And it's not because it's nerdy (as the summary opines).
Yeah, my favorite part was, "Is this because of the fact that Bill Gates has made the whole industry look nerdy?" Really? Bill Gates made it look nerdy? Like if not for Gates, the whole industry would be filled with badass cowboys and hot chicks or something?
Yeah, even the fact that "badass cowboys and hot chicks" popped into my head as the opposite of "nerdy" is probably an indication that I'm an IT nerd.
But yeah, I've found that at least the IT work that falls on the support/maintenance side (as opposed to the development side) is kind of boring crap-work. It's fixing problems that some other moron broke, and cleaning up problems caused by poor design. It's 2008, and we still don't even have decent backup/archive methods. Every product out there has huge problems and gaping holes in their functionality that should have been fixed 15 years ago, but instead everyone has been working on things like database-driven filesystems that never make it to market.
That's right, I'm looking at you, Microsoft.
InfoTech work isn't all science-fictiony and cool. Oddly, it's more like being a Fonzie in training. It's like all this technology amounts to a broken jukebox that has to be smacked in just the right way to get it going, and you're just hoping to learn how to do that so you can stand around looking cool until someone needs you.
Except that this Fonzie never ends up looking cool and everyone treats him like a trained monkey. "Slap the machine and play me a song!" they all yell.
Oh, yeah, I know I've jumped the tracks and gone into nonsense. Whatever. I work in IT. Making up random crap on Slashdot is the most interesting part of my day.
Create an intelligence that makes mistakes and jumps to conclusions
Create God
The third option would of course have benefits, but it would be rather daunting to undertake. Anything short of that will be an intelligence which makes mistakes, because that's the nature of intelligence-- to analyze situations and draw conclusions which aren't completely founded, based on incomplete information.
No, really, I'm purposefully saying that you need emotions *first*, and then you can think about growing intelligence. I know that's completely the opposite of what other people are saying, but my point is that they're approaching the thing backwards because they don't understand the problem.
I'm not saying their emotions have to be complicated or varied at first, but drives are a fundamental starting point of intelligence. I say don't even bother starting to create real AI unless you have some kind of a plan to make your AI want things.
The problem, if I had to guess, is that the people trying to create AI are the sort who like to think of their own intelligence as a something removed from their physical bodies and from the world they live in. They're the sort who try their best to convince themselves that their own minds behave like disinterested computers, and then use that misunderstanding of their own intelligence as a model of what they're trying to perceive.
So they're thinking they're going to create an intelligence first, and then give the intelligence orders on what to do, and tools in order to do it. I'm saying that won't work to create real artificial intelligence. You have to build it with desires, drives, aversions, and an existence in a world (even if it's a virtual world) where it can pursue its desires. Not just because it would be "something for the AI to do", but because those are fundamental components of intelligence.
Disney already has registered TLDs for the localized versions of it's site for other regions and any further categorical distinctions for content can be accomplished with subdomains.
I think his point is that squatters already try to get users through misspellings and typos and such. So open up TLDs, and you'll see "disney.con", "disney.cpm", "disney.c0m", etc. And it's innocuous enough when they're just serving ads, but it's a bit more annoying when it's phishing sites. Imagine you're typing your bank's URL and you accidentally type "citi.con", and load up a page that looks just like "citi.com". If you put in your account info, you're pretty much screwed.
I get the joke, but I do wonder whether it might make sense to reevaluate the use of TLDs and DNS.
The reason I say this is that TLDs *are* confusing to people. There's often no particular reason why a domain might be using.com,.net, or.org. I know there's supposed to be a reason, but people don't follow those rules. Some people even use.ws on the idea that it stands of "WebSite", even though it's really just another country code. But even without all that, it's still confusing-- you know, the old story about people visiting "whitehouse.com" while looking for the White House website, being unfamiliar with the.gov TLD.
I don't know. The system seems to be working fine for the time being, but I've been trying to think of whether there might be some other method for finding the site you want or sending e-mail to the person you want, but one that doesn't reward people for snapping up domains early. Imagine what it'll be like in 100 years (assuming we are still using this system), trying to find a decent domain name that's not already in use. If we get to the point of having to break out "jfdsodkfhdlkw-jhfjdihfdu.com", I'm not sure that's much better than using IP addresses.
I have a feeling that the reason AI has 'underdelivered' is merely due to not understanding our own intelligence first.
This is the primary point I came in here to say. Whenever I've read anything about AI, it seems to be based on cool science-fictiony ideas, or else it's actually a simpler method to use statistical analysis to approximate human decision-making for particular purposes. If you're talking about real self-aware thinking things, the approaches are all wrong.
People tend to act treat the subject as though dumping enough raw information into a fast enough processor will yield intelligence, and then as that intelligence grows and develops, things like "sensible responses to answers" or "appropriate emotional responses" will emerge. Or else they think grouping enough "appropriate responses" will eventually yield intelligence.
It seems to me that that's all backwards. If you want to design an artificial intelligence, you first need a good philosophical understanding of how intelligence works, which will tell you straight-off something that AI researchers don't seem to consider: intelligence is an animal trait.
I think the absolute first thing you need to do is to figure out how to give machines emotions, to approximate pleasure/desire and pain/aversion. The second thing you need to do is give it "senses", and the ability to draw a very basic sensory conception of its world based on those senses, which includes a sense of time and objects. Also, you'll have to give it the ability to interact with its world in such a way that it is able to pursue its desires, encounter obstacles, and experience "pain". Finally, you'll have to figure out a way to give it the ability to adapt, to "rewrite its programing", preferably in a way that allows it to reproduce and evolve.
