Re:Apple humiliates Gmome, KDE, Linux
on
Ask Robert Young
·
· Score: 1
Apple was completely unable to create a successor for old, old Apple OS for a very, very long time.
But what do you expect from a company that is headed by a Coca Cola manager?
I do not want to badmouth the software that Apple effectively shipped. The problem is with the software that they were unable to ship. Especially their next generation OS.
That pretty much all Apple users are still working with a cooperative "multitasking" OS is a shame.
OS X is too late.
Its a shame that Apple will slowly die. But they will. Microsoft is aggressive, Apple is not. And OS X is far from revolutionary.
And I do agree with you about the open source user interface design (most of the time).
Apple humiliates Gmome, KDE, Linux
on
Ask Robert Young
·
· Score: 1
because Apple is known for being really bad in creating software.
And nevertheless, they managed to create a UNIX based system that any beginner can use. Many considered this to be impossible.
In the light of this, now compare Apple OS X to you favorite Red Hat or SuSE or whatever Linux distribution (which usually includes KDE and Gnome).
See the point?
Red Hat, SuSE and the like completely failed to make Linux attractive to the masses. They didnt even try.
Linux has no future on the desktop.
This is a simple fact.
Open Source is governed by stupidity and egomania. For this reason, the open source people cannot compete with Apple or Microsoft. They may be more bright, but they are too stupid to overcome their misguided ego.
How odd; I actually got an intelligent response on Slashdot...
Well, Slashdot surely is not the perfect place for such discussions.
Your point that an artificial being can only exist if its host computer takes an interest in it is valid, but it applies to humans also. We cannnot live, and cannot be considered self aware, if our bodies do not support us in some way.
Modifying the support function of (our physical) body would effectively be a modification of the state of the object/person/self awareness, or its conciousness. This differs from the scenario I painted, since there the state of the AI is not modified in any way. Lets just assume that the AI has not input data channel from the outside, just an output data channel. No state change from the outside is possible then.
You also make another big leap. You just assume that anything that "can intentionally be killed without outside influence to it, at a specific time, and even without that 'artificial being' being able to notice any difference" is not self aware. Why not? Certainly it is not self aware after bring killed, but then again neither are we.
The interesting question is: Can something be equivalent to a human person, if its complete state, when the being does exist, is exactly the same, as when the being does not exist?
The problem is, the word "killed" does not really seem to make sense if the state of the object is not altered (which is impossible anyways if there is no input data delivered to this object/subexpression).
Furthermore, this artificial being's state is changed if you consider the entire computer system as a part of that being.
Yes, however, the computer may hold all kinds of additional objects, and we can isolate these from each other and the outside, and just leave some specific data connections in place. We have different beings within the computer then.
Of course, one could still state that anything, which is connected by information links, has to be viewed as parts of one entity/being. However, since the data link to the outside world is required, we would have to include this, too, and probably would end up with the entity "universe" then. This likely would lead to some kind of "information based pantheism".
So the string of bits is still there, but it isn't doing anything
When looking at the complete "mathematical equation" (the complete computer program, which includes the AI), then, after "loosing interest", the data comprising the AI no longer exists.
More precisely: The "equation" (complete computer program) that includes the AI is logically identical to the one that has the AI eliminated.
E.g., if a Java program cuts the last reference to an object ("loses interest", now and forever), then the Java virtual machine will simplify the equation. There is no recovery possible (otherwise it would be illegal to simplify the equation).
Our reality was not always the way it is now, how do we know it won't suddenly smash down to one particle?
This would be modifying the object (us). There is no logical contradiction then, since killing someone by modifying him is easy:)
In my scenario, the object/AI is not modified in any way. The outside is modified (it cuts the data flow from the object/AI). This is killing someone without changing his state (which seems logically contradictory).
How does any of this even relate to intelligence?
Interesting question:)
From my point of view it relates to a very strict but common definition of artificial intelligence: An AI can be all what a human person can be.
Another conclusion from the scenario is: For AIs, existence and nonexistence is the same.
The problem with causality, as I anderstand it, is that (most) physicists believe that the description of nature (at subatomic level) given by current quantum physics is (reasonably) complete. Specifically, that a particle has no location, but instead only the probability distribution curve/wave of a location.
But when the particle is observed, it is found at a specific place. However, without a _causal_ reason. Admittedly, thats sounds pretty strange.
