Latest post of the previous page:
Hmm, could debate the difference between thise things that fail because of physics and those that fail due to their complexity being beyond the ability of the human mind to truly predict their performance or vulnerability.Alan H wrote:It's essentially impossible. It's like asking for a car that never rusts or that never breaks down or a plane that never crashes. There have been great strides in making systems robust but it's always an ongoing battle.Dave B wrote:Thanks for that, Alan, will be doing an off-line backup soon!
Wonder if Microsoft will ever develop a truly secure system? Or is thst impossible perhaps? The more vomplex they bevome the more vulnerable it seems, oh, how I remember CP/M . . .
Cars are designed to a cost and for a profit - no material added that ensures that they last forever and, occasionaly, the scrimping and saving causes early failures that should have been expected, usually at a cost to the company and, sometimes, to life. Almost intended errors and failures. By now companies like Microsoft should have virtual AI systems that know every error made so far and can spot possible, similar, vulnerabilities - bit like the new ctime analysis system bring developed for the police. That would cover some problems but not all I admit - no real accounting for human fallibility! Maybe MS should have to run their systems online for a few years before releasing them? Maybe they should have to pay compensation for their design errors that cost their customers?
The more complex the system the more error/failure/hack prone it is. Once I knew what every bit under the car bonnet did and could tune, replace or fix them all...
What really worried me was that the NSA knew of the error but witheld that knowledge so they could possibly exploit it themselves.