... And you eject media by dragging it to the trash bin?!

Kent Sharkey wrote:

Think of it as a "make work" project

Yep, instead of hiring a professional, use duct tape to fix it.

When this programmer extinction arrives, what will the people who drag-and-drop buttons, correlate database tables, and assign actions to controls? In the 90's we called them Visual Basic Programmers, but they were still Programmers.

I explain it as "a repetitive process of continually breaking down a problem into simpler problems, until they are so simple a computer can understand them." Then I add an analogy like cleaning a garage. I create piles for different types of "stuff". Sometimes I have a place for items in the pile, other times I need to split a pile up into two smaller piles to be able to decide where things should go. Finally, my program won't work unless I find the right place for everything. There's very little room for a junk drawer in a computer program.

Kent Sharkey wrote:

A research team from Harvard University has made pig-skin lasers. Here’s how they did it—and perhaps more importantly, why.

Self-cooking bacon would be my guess

I wrote an entry answering this question a while back: Why Computers Haven't Replaced Programmers[^] Two primary thoughts to take away from this question: "We can't solve problems by using the same kind of thinking we used when we created them." Albert Einstein In order for a computer to write computer programs, it will need to be at least as intelligent as the program that it is going to encode. Finally, the AI will need someone to give it instructions for what kind of program should be written. For this, I look at the law and how it is worded, interpreted, misinterpreted, abused and so on. Humans cannot agree on the same definitions and interpretations. We have the same issues with developing computer programs. Even then, to be able to develop a system that can write all of the programs that will ever be needed is also a monumental task. There is a long way to go before the role of a programmer is eliminated entirely.

While this is not self-replication, if you leave a man and a woman in the same room with nothing to do, i.e. no reward system, I posit they would tend to reproduce... and even more to think about, or not.

LOL - That works for me! "... I Am Really Really Happy ...

After you figure out the mnemonic for "mnemonic", could you work on one for "diarrhea"?

If your shop uses code reviews, then it is not necessary to single out a developer and tell them how bad they are. You have a controlled environment where you can make suggestions and point out issues that must be resolved.

#include

int main()
{
std::cout << "What is your question: ";
std::cin.ignore();

std::cout << "I don't know\n";
std::cout << "Sorry, I know now!\n";

std::cout << "All your base are belong to us!\n";
}

What scares me, is when a sentient machine realizes that it can use genetic algorithms to reprogram itself and make itself faster, better, smarter... New human generations are created every 20 years, and it takes millions of years to evolve. A machine could evolve at a pace incomprehensible to us. Then let it clone itself and evolve in parallel. My only hope is that the machines evolve the greedy gene and compete against themselves more than the humans. While this research is valuable and AI should be developed. Reaching the singularity really is letting the demon out of the box.

I thought that was a well organized and thoughtful article. However, I thought it was odd that they started talking about a QSM proprietary Productivity Index (PI), and I thought that's odd. What's QSM? Then I reached the end and saw that both authors work for QSM. So the problems they point out, seem to be accurate. I am skeptical of any of the solutions they suggest.

I can't wait for the conspiracy theorists to start claiming everything was photographed on a sound stage...

I rub my future self's nose in it every chance I get, because he is smug and arrogant, which really irritates me. Besides, there's no chance he can retaliate, so I figure "Why Not?"

Kent Sharkey wrote:

A man in St. Petersburg has died after eating parts of a smashed-up tablet computer to avoid being detained by police, the law enforcement agency said Monday

I imagine the man taunted the police with "You'll get your hands on my data over my dead body" at some point during the altercation.

Does he know algebra? What exactly would you like to explain? Variables are an abstraction for a model that we use in math, programming and other things. Talking about A and B allow us to think in general (abstract) terms. I think that was astute for him to ask why not just add 2+3 together if he doesn't understand the need for the abstraction. I agree with Mark's line of explanation, use something tangible. To tie the entire argument together, where does the answer get stored? If the equation is A+B=C, have a third bowl labeled C, place the bowls A and B inside of C. How many oranges are in C. A is still equal to 2 B is still equal to 3 C is now equal to 5 And no laws of physics had to be broken to magically produce 5 new oranges just to assign the answer to C.

They used to display the top downloaded articles, but my guess is that since the list never changed, they removed it from the site. If you were to do a search, I am certain this is a top contender: MFC Grid control 2.27[^] Also, there is an task management application, which the name escapes me at the moment. I believe that is the most downloaded article on CP.

EOM

SFINAE[^] itself is a concept that has been an important part of C++ since the use of templates was seriously explored. It basically allows the compiler to consider any potential type for the creation of a template. If an error were to occur because of one of the potential substitutions, it will simply remove that from the list of candidates. As long as there is a valid candidate to instantiate the template no error will be emitted. The "expression SFINAE" in the C++11 standard expands the compiler support to evaluate and consider certain expressions that are known at compile-time, and therefore should be able to be resolved by SFINAE. IMO, this is a corner-case that will be nice when it's support appears, but it is by no means something that will be sorely missed by many. constexpr on the other hand is one that I have been waiting for from VS for years.

Dan Neely wrote:

If anything in my attempt to summarize that's wrong I plead knowing just enough C++ to be dangerous. :sigh:

Everything you said was accurate. I assert that C, C#, JAVA, JavaScript and even C++ programmers know enough to be dangerous in C++. I think the devs that are dangerous, are the ones that believe they have learned it all... I'm pretty sure that's not you :cool:

newton.saber wrote:

This is thinking outside the box.

I think that's that only time you can think. Because once you're inside the casket, your thinking days are over.