So let's give the users an education. Instead of trying to block all these botnets, hack them! They are designed to distribute malicious software, so use that capability! Write a payload that would erase the hard drive on every infected machine and send it out there. I guarantee you that in a few weeks the users will be educated.
Does anyone know the status of kernel modesetting for R6/700? As in, being able to run a regular framebuffer console without X. I can't find any mention of anyone working on this.
> I don't think you "get" the Chinese Room. You defer the problem to the virtual entity that is created, > but *there is no necessity for such a virtual entity*. One way to view the argument is Occam's Razor: > as there is no need for the virtual entity to exist and no reason for it to
I am not postulating the existance of some hypothetical virtual entity; I am pointing to the obviously existent entity in the statement of the problem and stating that it is virtual. That entity is the process of translation itself. Just as a running program stores its state in memory, the translation process stores its state in the mind of the man in the room. The stepid variable, for example, that indexes into the rulebook and determines the next rule to apply to the data. A computer would store this variable in a number, stored in a register, stored using a collection of physical transistor states. The Chinese room stores stepid in the mind of the man in the room, who stores the number as a virtual number entity in his mind, stored somehow in the physical cells inside his brain. This variable is a virtual entity stored as a virtual entity stored on a physical brain. The translation process is the collection of these variables, the rules themselves (when loaded into memory), the rulebook, and whatever mental subroutines the man uses for the physical motions of translation. This process is an entirely virtual entity, which changes as the translation progresses, and converts its input into its output by "thinking".
> For example, if you have a case where there *is* a purpose to ascribing the existence of a > virtual entity then the "useless" component cause the argument to fail.
The Chinese room argument is said to fail because it is impossible to determine who is actually "thinking" in Chinese. The man is not thinking because he does not know chinese, the room is a passive entity, and so is the rulebook, and neither can take any action. My answer is that the virtual entity of the translation program is "thinking". The man in the room is the active physical entity running the translator, and in this example the man is the hardware (along with the room, for storage), while the virtual "thinking" program he is running is the software.
> So if I translate a program into a circuit board (I believe there are compilers which > output circuit diagrams) you can point at a wire or transistor and say 'that one's virtual'?
Of course not. A transistor or a wire is never going to be virtual, by definition. A virtual entity, such as a number, exists in the information generated by a physical machine. The machine is real, the number is virtual. The number is represented by a collection of physical states, but its identity is whatever we want it to be. You must define how a physical representation maps to virtual concepts, just as you must define how words in a language map to physical objects around you. In contrast, physical relationships, like the law of gravity, already exist in reality, and do not require any arbitrary creative activity on your part. (It is important to emphasize that the concepts describing the law of gravity in your mind are virtual, mapping to the physical relationship that is the law of gravity. Gravity exists in reality, the word "gravity" exists only in your mind)
> You've suggested that it's quite obvious where the "hardware" of the brain ends and the > "software" begins -- but it doesn't seem at all obvious to me, for the exact reason that > wherever the boundary lay, the observable effect would be the same.
Observable to an ordinary person through ordinary means. To such superficial observation it indeed makes no difference how the brain works internally. But if you were to actually examine the brain cell by cell, to the deepest available physical detail, the difference would be very important. No matter how hard you search, you will not find memories and beliefs in any particular cell. You will only find physical representations of data encoding those memories and beliefs. These represenations are stored on the hardware - cell shape and connections, resistances of various interfaces, neurotransmitter density, etc. But the data that they represent is the software, having meaining, and thus existence, only to the machine that runs it - the physical brain.
> State is always represented physically somehow - be it a charge in a transistor, the position > of a cog, the firing of a neuron. Nothing is ever *entirely* abstract.
Physical representation is not the same as physical identity. A flipflop can be in two physical states, each with its own recognizable physical identity. That is, you can look at it and unambiguously distinguish the two states as two different voltage configurations. A number stored on a collection of flipflops is physically represented by their states, but its identity is not physically encoded anywhere. There is no physical connection that makes a flipflop state map to the logical value of zero or one; these are arbitrary, assigned by us for the purpose of computation. Likewise, there is no physical reason to group a collection of bits into a number, no physical reason to interpret them as a 1, 2, or 4 byte integer, no physical reason to interpret them in LSB or MSB format. The number does not physically exist: it is the way we choose to interpret a particular arrangement of physical states, and thus exists only in the mind of the interpreter, be it you or the computer. Contrast this with a physical relationship, say, "inside". If a ball is inside a box, the relationship is a physical fact, unchanged by what you want to think of it. While you can suddenly decide to read all numbers in LSB format and have them all come out as valid numbers, you can not decide that "inside" suddenly means "outside" and make the ball magically leave the box.
