Correction: at least two possibilities. There's no evidence yet ruling out more, so it's not neccesarily true that ID must be true if evolution is false and vice versa.
As far as meriting study on scientific grounds, whether ID merits it is completely independent of whether evolution has been disproven or not. Even if there's no evidence to disprove evolution, ID would still merit study on scientific grounds if it met the requirements of a scientific theory. The problem is that, by definition, ID doesn't meet those requirements. The fundamental requirement is that it must make predictions about the world that can be checked against observed evidence and determined to be correct or not correct. The problem with ID is that any observation can be handwaved with "Well, the Designer must have wanted it that way.". Given any set of observations, none of the explications of ID provide any useful way of making a testable prediction about things we haven't observed.
Evolution, OTOH, does provide ways of making predictions. For example, one of the direct implications of evolution is that if a normally contra-survival genetic trait happens to provide an advantage in one particular set of conditions then it should be rare everywhere except where those conditions prevail, where it'll be common. When we look at the real world we see examples like sickle-cell anemia, a genetically-linked trait that causes significant health risks but also provides high resistance to malaria. Sickle-cell anemia is a relatively rare condition in most populations in the world, but is very common in populations that originated in the tropics where malaria's common. Intelligent Design provides no way of predicting this, evolution does.
To make the situation worse, all the explications of ID I've heard also rest on fallacies. For example, one of the more common ones is that we don't have any good explanation for how life originated in the first place, that it's just too improbable that inorganic compounds miraculously combined in exactly the right way to form long-chain organic molecules like DNA. Problem is, we do have a good explanation for that. Further, when we replicate the conditions early in Earth's history when life was originating, we see exactly that "miraculous" combination occur in the laboratory. When you take ID and replace all those fallacious claims with what's been actually observed, ID's left hanging in mid-air with nothing left beneath it. Much, in fact, like the theories that said that heavier objects fall faster than light objects after observations in vacuum columns showed all objects falling at the same speed regardless of weight.
Reality check: the amount of spam my local filters are catching hasn't changed significantly over a year ago. I still get about the same (largish) number per day, and it's about the same percentage of my total mail count. That's only local, though. At my ISP, the spam filters are blocking at least 50% more than they were a year ago. I'm sorry, FTC, but if a 50% annual increase is your idea of "working", you really need to go read this book by Merriam-Webster...
The simple fact is, we don't know for sure that one creature became another.
This would be just plain wrong. We have seen creatures in the laboratory change into different species. It's easiest to see with microbes and certain insects because their lifespans are short enough we can feasibly observe hundreds or thousands of generations, but there's no fundamental reason anyone can explicate why the same thing shouldn't happen to larger creatures. So, do I assume that what I see happening has happened before, or do I assume that for some reason the rules changed somewhere in the past to go from prohibiting what I see now to allowing it? Simplicity says assume that the rules didn't suddenly and drastically change unless and until someone presents some sort of evidence to suggest such a change (surely something drastic enough to alter the way DNA itself works should leave some signs somewhere).
But that's the very definition of a scientific theory: that you can test and disprove it. As you noted, Intelligent Design falls into the category of things that can't be disproven. It's thus not a scientific theory and as the judge noted doesn't belong in a science class. This is the problem the ID people have: if it were treated like a scientific theory, it'd take about 15 minutes to utterly demolish it beyond resurrecting.
Okay, now tell me how blood clotting works and doesn't turn ALL my blood into one big clot.
Well, first you have to look at the mechanism for blood clotting. It doesn't need a "turn off" mechanism, it's default state is off. It has a mechanism for turning clotting on, which requires both a) exposure to oxygen and b) physical breakage of platelets. Get away from the immediate area of a cut and you wind up not having at least one of those, so clotting can't start on it's own and the clotting around the cut can't progress further. It's kind of like a car with the spark plugs removed: you don't need something to stop it, you just need the absence of something that makes it go.
As for how it evolved, probably from a mechanism to seal cell membranes when they were punctured. We know there's variation, we can see it today in people whose blood doesn't clot well. Doubtless there were variations the other way where clotting happened far more readily, but those branches probably dead-ended long ago as creatures with a tendency to have their blood solidify when they were injured wouldn't live long enough to propagate that variation.