So in a way, the most obvious answer is that if you want an artificial intelligence, you'll have to design an artificial/virtual animal and place it into an environment where it can evolve intelligence. There may be some shortcuts on growing/evolving it faster, but you shouldn't be quick to discount the animal nature of intelligence as we know it.
And the reason for these things are bound up with the fact that, like I said, the only model for real intelligence we have to base anything on is animal intelligence. Animals develop and express their intelligence by being self-motivated in a world that presents obstacles. If there's nothing you want, there's no point in figuring anything out. If there's no way to get what you want, then there's no point in figuring things out. If there are no obstacles in your way, then there's nothing to figure out.
So if you don't have a self-motivated desire and the ability to move towards achieving that desire, then you can't make self-determined intelligent decisions. If course, this also presents a scary twist to the whole AI thing, because it suggests one of the chief scifi fears of AI will turn out to be correct: If we're successful in creating AI, we may not be able to control it.
That doesn't seem like a very good reason to me. What I mean is, that's a good reason why you wouldn't have everything encrypted at the outset, but that's not a good reason not to build out the protocols/infrastructure with the option of security/encryption in mind. Because even in cases where encryption is computationally too expensive for every transaction, you might still want to offer the option of encryption for for cases where encryption is worth the expense.
I guess it just seems a bit like the Y2K bug-- sure, it probably made sense at the time, but in hindsight it wasn't the best decision. Even now, secure protocols can seem like a bit of an add-on hack instead of an integrated part of the system. FTP is a pretty good example, IMO. FTP is awful security-wise, and FTP with TLS/SSL can be a bit of a PITA. SFTP looks to be a good solution, but ends up having its own headaches. You generally either have to give people remote shell access or go through extra steps to disable the normal shell and jail the user. I guess OpenSSH is now trying to address the issue, but it's still not particularly straight-forward, IMO. Also, SSH keys have none of the benefits that SSL certs have, e.g. identity verification or the ability to transparently revoke keys and issue new ones.
Yeah, it seems to me that it was an oversight that networking wasn't encrypted in the first place. When lots of these protocols were being developed, security didn't seem to be much of a consideration.
It's about time that these things got rectified, but I'm not sure what the best course is. For example, using SSL concerns me in that we've accepted the convention that certificates should be issued by certain set organizations that require exorbitant fees. I mean, hundreds or thousands of dollars per year for an SSL cert? Seems a bit much to me. Yeah, I know you can generate your own, which will cause you to get complaints from your websites' users when they see what looks to them like an error message.
I'm not a security expert, but I get the sense someone needs to go back to square one and figure out how to build a coherent, open, and secure model for networking that doesn't rely on giving such control to a small number of companies.
Let's take a look at the charges levied against the student:
34 felony counts of altering public record
11 felony counts of stealing and secreting public records
7 felony counts of illegal computer access and fraud
6 felony counts of burglary
4 felony counts of identity theft
3 felony counts of altering book of records
2 felony counts of receiving stolen property
1 felony count of conspiracy
1 felony count of attempting altering of a public record
And, as the article says, he's not just being charged with breaking in to the school's computers, he also put in a remote access program (which others could have used to do a variety of things with the school's network with). Then there's the fact that he broke into the principal's office to steal a test he did badly on...
Yes, and who was hurt in the end? What was damaged? What was the result of his crimes?
As far as I can tell, the problems caused by his crimes are essentially the same as if he were cheating. Specifically, the people who might have been hurt if the guy hadn't been caught would be the other student who failed to get into college because this kid took a slot that would otherwise have been available.
But (correct me if I'm wrong) nobody was directly harmed. Nothing of particular value was stolen. His actions were self-serving, but not malicious. He's potentially dangerous if his life takes the wrong turn, or potentially harmless if he learns the right things. A lot happens to a person between the ages of 18 and 28, and this guy will probably be a different person at the end of the next 10 years.
So the question in my mind, primarily, is not whether this guy did "the wrong thing" and should theoretically be punished. He's already going to be punished. He'll have a hard time getting into any college, and for the rest of his life, whenever he's asked on an application as to whether he's been convicted of a felony, he'll have to check "yes". He's going to have a lot of trouble because of his actions.
So the primary question in my mind is "what will give this guy the best chance of becoming a productive member of society?" Like, what events are most likely to turn him into a person that won't cause further trouble, or even more serious problems?
I doubt that jail time will be terrifically helpful, and I'm sure that significant time in jail will be counter-productive. Community service may be helpful. The most helpful thing will probably be for someone to show him that he can still be successful if he works his ass off and earns back the trust that he's lost.
38 years in jail is way too steep, any jailtime would probably be.
But this is no where near "just a mistake". It is not a one-time break-in to prove that security is insufficient. He was deliberately and continously (34 times alledgedly) "altering public records", for his own personal benefit.
Even if your security is inadquate it doesn't mean that it's not a crime to break in, and even if it does, fraud is still fraud.
No, it's not "just a mistake". But in cases like this, I think it's important to keep perspective. The effective results of his actions weren't particularly criminal. I mean, yes, what he did was technically a crime, but the effects of the crime weren't particularly different from if he had just cheated outright.
