The XO already has USB ports. So the XO can be expanded with a USB memory stick - or even a hard drive for use at a desktop with power. The USB ports are hardened against dirt/water. I suspect it would require more effort to harden a SD slot. Furthermore, if they do add a flash slot, I read a showdown between the durability of various flash formats, and compact flash won hands down. The ultimate test was giving the memory sticks to 6 year old boys and instructing them to destroy the sticks. The boys pounded them with rocks, etc, in their efforts. Only the compact flash survived that test. CF isn't the smallest format, but I'm not sure that is the most important feature for where the XO needs to go, and it is small enough.
Good point. There is a big difference between Bugs Bunny violence and hyper realistic violence - whether movies or todays super 3D performance video games. There would also seem to be some survival value in the response. If there is a lot of violence going on around you, you need to be ready to respond in kind (or find a good hiding place).
Okay, again, I have to point out that a police state is not dependent on the number of laws. A police state is one where the police, or executive branch, can lock people up without charge outside of the law for undetermined amounts of time.
You know, like here.
In theory, you are right. However, in practice, bookshelves full of laws mean that the police or executive branch can always lock up anyone they want for any length of time. You are breaking many laws all the time, whether you realize it or not, no matter how law abiding you try to be. So finding something to charge you with is a trivial matter.
Except there's one major problem with your approach: Blacklisting IP addresses is a badly flawed way of blocking spam. There are many scenarios (all common) whereby you end up blacklisting valid IPs (e.g. zombies on dynamic IPs, shared hosting, etc.). Those are all very valid reasons to blacklist IPs. I certainly don't want most mail sent from a dynamic IP. If someone (like me) has a semi-stable dynamic IP, and doesn't send spam from it, then it doesn't get blacklisted. Mostly I blacklist domains anyway. I track reputations (rather than one strike and your out for a week) and start blacklisting rfc4408 authenticated mail from domains or rfc2821 authenticated helo domains or PTR authenticated non-dynamic connect IPs when they send mostly spam. Rejection for a new domain starts after about 20 spams in a row. I reject mail with none of those forms of identification. I'll accept DKIM eventually.
I've had to do that on many occasions because lusers tend to put their own email into their email client incorrectly. It is always something close, so trying variations of hyphen/underscore, period separators, etc, will often get it.
Honeypots have been a published anti-spam technique for a decade. The idea is to publish bogus mailboxes that are not close to any legit mailbox. Any message with a honeypot as any recipient is spam. 100% accurate. (And I blacklist the IP for a week for good measure.) I use a variation, where any message with 3 or more invalid recipients is spam (blacklist IP). That is a little risky since someone may legitimately be trying various mailboxes manually with a telnet session because they forgot the exact name.
This technique gives each recipient a score between 0 and 1 that reflects how close to a honeypot that recipient is, with actual honeypots (100% spam) being 1.0.
While this is a rare case of the algorithm actually being original (as opposed to rehashing an old idea "on the web"), it is yet another software patent. I'll lump it with RSA - the kind of software patent you might actually want to read if all software patents were that original.
But really, compared with the old AT&T, Standard Oil, and especially the British East India Company, Microsoft is an amateur.
I agree. I used to get really exercised about M$, but then I started reading about Monsanto. M$ is greedy and unethical, but their behavior doesn't usually kill or maim people. Monsanto is another story.
My biggest problem E-books is how easy they are (the DRMed ones) to centrally control. The Ministry of Truth was an expensive operation, what with collecting, incinerating, and reprinting books they wanted to change. E-books can be "updated" at the push of a button. WORM media and the kind of widespread copying publishers hate are our weapons against the rise of the Ministry of Truth.
... are the biggest roadblock to IT development. No entity, not even non-commercial open source, is safe from being sued to oblivion for the crime of not only having an idea, but implementing it. The risk is still low enough, that most of us are still taking it. But it is building like an epidemic. The only defense is a policy of Mutually Assured Destruction backed by a massive portfolio of your own asinine software patents.