There's a fairly simple way to avoid these attacks: never ever trust any link in any e-mail, period. If you think the e-mail is legitimate, ignore the links in it and use your own bookmarks to go to the relevant site and check your account or similar page there. If it really is legitimate, there'll be a way to find the information without depending on the e-mail links. It's not completely fool-proof, but for a phisher to fool you when you do this they'd have to vandalize the legitimate web-site to include their links on it's actual pages. That's harder than just faking an e-mail.
Why should I have to tell anyone this? It's received wisdom that if you receive a phone call from someone claiming to be your bank and asking to verify things like your PIN you should hang up, look up the bank's phone number in the phone book, call them yourself and ask Customer Service about the situation. First rule: never trust the identity of the other end unless you called them. Why should e-mail be any different?
I agree, typing posture seems to be one of the important things. I was always taught: sit straight, elbows directly under shoulders, forearms just below horizontal so your wrists are just lower than your elbows, the back of your hand should be level with the top of your forearm so your wrist isn't at an angle to your forearm, fingers should dangle down to touch the keys. If we let our wrists fall towards the desk, the typing teacher would give a smack and "Wrists up!".
Modern office desks have you sitting too low for correct position as described above. Combine with wrist-rests that're only useful if your wrists have collapsed down to the desk and you have a recipe for problems. Bring back the typist's desk.:)
There's one major problem with your scenario. It's actually fairly obvious: when you go looking through the e-mail, the only stuff identifiable as coming from an Abdullah won't have anything to do with the anthrax. Do you think the real Abdullah will be stupid enough to use an e-mail address clearly matching his name? No, his e-mail will come from something like hot18yo84172@hotmail.com or somesuch, and it'll be buried in the mountain of sex-spam e-mails your target receives and discards every day just like the rest of us. Now, if you have Abdullah and want to find out who he's been talking to, then this kind of retention might be useful. Unfortunately it's also unneccesary, since if you've already got enough to nail Abdullah you've got enough to go into court, get a warrant and tap his computer directly without having to mess with his ISP.
There's also the side-effects that've already been noted. While the retention may not be useful for tracking terrorists (it's purported justification), it'll be very useful to people whose investigations have nothing to do with terrorism and who've been unable to get anything like this on their own merits. That makes me thing the whole thing's an end-run, and "terrorism" is just an excuse.
3. Check your state laws. Several states have laws on the books that say that, for returns of defective products, the vendor cannot charge a restocking fee even on return for refund. Some may require the vendor to cover return shipping as well. Returns because of mere dissatisfaction with the item (that is, the item isn't defective, merely not what you wanted) aren't covered by this. Find out what the law is, then pay with a major credit card so you can do a charge-back if neccesary.
4. Again, check the law. Federal mail-order laws should apply here, and on those the price quoted at check-out is the final price. If the vendor wants to increase the price, he's got to contact you and get your approval first and must cancel the order if you don't pay the increase. If he's already charged you, he has to refund 100% of the charge, no deductions. Using a major credit card is again recommended, so you can do a charge-back in the event the merchant balks.
In short, know what the law requires of both you and the vendor and what parts of the law trump or are trumped by the vendor's claims, and use a form of payment where you can if neccesary charge the vendor back and make him fight your credit-card company to get the money instead of you fighting him to get a refund.
Of course, the above's no substitute for checking out the vendor before-hand. Look at reviews skeptically, and remember that if it looks too good a deal to be true it likely is.
By that logic, your name isn't property therefore you shouldn't have any problem with someone using it to order all sorts of illegal and incriminating things. Next!
You're making a bunch of assumptions here. Please check their validity before speaking. When an e-mail server at an IP address that reverse-resolves to a hostname in a given domain, and which is in the MX list for that domain, is sending out spam purporting in it's headers to come from that domain, the problem most certainly is originating from the domain and the domain owner needs to take responsibility. And no, you obviously can't trust a host to claim it's correct hostname, but then I don't trust the host to claim it's correct hostname and there's plenty of tools to find out both the IP address block and the DNS domain a given host is in.