The following page looks like a reasonably good introduction: Quantum Primer (http://www.sfu.ca/chemcai/QUANTUM/Quantum_Primer. html)
Quotes of the relevant parts:
"Q13. Exactly what is it that is "waving"?
We pointed out earlier that a wave is a change that varies with location in a periodic, repeating way. What kind of a change do the crests and hollows of a "matter wave" trace out? The answer is that the wave represents the value of a quantity whose square is a measure of the probability of finding the particle in that particular location. In other words, what is "waving" is the value of a mathematical probability function.
Q14. What is the uncertainty principle?
In 1927, Werner Heisenberg proposed that certain pairs of properties of a particle cannot simultaneously have exact values. In particular, the position and the momentum of a particle have associated with them uncertainties x and p given by [...]
As with the de Broglie particle wavelength, this has practical consequences only for electrons and other particles of very small mass. It is very important to understand that these "uncertainties" are not merely limitations related to experimental error or observational technique, but instead they express an underlying fact that Nature does not allow a particle to possess definite values of position and momentum at the same time. This principle (which would be better described by the term "indeterminacy" than "uncertainty") has been thoroughly verified and has far-reaching practical consequences which extend to chemical bonding and molecular structure.
Q15. Is the uncertainty principle consistent with particle waves?
Please see my other answer a (few minutes ago, "being digital has problems").
But regarding the subatomic level, physicists are pretty much sure that on this level predictability and causality do not exist. All what the (well proven) theory allows to exist, are probability distributions for the observation of specific properties of subatomic particles (like location, speed, etc.).
The free will issue is one of the questionable but inevitable conclusions of the idea that computer programs can be alive, or that human conciousness can be the product of a computer program.
Another such conclusion is that humans then are machines that are completely defined by an initial state and the subsequent input data.
This effectively means that a human is identical to a static string of bits of limited size (which simply contains initial state + subsequent input data). This surely looks philosophically interesting.
So, we have the "string of bits" in the computer, or, alternatively an "intelligent" subsystem that creates that string of bits "live" from outside input.
The specific problem here is (beyond others), that with this picture we have no handle to reality. This string of bits is only "real" relative to the computer it is stored.
The "string of bits" can only exist if there is an outside interest in it. Because, by inherent necessity of computer mathemathics/logic, for the state of that "string of bits", its existence and nonexistence are identical if the computer loses the "interest" in it.
"Losing interest", for a computer, means deleting the last reference to the data object (think Java). Now, in computer logic, the object which represented the "artificial intelligent being", can be removed without modifying the meaning of the overall formula/expression.
The result is: An "artificial intelligent being" can only exist as long as there is an outside interest in it (like a Java reference to the object). As soon as this "interest" vanishes, existence and nonexistence become logically identical.
Another logical problem here is that an "artificial intelligence" can, as shown above, brought from existence into nonexistence without changing its state (from the outside, we dont change a single bit of the objects state).
Should we call something a person, if it can intentionally be killed without outside influence to it, at a specific time, and even without that "artificial being" being able to notice any difference?
Not only that it cannot notice a difference, there effectively is no difference for the "being", as its own state was not changed at all.
The interesting approach of MojoNation is to split each file in many small parts, and then store these with RAID5-like redundancy on many different machines.
This is the way to go, since it properly abstracts all user-machines as a large but unreliable block storage.
It also allows, e.g., a client on DSL/Cable to saturate its bandwith even if it downloads a file from modem owners (it just uses many of them concurrently). This again is similar to RAID5.
Unfortunately, in my experience, MojoNation is still too early in development to be usable. It also seems to be too centralized.
The 2GHz part was a handpicked chip, cooled like hell, and is far from being available.
The 850 mW number is measured "the Intel way", and therefore some considerable spindoctoring is involved.
Of course one can buy into the Intel marketing, but I prefer to spare my enthusiasm until I see that stuff for real, in volume, and tested by independent and reliable publications.
What ct found in their test lately (using the ISO patch LAME version) was that in all disciplines LAME is as good or even slightly better than Fraunhofer.
Except for speech at 32kbit/s, where Fraunhofer was better (I suppose we can live with that).
Bottom line: Forget about all other encoders, and use LAME instead. Your (and potentially other peoples) ears will thank you for this!
Ok. I must accept that a matter of personal preference.