I was looking at solely whether or not GMO food is safe to eat. Some say it isn't, but you can't deny that most scientists working in this area consider GMO food safe for human consumption. Regardless of how safe something is for most people, there will be some who have a bad reaction to it. This is especially true for biological molecules. Even with non GMO food, people can have allergies or IBS or whatever (e.g. nuts, milk). Since the GMO food is different from regular food, it *must* be labeled, even if the changes are beneficial for most people. To do otherwise is to treat people like lab rats, where you don't care if.001% have a bad reaction and suffer or die.
Most scientists working in this area see no harmful effects from GM food, yet many in the general public think GM food is going to kill them, cause cancer, or other such nonsense. Or human cloning. Many people in the general public are absolutely terrified of human cloning, yet I'd bet most scientists see no problem with this from a biotech standpoint, except for a few ethical considerations.
That is a straw man. The issues with GM have to do with labelling (so you know you're not getting what the term "tomato" usually stands for, whether or not you like the alleged improvements) and stuff escaping and destroying ecosystems.
We already have a problem with BT corn escaping and contaminating crops of small/poor farmers. Surprise, surprise, the pollen gets blown into other fields. Fortunately, most people aren't highly allergic to the toxin, but then Monsanto might come along and sue them to take away their livelihood (I'm only aware of an actual example with GM canola, but its going to happen). There's nothing like having to beg giant faceless corporations for permission to plant your subsistence crops. And you thought proprietary computer software was bad...
Labelling of GMO food is important because our understanding of nutrition is always incomplete. Tang was cutting edge for the Apollo mission, but is a nutritional joke comparable to Koolaid now. The "improvements" to GMO food are worth trying, but *only* if there is a way for consumers with bad reactions to avoid them. A conventional example is MSG. Most people don't have a problem with it, it has been used for centuries, it improves the flavor of food. But I personally know several people for whom MSG triggers severe migraine headaches. They *really* need to know which packages have MSG without having to guess about cowardly disguises like "natural flavor". We carefully label "may contain traces of peanuts" because peanut allergy can be life threatening. But a severe migraine puts the sufferer out of commission for a day - a big hit to productivity.
When less informed people hear about problems they don't fully understand, they often don't describe them accurately or even recognizably, allowing those willfully ignoring the problems to attack straw men.
While the GPL licenses are some of the more complex in OSS, the provisions affect distribution, not use. Mr. Ball can *use* GPL stuff all he wants without worrying about licensing issues. It is only when he develops the Ball Instrument Manufacturing accounting package based on GPL software and wants to distribute it that he has to be careful.
So when the first scientist creates life that gives us three options (by my in-the-moment count): 1) there's nothing divine about creating life, 2) those scientists are the hand of God at work or 3) humanity has elevated itself to the point where it possesses at least some divine powers. Humanity has always been considered to already possess some divine powers by Jews and Christians (and other religions for different concepts of "god"). "Ye are gods." "We are made in the image of God." Classically, literary and mathematical works were pointed to as examples of "sub-creation". Humans creating self-consistent worlds that exist apart from their creator. In the computer age, we create dynamic virtual worlds as a matter of course, and people these virtual worlds with self-replicating agents. We interact with these virtual worlds via avatars, and occasionally work "miracles" by patching memory or using cheat codes. We "speak" (write software) worlds into existence. We are gods.
If you think about it, *our* origin is a historical, not a scientific question. It is a question of what *did* happen, not what we normally observe to happen. Consider this: if in the future, we succeed in creating artificial intelligence/life, we will have firmly established that we *could* have been designed. Our creations may go on in their turn to design intelligences. The source of that chain of created beings has to stop somewhere - at an uncreated intelligence. That stopping point is by definition "God". Why are we so sure that we are the ultimate uncreated intelligence? The application of Occam's razor to origins depends on your presuppositions.