That's one of the reasons I spoke of domains, not IP blocks. The IP block is typically owned by the company providing network connectivity, the domain is owned by the person actually responsible for the machine, and I need to contact the person responsible for the machine (although if that fails I'll take the connectivity provider and if my complaint causes the machine owner's entire business to be cut off from the 'net then so be it).
I've seen far too many problems where contact had to be made immediately, a delay of several days for postal mail simply wasn't acceptable. If you've got a domain and service provided through it, you need to have someone who can be contacted immediately. If you don't, you shouldn't be connecting machines where they can have contact with the world at large.
Only one problem: your property affects other people's property. If a machine (or a user) in your domain misbehaves, other people suffer the effects. They need some way of contacting the person responsible for the domain to report problems and get them dealt with, and it needs to not depend on proper operation of the Internet and/or your domain.
You have a right to do whatever you want with your property. However, by your own logic everybody else has that same right, and one of the things we've decided to do with our property is not let yours connect to ours unless you cough up certain information.
Google to European publishers: "OK, if you guys don't want your content indexed, we won't index it. And we'll remove it from our database while we're at it."
European publishers, a month later: "Why have visits to our sites dropped by 80% since Google stopped stealing uor content?!"
What gets me about this is that it's not new. Telephone scams of a similar nature have been around forever. And the defense is the same for both: never trust the other party if you didn't originate the call. Whether I'm getting an e-mail from PayPal about my account being locked or a phone call from American Express about potentially fraudulent activity on my card, my first reaction is to simply ignore everything the caller/sender tells me. I go to my own bookmarks and get to my account on the respective web sites from my own links, or I call the customer-service phone numbers I've already got for them. If the problem's real, there'll be a notice when I log in to my account or the customer service people will know about it. My URLs and phone numbers can't be fudged by the phishers, so I can be sure I got to the right site or company. This is simple, basic and easy. If people can't apply this simple rule, I have to ask "What's wrong with this picture?".
The city can use wi-fi access for their own purposes, eg. law-enforcement and public-safety datalinks. It's not as if wide areas of wi-fi coverage are useless to the government.
The public itself is looking for Internet access. It's not as if there's not a public demand for the service.
Companies like BellSouth are not providing the service. In most of the areas where public wi-fi's being considered or actually deployed, the telcos that oppose it have also steadfastly declined to provide Internet service themselves because it's not profitable for them to do so.
To me #3 is the clincher. Saying the government shouldn't compete with private business is one thing, but when said private business won't provide a service what justification is there for preventing the government from stepping in given both public demand and government usefulness?
Problem is, on most of the automated systems they depend on calendars being published. If I just decline, I get a call from the marketing guy wondering why I'm declining his high-priority meeting when there's nothing conflicting in my calendar. Since he's got visible evidence my time's free, I've a harder time getting him to give up. When my calendar's not visible, it's a lot easier to re-route him to my manager (who gets paid to sort out things like this and decide what should take priority). Yes, public visibility of calendars and automated entry of items don't neccesarily go hand-in-hand, but every corporate calendaring system I've seen treats them as if they did.
True, but I also think of the havoc that can be wreaked on that sort of system when
People have items on their schedule that, for policy reasons, can't be put on the publicly-visible calendar. This can come about, for example, when you've got a meeting scheduled for something that doesn't officially exist yet and that management has decreed isn't going to officially exist until management's ready to inform other departments of it. This can happen because technical management doesn't want Marketing screwing up a technical initiative before it's really stabilized, or simply because the project's going to obsolete one group or another and upper management wants to control when and how this is made public.
The priorities of your immediate management differ from the priorities of the people scheduling meetings. I don't know how often I had to manually override the priority on a meeting request from suits when, regardless of whether they considered it critical priority or not, my direct manager did not want me wasting time on something I wasn't involved in when I needed to be attending a meeting with a sane priority on a project that was ticking down to release.
Manual systems, or private calendaring with control in my hands, allows easy solutions to both those. Automated systems more often get in the way since the correct outcome tends to be the opposite of what the system's rules say the outcome should be.
E-mail. You e-mail me a note suggesting a meeting at a certain time, I look at my schedule and what the meeting's about and either shoot you an acceptance and mark it on my calendar, suggest a different time or just send back a regrets-decline note. The advantage is that this works no matter which e-mail clients and calendar system each of us is using, and works when I have priorities and things on my schedule that you aren't supposed to be aware of.
