No, a Microsoft solution is perfectly acceptable. That's not the problem. You can have central password administration in Linux, too. The problem is having that central administration under the control of a single individual.
You're simply stating that you have the same problem that the OP is trying to solve, but on a different platform. You have a single person holding the keys to the kingdom, who doles out access as he/she sees fit. If that sysadmin were to get honked off at you, he could cripple you and you'd be facing significant pain to recover from it. If he wanted to get access to secrets, he could easily do so. He might lose his job or spend quality time with Bubba in prison as a result of either action later, but the damage would be done. Your company would have lost access to your critical systems and/or be facing reporters asking how your customer database was sold to the highest bidder in East Bumfukistan.
The question is about spreading around the rights and responsibilities so your entire infrastructure is not under the absolute control of a single (hopefully trustworthy) sysadmin, just in case that person turns out not to be so trustworthy. You should pay attention to this topic, because the system you've described has the same vulnerabilities.
Nah. Because then cell companies would have to compete based on features rather than supposed technical advantages of their underlying technology. That's CRAZY talk. Next thing you know, people can use (gasp) UNLOCKED cell phones on any carrier they choose, then it's utter chaos when customers aren't locked into their comfortable multi-year agreements with multi-hundred-dollar early termination fees! Why, that'd be unAmerican!
Here in 'Merika, we believe in this thing called the "free market" which means companies are "free" to screw us over.
Not that I'm a big iPhone fan, but you did hear that the iPhone will be on Verizon, supposedly this month, right? So if you're happy with your appley goodness, you can still get it in a Verizon crust, almost certainly by August of this year. Yet to be seen is if Verizon locks the GPS or pulls other stunts like they did with the Blackberry...
At the moment, even if every car manufacturer made all of their cars with such a system and every consumer were forced to buy it on any new car, it would be over half a decade before there was a 50/50 chance the car who just tried to pull out in front of you had a computer yours could honk at.
The best systems are independently defensive and merely informative to those around you (they don't depend on the people around you to act).
Of course, if you want to really screw something like this up, an RF jammer would do the job quite nicely and with a lot less fuss about encryption and all that nonsense. You don't need to commandeer it, just make it fail.
Not making any suggestions, because of course that would be illegal, but such a thing would be relatively trivial to build. It doesn't need to broadcast anything specific, just a crapload of static at a shifting frequency until the altitude of the drone is observed to be decreasing at a rate approximating terminal velocity.
The sort of person who finds themselves under sudden drone scrutiny is statistically less likely to be concerned about the legality of their methods to elude capture or avoid observation, or who the drone might hurt in the resulting impact.
In fact, the resulting impact might slow the pursuers down, what with the possible road damage and possible injured civilians to care for.
10,500 is just the maximum operating altitude. Given its 40-minute operating limit, I expect they'll hover it up to about 500-1000 feet and use it there, probably launched from a vehicle because 40 minutes isn't a whole lot of range. 500-1000 is below minimums for flying over most populated areas anyway, so there won't be a lot of fixed-wing or whirlybird traffic down there. Once they enter the upside down wedding cake, though, they'll HAVE to be talking to a controller, it HAS to have a transponder, and the controller will be slicing them a piece of wedding cake to fly in.
I'm actually more concerned with something like this in "C" and higher airspaces, where there's generally no RADAR approach and avoidance is more often than not VFR-only. Unless it has a lot of side-view cameras and several operators, something like this is going to be pretty much incapable of see-and-avoid.
I suspect something like this will be more likely to be seen outside a skyscraper window (where planes are not allowed) than over a rural house.
[Glasses_Man (Cusco? something like that) takes control of the drone and solves a crime] [Local officer comes up and starts yelling at Glasses_Man while G_M is trying to land the drone].
Cusco: "What was that, I couldn't hear you over the..."
[puts on glasses] "...droning."
A magnetic compass is only accurate, as you say, if you are flying straight and level. For navigation, it's more convenient to use a gyro (which is a very reliable piece of kit, but also has gyroscopic precession issues so you need to recalibrate it to the mag compass occasionally).
Since the gyro has to be calibrated, and the mag compass is a cheap simple and reliable backup nav unit anyway, pilots use magnetic north.
You can get as complex as you want with gizmos, but at some point in your flying career there's a chance you're going to rely solely on a mag compass for navigation. If you end up in that mode, trust me, you DO NOT want to make your calculations any more complex than you need.
Aviation calibrates everything around your simplest instruments. For directions, this is the mag compass. The GPS can calculate mag north easily, the compass cannot calculate true north on its own. It's too simple a device, which is of course what makes it so reliable.
When you're under complete instrument failure and things are going pear-shaped on you, you really don't need any more work on your plate. Get to straight and level for a second, now you have an accurate compass reading.
Until it dies, then you either use a compass and have to adjust for declination, or use a compass and use mag north.