> I also don't see the "where" and "how" questions being all that separate. If we could successfully identify the "software", then we could codify it into an algorithm
You would not be able to implement Google's page ranking algorithm unless you know how it works, but that will not stop you from identifying the difference between Google's servers and the search engine that runs on them.
> I'm not sure what you mean mean by "thinking faster than the brain can change."
Ok, that was not well stated. I am trying to describe the difference between physical and virtual entities; I came up with a somewhat better explanation in another thread.
> I'm going to have to apologize as I'm not going to invest enough time in a random internet > discussion to read a book. But any points of his you want to bring up here I'll listen to.
If you are interested in how the brain works, you really should read the book. But the basic point made there is the same is mine: brain is the hardware, mind is the software. The book explains how the neocortex is almost entirely homogeneous (yes, there are references), and hypothesizes that all regions of the brain are actually generic hardware running different software. It has more on the physical brain structure, and how this software actually works.
> I have a friend with an EE background, who is fond of reminding us "anything that can be done in software, can be done in hardware".
That's not what I mean. When someone says "it can be done in hardware", he just means he will not need a CPU to do it. There are many electronic constructs out there that run "programs", but aren't processors because they are designed to run just one program. That one program can not be changed without changing the hardware, which is why it is not called software. There is, however, still a distinction between the program and the hardware that runs it. One is physical, the other virtual.
> Given a hardware implementation and a software implementation of the same algorithm, an end user wouldn't be able to tell the difference.
I am not talking about what is, not what it looks like. Just because you can't tell Diet Coke from regular, doesn't mean the they have the same composition.
> Here's a few that I've discussed with atheist friends:
Fruitlessly, no doubt, since your position is based on faith rather than evidence. The difference between religion and science is that the former states "believe this because I (or God) tell you so", while the latter states "here is what we see in reality, and here are our logical conclusions about what is, based on what we see"
> - What is a morally good or just action? Why?
A morally good action is one that creates more of what you value. Religions like Christianity usually specify the virtues directly, leaving the values unsaid. For example, consider the Christian virtues of Prudence, Justice, Temperance, Fortitude, Faith, Hope, and Charity. Prudence is a virtue because it increases your chances of coming up with correct decisions, an outcome you consider valuable. Justice, or rendering to each and all what is due to them, exists because you value the kind of world where you get what you deserve. Practicing temperance increases the time you spend on contemplating God (which you value) and decreases the time you spend on physical pleasures (which you don't). Fortitude is the expression of the valuable (to you) mindset of acceptance, that what happens to you is God's will and it's not your place to question it. Faith makes sure you don't think too much (which is bad for God, placing it in danger of extinction). Hope increases your desire for the afterlife. And Charity cultivates the valuable trait of dependence in other people.
My moral code contains different virtues because I value different things. I do not decide what to value based on what some nonexistent entity tells me (or, rather, what the Church says the nonexistent entity would tell me if it felt like doing so); my values are an expression of what I personally want the world to be like. Actions making the world more like what I want it are virtues, and actions contrary to that are vices. You went directly to the virtues, because the Church dictates it so, but the process of coming up with those virtues could be the same for both of us, and if you accept the Christian values that the virtues imply, you would generate the same moral code as the Church did. Both of us use the same methods, but achieve different results. If you wish to argue that your moral code is superior to mine, you need to demonstrate that your values are superior to mine; i.e. that the kind of world you want is "better" than the kind of world I want.
> - Why be charitable towards the less fortunate?
Why indeed? Christianity encourages charity so that people would become accustomed to dependence on each other. Going your own way has many dangers to a system of beliefs based mostly on "because I say so", and so should be discouraged as much as possible. Charity also encourages humility, which is makes for easier subjects to rule, and more pliable minds for indoctrination. And Charity emphasizes the Christian disdain for material possessions; the poor are easier to manipulate due to humility, lack of education, dependence on the community, and general intellectual apathy. Christianity considers charity a virtue. In my moral code, it is one of the most vicious things you can do to a person.
That is not to say that I would never help other people. I just wouldn't do it solely because they are "less fortunate". I might help someone if I like him (by which I mean that he has traits I value). I might help him if he can do something for me later, although I might not have any specific reciprocal action in mind, or imply it as a condition for my assistance (that would be an implicit contract, a serious vice). Or I just might do it for personal entertainment if I am bored.
> - What sorts of causes beyond yourself are worth serving?
The ones that create something you value greatly. You serve God's purpose, whatever you imagine it may be. I decide what I value for myself and then find the causes that achieve those values. A
> I think it's better to cut away the needless scaffolding of intricate arguments and just state > the conclusions we're trying to arrive at. This, at least, is academically honest.
If you do that, you have a religion. Science requires that your conclusions be supported by evidence.
> I would also prefer that "simpler" positions are the default.