So yes, all 3 possibilities have always been tendered in classic Jewish and Christian theology. The issue is not whether "only God" can create. The issue is whether we (humanity as a whole) are the top of the chain (or the top of one of many chains). The issue is whether we are God, or one of the gods.
Science can't prove or disprove there's a god or gods, but it can turn up an awful lot of evidence that a particular idea of what a god is like is unlikely to be correct.
Scientists creating life from inanimate matter in the lab has absolutely nothing to say about whether god exists, but it pretty much blows out of the water the idea that creating life is the exclusive province of the divine. [italics mine]
Actually, *scientists* creating life from inanimate matter in the lab is an example of intelligent design (in this case the designer being the scientists). Historically, many proponents of Judaism and Christianity have proposed intermediates in their interpretation of creation (e.g. angels). A major point of just about all flavors of Jewish and Christian theology is that God prefers to use human (and sometimes angelic) agents as opposed to directly working miracles.
For that matter, we *already* design machines exhibiting many of the characteristics of "simple" life (robots), and the primary missing factor (self replication) is doable in principle - but just isn't cost effective for current applications and technology. (The ability to extract needed atoms from a wide variety of food sources via nano-tech is crucial to making self-replication cheap. Mining, smelting, growing crystals, etc are not easy to do on a small scale.) When we observe life springing from inanimate matter *without* the help of scientists (other than a reasonable interpretation of ancient conditions - or even modern conditions on some lifeless planet), then we will have observed abiogenesis.
Ok, not an NAS solution, but with 300GB to 500GB USB drives going for $100US or so, that is what I use for backup. The drives are hot plug, so they are easily moved between all the machines needing backup and swapped offsite.
The biggest problem with USB is filesystem compatibility if using with multiple OSes. With all linux machines, I don't bother with a MBR or parititions, and just put a volume group on the drive. But Windows doesn't understand that without special drivers (which do exist, BTW). Conversely, putting NTFS on the drive is problematic for Linux (it being an undocumented format). FAT32 is compatible, but way too inefficient for such a large drive.
My goal for backup is a encrypted distributed filesystem, where you share your 500GB USB drives with random people over the internet in a tit for tat arrangement. Pieces of data are encrypted and distributed over many locations, such that it can be recovered and decrypted when only XX% of the pieces are obtainable (where XX is configurable). Nobody can read other peoples pieces stored on their drive, and people can join/drop out at any time without jeopardizing anyones backup (unless a large percentage drops out at once).
Services like Amazon S3 are ok for putting encrypted tarballs on their server (including differential tars).
It wouldn't be charging for payload, but for QoS. The difference is that the customer gets to pick which packets are high/low priority via router settings (a consumer router would have simple options like "optimize voip") rather than the ISP deciding for you based on back room deals. The rich are always going to be able to afford bigger bandwidth (and bigger house and bigger car...). QoS should be a tool to efficiently manage what you can afford.
There is a huge problem. If e.g. VoIP works "better" than, say bittorrent, the next p2p *will* use VoIP[1] for faster downloads. Software writers and users are willing to game the system (use incorrect QoS settings etc) in order to improve their data rate on the expense of everybody else.
Net neutrality will not succeed, it cannot.
Not if ISPs charge extra for packets with low latency or high bandwidth bits set. Sure, you can set the high bandwidth bit for your torrents, but then you gotta pay for it. Much cheaper for most people to leave torrents on the "as available" plan and use QoS for their Voip and the occasional huge file they gotta have now.
What might be an idea, although this is abusable, would be some way of having a site collect info from others. Say domains A, B, and C are getting hit from the same IP range and blacklist it. They communicate that to some server, so domain D and E would either blacklist or use tarpits/QoS or other precautionary measures until their spam/DoS filters get triggered. I do this for domains using the GOSSiP protocol as implemented by pygossip. Each MTA consults a gossip server for reputation, and provides spam/ham feedback on specific messages. The gossip server maintains its own reputation database, and also queries peers for their opinion of a domain, and combines scores to get a reputation and a confidence.