But I also want to state that in my opinion this preference is wrong as it allows to make the code proprietary which I consider abusing it...
If you have to decide about the license for =your= code, this is an entirely different matter (compared to bitching about the GPL of others code, which can regularly be seen on slashdor).
ok, the title looks a lot like offtopic, and you may rightly ask how Ill try to get on-topic again.
look:
something that Linux lacks right now is a log-structured filesystem (the newest Solaris has, if Im not misinformed). I consider this a requirement for an absolutely mission critical system. But Linux, the free software, is rapidly closing in to Solaris, its commercial counterpart (journaling reiserfs, later log-structured filesystems, too). It wont be too long until it is on par with Solaris.
Now, with Postgres I see the same. Its a lot less on par with Oracle that Linux is with Solaris. Perhaps this has to do with the fact that more people want a free Unix than a free DB. However, this is changing, since DBs get more and more important.
The point is: I suspect that Postgres takes the same course as Linux: The free software will get on par with its commercial counterpart. Postgres will just arrive there a little later.
Slashdot Mirror
remember, B&W is no high twitch game.
Apple was completely unable to create a successor for old, old Apple OS for a very, very long time.
But what do you expect from a company that is headed by a Coca Cola manager?
I do not want to badmouth the software that Apple effectively shipped. The problem is with the software that they were unable to ship. Especially their next generation OS.
That pretty much all Apple users are still working with a cooperative "multitasking" OS is a shame.
OS X is too late.
Its a shame that Apple will slowly die. But they will. Microsoft is aggressive, Apple is not. And OS X is far from revolutionary.
And I do agree with you about the open source user interface design (most of the time).
because Apple is known for being really bad in creating software.
And nevertheless, they managed to create a UNIX based system that any beginner can use. Many considered this to be impossible.
In the light of this, now compare Apple OS X to you favorite Red Hat or SuSE or whatever Linux distribution (which usually includes KDE and Gnome).
See the point?
Red Hat, SuSE and the like completely failed to make Linux attractive to the masses. They didnt even try.
Linux has no future on the desktop.
This is a simple fact.
Open Source is governed by stupidity and egomania. For this reason, the open source people cannot compete with Apple or Microsoft. They may be more bright, but they are too stupid to overcome their misguided ego.
Its pretty interesting to be unsure if one is bored or fascinated, throughout a whole movie.
This itself probably was the aspect that fascinated me most when I saw 2001 for the first time...
What impresses me nowadays is how very modern and smooth the visuals of this 33 year old movie are.
Anyway, this move surely has its place in my very own hall of fame.
surely no LSD era movie, but its script looks very much drug induced...
Must have been a really bad trip.
Well, Slashdot surely is not the perfect place for such discussions.
Your point that an artificial being can only exist if its host computer takes an interest in it is valid, but it applies to humans also. We cannnot live, and cannot be considered self aware, if our bodies do not support us in some way.
Modifying the support function of (our physical) body would effectively be a modification of the state of the object/person/self awareness, or its conciousness. This differs from the scenario I painted, since there the state of the AI is not modified in any way. Lets just assume that the AI has not input data channel from the outside, just an output data channel. No state change from the outside is possible then.
You also make another big leap. You just assume that anything that "can intentionally be killed without outside influence to it, at a specific time, and even without that 'artificial being' being able to notice any difference" is not self aware. Why not? Certainly it is not self aware after bring killed, but then again neither are we.
The interesting question is: Can something be equivalent to a human person, if its complete state, when the being does exist, is exactly the same, as when the being does not exist?
The problem is, the word "killed" does not really seem to make sense if the state of the object is not altered (which is impossible anyways if there is no input data delivered to this object/subexpression).
Furthermore, this artificial being's state is changed if you consider the entire computer system as a part of that being.
Yes, however, the computer may hold all kinds of additional objects, and we can isolate these from each other and the outside, and just leave some specific data connections in place. We have different beings within the computer then.
Of course, one could still state that anything, which is connected by information links, has to be viewed as parts of one entity/being. However, since the data link to the outside world is required, we would have to include this, too, and probably would end up with the entity "universe" then. This likely would lead to some kind of "information based pantheism".
When looking at the complete "mathematical equation" (the complete computer program, which includes the AI), then, after "loosing interest", the data comprising the AI no longer exists.