To be honest, I think that the software's the wrong thing to be looking at. Simply require an audit trail that's independent of the machine count and will let you verify whether the machine tallies are correct without having to assume any part of the machine side is accurate, eg. a paper ballot printed, inspected by the voter and deposited in a ballot box handled seperately from the machine's memory packs. If you can verify the machine results without having to depend on the machines at any point, then you can always catch any attempt to fudge the machine's numbers and you don't have to trust the machine at all to get good results. Then mandate random comparisons of a sample of the machine results with the audit trail, with any significant discrepancy triggering an automatic across-the-board audit.
Something that never fails is good, but better still is to assume it will fail despite all your efforts and design it so any failure's recoverable.
Evidence says otherwise. Take, for example, the latest "extremely critical" vulnerability in IE. It's a simple failure-to-initialize bug. With what's known about it, isolating the code problem should take one developer no more than a day or two. Fixing it shouldn't take more than another day of developer time, and the risk to other code should be minimal. It's the kind of bugfix that, where I work, we routinely fix as found and roll into the next scheduled release. Microsoft was told about the problem nearly 6 months ago, and they've done several major patch releases since then.
The bug is, as of today, still unfixed. And it turns out it allows a remote attack to execute arbitrary code as the logged-in user. And Microsoft is only now scrambling to fix the problem. Oops.
My first question to him would be: how do you figure ETs far enough away to need to infect us via SETI signals would figure out what they needed to put in to actually infect our computers? They'd need to know what software we were using, what hardware we were using, what instruction set to use in their code. How are they going to figure all that out from dozens or hundreds of light-years away? And if they're actually here, close enough to monitor our networks and figure out all they'd need to have figured out, why not bypass SETI entirely and just attack directly through our networks themselves? And finally, how are they supposed to have kept up with lag? The nearest star they could be attacking from is 4.3 light-years away. Assuming they can analyze signals and determine attack vectors instantly, that means any attack that gets back to us is going to be targeted at what we were using 8 and a half years ago. How likely is that to work against more modern systems?
I think the guy may have a Ph.D., but he's not thinking at all about some of the practical aspects.
The compiler can't do a lot of the optimization this kind of processor needs. It's not a matter of instruction scheduling or selection, it's a matter of algorithm selection. To make use of the CBE you need to design your program to combine massive amounts of threading with massively parallel data processing within those threads. You're trying to get it so there's enough happening simultaneously to keep all those cores fed with instruction streams. The compiler can't rewrite your program to do that, the programmer has to handle that part of it.
That's also why the CBE and in fact any supercomputer's a bitch to program for for most programmers: they aren't taught to think in terms of dozens of parallel sequences at once. It's a whole different and very alien worldview, and if you don't have it you can't make the hardware perform very well. OTOH, if you do grok how the hardware wants to do things and can start thinking in those terms, you can make the hardware do things most people won't believe even after they see them (see what Gran Turismo 4 did with the PlayStation 2 hardware for an example).
I wonder what would happen if someone created an original song that legitimately could match something in the filter list without being infringing, distributed it on Kazaa legally (since they're the copyright owner), and then sued the record labels for illegal interference with a copyright owner's rights? After all the only things the labels can legitimately ask to be filtered are things they own copyright on, and by filtering anything matching that list they're claiming to own anything which matches the list. What's the penalties for falsely claiming ownership of someone else's copyrighted work again?
Ideal? A setup where nobody attached to it can even search via Google is ideal? I don't see it as such. This isn't about the WLAN portion, no matter how big, the monthly costs you're saying you can avoid are for connecting that WLAN to the rest of the world.
Unfortunately you mistake what you're paying for. That $45/month isn't for the local link, it's mostly for the transport between your ISP and the rest of the Internet. Your scenario B has you paying $45/year for access to... well, nothing really. Only those computers associated with the same wireless access point as you. Basically exactly what you'd have if you set up an 802.11b access point of your own that wasn't plugged into a LAN. Now you have to go about getting your access points connected up to the rest of the Internet, and that costs money. Just like your home AP requires a router and connection to your ISP to let you talk to the Internet, a wireless service provider's access points require peering and transit service from a backbone provider or peering point to let the people attached to them talk to the Internet or they're just isolated LANs.