"Simpler" for whom? What a mathematician considers a simple idea might be utterly incomprehensible to a layman. Take the idea of virtual entities, which is the entire basis of the whole mind-brain controversy. A simple example of it might be this web page. It is currently stored in the cache of your browser as a file on your hard drive. While there are indeed physical bits on the hard drive that map directly onto the letters you are seeing on your screen, there is no physical entity or relationship corresponding to the file containing them, or to the directory containing the file, or the cache structure your browser maintains. All these entities are virtual, existent only in the simulated environment run on your computer, and there is no way to "see" them by examining the physical parts of the computer to any conceivable depth. Contrast that with a mechanical feedback system, like a flyball governor; the feedback is a concept, just like a file is, but it has a direct physical interpretation, evident from examination of the physical components of the system; any alien dropping onto our planet and seeing it will come up with the feedback concept just by looking. To find a file you need to extract the information stored on a physical device and then interpret that information in a specific way. If you don't know that way; for example, if you don't know how an ext2 filesystem is structured, or how the hard drive is partitioned into sectors, then you will not be able to come up with any files or even an idea of a file if you didn't have one before. Furthermore, virtual objects can depend on other virtual objects; the hard drive could be encrypted, adding yet another layer, and preventing anyone from discovering the virtual structures hidden in it.
Once you understand the concept of virtual objects and concepts and the distinction between them and physical objects and relations, it is very simple. Not only that, but it is something I deal with every day on my computer, and is a tangible and obvious part of my life experience. But most people find it completely incomprehensible, since it is not a part of their nontechnical everyday experience.
> The mind is just the brain. If it is not, we'd need a good reason to think this.
That reason is that ideas you are thinking right now do not have physical existence in your head. They are virtual entities, like the file on your hard drive. The distinction between the brain and the mind is necessary for the same reason we make a distinction between the computer and the web browser running on it: the former is physical, the latter virtual.
> The more fantastic the arguments I hear, the more weary I get. > Mental zombies and Chinese rooms are just some of the most egregious.
The Chinese room argument is a very good example of the misunderstanding of the physical/virtual distinction. The man in the room follows instructions like a computer, and is the physical part of the system. He can not translate Chinese any more than a computer running babelfish, but the virtual entity he creates by his operation can. The computer is not the program, the brain is not the mind, and the Chinese room is not the set of rules being executed in it. The translator "program" has no physical existence, and the reason people still bring up the argument is that they obviously can not find it. But that is so because they are looking in the wrong place; looking only at the physical components and ignoring the virtual objects created by their operation.
> So then what do you make of: "My favorite programming language is solder."?
A bad analogy. A complex mechanical system, like, say, a car engine, requires a lot of "programming" to work together correctly. Modern engines have computers with software controlling everything, but older engines only have mechanical components. The feedback loops in their operation are occasionally described in software terms, but because there is no conceptual entity "deciding" to do things, we don't call it that. Flow of control in a program has internal state and internal logic that is not embodied in any physical entity, and that is what makes it software.
> But the brain as forged by millions of years of evolution is very different than computational algorithms engineered in 100 or so years by humans. > We should learn much more neuroscience before we starting where, if anywhere, can we find the dividing line between the brain's "hardware" and "software".
Just because you don't know absolutely everything about the brain, doesn't mean you can not distinguish its hardware from its software. The line most definitely exists, as should be obvious to anyone from the fact that we think faster than the brain can physically change. "Where" you draw it is also pretty obvious if you have ever written a program. The only unknown, quite separate from the "where", is finding out how the hardware and the software work. Jeff Hawkins' On Intelligence is a very good book on the subject.
"Isn't it odd? When a politician or a movie star retires, we read front page stories about it. But when a philosopher retires, people do not even notice it." "They do, eventually."
> 3) The brain is still nothing more than a mass of cells.
As a programmer, I must point out the obvious analogy: "the computer is still nothing more than a collection of transistors", and reply that if that were so, nobody would have to argue whether it is better to run Linux or Vista. Philosophers would do much better once they explicitly state that there is a difference between hardware and software, that they are, respectively, the brain and the mind, and that anyone trying to conflate the two is either a con man or an idiot.
As an open source developer with half a dozen projects on SourceForge, I can promise you that if the government suddenly creates a possibility that I would be sued for them, I'm pulling each and every one of them off and closing the projects as permanently as I am allowed. Sure, I don't intentionally screw my users, but that doesn't mean I would bet that my code is perfect. Heck, I found a couple of small bugs just yesterday. These things happen. But if you want to punish me for them, you are welcome to live without my code. I'll pull it all off the net and keep it to myself.
Of course, nobody would notice if my little projects went off line. But I'm sure there are a few more important ones out there with commercial competitors who would just LOVE to see the competition disappear. A liability law wouldn't bother a large corporation, since it already has a legal department for blowing people off. I don't have a legal department, and I don't have millions of dollars to devote to defending myself in court. Guess who wins?