I keep IP blacklists and domain blacklists. IPs are blacklisted for 7 days. I experimented with various settings, measuring the diminishing returns (in saved bandwidth) from keeping them blacklisted longer and longer. 7 days is pretty optimal with about 500000 IPs blacklisted at any one time. This keeps spam bandwidth down to a continuous 100Kbps (400000 messages / day - for a one user domain!). Domains are auto-blacklisted based on reputation: total spams/total hams over the last 1024 messages. Reputation decays with time, so that a domain that finally purges their 'bot can send mail again in a week or so. Manually blacklisted domains are permanent, but are manually reviewed every year. There are some domain names that only people I don't want to hear from would buy.
There is no possible system designed around equal voting rights that won't succumb to the fact that half the voters are idiots and a significant proportion of those on the other side of the bell curve, the non-idiots if you like, are uninformed or misinformed. There are 3 roughly equal groups. The simple and ignorant, the smart and good hearted, and the smart and self-centered. The survival of civilization hinges on the decisions of the simple and ignorant. They are going to be led by the good/smart, or deceived by the evil/smart. It behooves the good/smart to be winsome, rather than cynical and pessimistic.
Republic: Where all three votes are irrelevant, because only uninformed "representatives" can vote on issues. The intent of a republic is that citizens vote for a representative that they know personally, not just from a campaign. In the US, there are no federal candidates that anyone knows personally - so federal elections are largely a joke (there need to be more reps or more levels to handle the population). Local elections, however, are a different matter. I get to go to parties and chat with local and state (delegate, representative) candidates. So I can vote with confidence. Governors and federal candidates generally come from the pool of local politicians. So give high priority to attending local primaries, political "hob nob" parties, and elections. That is the last great hope in the US.
Who did this programming? God. And the crazy thing is that beehives are only one tiny part of it. The overall program encompasses the entire universe. So ha ha ha... cuz you can study those bees all day long and it won't make you a better web programmer. It is the glory of God to conceal a matter, But the glory of kings is to search out a matter. So clearly, God *intended* for us to learn from His creation, contrary to the parent (which could be either a sarcastic dig against a straw man, or a troll). In fact, one of the criticisms offered by Intelligent Design against Materialism is that science is sometimes hindered when it is assumed that some aspect of the world has no design or function, when it turns out later that there is a well hidden design/function. "Junk" DNA comes to mind. (Of course, with ID, one is sometimes tempted to provide hand waving "God did it" explanations without actually trying to *understand how* He did it.)
It is the library, in particular event notification, that "caused" the problem. The language was fine. The same thing happens in Java when beginning programmers forget to unregister objects from event sources (or anything else requiring registration). I use the term "data cancer" to distinguish this type of bug from actual "memory leaks", which are in fact impossible with a working GC.
Slashdot Mirror
The XO already has USB ports. So the XO can be expanded with a USB memory stick - or even a hard drive for use at a desktop with power. The USB ports are hardened against dirt/water. I suspect it would require more effort to harden a SD slot. Furthermore, if they do add a flash slot, I read a showdown between the durability of various flash formats, and compact flash won hands down. The ultimate test was giving the memory sticks to 6 year old boys and instructing them to destroy the sticks. The boys pounded them with rocks, etc, in their efforts. Only the compact flash survived that test. CF isn't the smallest format, but I'm not sure that is the most important feature for where the XO needs to go, and it is small enough.
Good point. There is a big difference between Bugs Bunny violence and hyper realistic violence - whether movies or todays super 3D performance video games. There would also seem to be some survival value in the response. If there is a lot of violence going on around you, you need to be ready to respond in kind (or find a good hiding place).
Okay, again, I have to point out that a police state is not dependent on the number of laws. A police state is one where the police, or executive branch, can lock people up without charge outside of the law for undetermined amounts of time.
You know, like here.
