That's why you're the only one using the term in that way. The failure is IMO on your end for using the term "computer driven" incorrectly. Self driving (primitive systems were) can be completely analog. They tested driverless cars in Nebraska 60 years ago.

Probably those were "analog computers" that controlled the cars in those experiments. I recall a single course in analog computers, long after we were introduced to Fortran and writing code. But before the feedback systems three terms. So certainly an analog system could keep a car in it's lane very well, but pity whoever rides a bike close next to that lane.

 

You will notice that nobody except me calls them "Computer driven" cars. That is very deliberate, because we are mostly aware of how often computers are either unable to handle an exception, or else they just totally freeze for a while.
And of course, with the weekly required updates, who wants a car that does things differently every week. Even when it might be an improvement.

Hi,

Yes that's interesting and more to the point.

Let's just hope that none of them have to reboot during a time when they are in heavy traffic on a four-lane highway all moving at 70mph
Then on the other hand, they probably cause enough accidents that the traffic moves to a crawl, thus allowing them plenty of time. Maybe that's their game plan all along
Can't help but joke about some of this stuff.

 

There is another concern with computers driving cars, which is the intention of some folks to have them all communicating with both each other , AND the infrastructure.
That has multiple, show-stopping, issues. FIRST, that communication would demand 100% perfect performance 100% of the time, and SECOND, it would open the doors to all kinds of "bad actors." In addition to that, it would allow the government to decide who could drive where, and when.
Even the very best communication systems are not quite that good, even with dedicated upkeep. Which is not cheap. HOW would it stay functional when it was part of every vehicle on the roads?? WHO would pay for THAT!!!

 

There is another concern with computers driving cars, which is the intention of some folks to have them all communicating with both each other , AND the infrastructure.
That has multiple, show-stopping, issues. FIRST, that communication would demand 100% perfect performance 100% of the time, and SECOND, it would open the doors to all kinds of "bad actors." In addition to that, it would allow the government to decide who could drive where, and when.
Even the very best communication systems are not quite that good, even with dedicated upkeep. Which is not cheap. HOW would it stay functional when it was part of every vehicle on the roads?? WHO would pay for THAT!!!

Hi,

That's also an interesting take on this.

However, the government would be a proxy to the companies making the hardware and software, and you know what that means. It means now all of a sudden road surface area becomes a commodity with priorities depending on who is paying how much and where. We could end up with Verizon owning Rt 287 or some other route in your area, and forcing cars that don't pay as much to drive on the backroads.
The question is, could it make traffic jams less frequent. Some of highways around here get backed up every single day so instead of going 60mph your lucky to be going 5mph, no kidding.

 

I am not clear about how" the government would be a proxy to the companies making the hardware and software, ". Although what it would take to provide "100% reliability" would probably demand the government participation.
Of course, the whole concept of "total communication for safety" is really quite stupid, given that the very best communication schemes are both limited and costly to maintain. Just ask any good IT professional about what it takes to keep a well designed medium size, continuously maintained, wireless communications system at 100% functionality.
Then try to imagine the effort and resources that it would take to even make a network of all vehicles functional, much less reliable. So that pert of the concept of having a computer driving is a non-starter.

Then, consider the claimed benefit for all of improved safety. The fact is that most drivers are never involved in any collisions, and that only a quite small portion of drivers are involved in serious collisions. So there is not much cost benefit to the majority of drivers. Not one bit.

 

I am not clear about how" the government would be a proxy to the companies making the hardware and software, ". Although what it would take to provide "100% reliability" would probably demand the government participation.
Of course, the whole concept of "total communication for safety" is really quite stupid, given that the very best communication schemes are both limited and costly to maintain. Just ask any good IT professional about what it takes to keep a well designed medium size, continuously maintained, wireless communications system at 100% functionality.
Then try to imagine the effort and resources that it would take to even make a network of all vehicles functional, much less reliable. So that pert of the concept of having a computer driving is a non-starter.

Then, consider the claimed benefit for all of improved safety. The fact is that most drivers are never involved in any collisions, and that only a quite small portion of drivers are involved in serious collisions. So there is not much cost benefit to the majority of drivers. Not one bit.