So probably (I'm guessing, because I don't know the kid or all the details) the guy deserves to be expelled, and put in a position where he really has to work his ass off to earn his way into any college. Community service wouldn't be a bad idea.
But I agree, jail time sounds like a bad idea for a situation like this. It's not just "too harsh", but you have to wonder what the effects of sending him to jail would be.
When your 18 years old, you still have a lot of time ahead of you, and a lot of learning-- for better or for worse. This guy probably still has the potential to turn it all around and be a productive member of society, but if you send him to jail, you're probably going to diminish that potential. Instead of learning to be productive, he'll be socialized to the jail environment, which means learning the wrong things, ie things that will actually make it harder for him to function appropriately in general society.
I think we should generally be cautious about sending young people (even over the age of 18) to jail, unless we feel that they pose a significant continuing danger or that they absolutely cannot be rehabilitated.
Regardless of the methods you choose, I would highly recommend using at least two different media.
IMO, any decent backup entails putting the data in at least two distinct locations. It's hard for personal data because you either have to keep a copy at work (which isn't necessarily private or secure), at a relative's, pay for online storage (which can be expensive), or some other place.
The second measure I would take for long-term backups would be to make sure you're storing checksums for each file. That way, you can periodically check for corruption within any of the copies you're keeping at each location. If you find a corrupted file, you copy it from the other location to preserve both sets as uncorrupted. If you start seeing corruption as a problem in one of your sets of media, copy the whole backup to new media.
I would figure that you're unlikely to keep the same media over the course of 20 years-- or at least, it's risky to assume that you will be using the same media. Make sure you're not doing anything that makes it hard to move everything to new media. Also, it's probably not a great idea to use the exact same media for all of your backup sets. Like, if you're backing up to DVD, don't back up everything to the same brand of DVD, leave copies in two places, and assume your safe. If the brand you've bought happens to have a manufacturing flaw that year that causes the discs to decompose in 5 years, you might lose both copies at approximately the same time.
Personally, with todays prices, I would probably keep one copy on hard drive. It's convenient. I might also burn to DVD and ship those to a relative's house. It's diverse both in terms of media and location, which is key. Then periodically check both copies for corruption and replace corrupted media (mark your calendar so you don't forget). Change media as you go.
Having said that, I think it would be great if they contributed to future versions of ODF, as long as they aren't the only voice being heard.
That's all I'm saying.
Somewhat off-topic: I would love for Apple's iWork to support ODF, since I think it's hands-down the best office suite for OSX, but they use an Apple-developed XML format that nothing else reads. I would be surprised if Apple developed their own format for the purpose of vendor lock-in (since it is XML, after all, and they don't have a big enough share to lock in), but rather I would guess it's because they felt that ODF was somehow unsuitable for their purposes.
So I'd love to see Apple take what they've learned from developing their own file formats and apply it to helping to improve ODF, and then supporting ODF fully. I think they may very well do that, but I suspect they'll move at a snails pace on that particular project.
But granted, a patent is a better protection in case someone sues - a patent gives them a possible opporunity to countersue.
Yes, that's the whole point. Defensive patents aren't necessarily about protecting yourself from being sued for using the technology you're patenting (though it may be that it's easier to legally establish that you've patented it than it is to establish prior art). It's so when someone comes after you and claims you're infringing on their obscure/stupid little patent, you can say, "Oh yeah? Well you're infringing on 75 of my stupid little patents. Do you want to fight this out, or do you want to leave me alone?"
This is the basic way that big companies (Google, Apple, IBM, Novell, Microsoft, etc) protect themselves from each others' patents. Unfortunately, it doesn't really protect against real patent trolls, who have produced no product and therefore cannot be accused of infringing on a patent.
Seriously, hate the game, i.e. the patent system. It makes a lot of sense for companies to patent everything they possibly can, if only for defensive purposes. If the patent is accepted, cool, you have another patent to sit on. If it's not accepted, oh well, you lost some time and money. There's no real reason not to try to patent any little thought that passes through your mind, no matter how stupid or banal.
There are probably other reasons to hate this guy. Being even somewhat responsible for Facebook is probably enough. But filing for frivolous patents is just the way you do business these days, so nothing to get excited about there. It's just an indication that (in case you haven't figured it out) the patent system is in serious need of reform.
From the impression I got, we got thrown a bone, and ODF and OOXML are going to be merged in the next couple of years, and MS will have de facto control because OOXML allows for proprietary extensions.
You know, in abstract, I don't think this is all bad. If you ask me, companies like Microsoft and Apple (and anyone else making office suites) should be involved in making ODF v2. If it's really going to be the common, standard interchange format for office suites, everyone should have input.
Because I could see someone writing an office suite and saying, "Standards are good and all, but ODF doesn't do what we need it to do. It's too bulky, but doesn't allow us to support [feature X], doesn't support [feature Y] in a way that allows us to get good performance, and [feature Z] makes it too difficult for us to develop a good converter." (or whatever. I don't know what valid complaints someone might have about ODF) And ideally, all those people would have a place at the table to talk about making ODF better, so that no one has any reason not to support ODF.
Of course, the only real problem here is that we don't trust Microsoft to work well within the system in order to develop improvements that will remain free. I still hold out some hope that one of these days, Microsoft will get the idea that they shouldn't be sabotaging technological progress by keeping a strangle-hold on computing, but I share your distrust.
Aren't you facing the exact same risk whenever you cite any other source, too?
