easily seen and understood.] Or
2) part of the caliper body that is usually covered by the arm could be painted bright red or something, so that when the arm is reaching the end of its travel, there is a visible indication that this is happening before it causes trouble.
3) [less related to this particular issue] Many brake pads have locating tangs on them, that protrude from the caliper body. I think that it should be mandatory for the caliper design to be such that it allows the pad wear to be viewed directly, but failing that, there is no reason why the tangs shouldn't be set so that when the pads are worn, the protruding tang ends will touch one another, giving a clear indication of pad wear.
When it comes to making things safe, one of the yardsticks used is to see if the people involved have acted 'reasonably' or not. This means that your approach ought to compare well with that used by others, and furthermore if there are things that you could easily do to improve things, you ought to do them. So (for example) if you make the least safe brake on the market, that could easily be improved upon, you may be deemed not to have acted 'reasonably' in failing to improve it. Judged by this yardstick, the makers of very many disc brake calipers could be viewed as having behaved 'unreasonably' IMHO.