What? Come on now. I know/. is very pro-Open Source, con-MS, but thats ridiculous.
What? Come on now. I know/. was very pro-Open Source, con-MS, but thats ridiculous.
There, fixed that for you. Come on now, they've even changed SlashBack to Backslash. The readership has changed dramatically, as has the demographic. It's not necessarily a bad thing, they're still Open Source advocates, but, eh, there are a lot of MCSEs on this site. Then again, there are a lot more MCSEs running Linux these days too. It's culture at work.
Since JAXA doesn't currently have a 100 ton-class heavy lift rocket or a human transportation system perhaps now is a good time for JAXA to join in with NASA on the Project Constellation rocket program.
Don't be silly. They have 24 years to complete this. All of the technology to get them to the moon is already there, so they only need the additional tech to build a space station there. Some version of this is well within reach. The reasons that we don't do things like this have more to do with money, ambition, and government red tape than they do with our ability to do so, as it is with many worthwhile ventures.
Yes. There are companies offering solid state storage devices commercially as high-speed replacements for hard drives. BitMicro is one of them, and offer terrabyte RAIDS of the stuff.
People have been talking about doing exactly this technique for quite a while. It just never hit the mainstream. I even think that there were a couple commercial implementations of this, but I'm not sure on that last point. It is definitely talked about in research papers on filesystems that I have read.
Strongly recommended before submitting a resume...
I can just picture where this could go:
"Seeking a position as a full-time BDSM instructor" in one copy, "Seeing a position as full-time kindergarten teacher," in another.
Eventually, you'll probably see most operating systems implementing this, or this being implemented in a virtual machine. If you're concerned about privacy, you should be using crypography anyway (now, the question being, how do you isolate the entered passwords to unlock your keyring from the snapshots taken by a virtual machine hosting your operating system).
At any rate, there is more good to this than bad, and since this isn't even a real snapshotting mechanism (snapshotting your system memory) your crypto will protect you just fine.
The point of the chart isn't to point out what you think it is.
The point is that some branches of AI care nothing for emulating humans in any way, and ohers care only about emulating certain aspects of being human, human thought, or human behavior.
You'll notice that the same is true for rationality.
Anyway, it's not my chart, it's Stuart Russell and Peter Norvig's.
The problem with purchasing his sister a brain is the installation process. Most users don't want to open the cases on their bodies, or have the requisite expertise. I'm sure that his sister could be sent back to the factory, they'll refurbish her put in the brain if necessary, and sell her at a discount, and he'll get a nice new sister.
Your sister is an ingratious person who doesn't appreciate the effort that you put into furnishing her a computer in the first place. My advice to you is that, if she wants a store-bought computer, and feels that it will be better, let her have one, and stop giving her tech support. On the other hand, you might also consider getting a new, more appreciative sister.
There isn't anything terribly new or sinister sounding about this. They always have, and always will, monitor media coming out of war zones. Soldiers and contractors are informed of this going in, and consent to far more intrusive things. They always have, they always will.
You're reading too much into why people download beta software.
They don't download it to test it. They download it to be cooler than everyone else. To have the new, bleeding-edge stuff.
So, MS probably isn't getting much useful data about bugs, certainly, if it's this many people, they only need a fraction of them. Instead, they have thousands of users of buggy software, and since they're chasing off a reputation for buggy software, they probably don't really want this.
So, $1.50. You get software really cheaply (minus support, though, they'll probably be nagged into it), and they get fewer yahoos, a laughable amount of money, and justification for this.
Don't forget, a lot of the beta testers will just run the betas, and not purchase the actual product. Why get the newest version of office for a couple hundred? You can get the beta for free. Now that it's $1.50, most people will probably stick to the version that came with their computer.
That's why. Even a small company can appreciate that this many beta testers is not a favor of any kind, except perhaps for publicity's sake.
With all due respect, a wiretap is something different.
You're free to be upset about that, as many people are, however, it's not really "wiretapping."
In wiretapping, they look specifically at your phone, and record the transmissions over it. Records of who called who are just a normal part of the audit trail generated by the network. The government requisition of these has raised a number of eyebrows, however, it's not wiretapping.
MS seeks to keep 90% of the market 90% happy. That's all that you need to dominate the market. They don't need to appease the open source community in order to do this.
As for competition from Macs. Maybe, but people still have their reasons to buy PCs. Granted, Macs are cheaper now, over what they used to be, OSX is nifty.
MS, I'm sure, has their reasons for opening some sort of diplomatic relations with OSS, but fear of collapse in anything resembling the near future isn't one of them.