I don't think he's opening himself. As I understand it from the article, he's not going looking for such material. What he's doing is conforming to the DMCA takedown procedure. If someone notifies him of infringing material, he'll remove it pending an objection by the person who posted the material. The DMCA provides explicit protection for someone who follows that procedure which he can then use as a shield. And meanwhile he can use the fact that he entered into this agreement to remove demonstrated infringing content as a counter to anyone (like the RIAA) trying to argue that he's encouraging infringing uses.
Slashdot Mirror
Correction: at least two possibilities. There's no evidence yet ruling out more, so it's not neccesarily true that ID must be true if evolution is false and vice versa.
As far as meriting study on scientific grounds, whether ID merits it is completely independent of whether evolution has been disproven or not. Even if there's no evidence to disprove evolution, ID would still merit study on scientific grounds if it met the requirements of a scientific theory. The problem is that, by definition, ID doesn't meet those requirements. The fundamental requirement is that it must make predictions about the world that can be checked against observed evidence and determined to be correct or not correct. The problem with ID is that any observation can be handwaved with "Well, the Designer must have wanted it that way.". Given any set of observations, none of the explications of ID provide any useful way of making a testable prediction about things we haven't observed.
Evolution, OTOH, does provide ways of making predictions. For example, one of the direct implications of evolution is that if a normally contra-survival genetic trait happens to provide an advantage in one particular set of conditions then it should be rare everywhere except where those conditions prevail, where it'll be common. When we look at the real world we see examples like sickle-cell anemia, a genetically-linked trait that causes significant health risks but also provides high resistance to malaria. Sickle-cell anemia is a relatively rare condition in most populations in the world, but is very common in populations that originated in the tropics where malaria's common. Intelligent Design provides no way of predicting this, evolution does.
To make the situation worse, all the explications of ID I've heard also rest on fallacies. For example, one of the more common ones is that we don't have any good explanation for how life originated in the first place, that it's just too improbable that inorganic compounds miraculously combined in exactly the right way to form long-chain organic molecules like DNA. Problem is, we do have a good explanation for that. Further, when we replicate the conditions early in Earth's history when life was originating, we see exactly that "miraculous" combination occur in the laboratory. When you take ID and replace all those fallacious claims with what's been actually observed, ID's left hanging in mid-air with nothing left beneath it. Much, in fact, like the theories that said that heavier objects fall faster than light objects after observations in vacuum columns showed all objects falling at the same speed regardless of weight.
Reality check: the amount of spam my local filters are catching hasn't changed significantly over a year ago. I still get about the same (largish) number per day, and it's about the same percentage of my total mail count. That's only local, though. At my ISP, the spam filters are blocking at least 50% more than they were a year ago. I'm sorry, FTC, but if a 50% annual increase is your idea of "working", you really need to go read this book by Merriam-Webster...
The simple fact is, we don't know for sure that one creature became another.
This would be just plain wrong. We have seen creatures in the laboratory change into different species. It's easiest to see with microbes and certain insects because their lifespans are short enough we can feasibly observe hundreds or thousands of generations, but there's no fundamental reason anyone can explicate why the same thing shouldn't happen to larger creatures. So, do I assume that what I see happening has happened before, or do I assume that for some reason the rules changed somewhere in the past to go from prohibiting what I see now to allowing it? Simplicity says assume that the rules didn't suddenly and drastically change unless and until someone presents some sort of evidence to suggest such a change (surely something drastic enough to alter the way DNA itself works should leave some signs somewhere).
But that's the very definition of a scientific theory: that you can test and disprove it. As you noted, Intelligent Design falls into the category of things that can't be disproven. It's thus not a scientific theory and as the judge noted doesn't belong in a science class. This is the problem the ID people have: if it were treated like a scientific theory, it'd take about 15 minutes to utterly demolish it beyond resurrecting.
Okay, now tell me how blood clotting works and doesn't turn ALL my blood into one big clot.