Yes, anything can be misleading or inaccurate. That's not why citation matters. The purpose of citation is so that the reader can refer to the source to see (a) whether the source supports the interpretation you offer; and (b) how the source supports itself.
The second reason (b) is why you should always cite primary sources. The point isn't that primary sources are infallible, but that if they're truly primary sources, they'll support themselves. They'll give examples, evidence, etc. as to why the claims they're making are true, and the reader is then able to evaluate the claims on the basis of the person who originated those claims.
If you cite a secondary source, then you're leading the reader on a trail of citations that might go nowhere. I could cite you, you could cite someone else, that someone else cite yet another person, and off we go. You're essentially setting up a research project for the reader to figure out where the information actually came from.
Also, by the time the information comes through so many people, it can be distorted. It can be like a game of telephone, where what started out as a fact gets interpreted, and the interpretation gets interpreted, and that interpretation gets interpreted, ad nauseam. So by the end, you have no idea how distorted the truth is.
So seriously, if your research paper is relying on certain facts, try to find the original piece of writing that asserted those facts, and read that work for yourself. If you can't do that (in the case of a lost work that no longer exists, but is cited elsewhere), find the source that is as close as possible to the original, and cite that. Always go to the most original point, and always cite the primary work.
Wikipedia is a perfectly good place to start, and luckily they've started to encourage people to cite sources so that you can find the primary source for yourself. So when you want to use a fact from Wikipedia, follow their citation, read the work for yourself, and then you can cite *that* work as your primary source.
Ok, so why bother having smart kids go to school at all? We should let them sit at home, or meet on their own accord for classes with no teacher. They can work it out for themselves, right?
So if you disagree with that, then you should be willing to admit that even smart kids need the attention of a teacher. But all this still ignores the possibility of smart kids (and well-behaved kids) who don't have exceptionally healthy home-lives. And who has a perfect home-life, after all?
Not that I think we shouldn't give attention to kids who are struggling, whether they have interested parents or not.
I'm just saying that if you have 1 teacher and 30 kids, paying special attention to 5 kids will mean the teacher is paying less attention to the remaining 25. Smart kids need help too, so there you go, focussing on struggling kids hurts the education of high-achievers.
Ideally, we would have enough teachers that every kid could get some special attention. But if you have to run big worker-factories like today's public schools, you have to accept that some kids will fail. Failing to accept that means that we'll fail to develop our gifted children to their potential.
I agree, and that's the phenomenon I was trying to indicate. I think partial blame (at least partial) has to go to the media. The evening news will put up a story that says, "Scientists have proven that drinking green tea cures cancer!" and they don't really provide any context.
Of course, when you look into the study, that's not even what the study was about. So it might turn out to be a small study as to whether the effects of some particular extract that exists in green tea (but in small quantities), when injected directly into the tumor, can help radiation therapy be more successful, and they found that it was helpful to a small but statistically relevant degree.
So I'm just making that example up off the top of my head, but stuff like that happens, where the study is fine, but the media outlets don't do their research, misunderstand the nature of the study, and then try to make it as sensationalistic as possible.
But of course the main problem is that people generally believe it without any investigation. Of course, who expects people in general to go around reading research papers?
Here's the problem with your assessment: If you tell kids they're dumb, they tend to live down to your expectations. Tell kids they're gifted, and their scores go up.
Well I do believe that that's somewhat true, but also not so simple. If I tell kids they're useless, they'll lose interest. If I tell them they're perfect just the way they are, they don't tend to try very hard either. There's something in between, trying to make them feel like they're doing well, but that there is much left to do, that works a bit better.
But I do think people should recognize that attention is not an unlimited resource. It's not about which kids are smart and which are dumb, but that you can only pay so much attention to any one child before you're not paying enough attention to the others. And when you're teaching a class with 40 kids, your attention is divided pretty thinly already.
But kids really soak up attention. I'm not a teacher, but I've spent enough time with kids to know that it's terribly important that someone has taken an interest in them, spent time with them, and bothered to try to help them grow up. If you ignore them, they'll turn to mischief just to get your attention.
Garlic cures lots of ills (seems obvious to some people) --> Rub garlic on your feet and put socks on ("When I wake up I get garlic breath so I know it works!" Obvious, right?) --> PROFIT
What, are you crazy? You think it's obvious that rubbing garlic on your feet will cure headaches, because you get garlic on your breath? That doesn't seem the slightest bit obvious. In fact, it doesn't make the slightest bit of sense.
There are no studies that have shown this works for anything, but people still believe it. In the same sense that there are no studies that have shown that the NCLB act benefits society -- but some people still believe it.
Er... so? What does that have to do with anything? I'll repeat myself: I said *some* people won't believe obvious things until you say that "studies have proven it". You respond by providing an example where other people have believed something that's not at all obvious without having evaluated any studies. So your example obviously doesn't serve to rebut but claim.
Studies = good. Paying attention to studies = good.
And "taking random studies, failing to evaluate the quality of the study, taking the results of the study out of context, and using those out-of-context interpretations as indisputable proof" = bad.
But more to my point, "failing to assume blatantly obvious things when you have no studies whatsoever to rely on" = stupid, and "Failing to use your own mind and instead assuming that a statement is true simply because someone else prefixed the statement with 'studies have proven'" = stupid.
Slashdot Mirror
And it's not because it's nerdy (as the summary opines).
