This is a great point. Additionally, in a world where there are many competitors with self-driving cars, I can't imagine roads not having and API of sorts to simplify the work cars have to do. Instead of painted line markers, why not IR beacons or something similar. Could be the same for road work signs or temporary barriers.
This issue brings up a larger point that many in the tech world have ignored for awhile. Now that end to end encryption is easily accessible by everyone, how do the authorities figure out beforehand when a terrorist attack will take place? The answer is not banning encryption technologies or giving authorities a 'key' as the FBI director has proposed. But now authorities have a very difficult problem on their hands. Encryption can't be contained, and the authorities need to figure out how to still catch the bad guys.
It takes what it has always taken: good police work.
Encryption is like envelopes for the mail. They can be broken into with enough force, but their purpose it to shield the contents of the message from those handling it in transit. A post card offers no such protection.
So suppose the banned encryption (envelopes). Im sure the government would still reauire envelopes. As would anybody dealing with health records. Or banking, financial stuff has a legitimate need for secrecy. So they ned envelopes.
Pretty soon what you have is a class of people allowed to USE envelopes and a class of people not allowed it use them.
Like envelopes, encryption is an idea, not an inplementation. They may block you from using this tool or that tool, just like a government could make selling pre-made envelopes illegal - but the IDEA of encryption is simple just like the idea of an envelope. Anybody with paper, scissors, and glue can fashion their own envelope. Encyption is the same, one person can dream up an build their own encoder and you can never ban or prevent that idea from being used.
We really need to emphasize the envelope metaphor a lot more. It's much easier for non-technical propel to understand than any discussion of key/signing/etc.
The "smart" bad guys (most aren't, but some are) have been communicating covertly all the time. Avoiding electronic communictation entirely, using dead drops, trusted couriers, etc. You catch them by infiltrating their organizations, turning members into double-agents by bribery or blackmail, etc. the way it's always been done.
That's interesting, I don't remember the jobs of law enforcement and intelligence ever having supposed to been easy. In fact, we nominally have plenty of safeguards meant to make their job difficult. That's the point. If they don't feel like devoting effort to get past barriers, then the case is not worthwhile to pursue.
Catching the bad guys is easier than ever, encryption or not.
It's not like they were particularly good at stopping terrorism before encryption became widespread. The problem is that they think they can stop all terrorist attacks, which is impossible.
The right answer is intelligence gathering in the field, the old fashioned way. It's the only method that's compatible with freedom of expression and the 4th amendment.
The company I work for does something similar. Everyone at the same level gets the same salary, with percentage variations based on geography. It completely removes the competitive and secretive nature I've seen elsewhere. And it also pushes each individual to live up to their title due to the extra transparency.
I work for a large (30K employees) company that does something similar. For each job title, there is a salary and bonus range. So from someone's title in Outlook, you can pretty much guess their salary within 10% or so. This information is all readily available to all employees, though I don't think it is shared externally.
I work for a large company that does this - except it's pretty much useless for guessing someone's salary, because the salary range for each job title are something like $X to $2.25X.
Interesting read. Most blog posts I've seen about the CAP theorem have an academic feel, this one is more approachable. Really, the CAP triangle is outdated. The trade off in a distributed system is between consistency and availability - you choose partition tolerance when you have nodes networked together.