Hi,

What I meant by the gov/proxy idea is that at first it looks like the government is controlling the technology so that big tech has to follow the rules, but then big tech asks for favors and suddenly it's big tech making all the rules anyway. This is the state of the Union now apparently.

 

Regardless of who sets whatever standard, I see no way that inter-vehicle communication would ever be reliable enough to provide any believable safety function, if any realistic cost were possible.

 

Regardless of who sets whatever standard, I see no way that inter-vehicle communication would ever be reliable enough to provide any believable safety function, if any realistic cost were possible.

Hi,

I agree with you in part because they don't do things right in many cases. However, I also try to convince people not to try to tell the future because the future is always unknown even when it looks completely sure. There could be some groundbreaking technology that comes about that makes things that were impossible suddenly possible. And to that end, it's impossible, utterly and completely, to predict that.

Not sure if you knew this or not, but the very first airplane flight was ignored because at the time everybody thought that there could be no such thing as "heavier than air" flight. If I ever go back in time (ha ha) I'll show them a picture of a 747 or perhaps the Space Shuttle which would really rattle their cages.

I don't know what will come about in the future, but I am not liking quantum computing very much so far because they are taking so long with getting a powerful home unit.
Not as much as self-driving cars however

 

Hi,

I agree with you in part because they don't do things right in many cases. However, I also try to convince people not to try to tell the future because the future is always unknown even when it looks completely sure. There could be some groundbreaking technology that comes about that makes things that were impossible suddenly possible. And to that end, it's impossible, utterly and completely, to predict that.

Not sure if you knew this or not, but the very first airplane flight was ignored because at the time everybody thought that there could be no such thing as "heavier than air" flight. If I ever go back in time (ha ha) I'll show them a picture of a 747 or perhaps the Space Shuttle which would really rattle their cages.

I don't know what will come about in the future, but I am not liking quantum computing very much so far because they are taking so long with getting a powerful home unit.
Not as much as self-driving cars however

Understand that the "self" in a "self driving car" IS A COMPUTER running a program of several million lines of code. Then, know that those lines of code have not been properly validated.

 

Understand that the "self" in a "self driving car" IS A COMPUTER running a program of several million lines of code. Then, know that those lines of code have not been properly validated.

They're validated by the beta testers.

Only 83 deaths through 2024.

 

They're validated by the beta testers.

Only 83 deaths through 2024.

NO!!! They are checked to see that they work , WHICH IS NOT EVEN CLOSE TO AN ACTUAL VALIDATION. An actual validation,as done with NASA spacecraft software, is a line by line examination checking what each line does. That may take up to an hour per line for an examination by those who actually know what all of the instructions actually do. Masters, not "programmers".

 

NO!!! They are checked to see that they work , WHICH IS NOT EVEN CLOSE TO AN ACTUAL VALIDATION. An actual validation,as done with NASA spacecraft software, is a line by line examination checking what each line does. That may take up to an hour per line for an examination by those who actually know what all of the instructions actually do. Masters, not "programmers".

Oh! Yeah. They did that with AI.

No worries.

 

Hi,

I agree with you in part because they don't do things right in many cases. However, I also try to convince people not to try to tell the future because the future is always unknown even when it looks completely sure. There could be some groundbreaking technology that comes about that makes things that were impossible suddenly possible. And to that end, it's impossible, utterly and completely, to predict that.

Not sure if you knew this or not, but the very first airplane flight was ignored because at the time everybody thought that there could be no such thing as "heavier than air" flight. If I ever go back in time (ha ha) I'll show them a picture of a 747 or perhaps the Space Shuttle which would really rattle their cages.

I don't know what will come about in the future, but I am not liking quantum computing very much so far because they are taking so long with getting a powerful home unit.
Not as much as self-driving cars however

"Heavier-than-air flying machines are impossible." - Lord Kelvin

Not impossible due to the physics of powered flight, impossible with the current engine technology for the foreseeable future. Most of the things people say today are impossible are due to current physics, not technology or any possible technology using current physics (classical or QM). The limitation is not advances in technology, we would need new physics of a radical (changing basic laws like conservation of energy and momentum) nature for most of the impossible things to happen today.