What can I say? An unsecured wireless router seems pretty much like an open invite. If you're savvy enough to have a reason to care, you're probably savvy enough to secure it (since they are designed for users who aren't savvy). Why else would you leave it unsecured aside from a view that others should be able to use it?
Perhaps the neighbor doesn't realize that you care. Perhaps the other neighbors don't. Maybe they're likely to go jack the connection of somebody who doesn't mind.
I've witnessed a CMM process developing, but never suffered through ISO-9000.
CMM is interesting in that it has the possibility of being fairly lightweight (at least, level 2 or 3). The problem is that people never get the "point", accountability for ones actions and ability to find out what's going on, and see lots of documents.
It gets worse, because there are always people in charge who practically live for documents, that's why they do things that involve writing lots of them, rather than developing software. Then, you have sycophants who want to ingratiate themselves to these people, they just make lots of "helpful suggestions."
Then, you have a big group of people, kind of lost in what they're doing, but they're going to do it anyway.
In a meeting once, I could sense that our process was getting out of control, and suggested that we have a checklist, so a developer, like I was, at the time, could keep track of all of the pointless crap that he had to do. The checklist idea was added to our process... so, now, the checklist was another thing to do. It had a place for the developer had his manager to sign that he had done each step in the process. My bullets telling me what I had to do to make these people happy became another insane requirement.
When the org in question passed their audit (an external organization that I had been contracted to), they had a celebration touting how many documents they had produced in order to pass. In the end, there was a flood of inconvenient documents that provided no more transparency than there would be in their absence.
I hope that I didn't offend any of my former coworkers with this.
Certainly, but my philosophy on the topic is that you never come closer to any given goal without trying, and science would never make progress toward the unattainable would trying to reach it. A journey of a thousand miles begins with the first step and what-not.
With all due respect, I think that we're all quite aware of this. The early papers in the 1960s were quite enthusiastic about the prospects of developing human-competitive in intelligence, but this was before there was a well developed theory of computational complexity. With its advent, it became easy to observe that the methods discussed to solve a number of problems were computationally intractible, and enthusiasm that we would develop human-competitive AI waned, however, the early writers on the topic had no hangups discussing the possibility of machines with incredibly powerful intelligence.
I understand what you're driving at, and I mean no offense in what I'm stating. I'm just saying that there are a number of people in the field who do pursue these topics, even today. I don't mean offense by anything.
The table breaks down like this. (I know, variable width font, but bear with me).
"Systems that think like humans" + "Systems that think rationally"
"Systems that act like humans" + "Systems that act rationally"
These are all, of course, different approaches, so, certainly, you're right, there are a number of scientists whose goal is an immediate analysis to a particular phenomenon, or, in an application, an immediate solution to a particular problem, in fact, it would be impossible to carry out good science without crafting one's experiments in such a manner. However, in terms of long-term goals, there still exist a significant camp of people whose goal it is to emulate human behavior, and even a camp trying to emulate human thought. They even do things such as study psychology literature, and scan the working brain with technologies such as FMRI in order to build ideas and models of how the brain works. There isn't always a lot of interplay between these groups, and even insiders in a particular field may be barely cognizant of the fact that the others exist, but, they still hold a place in the field, at least as it is taught. In practice, classroom exercises do tend towards the work of the nature that you cite though, and it does make up the bulk of modern research.
There is interplay though. Mitchell, in his NESCAI talk discussed the possibility of furthering machine learning through the study of brain activity, and perhaps vice versa. There was a talk at AAAI-05 regarding a machine learning model based on current knowledge of the workings of the brain. Minsky addresses human-like thought in "The Society of Mind," there are "Theories of Mind," humanoid and social robotics experiments. To pick the highlights of these really does reveal a full-spectrum in current research, some more controversial than other research, but all actively pursued in the current scientific community.
Certainly, I can buy the philosophical side of what you're saying. On the other hand then there are people who have written specifically about the concept of "consciousness." For instance, I have not yet taken the opportunity to read it, but Minsky's new book, "Emotional Machines," is supposed to address many of the issues encompassed by it. Though, his talk on the topic leads one to believe that he explains away the concept for consciousness rather than addressing it as a topic in and of itself.
Slashdot Mirror
http://www.psywarrior.com/rockmusic.html
What? Come on now. I know /. is very pro-Open Source, con-MS, but thats ridiculous.
/. was very pro-Open Source, con-MS, but thats ridiculous.
What? Come on now. I know
There, fixed that for you. Come on now, they've even changed SlashBack to Backslash. The readership has changed dramatically, as has the demographic. It's not necessarily a bad thing, they're still Open Source advocates, but, eh, there are a lot of MCSEs on this site. Then again, there are a lot more MCSEs running Linux these days too. It's culture at work.