I had a problem with bad RAM on a brand new machine and nearly ended up sending the DIMMs back. But I wondered how that could happen, since they were Corsair Dominator, supposedly a reliable brand. Turns out that when my BIOS "loaded" the "extreme" profile to run the memory at 1600 instead of 1333, it didn't actually set the correct voltages, but only displayed them. Since the values were grayed out, I had assumed it would just set them, and so didn't check further at first. Later, after some corrupted data and many many memtest errors it occured to me to look at the BIOS settings again. Raising the voltage on the DIMM to the correct value (which BIOS was displaying as its "extreme profile") eliminated all the errors right away. (This was on GA-EX58-UD5, by the way)
> We need a gradual level of tiers ranging from a database that does its own journaling > and needs to know that data is fully written to disk to an application swapfile that if > it never hits the disk isn't a big deal (granted, such an app should just use kernel swap, > but that is another issue).
Actually there already is a syscall for telling the kernel how the file will be used.
posix_fadvise (int fd, off_t offset, off_t len, int advice)
POSIX_FADV_DONTNEED sounds like what you would use for your swapfile case.
I don't know if the kernel actually does anything with this information, but it looks like this would be a good place to implement any new interfaces for what you are suggesting.
No, without a power failure or a crash nothing bad happens. If the data all makes it to the disk, it doesn't really matter in which order it was written. The corruption happens only when the sequence of writes is interrupted. A careful programmer will order writes to a file to keep the data readable at every point, tolerating an interruption anywhere. For example, the new data could be appended to the end of the file before overwriting any data in the middle, which is what ext3 does with the journal. If a crash happens, either there would be some extra data at the end of the file, or the overwrite would be partial. Both cases are recoverable; the first, by truncating the file and eliminating the new data, the second, by redoing the write with the new data. This way you can only lose whatever you were trying to write at the moment of the crash.
Yes, but it's a more general issue of changing the order of operations. If I write a file and then rename it, I expect the writing to happen before the renaming. If I write to the file in two different places, I expect the actual writes to occur in the order I made them. ext4 designers evidently decided that they knew better how to order the writes, breaking everyone's expectations in the process. Sure, it probably improves performance because you can minimize seeks by ordering writes in a particular order, but the KDE problem illustrates that if you do that, you must at the very least ensure that the dependencies between files are honored. In KDE's case the dependency is between the file and the directory in which it is contained. When such a dependency exists, the filesystem MUST NOT reorder the writes! It isn't just KDE config files. The Firebird database engine, for example, does not fsync at all, relying instead on correct write ordering to preserve data integrity. If the ordering is broken, all Firebird databases will become corrupted.
> I don't want to hear about Ext4, fsync(), or data loss again.
Well, tough for you. Those discussions are still missing the poing that using fsync is the bloody wrong thing to do. It is not my job to ensure that the filesystem operations are ordered properly. It is the filesystem's job. Period. And all the arguments you link to for using fsync miss this point entirely.
The difference between games and movies is that the former is replayed often. I would never bother buying a game that is designed to be played exactly once, if only because the plot in games is always so boring and poorly written, that it is not even worth knowing. If Fallout 3 were a book, I wouldn't buy it. If Bioshock were a book, I'd burn it. If Half-Life 2 were a book, well, it wouldn't be much of a book. With that in mind, it is obvious that knowing the plot is pretty much irrelevant, since most of the time I would be replaying and know it anyway. So I always read every walkthrough I can find even before I get the game, to make sure it will be worth playing more than once.
> The development and legal communities would, I assume, vehemently oppose this idea, but it is possible. > Basically, the FTC would come up with a list of things all EULAs include, then a list of optional provisions > that the licensor (the game company) could include.
Why would anyone oppose that? Standardizing license clauses reduces everyone's workload. A small company would not need a team of lawyers to write a EULA. And the user will be able to tell what each license means without having to read the whole bloody lawyer-speak thing. This is good.
So where can I see some benchmarks showing just how much of a slowdown I can expect after switching from ext2 to ext4? All the benchmarks I see around here compare it to ext3 and to ReiserFS only. Also, is it possible to run ext4 without the journal? Any benchmarks on that? (Oh, and please, don't bother with the reliability lectures. I couldn't care less.)
I don't want to click on your stupid big buttons. If I want to do something fast, I use the keyboard. You do provide keyboard shortcuts for your buttons, right? And if you say that you really really need an 800 pixel wide dialog, I say bullshit. We got by just fine ten years ago with 640x480 screens, and before you can say "we have more features now", I'll tell you to get rid of them and fix the bugs first. Call me bitter, but after a week of trying to play Fallout 3 with the screen freezing every five minutes (or anytime more than two sounds are playing), I have a particularly sharp axe to grind.