Yeah, my favorite part was, "Is this because of the fact that Bill Gates has made the whole industry look nerdy?" Really? Bill Gates made it look nerdy? Like if not for Gates, the whole industry would be filled with badass cowboys and hot chicks or something?
Yeah, even the fact that "badass cowboys and hot chicks" popped into my head as the opposite of "nerdy" is probably an indication that I'm an IT nerd.
But yeah, I've found that at least the IT work that falls on the support/maintenance side (as opposed to the development side) is kind of boring crap-work. It's fixing problems that some other moron broke, and cleaning up problems caused by poor design. It's 2008, and we still don't even have decent backup/archive methods. Every product out there has huge problems and gaping holes in their functionality that should have been fixed 15 years ago, but instead everyone has been working on things like database-driven filesystems that never make it to market.
That's right, I'm looking at you, Microsoft.
InfoTech work isn't all science-fictiony and cool. Oddly, it's more like being a Fonzie in training. It's like all this technology amounts to a broken jukebox that has to be smacked in just the right way to get it going, and you're just hoping to learn how to do that so you can stand around looking cool until someone needs you.
Except that this Fonzie never ends up looking cool and everyone treats him like a trained monkey. "Slap the machine and play me a song!" they all yell.
Oh, yeah, I know I've jumped the tracks and gone into nonsense. Whatever. I work in IT. Making up random crap on Slashdot is the most interesting part of my day.
Funny, I was just posting this in the other recent AI thread. So how outside the mainstream are you?
The third option would of course have benefits, but it would be rather daunting to undertake. Anything short of that will be an intelligence which makes mistakes, because that's the nature of intelligence-- to analyze situations and draw conclusions which aren't completely founded, based on incomplete information.
No, really, I'm purposefully saying that you need emotions *first*, and then you can think about growing intelligence. I know that's completely the opposite of what other people are saying, but my point is that they're approaching the thing backwards because they don't understand the problem.
I'm not saying their emotions have to be complicated or varied at first, but drives are a fundamental starting point of intelligence. I say don't even bother starting to create real AI unless you have some kind of a plan to make your AI want things.
The problem, if I had to guess, is that the people trying to create AI are the sort who like to think of their own intelligence as a something removed from their physical bodies and from the world they live in. They're the sort who try their best to convince themselves that their own minds behave like disinterested computers, and then use that misunderstanding of their own intelligence as a model of what they're trying to perceive.
So they're thinking they're going to create an intelligence first, and then give the intelligence orders on what to do, and tools in order to do it. I'm saying that won't work to create real artificial intelligence. You have to build it with desires, drives, aversions, and an existence in a world (even if it's a virtual world) where it can pursue its desires. Not just because it would be "something for the AI to do", but because those are fundamental components of intelligence.
I think his point is that squatters already try to get users through misspellings and typos and such. So open up TLDs, and you'll see "disney.con", "disney.cpm", "disney.c0m", etc. And it's innocuous enough when they're just serving ads, but it's a bit more annoying when it's phishing sites. Imagine you're typing your bank's URL and you accidentally type "citi.con", and load up a page that looks just like "citi.com". If you put in your account info, you're pretty much screwed.
"isgay.mobi" isn't taken yet.
I get the joke, but I do wonder whether it might make sense to reevaluate the use of TLDs and DNS.
The reason I say this is that TLDs *are* confusing to people. There's often no particular reason why a domain might be using .com, .net, or .org. I know there's supposed to be a reason, but people don't follow those rules. Some people even use .ws on the idea that it stands of "WebSite", even though it's really just another country code. But even without all that, it's still confusing-- you know, the old story about people visiting "whitehouse.com" while looking for the White House website, being unfamiliar with the .gov TLD.
I don't know. The system seems to be working fine for the time being, but I've been trying to think of whether there might be some other method for finding the site you want or sending e-mail to the person you want, but one that doesn't reward people for snapping up domains early. Imagine what it'll be like in 100 years (assuming we are still using this system), trying to find a decent domain name that's not already in use. If we get to the point of having to break out "jfdsodkfhdlkw-jhfjdihfdu.com", I'm not sure that's much better than using IP addresses.
This is the primary point I came in here to say. Whenever I've read anything about AI, it seems to be based on cool science-fictiony ideas, or else it's actually a simpler method to use statistical analysis to approximate human decision-making for particular purposes. If you're talking about real self-aware thinking things, the approaches are all wrong.
People tend to act treat the subject as though dumping enough raw information into a fast enough processor will yield intelligence, and then as that intelligence grows and develops, things like "sensible responses to answers" or "appropriate emotional responses" will emerge. Or else they think grouping enough "appropriate responses" will eventually yield intelligence.
It seems to me that that's all backwards. If you want to design an artificial intelligence, you first need a good philosophical understanding of how intelligence works, which will tell you straight-off something that AI researchers don't seem to consider: intelligence is an animal trait.
I think the absolute first thing you need to do is to figure out how to give machines emotions, to approximate pleasure/desire and pain/aversion. The second thing you need to do is give it "senses", and the ability to draw a very basic sensory conception of its world based on those senses, which includes a sense of time and objects. Also, you'll have to give it the ability to interact with its world in such a way that it is able to pursue its desires, encounter obstacles, and experience "pain". Finally, you'll have to figure out a way to give it the ability to adapt, to "rewrite its programing", preferably in a way that allows it to reproduce and evolve.