Since JAXA doesn't currently have a 100 ton-class heavy lift rocket or a human transportation system perhaps now is a good time for JAXA to join in with NASA on the Project Constellation rocket program.
Don't be silly. They have 24 years to complete this. All of the technology to get them to the moon is already there, so they only need the additional tech to build a space station there. Some version of this is well within reach. The reasons that we don't do things like this have more to do with money, ambition, and government red tape than they do with our ability to do so, as it is with many worthwhile ventures.
Well, they say that they are.
http://www.bitmicro.com/products_edisk_35_ide.php
Yes. There are companies offering solid state storage devices commercially as high-speed replacements for hard drives. BitMicro is one of them, and offer terrabyte RAIDS of the stuff.
I don't think that's what he's driving at.
People have been talking about doing exactly this technique for quite a while. It just never hit the mainstream. I even think that there were a couple commercial implementations of this, but I'm not sure on that last point. It is definitely talked about in research papers on filesystems that I have read.
Strongly recommended before submitting a resume... I can just picture where this could go: "Seeking a position as a full-time BDSM instructor" in one copy, "Seeing a position as full-time kindergarten teacher," in another.
In the above post, "virtual machine" should really be "virtual machine monitor." For those of you playing along at home.
Eventually, you'll probably see most operating systems implementing this, or this being implemented in a virtual machine. If you're concerned about privacy, you should be using crypography anyway (now, the question being, how do you isolate the entered passwords to unlock your keyring from the snapshots taken by a virtual machine hosting your operating system).
At any rate, there is more good to this than bad, and since this isn't even a real snapshotting mechanism (snapshotting your system memory) your crypto will protect you just fine.
The point of the chart isn't to point out what you think it is.
The point is that some branches of AI care nothing for emulating humans in any way, and ohers care only about emulating certain aspects of being human, human thought, or human behavior.
You'll notice that the same is true for rationality.
Anyway, it's not my chart, it's Stuart Russell and Peter Norvig's.
The problem with purchasing his sister a brain is the installation process. Most users don't want to open the cases on their bodies, or have the requisite expertise. I'm sure that his sister could be sent back to the factory, they'll refurbish her put in the brain if necessary, and sell her at a discount, and he'll get a nice new sister.
Your sister is an ingratious person who doesn't appreciate the effort that you put into furnishing her a computer in the first place. My advice to you is that, if she wants a store-bought computer, and feels that it will be better, let her have one, and stop giving her tech support. On the other hand, you might also consider getting a new, more appreciative sister.
There isn't anything terribly new or sinister sounding about this. They always have, and always will, monitor media coming out of war zones. Soldiers and contractors are informed of this going in, and consent to far more intrusive things. They always have, they always will.
You're reading too much into why people download beta software.
They don't download it to test it. They download it to be cooler than everyone else. To have the new, bleeding-edge stuff.
So, MS probably isn't getting much useful data about bugs, certainly, if it's this many people, they only need a fraction of them. Instead, they have thousands of users of buggy software, and since they're chasing off a reputation for buggy software, they probably don't really want this.
So, $1.50. You get software really cheaply (minus support, though, they'll probably be nagged into it), and they get fewer yahoos, a laughable amount of money, and justification for this.
Don't forget, a lot of the beta testers will just run the betas, and not purchase the actual product. Why get the newest version of office for a couple hundred? You can get the beta for free. Now that it's $1.50, most people will probably stick to the version that came with their computer.
That's why. Even a small company can appreciate that this many beta testers is not a favor of any kind, except perhaps for publicity's sake.
With all due respect, a wiretap is something different.
You're free to be upset about that, as many people are, however, it's not really "wiretapping."
In wiretapping, they look specifically at your phone, and record the transmissions over it. Records of who called who are just a normal part of the audit trail generated by the network. The government requisition of these has raised a number of eyebrows, however, it's not wiretapping.
You know. I got so busy noticing my point at the top and the overarching message of the post that I completely missed that entire bit of nonsense!
MS seeks to keep 90% of the market 90% happy. That's all that you need to dominate the market. They don't need to appease the open source community in order to do this.
As for competition from Macs. Maybe, but people still have their reasons to buy PCs. Granted, Macs are cheaper now, over what they used to be, OSX is nifty.
MS, I'm sure, has their reasons for opening some sort of diplomatic relations with OSS, but fear of collapse in anything resembling the near future isn't one of them.