Slashdot Mirror
So let's give the users an education. Instead of trying to block all these botnets, hack them! They are designed to distribute malicious software, so use that capability! Write a payload that would erase the hard drive on every infected machine and send it out there. I guarantee you that in a few weeks the users will be educated.
Does anyone know the status of kernel modesetting for R6/700? As in, being able to run a regular framebuffer console without X. I can't find any mention of anyone working on this.
> I don't think you "get" the Chinese Room. You defer the problem to the virtual entity that is created,
> but *there is no necessity for such a virtual entity*. One way to view the argument is Occam's Razor:
> as there is no need for the virtual entity to exist and no reason for it to
I am not postulating the existance of some hypothetical virtual entity; I am pointing to the obviously existent entity in the statement of the problem and stating that it is virtual. That entity is the process of translation itself. Just as a running program stores its state in memory, the translation process stores its state in the mind of the man in the room. The stepid variable, for example, that indexes into the rulebook and determines the next rule to apply to the data. A computer would store this variable in a number, stored in a register, stored using a collection of physical transistor states. The Chinese room stores stepid in the mind of the man in the room, who stores the number as a virtual number entity in his mind, stored somehow in the physical cells inside his brain. This variable is a virtual entity stored as a virtual entity stored on a physical brain. The translation process is the collection of these variables, the rules themselves (when loaded into memory), the rulebook, and whatever mental subroutines the man uses for the physical motions of translation. This process is an entirely virtual entity, which changes as the translation progresses, and converts its input into its output by "thinking".
> For example, if you have a case where there *is* a purpose to ascribing the existence of a
> virtual entity then the "useless" component cause the argument to fail.
The Chinese room argument is said to fail because it is impossible to determine who is actually "thinking" in Chinese. The man is not thinking because he does not know chinese, the room is a passive entity, and so is the rulebook, and neither can take any action. My answer is that the virtual entity of the translation program is "thinking". The man in the room is the active physical entity running the translator, and in this example the man is the hardware (along with the room, for storage), while the virtual "thinking" program he is running is the software.
> So if I translate a program into a circuit board (I believe there are compilers which
> output circuit diagrams) you can point at a wire or transistor and say 'that one's virtual'?
Of course not. A transistor or a wire is never going to be virtual, by definition. A virtual entity, such as a number, exists in the information generated by a physical machine. The machine is real, the number is virtual. The number is represented by a collection of physical states, but its identity is whatever we want it to be. You must define how a physical representation maps to virtual concepts, just as you must define how words in a language map to physical objects around you. In contrast, physical relationships, like the law of gravity, already exist in reality, and do not require any arbitrary creative activity on your part. (It is important to emphasize that the concepts describing the law of gravity in your mind are virtual, mapping to the physical relationship that is the law of gravity. Gravity exists in reality, the word "gravity" exists only in your mind)
> You've suggested that it's quite obvious where the "hardware" of the brain ends and the
> "software" begins -- but it doesn't seem at all obvious to me, for the exact reason that
> wherever the boundary lay, the observable effect would be the same.
Observable to an ordinary person through ordinary means. To such superficial observation it indeed makes no difference how the brain works internally. But if you were to actually examine the brain cell by cell, to the deepest available physical detail, the difference would be very important. No matter how hard you search, you will not find memories and beliefs in any particular cell. You will only find physical representations of data encoding those memories and beliefs. These represenations are stored on the hardware - cell shape and connections, resistances of various interfaces, neurotransmitter density, etc. But the data that they represent is the software, having meaining, and thus existence, only to the machine that runs it - the physical brain.
> State is always represented physically somehow - be it a charge in a transistor, the position
> of a cog, the firing of a neuron. Nothing is ever *entirely* abstract.
Physical representation is not the same as physical identity. A flipflop can be in two physical states, each with its own recognizable physical identity. That is, you can look at it and unambiguously distinguish the two states as two different voltage configurations. A number stored on a collection of flipflops is physically represented by their states, but its identity is not physically encoded anywhere. There is no physical connection that makes a flipflop state map to the logical value of zero or one; these are arbitrary, assigned by us for the purpose of computation. Likewise, there is no physical reason to group a collection of bits into a number, no physical reason to interpret them as a 1, 2, or 4 byte integer, no physical reason to interpret them in LSB or MSB format. The number does not physically exist: it is the way we choose to interpret a particular arrangement of physical states, and thus exists only in the mind of the interpreter, be it you or the computer. Contrast this with a physical relationship, say, "inside". If a ball is inside a box, the relationship is a physical fact, unchanged by what you want to think of it. While you can suddenly decide to read all numbers in LSB format and have them all come out as valid numbers, you can not decide that "inside" suddenly means "outside" and make the ball magically leave the box.
> I also don't see the "where" and "how" questions being all that separate. If we could successfully identify the "software", then we could codify it into an algorithm
You would not be able to implement Google's page ranking algorithm unless you know how it works, but that will not stop you from identifying the difference between Google's servers and the search engine that runs on them.