More precisely: The "equation" (complete computer program) that includes the AI is logically identical to the one that has the AI eliminated.
E.g., if a Java program cuts the last reference to an object ("loses interest", now and forever), then the Java virtual machine will simplify the equation. There is no recovery possible (otherwise it would be illegal to simplify the equation).
Our reality was not always the way it is now, how do we know it won't suddenly smash down to one particle?
This would be modifying the object (us). There is no logical contradiction then, since killing someone by modifying him is easy :)
In my scenario, the object/AI is not modified in any way. The outside is modified (it cuts the data flow from the object/AI). This is killing someone without changing his state (which seems logically contradictory).
How does any of this even relate to intelligence?
Interesting question :)
From my point of view it relates to a very strict but common definition of artificial intelligence: An AI can be all what a human person can be.
Another conclusion from the scenario is: For AIs, existence and nonexistence is the same.
The problem with causality, as I anderstand it, is that (most) physicists believe that the description of nature (at subatomic level) given by current quantum physics is (reasonably) complete. Specifically, that a particle has no location, but instead only the probability distribution curve/wave of a location.
But when the particle is observed, it is found at a specific place. However, without a _causal_ reason. Admittedly, thats sounds pretty strange.
The following page looks like a reasonably good introduction:. html)
Quantum Primer (http://www.sfu.ca/chemcai/QUANTUM/Quantum_Primer
Quotes of the relevant parts:
"Q13. Exactly what is it that is "waving"?
We pointed out earlier that a wave is a change that varies with location in a periodic, repeating way. What kind of a change do the crests and hollows of a "matter wave" trace out? The answer is that the wave represents the value of a quantity whose square is a measure of the probability of finding the particle in that particular location. In other words, what is "waving" is the value of a mathematical probability function.
Q14. What is the uncertainty principle?
In 1927, Werner Heisenberg proposed that certain pairs of properties of a particle cannot simultaneously have exact values. In particular, the position and the momentum of a particle have associated with them uncertainties x and p given by [...]
As with the de Broglie particle wavelength, this has practical consequences only for electrons and other particles of very small mass. It is very important to understand that these "uncertainties" are not merely limitations related to experimental error or observational technique, but instead they express an underlying fact that Nature does not allow a particle to possess definite values of position and momentum at the same time. This principle (which would be better described by the term "indeterminacy" than "uncertainty") has been thoroughly verified and has far-reaching practical consequences which extend to chemical bonding and molecular structure.
Q15. Is the uncertainty principle consistent with particle waves?
Yes; either one really implies the other. [...]
Please see my other answer a (few minutes ago, "being digital has problems").
But regarding the subatomic level, physicists are pretty much sure that on this level predictability and causality do not exist. All what the (well proven) theory allows to exist, are probability distributions for the observation of specific properties of subatomic particles (like location, speed, etc.).
The free will issue is one of the questionable but inevitable conclusions of the idea that computer programs can be alive, or that human conciousness can be the product of a computer program.
Another such conclusion is that humans then are machines that are completely defined by an initial state and the subsequent input data.
This effectively means that a human is identical to a static string of bits of limited size (which simply contains initial state + subsequent input data). This surely looks philosophically interesting.
So, we have the "string of bits" in the computer, or, alternatively an "intelligent" subsystem that creates that string of bits "live" from outside input.
The specific problem here is (beyond others), that with this picture we have no handle to reality. This string of bits is only "real" relative to the computer it is stored.
The "string of bits" can only exist if there is an outside interest in it. Because, by inherent necessity of computer mathemathics/logic, for the state of that "string of bits", its existence and nonexistence are identical if the computer loses the "interest" in it.
"Losing interest", for a computer, means deleting the last reference to the data object (think Java). Now, in computer logic, the object which represented the "artificial intelligent being", can be removed without modifying the meaning of the overall formula/expression.
The result is: An "artificial intelligent being" can only exist as long as there is an outside interest in it (like a Java reference to the object). As soon as this "interest" vanishes, existence and nonexistence become logically identical.
Another logical problem here is that an "artificial intelligence" can, as shown above, brought from existence into nonexistence without changing its state (from the outside, we dont change a single bit of the objects state).
Should we call something a person, if it can intentionally be killed without outside influence to it, at a specific time, and even without that "artificial being" being able to notice any difference?
Not only that it cannot notice a difference, there effectively is no difference for the "being", as its own state was not changed at all.