What can I say? An unsecured wireless router seems pretty much like an open invite. If you're savvy enough to have a reason to care, you're probably savvy enough to secure it (since they are designed for users who aren't savvy). Why else would you leave it unsecured aside from a view that others should be able to use it?
Perhaps the neighbor doesn't realize that you care. Perhaps the other neighbors don't. Maybe they're likely to go jack the connection of somebody who doesn't mind.
He used the money to buy a hamster... So, what, he taught for 1 hour at $25 an hour?
Improve is a relative term, but this is certainly gentler than certain other approaches.
I don't really see the point. It's funny as a practical joke. In terms of protecting your network... why not just secure it instead?
You are wise.
I've witnessed a CMM process developing, but never suffered through ISO-9000.
CMM is interesting in that it has the possibility of being fairly lightweight (at least, level 2 or 3). The problem is that people never get the "point", accountability for ones actions and ability to find out what's going on, and see lots of documents.
It gets worse, because there are always people in charge who practically live for documents, that's why they do things that involve writing lots of them, rather than developing software. Then, you have sycophants who want to ingratiate themselves to these people, they just make lots of "helpful suggestions."
Then, you have a big group of people, kind of lost in what they're doing, but they're going to do it anyway.
In a meeting once, I could sense that our process was getting out of control, and suggested that we have a checklist, so a developer, like I was, at the time, could keep track of all of the pointless crap that he had to do. The checklist idea was added to our process... so, now, the checklist was another thing to do. It had a place for the developer had his manager to sign that he had done each step in the process. My bullets telling me what I had to do to make these people happy became another insane requirement.
When the org in question passed their audit (an external organization that I had been contracted to), they had a celebration touting how many documents they had produced in order to pass. In the end, there was a flood of inconvenient documents that provided no more transparency than there would be in their absence.
I hope that I didn't offend any of my former coworkers with this.
Certainly, but my philosophy on the topic is that you never come closer to any given goal without trying, and science would never make progress toward the unattainable would trying to reach it. A journey of a thousand miles begins with the first step and what-not.
With all due respect, I think that we're all quite aware of this. The early papers in the 1960s were quite enthusiastic about the prospects of developing human-competitive in intelligence, but this was before there was a well developed theory of computational complexity. With its advent, it became easy to observe that the methods discussed to solve a number of problems were computationally intractible, and enthusiasm that we would develop human-competitive AI waned, however, the early writers on the topic had no hangups discussing the possibility of machines with incredibly powerful intelligence.
I understand what you're driving at, and I mean no offense in what I'm stating. I'm just saying that there are a number of people in the field who do pursue these topics, even today. I don't mean offense by anything.
The table breaks down like this. (I know, variable width font, but bear with me).
"Systems that think like humans" + "Systems that think rationally"
"Systems that act like humans" + "Systems that act rationally"
These are all, of course, different approaches, so, certainly, you're right, there are a number of scientists whose goal is an immediate analysis to a particular phenomenon, or, in an application, an immediate solution to a particular problem, in fact, it would be impossible to carry out good science without crafting one's experiments in such a manner. However, in terms of long-term goals, there still exist a significant camp of people whose goal it is to emulate human behavior, and even a camp trying to emulate human thought. They even do things such as study psychology literature, and scan the working brain with technologies such as FMRI in order to build ideas and models of how the brain works. There isn't always a lot of interplay between these groups, and even insiders in a particular field may be barely cognizant of the fact that the others exist, but, they still hold a place in the field, at least as it is taught. In practice, classroom exercises do tend towards the work of the nature that you cite though, and it does make up the bulk of modern research.
There is interplay though. Mitchell, in his NESCAI talk discussed the possibility of furthering machine learning through the study of brain activity, and perhaps vice versa. There was a talk at AAAI-05 regarding a machine learning model based on current knowledge of the workings of the brain. Minsky addresses human-like thought in "The Society of Mind," there are "Theories of Mind," humanoid and social robotics experiments. To pick the highlights of these really does reveal a full-spectrum in current research, some more controversial than other research, but all actively pursued in the current scientific community.
Certainly, I can buy the philosophical side of what you're saying. On the other hand then there are people who have written specifically about the concept of "consciousness." For instance, I have not yet taken the opportunity to read it, but Minsky's new book, "Emotional Machines," is supposed to address many of the issues encompassed by it. Though, his talk on the topic leads one to believe that he explains away the concept for consciousness rather than addressing it as a topic in and of itself.
"The question of whether a computer can think is no more interesting than the question of whether a submarine can swim."
-Edsger Dijkstra
Thanks, I've been wondering the source ever since he brought it up.