> I'm not sure what you mean mean by "thinking faster than the brain can change."
Ok, that was not well stated. I am trying to describe the difference between physical and virtual entities; I came up with a somewhat better explanation in another thread.
> I'm going to have to apologize as I'm not going to invest enough time in a random internet
> discussion to read a book. But any points of his you want to bring up here I'll listen to.
If you are interested in how the brain works, you really should read the book. But the basic point made there is the same is mine: brain is the hardware, mind is the software. The book explains how the neocortex is almost entirely homogeneous (yes, there are references), and hypothesizes that all regions of the brain are actually generic hardware running different software. It has more on the physical brain structure, and how this software actually works.
> I have a friend with an EE background, who is fond of reminding us "anything that can be done in software, can be done in hardware".
That's not what I mean. When someone says "it can be done in hardware", he just means he will not need a CPU to do it. There are many electronic constructs out there that run "programs", but aren't processors because they are designed to run just one program. That one program can not be changed without changing the hardware, which is why it is not called software. There is, however, still a distinction between the program and the hardware that runs it. One is physical, the other virtual.
> Given a hardware implementation and a software implementation of the same algorithm, an end user wouldn't be able to tell the difference.
I am not talking about what is, not what it looks like. Just because you can't tell Diet Coke from regular, doesn't mean the they have the same composition.
> Here's a few that I've discussed with atheist friends:
Fruitlessly, no doubt, since your position is based on faith rather than evidence. The difference between religion and science is that the former states "believe this because I (or God) tell you so", while the latter states "here is what we see in reality, and here are our logical conclusions about what is, based on what we see"
> - What is a morally good or just action? Why?
A morally good action is one that creates more of what you value. Religions like Christianity usually specify the virtues directly, leaving the values unsaid. For example, consider the Christian virtues of Prudence, Justice, Temperance, Fortitude, Faith, Hope, and Charity. Prudence is a virtue because it increases your chances of coming up with correct decisions, an outcome you consider valuable. Justice, or rendering to each and all what is due to them, exists because you value the kind of world where you get what you deserve. Practicing temperance increases the time you spend on contemplating God (which you value) and decreases the time you spend on physical pleasures (which you don't). Fortitude is the expression of the valuable (to you) mindset of acceptance, that what happens to you is God's will and it's not your place to question it. Faith makes sure you don't think too much (which is bad for God, placing it in danger of extinction). Hope increases your desire for the afterlife. And Charity cultivates the valuable trait of dependence in other people.
My moral code contains different virtues because I value different things. I do not decide what to value based on what some nonexistent entity tells me (or, rather, what the Church says the nonexistent entity would tell me if it felt like doing so); my values are an expression of what I personally want the world to be like. Actions making the world more like what I want it are virtues, and actions contrary to that are vices. You went directly to the virtues, because the Church dictates it so, but the process of coming up with those virtues could be the same for both of us, and if you accept the Christian values that the virtues imply, you would generate the same moral code as the Church did. Both of us use the same methods, but achieve different results. If you wish to argue that your moral code is superior to mine, you need to demonstrate that your values are superior to mine; i.e. that the kind of world you want is "better" than the kind of world I want.
> - Why be charitable towards the less fortunate?
Why indeed? Christianity encourages charity so that people would become accustomed to dependence on each other. Going your own way has many dangers to a system of beliefs based mostly on "because I say so", and so should be discouraged as much as possible. Charity also encourages humility, which is makes for easier subjects to rule, and more pliable minds for indoctrination. And Charity emphasizes the Christian disdain for material possessions; the poor are easier to manipulate due to humility, lack of education, dependence on the community, and general intellectual apathy. Christianity considers charity a virtue. In my moral code, it is one of the most vicious things you can do to a person.
That is not to say that I would never help other people. I just wouldn't do it solely because they are "less fortunate". I might help someone if I like him (by which I mean that he has traits I value). I might help him if he can do something for me later, although I might not have any specific reciprocal action in mind, or imply it as a condition for my assistance (that would be an implicit contract, a serious vice). Or I just might do it for personal entertainment if I am bored.
> - What sorts of causes beyond yourself are worth serving?
The ones that create something you value greatly. You serve God's purpose, whatever you imagine it may be. I decide what I value for myself and then find the causes that achieve those values. A
> I think it's better to cut away the needless scaffolding of intricate arguments and just state
> the conclusions we're trying to arrive at. This, at least, is academically honest.
If you do that, you have a religion. Science requires that your conclusions be supported by evidence.
> I would also prefer that "simpler" positions are the default.
