you're correct that a free workforce can heighten productivity immensly. but that doesn't fly in our current economic model. when using (robotic) slaves, it has only ever truly benefit the rich.
That's only because wealth is not redistributed properly.
eg. If you're Doing It Right then it's impossible to get a "buffer overflow" in C++. Most of the exploits you see are down to buffer overflows so I leave you to draw your own conclusions about the programmers.
Arrays. Arrays are evil. C++ should have skipped arrays and gone directly to std::vector.
So is this correct?
vector<char> a; int i; cin >> i; a[i] = 0;
Let me give you a hint: it's not. It still allows for buffer overflows. Why is the operator [] not bounds-checked is beyond me. They could have had the operator [] safe and the function at() unsafe, if you really wanted such fast access.
The second big part is rvalue references. They really do give major performance benefits for STL containers, especially of user-defined types (as smart implementations of C++03 already optimize their containers for standard types to avoid copy by using swap instead - but they can only legally do it for types they control...).
RValue references do not give any major performance benefits - it's the move operation that gives the performance benefit. RValue references can be emulated by templates in the current version of the language. RValue references is a notation for making move semantics available to the compiler. Move semantics are not magically supported. You still have to code the move constructor and the move assignment operator.
Moving data around is a dangerous practice. Auto_ptr was considered bad exactly because of this - you don't know where your data may end up to. Moving data means more destructive updating. Modern programming language theory is slowly moving away from destructive updating, as it is much easier to do better optimizations when things are not allowed to be destructively updated. I expect that there would be a lot of problems in production code from this, when non-const variables will be passed to functions that reset the variables to their initial state; it would be the auto_ptr fiasco all over again.
But the compiler has all the instantiation tree at hand - it could simply reverse the reporting order and project the missing or incompatible code at the point of the declaration of the template instantiation.
As a long time C++ developer, I can say the lack of proper lambda functions is a big issue, but why do you say rvalues is an equally big issue? I never ever needed something like rvalues, nor have I seen somewhere someone claiming so (unlike lambda functions which is an often requested feature).
Don't hold your breath while waiting for that to happen - it requires a big shift in the way of thinking, in tools and programming languages.
We are still fighting against buffer overflows, for Christ's sake. When one mentions "why not use another programming language then other than C", most people say it's not C's fault, it's the programmer's fault.
With mentality like that, don't expect any progress soon.
The classic mistake in AI is that it is so simple to make a brain, that it has eluded scientists.
All that the brain does is pattern matching. The brain does not run algorithms; it matches experiences from external stimuli to stored ones in the brain and then recalls the response that is stored along the mostly matched experience.
The real technological problem is how to load/store visual/audio/touch/smell/taste experiences in such a way that the process is continuous and quick. Nature has developed an analog mechanism with a very high degree of parallelism which is slow at sequential tasks but very fast in parallel tasks. The brain's pattern matching is extremely fast, because it is performed in parallel.
The fact that the brain grows is not significant. A machine with enough capacity can easily simulate a brain; the machine does not need to grow.
For an experiment of such magnitude, a delay of a few years is not very important...it's way more important to make the experiment in a good way, above anything else.
Is a big spaceship assembled on orbit, with artificial gravity from huge rotating sections, nuclear propulsion and landing craft viable within a trillion dollars budget?
That's what NASA should be planning for. Such a beast would make the trip Mars - Earth a commodity, opening up space for good.
When it comes to space, there is no public or private sector. There is only the human sector. NASA should collaborate with anyone willing to invest enough money for such projects.
Thank you for the explanation. Here is a shorter explanation: using homomorphic encryption, mathematical operations on encrypted data can produce results which are themselves encrypted by the same encryption code.
If the data are stored in computer files, then it would be relatively easy to write a program that copies the file and converts the units from the imperial to the metric system.
You are correct, and you have just revealed the hidden catch of free trade: if salaries are not equal across areas, free trade between those areas does not work.
What is strange is that people do not realize that the melting of individual cultures into one Earth culture is unavoidable. The explosion of communication all over the globe means breaking down the cultural barriers between countries. Using tax breaks for maintaining one's culture may help the culture survive for a few more years, but the melting is inevitable.
Example #1: it may be that each observer sees the order of events differently, but that could not be used to change the state of the other observer. It's only an observational problem.
Example #2: the twin paradox is no paradox at all. The faster an object moves, the slower time passes. It has been proven with atomic clocks.
In conclusion: trips to the future are allowed by traveling at relativistic speeds. Trips to the past are not.
Time travel to the past is not possible - the past does not exist. Alice can sent information to Bob when Bob sees a past image of Alice, but when he tries to affect Alice, nothing will happen: Alice is not there, it's only a ghost image of Alice.
If I had a faster-than-light Star destruction beam and shoot it directly to the Sun, would the Sun be destroyed? nope, because I would have shot my ray not to the current position of the Sun, but to the position of the Sun 8 minutes ago.