A program is, regardless of its size, is a purely mathematical/logical formula/expression.
Computers simply process precisely defined instructions, they do not think.
There arise some logical problems if we assume that a human personality also simply is a purely mathematical/logical formula/expression.
Freenet seems to always store the whole files.
The interesting approach of MojoNation is to split each file in many small parts, and then store these with RAID5-like redundancy on many different machines.
This is the way to go, since it properly abstracts all user-machines as a large but unreliable block storage.
It also allows, e.g., a client on DSL/Cable to saturate its bandwith even if it downloads a file from modem owners (it just uses many of them concurrently). This again is similar to RAID5.
Unfortunately, in my experience, MojoNation is still too early in development to be usable. It also seems to be too centralized.
Well, no chip that comes off the line is untested...
What I want? I want make it very clear that this thing is a demonstration of a handpicked chip.
Room temperature? I read something different.
The "mere existence of the 1.4GHZ P4"?
Hopefully this "existence" will be somewhat more real than the existence of the 1GHZ P3s.
Anyway. I dont really see what your problem is.
I agree completely.
The point is that companies will lie to you. Always. They are whores of their shareholders or investors.
And surely Transmeta practises in spindoctoring, too.
The 2GHz part was a handpicked chip, cooled like hell, and is far from being available.
The 850 mW number is measured "the Intel way", and therefore some considerable spindoctoring is involved.
Of course one can buy into the Intel marketing, but I prefer to spare my enthusiasm until I see that stuff for real, in volume, and tested by independent and reliable publications.
Slashdot is his site.
He stated his opinion, and its obvious that this simply is his opinion.
Twisting of truth is a problem, stating an opinion is not.
Slashdot is not CNN. If the maintainers of Slashdot feel strong about something and express this in clear words, thats perfectly ok.
People simply should go to CNN or ZDNET or whatever if they want anonymous news from seemingly unbiased droids.
since 2 years generally is the amount of time id wants to spend for new games.
You are heading into the FUD department with your message...
If you want non-GPL proprietary extensions to reiserfs, you can purchase them, too.
so no reason to blame them for this.
using a devel fs for anything not completely redundant is just asking for trouble...
ok, thats just half of the truth.
What ct found in their test lately (using the ISO patch LAME version) was that in all disciplines LAME is as good or even slightly better than Fraunhofer.
Except for speech at 32kbit/s, where Fraunhofer was better (I suppose we can live with that).
Bottom line: Forget about all other encoders, and use LAME instead. Your (and potentially other peoples) ears will thank you for this!
Ok. I must accept that a matter of personal preference.
But I also want to state that in my opinion this preference is wrong as it allows to make the code proprietary which I consider abusing it...
If you have to decide about the license for =your= code, this is an entirely different matter (compared to bitching about the GPL of others code, which can regularly be seen on slashdor).
believe me, they will make you happy.
You seem to have a problem with honesty. Just as most people out there.
Anyway, as a data point, Postgres 6.5 is running perfectly stable here, for years, hammered daily.
No compromises. If its free, it must stay free, and noone is permitted to make it proprietary.
And if someone calls the GPL a "virus", he obviously wants to unfairly take advantage of other peoples work without contributing anything by himself.
then it can work.
It seems to me that PostgreSQL is the only real long-term threat to Oracle.
If "Great Bridge" behaves well, and keeps the developers as well as the Postgres users happy, and hires as many Postgres developers as it can...
Then it can rightfully claim to be the best suited supplier of commercial Postgres support and commercial Postgres-related consulting services.
ok, the title looks a lot like offtopic, and you may rightly ask how Ill try to get on-topic again.
look:
something that Linux lacks right now is a log-structured filesystem (the newest Solaris has, if Im not misinformed). I consider this a requirement for an absolutely mission critical system. But Linux, the free software, is rapidly closing in to Solaris, its commercial counterpart (journaling reiserfs, later log-structured filesystems, too). It wont be too long until it is on par with Solaris.
Now, with Postgres I see the same. Its a lot less on par with Oracle that Linux is with Solaris. Perhaps this has to do with the fact that more people want a free Unix than a free DB. However, this is changing, since DBs get more and more important.
The point is: I suspect that Postgres takes the same course as Linux: The free software will get on par with its commercial counterpart. Postgres will just arrive there a little later.