"Simpler" for whom? What a mathematician considers a simple idea might be utterly incomprehensible to a layman. Take the idea of virtual entities, which is the entire basis of the whole mind-brain controversy. A simple example of it might be this web page. It is currently stored in the cache of your browser as a file on your hard drive. While there are indeed physical bits on the hard drive that map directly onto the letters you are seeing on your screen, there is no physical entity or relationship corresponding to the file containing them, or to the directory containing the file, or the cache structure your browser maintains. All these entities are virtual, existent only in the simulated environment run on your computer, and there is no way to "see" them by examining the physical parts of the computer to any conceivable depth. Contrast that with a mechanical feedback system, like a flyball governor; the feedback is a concept, just like a file is, but it has a direct physical interpretation, evident from examination of the physical components of the system; any alien dropping onto our planet and seeing it will come up with the feedback concept just by looking. To find a file you need to extract the information stored on a physical device and then interpret that information in a specific way. If you don't know that way; for example, if you don't know how an ext2 filesystem is structured, or how the hard drive is partitioned into sectors, then you will not be able to come up with any files or even an idea of a file if you didn't have one before. Furthermore, virtual objects can depend on other virtual objects; the hard drive could be encrypted, adding yet another layer, and preventing anyone from discovering the virtual structures hidden in it.
Once you understand the concept of virtual objects and concepts and the distinction between them and physical objects and relations, it is very simple. Not only that, but it is something I deal with every day on my computer, and is a tangible and obvious part of my life experience. But most people find it completely incomprehensible, since it is not a part of their nontechnical everyday experience.
> The mind is just the brain. If it is not, we'd need a good reason to think this.
That reason is that ideas you are thinking right now do not have physical existence in your head. They are virtual entities, like the file on your hard drive. The distinction between the brain and the mind is necessary for the same reason we make a distinction between the computer and the web browser running on it: the former is physical, the latter virtual.
> The more fantastic the arguments I hear, the more weary I get.
> Mental zombies and Chinese rooms are just some of the most egregious.
The Chinese room argument is a very good example of the misunderstanding of the physical/virtual distinction. The man in the room follows instructions like a computer, and is the physical part of the system. He can not translate Chinese any more than a computer running babelfish, but the virtual entity he creates by his operation can. The computer is not the program, the brain is not the mind, and the Chinese room is not the set of rules being executed in it. The translator "program" has no physical existence, and the reason people still bring up the argument is that they obviously can not find it. But that is so because they are looking in the wrong place; looking only at the physical components and ignoring the virtual objects created by their operation.
> So then what do you make of: "My favorite programming language is solder."?
A bad analogy. A complex mechanical system, like, say, a car engine, requires a lot of "programming" to work together correctly. Modern engines have computers with software controlling everything, but older engines only have mechanical components. The feedback loops in their operation are occasionally described in software terms, but because there is no conceptual entity "deciding" to do things, we don't call it that. Flow of control in a program has internal state and internal logic that is not embodied in any physical entity, and that is what makes it software.
> But the brain as forged by millions of years of evolution is very different than computational algorithms engineered in 100 or so years by humans.
> We should learn much more neuroscience before we starting where, if anywhere, can we find the dividing line between the brain's "hardware" and "software".
Just because you don't know absolutely everything about the brain, doesn't mean you can not distinguish its hardware from its software. The line most definitely exists, as should be obvious to anyone from the fact that we think faster than the brain can physically change. "Where" you draw it is also pretty obvious if you have ever written a program. The only unknown, quite separate from the "where", is finding out how the hardware and the software work. Jeff Hawkins' On Intelligence is a very good book on the subject.
"Isn't it odd? When a politician or a movie star retires, we read front page stories about it. But when a philosopher retires, people do not even notice it."
"They do, eventually."
> 3) The brain is still nothing more than a mass of cells.
As a programmer, I must point out the obvious analogy: "the computer is still nothing more than a collection of transistors", and reply that if that were so, nobody would have to argue whether it is better to run Linux or Vista. Philosophers would do much better once they explicitly state that there is a difference between hardware and software, that they are, respectively, the brain and the mind, and that anyone trying to conflate the two is either a con man or an idiot.
As an open source developer with half a dozen projects on SourceForge, I can promise you that if the government suddenly creates a possibility that I would be sued for them, I'm pulling each and every one of them off and closing the projects as permanently as I am allowed. Sure, I don't intentionally screw my users, but that doesn't mean I would bet that my code is perfect. Heck, I found a couple of small bugs just yesterday. These things happen. But if you want to punish me for them, you are welcome to live without my code. I'll pull it all off the net and keep it to myself.
Of course, nobody would notice if my little projects went off line. But I'm sure there are a few more important ones out there with commercial competitors who would just LOVE to see the competition disappear. A liability law wouldn't bother a large corporation, since it already has a legal department for blowing people off. I don't have a legal department, and I don't have millions of dollars to devote to defending myself in court. Guess who wins?