We also need these new physics of a radical nature to be completely compatible within the limits of current classical and QM physics.

From a reddit article: https://www.reddit.com/r/AskHistorians/comments/nm97ai

Heavier-than-air flight had been “around the corner” for decades, vaunted breathlessly in books and magazines, with little to show for it. Technology was hitting a wall. That people had doubts was thus not surprising.

Of the several quotes claiming the heavier-than-air flight was impossible, that of Joseph LeConte is particularly interesting, because he wrote a long paper in The Popular Science Monthly explaining his views. Six years later, he write another paper in the same magazine where he partly retracted them.

The core argument of LeConte in 1888 is actually sensible (we should note that he was a geologist, not an engineer or a physicist). Powered flight requires energy, and birds, through millenia of evolution (he was a strong proponent of Darwinism), have become perfect flying machines. Not only their bodies are fully adapted to flight, but they are supremely efficient in turning food energy into mechanical energy.

See how this machine has been gradually perfected throughout infinite ages, especially in birds. During the whole geological history of the earth this machine has been steadily improving in structure of skeleton, energy of muscle, and rapidity of combustion of fuel, by struggle for life and survival of only the swiftest, the most energetic, and the hottest-blooded, until an almost incredible intensity is reached in birds. Moreover, in them everything is sacrificed to the supreme necessity of flight.

But yet, LeConte concludes, “this machine thus perfected through infinite ages by a ruthless process of natural selection, reaches its limit of weight at about fifty pounds!” It was thus impossible to make an engine powerful enough to propel a flying machine that would weigh at least 300-400 pounds (he was not too far off: the Wright Flyer weighed 600 pounds unoccupied).

In 1894, LeConte published a new article in Popular Science. By then, he had been in contact with Samuel Langley, from the Smithsonian, and had heard about the experiments of Hiram Maxim in London. These attempts convinced him that 1) the limit of weight could be raised above 50 pounds, 2) that an artificial engine could be more efficient at producing force than an animal body, and 3) that new theories in flight mechanics showed that it was possible to sustain flight for a machine, “even one of great weight”. He concluded by retracting his previous opinion:

In the distant future, and by means of such gradual approaches, the engineering difficulties in the way of a true flying machine may be finally overcome.

LeConte was even more optimistic after talking with Langley a few months later.

Neither Langley nor Maxim were successful and their projects were seen as failures. Langley's unmanned flights proved that heavier-than-air flight was feasible, but his manned flights ended up with his test pilot taking unwanted baths in the Potomac in 1903, much to the mirth of the US press. Maxim’s enormous “captive flying-machines” did achieve lift-off: Maxim once found himself “floating in the air with the feeling of being in a boat” for a few seconds. However, their steam engine consumed considerable amounts of water, making them useless for actual flight (Maxim, 1915).

As for Lord Kelvin, he had actually been able to take a "flight" in Maxim's flying machine in July 1894. He had been unconvinced, calling it a "kind of child's perambulator with a sunshade magnified eight times". According to his biographer S.P. Thompson (Thompson, 1911):

He did not believe in the aeroplane and thought that the problem of flight might be better solved with a platform having a vertically working propeller at each corner.

Leaving aside the fact that Kelvin casually invented our modern quadcopters, this shows how difficult it was to convince people with such immature technology. Maxim wrote about this period:

Many years ago scientific engineers said, "Give us an engine that will develop the power of a man and will not weigh more than a barn-yard fowl, and we will very soon give you a flying-machine."

But that was not yet possible, and Maxim tried to recoup his losses by turning his flying machines into fairground rides in Blackpool and Sydenham.

Simon Newcomb’s quote from 1903 is particularly damning since it was made a few weeks before the Wrights’ flight. He was not totally negative, as he also wrote in the same article:

Quite likely the twentieth century is destined to see the natural forces which will enable us to fly from continent to continent with a speed far exceeding that of the bird. But when we inquire whether aerial flight is possible in the present state of our knowledge; whether, with such materials as we possess, a combination of steel, cloth and wire can he made which, moved by the power of electricity or steam, shall form a successful flying machine, the outlook may be altogether different.