So in a way, the most obvious answer is that if you want an artificial intelligence, you'll have to design an artificial/virtual animal and place it into an environment where it can evolve intelligence. There may be some shortcuts on growing/evolving it faster, but you shouldn't be quick to discount the animal nature of intelligence as we know it.
And the reason for these things are bound up with the fact that, like I said, the only model for real intelligence we have to base anything on is animal intelligence. Animals develop and express their intelligence by being self-motivated in a world that presents obstacles. If there's nothing you want, there's no point in figuring anything out. If there's no way to get what you want, then there's no point in figuring things out. If there are no obstacles in your way, then there's nothing to figure out.
So if you don't have a self-motivated desire and the ability to move towards achieving that desire, then you can't make self-determined intelligent decisions. If course, this also presents a scary twist to the whole AI thing, because it suggests one of the chief scifi fears of AI will turn out to be correct: If we're successful in creating AI, we may not be able to control it.
But what's the advantage of getting a cert issued by someone whose root certificate isn't in browsers, vs. just generating my own?
That doesn't seem like a very good reason to me. What I mean is, that's a good reason why you wouldn't have everything encrypted at the outset, but that's not a good reason not to build out the protocols/infrastructure with the option of security/encryption in mind. Because even in cases where encryption is computationally too expensive for every transaction, you might still want to offer the option of encryption for for cases where encryption is worth the expense.
I guess it just seems a bit like the Y2K bug-- sure, it probably made sense at the time, but in hindsight it wasn't the best decision. Even now, secure protocols can seem like a bit of an add-on hack instead of an integrated part of the system. FTP is a pretty good example, IMO. FTP is awful security-wise, and FTP with TLS/SSL can be a bit of a PITA. SFTP looks to be a good solution, but ends up having its own headaches. You generally either have to give people remote shell access or go through extra steps to disable the normal shell and jail the user. I guess OpenSSH is now trying to address the issue, but it's still not particularly straight-forward, IMO. Also, SSH keys have none of the benefits that SSL certs have, e.g. identity verification or the ability to transparently revoke keys and issue new ones.
Yeah, it seems to me that it was an oversight that networking wasn't encrypted in the first place. When lots of these protocols were being developed, security didn't seem to be much of a consideration.
It's about time that these things got rectified, but I'm not sure what the best course is. For example, using SSL concerns me in that we've accepted the convention that certificates should be issued by certain set organizations that require exorbitant fees. I mean, hundreds or thousands of dollars per year for an SSL cert? Seems a bit much to me. Yeah, I know you can generate your own, which will cause you to get complaints from your websites' users when they see what looks to them like an error message.
I'm not a security expert, but I get the sense someone needs to go back to square one and figure out how to build a coherent, open, and secure model for networking that doesn't rely on giving such control to a small number of companies.
Let's take a look at the charges levied against the student:
34 felony counts of altering public record 11 felony counts of stealing and secreting public records 7 felony counts of illegal computer access and fraud 6 felony counts of burglary 4 felony counts of identity theft 3 felony counts of altering book of records 2 felony counts of receiving stolen property 1 felony count of conspiracy 1 felony count of attempting altering of a public record
And, as the article says, he's not just being charged with breaking in to the school's computers, he also put in a remote access program (which others could have used to do a variety of things with the school's network with). Then there's the fact that he broke into the principal's office to steal a test he did badly on...
Yes, and who was hurt in the end? What was damaged? What was the result of his crimes?
As far as I can tell, the problems caused by his crimes are essentially the same as if he were cheating. Specifically, the people who might have been hurt if the guy hadn't been caught would be the other student who failed to get into college because this kid took a slot that would otherwise have been available.
But (correct me if I'm wrong) nobody was directly harmed. Nothing of particular value was stolen. His actions were self-serving, but not malicious. He's potentially dangerous if his life takes the wrong turn, or potentially harmless if he learns the right things. A lot happens to a person between the ages of 18 and 28, and this guy will probably be a different person at the end of the next 10 years.
So the question in my mind, primarily, is not whether this guy did "the wrong thing" and should theoretically be punished. He's already going to be punished. He'll have a hard time getting into any college, and for the rest of his life, whenever he's asked on an application as to whether he's been convicted of a felony, he'll have to check "yes". He's going to have a lot of trouble because of his actions.
So the primary question in my mind is "what will give this guy the best chance of becoming a productive member of society?" Like, what events are most likely to turn him into a person that won't cause further trouble, or even more serious problems?
I doubt that jail time will be terrifically helpful, and I'm sure that significant time in jail will be counter-productive. Community service may be helpful. The most helpful thing will probably be for someone to show him that he can still be successful if he works his ass off and earns back the trust that he's lost.
38 years in jail is way too steep, any jailtime would probably be. But this is no where near "just a mistake". It is not a one-time break-in to prove that security is insufficient. He was deliberately and continously (34 times alledgedly) "altering public records", for his own personal benefit. Even if your security is inadquate it doesn't mean that it's not a crime to break in, and even if it does, fraud is still fraud.
No, it's not "just a mistake". But in cases like this, I think it's important to keep perspective. The effective results of his actions weren't particularly criminal. I mean, yes, what he did was technically a crime, but the effects of the crime weren't particularly different from if he had just cheated outright.
So probably (I'm guessing, because I don't know the kid or all the details) the guy deserves to be expelled, and put in a position where he really has to work his ass off to earn his way into any college. Community service wouldn't be a bad idea.