Slashdot Mirror
What i d like to see is the USS Enterprise thrown by Q to the planet you describe. It would be extremely funny (for the aliens)...
That's only because wealth is not redistributed properly.
Even your line of reasoning has emotions; the emotion of fear of abandonment.
So is this correct?
Let me give you a hint: it's not. It still allows for buffer overflows. Why is the operator [] not bounds-checked is beyond me. They could have had the operator [] safe and the function at() unsafe, if you really wanted such fast access.
RValue references do not give any major performance benefits - it's the move operation that gives the performance benefit. RValue references can be emulated by templates in the current version of the language. RValue references is a notation for making move semantics available to the compiler. Move semantics are not magically supported. You still have to code the move constructor and the move assignment operator.
Moving data around is a dangerous practice. Auto_ptr was considered bad exactly because of this - you don't know where your data may end up to. Moving data means more destructive updating. Modern programming language theory is slowly moving away from destructive updating, as it is much easier to do better optimizations when things are not allowed to be destructively updated. I expect that there would be a lot of problems in production code from this, when non-const variables will be passed to functions that reset the variables to their initial state; it would be the auto_ptr fiasco all over again.
But the compiler has all the instantiation tree at hand - it could simply reverse the reporting order and project the missing or incompatible code at the point of the declaration of the template instantiation.
As a long time C++ developer, I can say the lack of proper lambda functions is a big issue, but why do you say rvalues is an equally big issue? I never ever needed something like rvalues, nor have I seen somewhere someone claiming so (unlike lambda functions which is an often requested feature).
Don't hold your breath while waiting for that to happen - it requires a big shift in the way of thinking, in tools and programming languages.
We are still fighting against buffer overflows, for Christ's sake. When one mentions "why not use another programming language then other than C", most people say it's not C's fault, it's the programmer's fault.
With mentality like that, don't expect any progress soon.
The classic mistake in AI is that it is so simple to make a brain, that it has eluded scientists.
All that the brain does is pattern matching. The brain does not run algorithms; it matches experiences from external stimuli to stored ones in the brain and then recalls the response that is stored along the mostly matched experience.
The real technological problem is how to load/store visual/audio/touch/smell/taste experiences in such a way that the process is continuous and quick. Nature has developed an analog mechanism with a very high degree of parallelism which is slow at sequential tasks but very fast in parallel tasks. The brain's pattern matching is extremely fast, because it is performed in parallel.
The fact that the brain grows is not significant. A machine with enough capacity can easily simulate a brain; the machine does not need to grow.
For an experiment of such magnitude, a delay of a few years is not very important...it's way more important to make the experiment in a good way, above anything else.
Lot's of resources are spent in experimentation. Stop experimenting and then game development will not exceed one year.
Is a big spaceship assembled on orbit, with artificial gravity from huge rotating sections, nuclear propulsion and landing craft viable within a trillion dollars budget?
That's what NASA should be planning for. Such a beast would make the trip Mars - Earth a commodity, opening up space for good.
When it comes to space, there is no public or private sector. There is only the human sector. NASA should collaborate with anyone willing to invest enough money for such projects.
Why should the human race survive?
Thank you for the explanation. Here is a shorter explanation: using homomorphic encryption, mathematical operations on encrypted data can produce results which are themselves encrypted by the same encryption code.
How is it possible for them to calculate the tax return if they do not analyze the data?
If humanity started 35,000 or more years ago...
If the data are stored in computer files, then it would be relatively easy to write a program that copies the file and converts the units from the imperial to the metric system.
Yes, pretend you are not you. How nice...and soon enough the bike gimmick will wear off...
The best advice is to be yourself, the nice version. There are thousands of women out there that want a good relationship.
You are correct, and you have just revealed the hidden catch of free trade: if salaries are not equal across areas, free trade between those areas does not work.
It does not work. You have to press 'change' in order to get rid of the problems.
Indeed.
What is strange is that people do not realize that the melting of individual cultures into one Earth culture is unavoidable. The explosion of communication all over the globe means breaking down the cultural barriers between countries. Using tax breaks for maintaining one's culture may help the culture survive for a few more years, but the melting is inevitable.
Apparently (according to the article), the cosmological constant can be altered at will.
Example #1: it may be that each observer sees the order of events differently, but that could not be used to change the state of the other observer. It's only an observational problem.
Example #2: the twin paradox is no paradox at all. The faster an object moves, the slower time passes. It has been proven with atomic clocks.
In conclusion: trips to the future are allowed by traveling at relativistic speeds. Trips to the past are not.
Time travel to the past is not possible - the past does not exist. Alice can sent information to Bob when Bob sees a past image of Alice, but when he tries to affect Alice, nothing will happen: Alice is not there, it's only a ghost image of Alice.
If I had a faster-than-light Star destruction beam and shoot it directly to the Sun, would the Sun be destroyed? nope, because I would have shot my ray not to the current position of the Sun, but to the position of the Sun 8 minutes ago.
There is no need for such philosophical twists.