It is easy to blame mass production for repeated product defects - but this really is a failure to check quality at the point of assembly. When a flaw is detected at assembly, the line should be stopped until the defective parts are replaced. As it is, a lot of defective product ends up in the supply pipeline ... and Apple incurs the cost of duplicate quality inspections. Costs multiply.
Friends of mine in QA always tell me the same story, that they are always under pressure to pass the products, no matter what their condition, just so they can ship product. I'm sure there have to be some places that practice good quality control, but I've never met anyone who worked at one yet.
One place I worked at used material that we knew we were going to get 10% returns from. Management was counting on the 90% immediate revenue to replace the 10% whenever it came in.
If it's not working, it's not shipped. If it's not shipped, it's not working. This should be the mantra of everyone that builds and sells things to people.
BrainiacV is right, all too often, companies get stuck focusing on themselves and forget who is really paying their salaries and what is really important to those people.
I don't know the ratio of truth vs. urban legend, but back in the early days of the personal computer industry (1980's), there were rumors of some companies shipping bricks in their boxes instead of products, just so they could book the revenue in that particular month. If that's true, it's pretty much the ultimate statement about not caring about customers.
Not every business is viable, but skimping on quality is not the way to keep it going.
It's not a rumor, the company was called Quantum and the products were hard disk drives. The storage industry then was caught in a vicious development cycle (I know, when isn't it?) and it took too long and cost too much to develop products of the next larger capacity before the prior
product became obsolete and was returned from the warehouse unopened because it was no longer "competitive". So the operations guy decided to "skip a generation" hoping to get the unopened "bricks" back and save the development bucks. He got caught and the company went under along with his management career. And in case you're wondering, no I DIDN'T work there but I was at a competing company at about that time!
Blog Doing Math in FPGAs Tom Burke 3 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...