I had a problem with bad RAM on a brand new machine and nearly ended up sending the DIMMs back. But I wondered how that could happen, since they were Corsair Dominator, supposedly a reliable brand. Turns out that when my BIOS "loaded" the "extreme" profile to run the memory at 1600 instead of 1333, it didn't actually set the correct voltages, but only displayed them. Since the values were grayed out, I had assumed it would just set them, and so didn't check further at first. Later, after some corrupted data and many many memtest errors it occured to me to look at the BIOS settings again. Raising the voltage on the DIMM to the correct value (which BIOS was displaying as its "extreme profile") eliminated all the errors right away. (This was on GA-EX58-UD5, by the way)
> We need a gradual level of tiers ranging from a database that does its own journaling
> and needs to know that data is fully written to disk to an application swapfile that if
> it never hits the disk isn't a big deal (granted, such an app should just use kernel swap,
> but that is another issue).
Actually there already is a syscall for telling the kernel how the file will be used.
posix_fadvise (int fd, off_t offset, off_t len, int advice)
POSIX_FADV_DONTNEED sounds like what you would use for your swapfile case.
I don't know if the kernel actually does anything with this information, but it looks like
this would be a good place to implement any new interfaces for what you are suggesting.
No, without a power failure or a crash nothing bad happens. If the data all makes it to the disk, it doesn't really matter in which order it was written. The corruption happens only when the sequence of writes is interrupted. A careful programmer will order writes to a file to keep the data readable at every point, tolerating an interruption anywhere. For example, the new data could be appended to the end of the file before overwriting any data in the middle, which is what ext3 does with the journal. If a crash happens, either there would be some extra data at the end of the file, or the overwrite would be partial. Both cases are recoverable; the first, by truncating the file and eliminating the new data, the second, by redoing the write with the new data. This way you can only lose whatever you were trying to write at the moment of the crash.
Yes, but it's a more general issue of changing the order of operations. If I write a file and then rename it, I expect the writing to happen before the renaming. If I write to the file in two different places, I expect the actual writes to occur in the order I made them. ext4 designers evidently decided that they knew better how to order the writes, breaking everyone's expectations in the process. Sure, it probably improves performance because you can minimize seeks by ordering writes in a particular order, but the KDE problem illustrates that if you do that, you must at the very least ensure that the dependencies between files are honored. In KDE's case the dependency is between the file and the directory in which it is contained. When such a dependency exists, the filesystem MUST NOT reorder the writes! It isn't just KDE config files. The Firebird database engine, for example, does not fsync at all, relying instead on correct write ordering to preserve data integrity. If the ordering is broken, all Firebird databases will become corrupted.
> I don't want to hear about Ext4, fsync(), or data loss again.
Well, tough for you. Those discussions are still missing the poing that using fsync is the bloody wrong thing to do. It is not my job to ensure that the filesystem operations are ordered properly. It is the filesystem's job. Period. And all the arguments you link to for using fsync miss this point entirely.
In Zoviet Germany ze vikipedia edits you!
The difference between games and movies is that the former is replayed often. I would never bother buying a game that is designed to be played exactly once, if only because the plot in games is always so boring and poorly written, that it is not even worth knowing. If Fallout 3 were a book, I wouldn't buy it. If Bioshock were a book, I'd burn it. If Half-Life 2 were a book, well, it wouldn't be much of a book. With that in mind, it is obvious that knowing the plot is pretty much irrelevant, since most of the time I would be replaying and know it anyway. So I always read every walkthrough I can find even before I get the game, to make sure it will be worth playing more than once.
> The development and legal communities would, I assume, vehemently oppose this idea, but it is possible.
> Basically, the FTC would come up with a list of things all EULAs include, then a list of optional provisions
> that the licensor (the game company) could include.
Why would anyone oppose that? Standardizing license clauses reduces everyone's workload. A small company would not need a team of lawyers to write a EULA. And the user will be able to tell what each license means without having to read the whole bloody lawyer-speak thing. This is good.
So where can I see some benchmarks showing just how much of a slowdown I can expect after switching from ext2 to ext4? All the benchmarks I see around here compare it to ext3 and to ReiserFS only. Also, is it possible to run ext4 without the journal? Any benchmarks on that? (Oh, and please, don't bother with the reliability lectures. I couldn't care less.)
I don't want to click on your stupid big buttons. If I want to do something fast, I use the keyboard. You do provide keyboard shortcuts for your buttons, right? And if you say that you really really need an 800 pixel wide dialog, I say bullshit. We got by just fine ten years ago with 640x480 screens, and before you can say "we have more features now", I'll tell you to get rid of them and fix the bugs first. Call me bitter, but after a week of trying to play Fallout 3 with the screen freezing every five minutes (or anytime more than two sounds are playing), I have a particularly sharp axe to grind.
If you are close enough for the enemy to see your face, you need neither the sniper rifle, nor complex distance calculations.