But I agree, jail time sounds like a bad idea for a situation like this. It's not just "too harsh", but you have to wonder what the effects of sending him to jail would be.
When your 18 years old, you still have a lot of time ahead of you, and a lot of learning-- for better or for worse. This guy probably still has the potential to turn it all around and be a productive member of society, but if you send him to jail, you're probably going to diminish that potential. Instead of learning to be productive, he'll be socialized to the jail environment, which means learning the wrong things, ie things that will actually make it harder for him to function appropriately in general society.
I think we should generally be cautious about sending young people (even over the age of 18) to jail, unless we feel that they pose a significant continuing danger or that they absolutely cannot be rehabilitated.
Regardless of the methods you choose, I would highly recommend using at least two different media.
IMO, any decent backup entails putting the data in at least two distinct locations. It's hard for personal data because you either have to keep a copy at work (which isn't necessarily private or secure), at a relative's, pay for online storage (which can be expensive), or some other place.
The second measure I would take for long-term backups would be to make sure you're storing checksums for each file. That way, you can periodically check for corruption within any of the copies you're keeping at each location. If you find a corrupted file, you copy it from the other location to preserve both sets as uncorrupted. If you start seeing corruption as a problem in one of your sets of media, copy the whole backup to new media.
I would figure that you're unlikely to keep the same media over the course of 20 years-- or at least, it's risky to assume that you will be using the same media. Make sure you're not doing anything that makes it hard to move everything to new media. Also, it's probably not a great idea to use the exact same media for all of your backup sets. Like, if you're backing up to DVD, don't back up everything to the same brand of DVD, leave copies in two places, and assume your safe. If the brand you've bought happens to have a manufacturing flaw that year that causes the discs to decompose in 5 years, you might lose both copies at approximately the same time.
Personally, with todays prices, I would probably keep one copy on hard drive. It's convenient. I might also burn to DVD and ship those to a relative's house. It's diverse both in terms of media and location, which is key. Then periodically check both copies for corruption and replace corrupted media (mark your calendar so you don't forget). Change media as you go.
Having said that, I think it would be great if they contributed to future versions of ODF, as long as they aren't the only voice being heard.
That's all I'm saying.
Somewhat off-topic: I would love for Apple's iWork to support ODF, since I think it's hands-down the best office suite for OSX, but they use an Apple-developed XML format that nothing else reads. I would be surprised if Apple developed their own format for the purpose of vendor lock-in (since it is XML, after all, and they don't have a big enough share to lock in), but rather I would guess it's because they felt that ODF was somehow unsuitable for their purposes.
So I'd love to see Apple take what they've learned from developing their own file formats and apply it to helping to improve ODF, and then supporting ODF fully. I think they may very well do that, but I suspect they'll move at a snails pace on that particular project.
Yes, that's the whole point. Defensive patents aren't necessarily about protecting yourself from being sued for using the technology you're patenting (though it may be that it's easier to legally establish that you've patented it than it is to establish prior art). It's so when someone comes after you and claims you're infringing on their obscure/stupid little patent, you can say, "Oh yeah? Well you're infringing on 75 of my stupid little patents. Do you want to fight this out, or do you want to leave me alone?"
This is the basic way that big companies (Google, Apple, IBM, Novell, Microsoft, etc) protect themselves from each others' patents. Unfortunately, it doesn't really protect against real patent trolls, who have produced no product and therefore cannot be accused of infringing on a patent.
Seriously, hate the game, i.e. the patent system. It makes a lot of sense for companies to patent everything they possibly can, if only for defensive purposes. If the patent is accepted, cool, you have another patent to sit on. If it's not accepted, oh well, you lost some time and money. There's no real reason not to try to patent any little thought that passes through your mind, no matter how stupid or banal.
There are probably other reasons to hate this guy. Being even somewhat responsible for Facebook is probably enough. But filing for frivolous patents is just the way you do business these days, so nothing to get excited about there. It's just an indication that (in case you haven't figured it out) the patent system is in serious need of reform.
You know, in abstract, I don't think this is all bad. If you ask me, companies like Microsoft and Apple (and anyone else making office suites) should be involved in making ODF v2. If it's really going to be the common, standard interchange format for office suites, everyone should have input.
Because I could see someone writing an office suite and saying, "Standards are good and all, but ODF doesn't do what we need it to do. It's too bulky, but doesn't allow us to support [feature X], doesn't support [feature Y] in a way that allows us to get good performance, and [feature Z] makes it too difficult for us to develop a good converter." (or whatever. I don't know what valid complaints someone might have about ODF) And ideally, all those people would have a place at the table to talk about making ODF better, so that no one has any reason not to support ODF.
Of course, the only real problem here is that we don't trust Microsoft to work well within the system in order to develop improvements that will remain free. I still hold out some hope that one of these days, Microsoft will get the idea that they shouldn't be sabotaging technological progress by keeping a strangle-hold on computing, but I share your distrust.
Remember how Micorsoft "supported" Java with their non-compliant, buggy implementation?
I really wonder sometimes whether that was purposeful, or if they're just unable to make a non-buggy implementation of... well, anything.
After trying out Windows Vista, it almost seems like they're trying to "embrace, extend, extinguish" their own products.
Aren't you facing the exact same risk whenever you cite any other source, too?