His text came right after Langley’s Aerodrome’s plunge in the Potomac, and he referenced Maxim’s failures. Newcomb, a mathematician and astronomer, ignored the concept of airfoil, and he completely failed to take into account the developments of the combustion engine (Langley, Maxim, and Ader had used steam engines). As Newcomb’s daughter wrote in his defense in 1919, after he had been mocked for his ill-timed prediction:

The mistake, and the only mistake, which Newcomb made consisted in doubting the ability of inventors to overcome such a great difficulty so soon, but this, while it produced a too conservative viewpoint, was not a scientific matter at all.

 

Understand that the "self" in a "self driving car" IS A COMPUTER running a program of several million lines of code. Then, know that those lines of code have not been properly validated.

Hi there,

I am not sure what you are trying to say here. All I am saying is that in the FUTURE we don't know what will come up. For the time in the here and now, I have to agree I don't like it either.
The future is just too hard to predict. Some things work out, some things don't. The more complicated things will be questioned more severely, and false judgements both pro and con will come up, and in the end only one will come out the winner. That's pretty much where our understanding lies for now, until we get time travel to the future maybe if that is ever possible.

It's kind of like a horse race where everybody thinks they know which horse will win, but then at the finish they get a surprise
Right now we are like in the middle of the race somewhere, where one horse may be a little ahead of another, but we still can't be sure which one will win.

We all have our suspicions about what will fly and what will not fly. To say with absolute certainty which is which is just an attempt to prove psychic abilities and does not say anything certain about the future (psychic in this context means someone that can foretell the future, or at least believes they can) .

I could make a list of the things I think might happen in the future, but I'd be a nut if I tried to push any of them with absolute certainty.
We do sometimes talk in the "future sense" in that we say what we believe at the moment, and I don't really knock that too much because it's a more casual thing. For example, "Oh yeah, that's NEVER going to happen!"

Some predictions are qualified, and that seems to be an actual good type of prediction. For example, we might say that more bridges in the United States will collapse within the next 20 years, but it's better if we say more bridges in the US will collapse within the next 20 years IF WE DO NOT DO MORE WIDESPREAD INSPECTIONS.
I think that kind of statement can be very helpful.
In the case at hand, we might say that really good quantum computing will not come about unless the technology advances.
In the case of self-driving cars, we might say that they may never be completely safe unless we improve the programming.
It is kind of interesting that there is some quantum computing going on already.
As to the future, and again this is not an outright prediction but a guess, would it be possible to have a self-driving car that is extremely safe with the use of 'ai' combined with quantum computing.

 

For the case of computer driven cars, for the foreseeable future, the requirements exceed the capabilities by a large amount. THAT is the current situation. In addition, the costs presently exceed the benefits by a large amount, considering that the benefits to most people would be minimal at best.
IN ADDITION, presently the ability o produce bug-free software at a price that would be accepted simply does not exist. The testimony to that reality is the current proposed concept of automated updates for the software running the computerized cars.!!

The very acceptance of the concept of repeated automated updates admits that the software is not anticipated to be adequately reliable. So compare the inconvenience of a computer crash with the "inconvenience" of an automotive crash.

Then consider the probability of producing a few million lines of totally bug-free code.

 

Quantum computing has a very narrow range of usefulness over classical computing today and likely in the near future. The real usefulness of current Quantum computing research is finding ways to use classical computers to solve previously thought to be intractable problems.
https://www.simonsfoundation.org/20...uter-beat-a-quantum-computer-at-its-own-game/
The Surprising Reason a Classical Computer Beat a Quantum Computer at Its Own Game