In theory, you are right. However, in practice, bookshelves full of laws mean that the police or executive branch can always lock up anyone they want for any length of time. You are breaking many laws all the time, whether you realize it or not, no matter how law abiding you try to be. So finding something to charge you with is a trivial matter.Those are all very valid reasons to blacklist IPs. I certainly don't want most mail sent from a dynamic IP. If someone (like me) has a semi-stable dynamic IP, and doesn't send spam from it, then it doesn't get blacklisted. Mostly I blacklist domains anyway. I track reputations (rather than one strike and your out for a week) and start blacklisting rfc4408 authenticated mail from domains or rfc2821 authenticated helo domains or PTR authenticated non-dynamic connect IPs when they send mostly spam. Rejection for a new domain starts after about 20 spams in a row. I reject mail with none of those forms of identification. I'll accept DKIM eventually.
I've had to do that on many occasions because lusers tend to put their own email into their email client incorrectly. It is always something close, so trying variations of hyphen/underscore, period separators, etc, will often get it.
Honeypots have been a published anti-spam technique for a decade. The idea is to publish bogus mailboxes that are not close to any legit mailbox. Any message with a honeypot as any recipient is spam. 100% accurate. (And I blacklist the IP for a week for good measure.) I use a variation, where any message with 3 or more invalid recipients is spam (blacklist IP). That is a little risky since someone may legitimately be trying various mailboxes manually with a telnet session because they forgot the exact name. This technique gives each recipient a score between 0 and 1 that reflects how close to a honeypot that recipient is, with actual honeypots (100% spam) being 1.0.
While this is a rare case of the algorithm actually being original (as opposed to rehashing an old idea "on the web"), it is yet another software patent. I'll lump it with RSA - the kind of software patent you might actually want to read if all software patents were that original.
I agree. I used to get really exercised about M$, but then I started reading about Monsanto. M$ is greedy and unethical, but their behavior doesn't usually kill or maim people. Monsanto is another story.
My biggest problem E-books is how easy they are (the DRMed ones) to centrally control. The Ministry of Truth was an expensive operation, what with collecting, incinerating, and reprinting books they wanted to change. E-books can be "updated" at the push of a button. WORM media and the kind of widespread copying publishers hate are our weapons against the rise of the Ministry of Truth.
... are the biggest roadblock to IT development. No entity, not even non-commercial open source, is safe from being sued to oblivion for the crime of not only having an idea, but implementing it. The risk is still low enough, that most of us are still taking it. But it is building like an epidemic. The only defense is a policy of Mutually Assured Destruction backed by a massive portfolio of your own asinine software patents.
That is a straw man. The issues with GM have to do with labelling (so you know you're not getting what the term "tomato" usually stands for, whether or not you like the alleged improvements) and stuff escaping and destroying ecosystems.
We already have a problem with BT corn escaping and contaminating crops of small/poor farmers. Surprise, surprise, the pollen gets blown into other fields. Fortunately, most people aren't highly allergic to the toxin, but then Monsanto might come along and sue them to take away their livelihood (I'm only aware of an actual example with GM canola, but its going to happen). There's nothing like having to beg giant faceless corporations for permission to plant your subsistence crops. And you thought proprietary computer software was bad...
Labelling of GMO food is important because our understanding of nutrition is always incomplete. Tang was cutting edge for the Apollo mission, but is a nutritional joke comparable to Koolaid now. The "improvements" to GMO food are worth trying, but *only* if there is a way for consumers with bad reactions to avoid them. A conventional example is MSG. Most people don't have a problem with it, it has been used for centuries, it improves the flavor of food. But I personally know several people for whom MSG triggers severe migraine headaches. They *really* need to know which packages have MSG without having to guess about cowardly disguises like "natural flavor". We carefully label "may contain traces of peanuts" because peanut allergy can be life threatening. But a severe migraine puts the sufferer out of commission for a day - a big hit to productivity.
When less informed people hear about problems they don't fully understand, they often don't describe them accurately or even recognizably, allowing those willfully ignoring the problems to attack straw men.