Well, first you have to look at the mechanism for blood clotting. It doesn't need a "turn off" mechanism, it's default state is off. It has a mechanism for turning clotting on, which requires both a) exposure to oxygen and b) physical breakage of platelets. Get away from the immediate area of a cut and you wind up not having at least one of those, so clotting can't start on it's own and the clotting around the cut can't progress further. It's kind of like a car with the spark plugs removed: you don't need something to stop it, you just need the absence of something that makes it go.
As for how it evolved, probably from a mechanism to seal cell membranes when they were punctured. We know there's variation, we can see it today in people whose blood doesn't clot well. Doubtless there were variations the other way where clotting happened far more readily, but those branches probably dead-ended long ago as creatures with a tendency to have their blood solidify when they were injured wouldn't live long enough to propagate that variation.
There's a fairly simple way to avoid these attacks: never ever trust any link in any e-mail, period. If you think the e-mail is legitimate, ignore the links in it and use your own bookmarks to go to the relevant site and check your account or similar page there. If it really is legitimate, there'll be a way to find the information without depending on the e-mail links. It's not completely fool-proof, but for a phisher to fool you when you do this they'd have to vandalize the legitimate web-site to include their links on it's actual pages. That's harder than just faking an e-mail.
Why should I have to tell anyone this? It's received wisdom that if you receive a phone call from someone claiming to be your bank and asking to verify things like your PIN you should hang up, look up the bank's phone number in the phone book, call them yourself and ask Customer Service about the situation. First rule: never trust the identity of the other end unless you called them. Why should e-mail be any different?
I agree, typing posture seems to be one of the important things. I was always taught: sit straight, elbows directly under shoulders, forearms just below horizontal so your wrists are just lower than your elbows, the back of your hand should be level with the top of your forearm so your wrist isn't at an angle to your forearm, fingers should dangle down to touch the keys. If we let our wrists fall towards the desk, the typing teacher would give a smack and "Wrists up!".
Modern office desks have you sitting too low for correct position as described above. Combine with wrist-rests that're only useful if your wrists have collapsed down to the desk and you have a recipe for problems. Bring back the typist's desk. :)
There's one major problem with your scenario. It's actually fairly obvious: when you go looking through the e-mail, the only stuff identifiable as coming from an Abdullah won't have anything to do with the anthrax. Do you think the real Abdullah will be stupid enough to use an e-mail address clearly matching his name? No, his e-mail will come from something like hot18yo84172@hotmail.com or somesuch, and it'll be buried in the mountain of sex-spam e-mails your target receives and discards every day just like the rest of us. Now, if you have Abdullah and want to find out who he's been talking to, then this kind of retention might be useful. Unfortunately it's also unneccesary, since if you've already got enough to nail Abdullah you've got enough to go into court, get a warrant and tap his computer directly without having to mess with his ISP.
There's also the side-effects that've already been noted. While the retention may not be useful for tracking terrorists (it's purported justification), it'll be very useful to people whose investigations have nothing to do with terrorism and who've been unable to get anything like this on their own merits. That makes me thing the whole thing's an end-run, and "terrorism" is just an excuse.
A couple of points on numbers 3 and 4:
3. Check your state laws. Several states have laws on the books that say that, for returns of defective products, the vendor cannot charge a restocking fee even on return for refund. Some may require the vendor to cover return shipping as well. Returns because of mere dissatisfaction with the item (that is, the item isn't defective, merely not what you wanted) aren't covered by this. Find out what the law is, then pay with a major credit card so you can do a charge-back if neccesary.
4. Again, check the law. Federal mail-order laws should apply here, and on those the price quoted at check-out is the final price. If the vendor wants to increase the price, he's got to contact you and get your approval first and must cancel the order if you don't pay the increase. If he's already charged you, he has to refund 100% of the charge, no deductions. Using a major credit card is again recommended, so you can do a charge-back in the event the merchant balks.
In short, know what the law requires of both you and the vendor and what parts of the law trump or are trumped by the vendor's claims, and use a form of payment where you can if neccesary charge the vendor back and make him fight your credit-card company to get the money instead of you fighting him to get a refund.
Of course, the above's no substitute for checking out the vendor before-hand. Look at reviews skeptically, and remember that if it looks too good a deal to be true it likely is.
Only one problem: your property affects other people's property. If a machine (or a user) in your domain misbehaves, other people suffer the effects. They need some way of contacting the person responsible for the domain to report problems and get them dealt with, and it needs to not depend on proper operation of the Internet and/or your domain.