The results of one recent test of quantum computers came out in June 2023, when IBM researchers published a paper in the journal Nature. Their paper detailed an experiment simulating a system with an array of tiny flipping magnets evolving over time. The researchers claimed that this simulation was only feasible with a quantum computer, not a classical one. After learning about the new paper through press coverage, Tindall decided to take up the challenge.
Tindall has been working with colleagues over the last several years to develop better algorithms and codes for solving complex quantum problems with classical computers. He applied these methods to IBM’s simulation, and in just two weeks he proved he could solve the problem with very little computing power — it could even be done on a smartphone.
“We didn’t really introduce any cutting-edge techniques,” Tindall says. “We brought a lot of ideas together in a concise and elegant way that made the problem solvable. It was a method that IBM had overlooked and was not easily implemented without well-written software and codes.”
Tindall and his colleagues published their findings in the journal PRX Quantum in January 2024, but Tindall didn’t stop there. Inspired by the simplicity of the results, he and his co-author Dries Sels of the Flatiron Institute and New York University set out to determine why this system could be so easily solved with a classical computer when, on the surface, it appeared to be a very complex problem.
“We started thinking about this question and noticed a number of similarities in the system’s behavior to something people had seen in one dimension called confinement,” Tindall says.

 

https://jackkrupansky.medium.com/wh...r-be-able-to-fulfill-most-of-the-d8f45a47b221

Never say never = 1 in 1,000 chance
Even when I make a fairly definitive statement, I tend to offer the caveat of my philosophy of never say never.

What I mean by never say never is that there may be very little chance, but there is at least some chance. How much of a chance? I use the model of never say never meaning no more than a 1 in 1,000 chance. Significantly less than a 1% chance. But still at least some possibility.

Even so, sometimes never really is never. It’s just that we can never know that in advance with any absolute sense of certainty.

 

https://jackkrupansky.medium.com/wh...r-be-able-to-fulfill-most-of-the-d8f45a47b221

Never say never = 1 in 1,000 chance
Even when I make a fairly definitive statement, I tend to offer the caveat of my philosophy of never say never.

What I mean by never say never is that there may be very little chance, but there is at least some chance. How much of a chance? I use the model of never say never meaning no more than a 1 in 1,000 chance. Significantly less than a 1% chance. But still at least some possibility.

Even so, sometimes never really is never. It’s just that we can never know that in advance with any absolute sense of certainty.

1. Practical quantum computing is the ultimate goal. But we are not even close.
2. Expectations for significant quantum advantage. And deliver dramatic business value. Again, we are not even close.

The article is two years old. Would you say that the previous two statements are still valid?

 

1. Practical quantum computing is the ultimate goal. But we are not even close.
2. Expectations for significant quantum advantage. And deliver dramatic business value. Again, we are not even close.

The article is two years old. Would you say that the previous two statements are still valid?

What's actually changed in those two years (thread started Dec 3, 2018 ) about the actual ability of current quantum computing for solving practical problems over classical computers?

https://www.sciencenews.org/article/quantum-computing-milestone-challenged
A quantum computing milestone is immediately challenged by a supercomputer
There’s no clear quantum advantage yet in tests of real-world problems

The tug-of-war between quantum computers and classical computers is intensifying.

In just minutes, a special quantum processor, called a quantum annealing processor, solved a complex real-world problem that a classical supercomputer would take millions of years to complete, researchers claim March 12 in Science. And that supercomputer, the team reports, would consume more energy to run the whole computation than the entire globe uses in a year. However, another group of researchers claims to have already found a way for a classical supercomputer to solve a subset of the same problem in just over two hours.

The QC research is pushing the limits of classical computers to handle the same types of problem, sometimes faster than the original QC or at least makes the time it takes to solve it, reasonable.

The new, conflicting results follow similar claims made in recent years. The nascent field of quantum computing has been advancing in lockstep with techniques to make supercomputers more efficient, resulting in a closely matched rivalry. While quantum computers have demonstrated the ability to solve truly random problems faster than classical computers, they have yet to come out on top for physical problems relevant to real-world systems.
...
Quantum computer scientist Joseph Tindall of the Flatiron Institute in New York City and colleagues simulated part of the same problem using a classical computer. They developed a method that repurposed a 40-year-old algorithm called belief propagation, commonly used in artificial intelligence. Their results, submitted to arXiv.org on March 7 but not yet peer-reviewed, claim to be more accurate than the quantum computer’s for certain cases of the two- and three-dimensional systems.

 

Having seen what I think may be an example of quantum computing in post #516, I am wondering three things: First, who would create the software for quantum computations?, second, How would any who wanted to use it, define the data to be entered?, and third, how would the results be understood??
And now the brutal question: Just what actual value could this actually provide????