While the GPL licenses are some of the more complex in OSS, the provisions affect distribution, not use. Mr. Ball can *use* GPL stuff all he wants without worrying about licensing issues. It is only when he develops the Ball Instrument Manufacturing accounting package based on GPL software and wants to distribute it that he has to be careful.
If you think about it, *our* origin is a historical, not a scientific question. It is a question of what *did* happen, not what we normally observe to happen. Consider this: if in the future, we succeed in creating artificial intelligence/life, we will have firmly established that we *could* have been designed. Our creations may go on in their turn to design intelligences. The source of that chain of created beings has to stop somewhere - at an uncreated intelligence. That stopping point is by definition "God". Why are we so sure that we are the ultimate uncreated intelligence? The application of Occam's razor to origins depends on your presuppositions.
So yes, all 3 possibilities have always been tendered in classic Jewish and Christian theology. The issue is not whether "only God" can create. The issue is whether we (humanity as a whole) are the top of the chain (or the top of one of many chains). The issue is whether we are God, or one of the gods.
Scientists creating life from inanimate matter in the lab has absolutely nothing to say about whether god exists, but it pretty much blows out of the water the idea that creating life is the exclusive province of the divine. [italics mine]
Actually, *scientists* creating life from inanimate matter in the lab is an example of intelligent design (in this case the designer being the scientists). Historically, many proponents of Judaism and Christianity have proposed intermediates in their interpretation of creation (e.g. angels). A major point of just about all flavors of Jewish and Christian theology is that God prefers to use human (and sometimes angelic) agents as opposed to directly working miracles.
For that matter, we *already* design machines exhibiting many of the characteristics of "simple" life (robots), and the primary missing factor (self replication) is doable in principle - but just isn't cost effective for current applications and technology. (The ability to extract needed atoms from a wide variety of food sources via nano-tech is crucial to making self-replication cheap. Mining, smelting, growing crystals, etc are not easy to do on a small scale.) When we observe life springing from inanimate matter *without* the help of scientists (other than a reasonable interpretation of ancient conditions - or even modern conditions on some lifeless planet), then we will have observed abiogenesis.
The biggest problem with USB is filesystem compatibility if using with multiple OSes. With all linux machines, I don't bother with a MBR or parititions, and just put a volume group on the drive. But Windows doesn't understand that without special drivers (which do exist, BTW). Conversely, putting NTFS on the drive is problematic for Linux (it being an undocumented format). FAT32 is compatible, but way too inefficient for such a large drive.
My goal for backup is a encrypted distributed filesystem, where you share your 500GB USB drives with random people over the internet in a tit for tat arrangement. Pieces of data are encrypted and distributed over many locations, such that it can be recovered and decrypted when only XX% of the pieces are obtainable (where XX is configurable). Nobody can read other peoples pieces stored on their drive, and people can join/drop out at any time without jeopardizing anyones backup (unless a large percentage drops out at once).
Services like Amazon S3 are ok for putting encrypted tarballs on their server (including differential tars).
It wouldn't be charging for payload, but for QoS. The difference is that the customer gets to pick which packets are high/low priority via router settings (a consumer router would have simple options like "optimize voip") rather than the ISP deciding for you based on back room deals. The rich are always going to be able to afford bigger bandwidth (and bigger house and bigger car ...). QoS should be a tool to efficiently manage what you can afford.
Net neutrality will not succeed, it cannot.
Not if ISPs charge extra for packets with low latency or high bandwidth bits set. Sure, you can set the high bandwidth bit for your torrents, but then you gotta pay for it. Much cheaper for most people to leave torrents on the "as available" plan and use QoS for their Voip and the occasional huge file they gotta have now.
The software is pymilter.
It is the library, in particular event notification, that "caused" the problem. The language was fine. The same thing happens in Java when beginning programmers forget to unregister objects from event sources (or anything else requiring registration). I use the term "data cancer" to distinguish this type of bug from actual "memory leaks", which are in fact impossible with a working GC.
We knew a teacher who was accused of something (which was completely false AFAIK), and someone slashed her tires.