You have a right to do whatever you want with your property. However, by your own logic everybody else has that same right, and one of the things we've decided to do with our property is not let yours connect to ours unless you cough up certain information.
Google to European publishers: "OK, if you guys don't want your content indexed, we won't index it. And we'll remove it from our database while we're at it."
European publishers, a month later: "Why have visits to our sites dropped by 80% since Google stopped stealing uor content?!"
What gets me about this is that it's not new. Telephone scams of a similar nature have been around forever. And the defense is the same for both: never trust the other party if you didn't originate the call. Whether I'm getting an e-mail from PayPal about my account being locked or a phone call from American Express about potentially fraudulent activity on my card, my first reaction is to simply ignore everything the caller/sender tells me. I go to my own bookmarks and get to my account on the respective web sites from my own links, or I call the customer-service phone numbers I've already got for them. If the problem's real, there'll be a notice when I log in to my account or the customer service people will know about it. My URLs and phone numbers can't be fudged by the phishers, so I can be sure I got to the right site or company. This is simple, basic and easy. If people can't apply this simple rule, I have to ask "What's wrong with this picture?".
Well, three points:
- The city can use wi-fi access for their own purposes, eg. law-enforcement and public-safety datalinks. It's not as if wide areas of wi-fi coverage are useless to the government.
- The public itself is looking for Internet access. It's not as if there's not a public demand for the service.
- Companies like BellSouth are not providing the service. In most of the areas where public wi-fi's being considered or actually deployed, the telcos that oppose it have also steadfastly declined to provide Internet service themselves because it's not profitable for them to do so.
To me #3 is the clincher. Saying the government shouldn't compete with private business is one thing, but when said private business won't provide a service what justification is there for preventing the government from stepping in given both public demand and government usefulness?Problem is, on most of the automated systems they depend on calendars being published. If I just decline, I get a call from the marketing guy wondering why I'm declining his high-priority meeting when there's nothing conflicting in my calendar. Since he's got visible evidence my time's free, I've a harder time getting him to give up. When my calendar's not visible, it's a lot easier to re-route him to my manager (who gets paid to sort out things like this and decide what should take priority). Yes, public visibility of calendars and automated entry of items don't neccesarily go hand-in-hand, but every corporate calendaring system I've seen treats them as if they did.
True, but I also think of the havoc that can be wreaked on that sort of system when
- People have items on their schedule that, for policy reasons, can't be put on the publicly-visible calendar. This can come about, for example, when you've got a meeting scheduled for something that doesn't officially exist yet and that management has decreed isn't going to officially exist until management's ready to inform other departments of it. This can happen because technical management doesn't want Marketing screwing up a technical initiative before it's really stabilized, or simply because the project's going to obsolete one group or another and upper management wants to control when and how this is made public.
- The priorities of your immediate management differ from the priorities of the people scheduling meetings. I don't know how often I had to manually override the priority on a meeting request from suits when, regardless of whether they considered it critical priority or not, my direct manager did not want me wasting time on something I wasn't involved in when I needed to be attending a meeting with a sane priority on a project that was ticking down to release.
Manual systems, or private calendaring with control in my hands, allows easy solutions to both those. Automated systems more often get in the way since the correct outcome tends to be the opposite of what the system's rules say the outcome should be.E-mail. You e-mail me a note suggesting a meeting at a certain time, I look at my schedule and what the meeting's about and either shoot you an acceptance and mark it on my calendar, suggest a different time or just send back a regrets-decline note. The advantage is that this works no matter which e-mail clients and calendar system each of us is using, and works when I have priorities and things on my schedule that you aren't supposed to be aware of.
To be honest, I think that the software's the wrong thing to be looking at. Simply require an audit trail that's independent of the machine count and will let you verify whether the machine tallies are correct without having to assume any part of the machine side is accurate, eg. a paper ballot printed, inspected by the voter and deposited in a ballot box handled seperately from the machine's memory packs. If you can verify the machine results without having to depend on the machines at any point, then you can always catch any attempt to fudge the machine's numbers and you don't have to trust the machine at all to get good results. Then mandate random comparisons of a sample of the machine results with the audit trail, with any significant discrepancy triggering an automatic across-the-board audit.