Yes, anything can be misleading or inaccurate. That's not why citation matters. The purpose of citation is so that the reader can refer to the source to see (a) whether the source supports the interpretation you offer; and (b) how the source supports itself.
The second reason (b) is why you should always cite primary sources. The point isn't that primary sources are infallible, but that if they're truly primary sources, they'll support themselves. They'll give examples, evidence, etc. as to why the claims they're making are true, and the reader is then able to evaluate the claims on the basis of the person who originated those claims.
If you cite a secondary source, then you're leading the reader on a trail of citations that might go nowhere. I could cite you, you could cite someone else, that someone else cite yet another person, and off we go. You're essentially setting up a research project for the reader to figure out where the information actually came from.
Also, by the time the information comes through so many people, it can be distorted. It can be like a game of telephone, where what started out as a fact gets interpreted, and the interpretation gets interpreted, and that interpretation gets interpreted, ad nauseam. So by the end, you have no idea how distorted the truth is.
So seriously, if your research paper is relying on certain facts, try to find the original piece of writing that asserted those facts, and read that work for yourself. If you can't do that (in the case of a lost work that no longer exists, but is cited elsewhere), find the source that is as close as possible to the original, and cite that. Always go to the most original point, and always cite the primary work.
Wikipedia is a perfectly good place to start, and luckily they've started to encourage people to cite sources so that you can find the primary source for yourself. So when you want to use a fact from Wikipedia, follow their citation, read the work for yourself, and then you can cite *that* work as your primary source.
Ok, so why bother having smart kids go to school at all? We should let them sit at home, or meet on their own accord for classes with no teacher. They can work it out for themselves, right?
So if you disagree with that, then you should be willing to admit that even smart kids need the attention of a teacher. But all this still ignores the possibility of smart kids (and well-behaved kids) who don't have exceptionally healthy home-lives. And who has a perfect home-life, after all?
Not that I think we shouldn't give attention to kids who are struggling, whether they have interested parents or not.
I'm just saying that if you have 1 teacher and 30 kids, paying special attention to 5 kids will mean the teacher is paying less attention to the remaining 25. Smart kids need help too, so there you go, focussing on struggling kids hurts the education of high-achievers.
Ideally, we would have enough teachers that every kid could get some special attention. But if you have to run big worker-factories like today's public schools, you have to accept that some kids will fail. Failing to accept that means that we'll fail to develop our gifted children to their potential.
...where did we get the notion that "No Child Left Behind" was actually helping anyone?I'm confused too. Why do you think my post is saying that "no child left behind" is helping people?
I agree, and that's the phenomenon I was trying to indicate. I think partial blame (at least partial) has to go to the media. The evening news will put up a story that says, "Scientists have proven that drinking green tea cures cancer!" and they don't really provide any context.
Of course, when you look into the study, that's not even what the study was about. So it might turn out to be a small study as to whether the effects of some particular extract that exists in green tea (but in small quantities), when injected directly into the tumor, can help radiation therapy be more successful, and they found that it was helpful to a small but statistically relevant degree.
So I'm just making that example up off the top of my head, but stuff like that happens, where the study is fine, but the media outlets don't do their research, misunderstand the nature of the study, and then try to make it as sensationalistic as possible.
But of course the main problem is that people generally believe it without any investigation. Of course, who expects people in general to go around reading research papers?
Here's the problem with your assessment: If you tell kids they're dumb, they tend to live down to your expectations. Tell kids they're gifted, and their scores go up.
Well I do believe that that's somewhat true, but also not so simple. If I tell kids they're useless, they'll lose interest. If I tell them they're perfect just the way they are, they don't tend to try very hard either. There's something in between, trying to make them feel like they're doing well, but that there is much left to do, that works a bit better.
But I do think people should recognize that attention is not an unlimited resource. It's not about which kids are smart and which are dumb, but that you can only pay so much attention to any one child before you're not paying enough attention to the others. And when you're teaching a class with 40 kids, your attention is divided pretty thinly already.
But kids really soak up attention. I'm not a teacher, but I've spent enough time with kids to know that it's terribly important that someone has taken an interest in them, spent time with them, and bothered to try to help them grow up. If you ignore them, they'll turn to mischief just to get your attention.
Garlic cures lots of ills (seems obvious to some people) --> Rub garlic on your feet and put socks on ("When I wake up I get garlic breath so I know it works!" Obvious, right?) --> PROFIT
What, are you crazy? You think it's obvious that rubbing garlic on your feet will cure headaches, because you get garlic on your breath? That doesn't seem the slightest bit obvious. In fact, it doesn't make the slightest bit of sense.
There are no studies that have shown this works for anything, but people still believe it. In the same sense that there are no studies that have shown that the NCLB act benefits society -- but some people still believe it.
Er... so? What does that have to do with anything? I'll repeat myself: I said *some* people won't believe obvious things until you say that "studies have proven it". You respond by providing an example where other people have believed something that's not at all obvious without having evaluated any studies. So your example obviously doesn't serve to rebut but claim.
Studies = good. Paying attention to studies = good.
And "taking random studies, failing to evaluate the quality of the study, taking the results of the study out of context, and using those out-of-context interpretations as indisputable proof" = bad.
But more to my point, "failing to assume blatantly obvious things when you have no studies whatsoever to rely on" = stupid, and "Failing to use your own mind and instead assuming that a statement is true simply because someone else prefixed the statement with 'studies have proven'" = stupid.