What's worse is that compared to all other disciplines in the sciences, arts, and trades, computer science is not a "science."

In the early days, before PC's, the action was Power Off and then Start. This is a universal solution for many devices today besides the computer. The problem necessitating this action however implemented on keyboards or directly on the computer hasn't changed.

Programming is faulted by a inconsistent, incomplete analysis of the problem domain, and this happens in large part by a lack of understanding for the language being used to "talk-about" things that compose that domain. Mathematicians do not care about what "X" refers to until they apply their conclusions to real things. Then it matters whether "X" refers to locomotives or cans of soup. What needs to be taught to young children is how we get from locomotives to an "X" that always means "locomotives" and not "eggs", "cans of soup"or some other; AND as well, if someone does mean "X" = "locomotives", "X" cannot be arbitrarily changed to mean eggs or cans of soup. What is missing cannot even be taught because it doesn't exist; that is a discipline for the science of designing computer systems like all others in the arts, sciences and trades. Imagine two doctors who have never met from different parts of the world entering an autopsy room. There would be no question that the body resting on the autopsy table was the same one each had learned about in medical school. Put two business people or business analysts in the same room and it would be a scene mindful of debates in our current Congress over what the body should look like.

Correct, before CTL-ALT-DEL there was a reboot key and it was invented by an IBM engineer to do what programming minds are incapable of doing quickly or easily, solving for the "all I did was" problem. Functionally, CTL-ALT-DEL is something done with the clear key on ordinary calculators and functionally in many other things we do, including in our last gasps before leaving this earth. For example, how many backspaces does it take in the course of writing an email to get all the words right--remember how corrections were made using a manual typewriter? Backspacing is a kind of "delete" key and most of us would be lost without it. Rather than to bemoan its existence, it should act as a reminder that we are not perfect in very absolute ways. The three-finger salute is symbolic of Gates' true feelings for his customers and those end users who are held hostage by his death-by-a-thousand cuts mistake-making operating system software, It would be reasonable to believe that he would embed his lack of compassion for people in such a keyboard artifact. His brilliance is in his ability to sell shares in a bridge he doesn't own and play the shell game without a pea. Gates was in large part responsible for aiding and abetting an industry responsible for the other half of Dickens' phrase "It was the best of times, it was the worst of times." What he did mostly is send the programmers who could have eliminated the need for a CTL-ALT-DEL key below deck to row his ship in circles.

Imagine if we had a way to start over each time we made a mistake, clear our heads of all the bad experiences and move ahead without the past baggage. It isn't the CTL-ALT-DEL keys that are important, (first off they were initially designed to reboot the computer), it is the fact that we as programmers design faults in to our systems and the complexity of having so many different programmers designing similar systems that are slightly different fosters anarchy. The real problem that has yet to be overcome is that computer science is not a science in the same way as all other established disciplines in the arts, sciences and trades. Until that happens several hundred years from now, we need a way to reboot from the errors in those experiments we call "programs" or "algorithms" or whatever.

Think of workstations as US citizens to put this problem in perspective. Then as people who live in the US; then in US and Canada, and so on. The complicated problem is more generally stated. Language does not possess the capability of referring to things uniquely. It will ultimately fail if for no other reason than we will forget or confuse the names being used and more likely, run out of names. To assure the uniqueness of the workstation identification, one has to first devise a list of numbers for workstations and then paste each number on the workstation. Since there is more than one list, the minimum number of numbers is 1-n for the lists containing workstations (even assuming all workstations are of the same kind)and 1-n for all other lists. This implies that all things being numbered take their place along side of all other things being numbered. To be perfectly correct, we must model the universe of things so that each is assigned a unique place in a set of all sets of things, numbered as 1.1.1.1.1..... In short, naming conventions are like pumps in boats, necessary else the boat will sink. That is pretty much the state of art for relational database naming conventions et al; a sinking ship.