Something that never fails is good, but better still is to assume it will fail despite all your efforts and design it so any failure's recoverable.
Evidence says otherwise. Take, for example, the latest "extremely critical" vulnerability in IE. It's a simple failure-to-initialize bug. With what's known about it, isolating the code problem should take one developer no more than a day or two. Fixing it shouldn't take more than another day of developer time, and the risk to other code should be minimal. It's the kind of bugfix that, where I work, we routinely fix as found and roll into the next scheduled release. Microsoft was told about the problem nearly 6 months ago, and they've done several major patch releases since then.
The bug is, as of today, still unfixed. And it turns out it allows a remote attack to execute arbitrary code as the logged-in user. And Microsoft is only now scrambling to fix the problem. Oops.
My first question to him would be: how do you figure ETs far enough away to need to infect us via SETI signals would figure out what they needed to put in to actually infect our computers? They'd need to know what software we were using, what hardware we were using, what instruction set to use in their code. How are they going to figure all that out from dozens or hundreds of light-years away? And if they're actually here, close enough to monitor our networks and figure out all they'd need to have figured out, why not bypass SETI entirely and just attack directly through our networks themselves? And finally, how are they supposed to have kept up with lag? The nearest star they could be attacking from is 4.3 light-years away. Assuming they can analyze signals and determine attack vectors instantly, that means any attack that gets back to us is going to be targeted at what we were using 8 and a half years ago. How likely is that to work against more modern systems?
I think the guy may have a Ph.D., but he's not thinking at all about some of the practical aspects.
The compiler can't do a lot of the optimization this kind of processor needs. It's not a matter of instruction scheduling or selection, it's a matter of algorithm selection. To make use of the CBE you need to design your program to combine massive amounts of threading with massively parallel data processing within those threads. You're trying to get it so there's enough happening simultaneously to keep all those cores fed with instruction streams. The compiler can't rewrite your program to do that, the programmer has to handle that part of it.
That's also why the CBE and in fact any supercomputer's a bitch to program for for most programmers: they aren't taught to think in terms of dozens of parallel sequences at once. It's a whole different and very alien worldview, and if you don't have it you can't make the hardware perform very well. OTOH, if you do grok how the hardware wants to do things and can start thinking in those terms, you can make the hardware do things most people won't believe even after they see them (see what Gran Turismo 4 did with the PlayStation 2 hardware for an example).
I wonder what would happen if someone created an original song that legitimately could match something in the filter list without being infringing, distributed it on Kazaa legally (since they're the copyright owner), and then sued the record labels for illegal interference with a copyright owner's rights? After all the only things the labels can legitimately ask to be filtered are things they own copyright on, and by filtering anything matching that list they're claiming to own anything which matches the list. What's the penalties for falsely claiming ownership of someone else's copyrighted work again?
Ideal? A setup where nobody attached to it can even search via Google is ideal? I don't see it as such. This isn't about the WLAN portion, no matter how big, the monthly costs you're saying you can avoid are for connecting that WLAN to the rest of the world.
Unfortunately you mistake what you're paying for. That $45/month isn't for the local link, it's mostly for the transport between your ISP and the rest of the Internet. Your scenario B has you paying $45/year for access to... well, nothing really. Only those computers associated with the same wireless access point as you. Basically exactly what you'd have if you set up an 802.11b access point of your own that wasn't plugged into a LAN. Now you have to go about getting your access points connected up to the rest of the Internet, and that costs money. Just like your home AP requires a router and connection to your ISP to let you talk to the Internet, a wireless service provider's access points require peering and transit service from a backbone provider or peering point to let the people attached to them talk to the Internet or they're just isolated LANs.
I don't think he's opening himself. As I understand it from the article, he's not going looking for such material. What he's doing is conforming to the DMCA takedown procedure. If someone notifies him of infringing material, he'll remove it pending an objection by the person who posted the material. The DMCA provides explicit protection for someone who follows that procedure which he can then use as a shield. And meanwhile he can use the fact that he entered into this agreement to remove demonstrated infringing content as a counter to anyone (like the RIAA) trying to argue that he's encouraging infringing uses.