A GPS can read in mag north easily. They're complex little gewgaws with computers. Why make a pilot keep one more bit of data in his head that could potentially be inconsistent with what he's observing and make him adjust for it when the computers can calculate mag north easily?
Maybe only a dozen or so pilots have to navigate solely off a mag compass because their glass cockpit or directional gyro went kapoof, but that's a dozen people a year who already have a lot on their plates when trying to land an aircraft after a major instrument failure. Why make that any harder than it has to be? Because we don't want to write declination adjustment software (stuff that's been in GPS receivers since the very early days of GPS) in the GPS units and fancy gewgaws?
Which is why you have a directional gyro on board. The DG, which runs off very simple gizmos and is very reliable, gives you a convenient reading of direction most of the time, the compass is there as the base frame of reference (because DGs have to be reset occasionally), and just in case the DG goes kerbonk. You may only know your direction with precision when flying straight and level if the DG goes bye-bye, but you can always level off for a few seconds and get a directional reading.
Even cockpits with all the fancy glass and such usually have a basic compass, a simple electrically-driven gyro, a vacuum-driven airspeed indicator, and a pressure-sensitive altimeter. That way, if you forgot to load software version 4.334.84.779.83.56.23885.2341a revision 12b, and the screen goes black taking all your radios and GPS and weather reporting and $18,000 hiney wipers, you know how high you are, what direction you are going, and how fast you are traveling.
We use mag north because the simplest systems can accommodate it easily. And, guess what? The most complicated systems can accommodate it easily, too. Name me one model of GPS that does not have a selection between "mag north" and "true north".
When flying, no matter where you are in the world, you use Zulu (GMT) because everyone always knows exactly what time it is. When flying, you use mag north because the simplest instruments always know where "North" is. When flying, you use English because every pilot can talk to every other pilot and every controller across the planet. When flying, you keep the simplest instruments available on board as backups to the complex stuff that breaks.
Flyin's complicated enough. Keep the basics simple.
The regulations are there for a very good reason. Every airplane needs a magnetic compass on board, because in all the years of using magnetic compasses, they aren't known for running out of batteries or needing to be rebooted at a critical moment or failing because a programmer fucked up and you just crossed time zones. Electronics can easily accommodate using mag north, magnetics cannot as easily accommodate using true north. So we use the system the simplest system can accommodate easily, and the more complex systems that are MORE than capable of adjusting for it do so.
Hobbyist planes are where professional pilots learn to fly, and most of them start out using a magnetic compass as their primary directional instrument (or at least use that as input to set a a directional gyroscope that's easier to read). That way, when the pro pilot is up there and the instrument panel suddenly goes dead in a puff of smoke, there's no reason to write off the lives of the passengers on board. The pilot knows where he's going, and this is due in part to the simple instruments that are on board, and in part to the fact that he's prepared in case this happens.
You learn to fly using shit that don't break, then you get to play with the fancy doodads later, but you never forget how to use the shit that don't break. People's lives depend on that.
I think it's rather the opposite. It's the freedom from copyright that allows some people to censor it as they like, but also allows the original, uncensored work to continue to exist and be copied by people who prefer that version.
If the copyright holder decided to censor the work, they could simply stop distribution of any previous versions and the new censored version would be the only new one available. Your only chance of getting a copy of the uncensored work would be to purchase a used copy (which would quickly become desirable and therefore expensive) or pirate it.
Switching to an equally (among certain geeks) controversial but far more recent censorship issue, when Star Wars IV was re-released with the "Greedo shot first" bit, it was before the DVD player existed. Lucas stopped distribution of the original work, which was only available on VCR tape that could wear out. Within legal copyright, there is no way to possess anything but an original VHS tape of the movie, which are a rare and vanishing breed. If copyright was anything approaching sane, that movie would have fallen out of copyright several years ago and would be freely available for distribution and copying in its original form.
If a work is still under copyright, no one else can mess with it. But if the copyright owner decides to mess with it, no one else can stop them.
Lack of copyright means everyone owns it. They can do what they want with it, copy it, distribute it, and censor their own version for re-distribution.
Copyright means only one person can do anything with it. If they decide not to censor it, then it cannot be censored. If they decide to censor it, then it must be censored.
Copyright is only beneficial in this way if everyone agrees that there should be only one version of a work, and everyone agrees with the copyright holder as to which version that is (in which case the work protects itself).
In the same way, dictatorship is the most efficient form of government if you have a truly benevolent dictator, and having an unregulated monopoly is always the cheapest and best way of making a product if you assume the monopolistic company would have a natural inclination to serve their community and employees in the most efficient and mutually beneficial manner possible.
For those keeping score at home, none of the three will ever work out that way.
I think you may have missed my point, or I made it poorly (which is likely). Let me try again.
The issue at hand is network neutrality.
A caching service like this can easily benefit both the ISP and the customer by reducing unnecessary bandwidth, as long as it isn't abused. The gist of the article is that "caching servers bad", and I disagree. I agree that caching servers can be abused, but let's not throw the baby out with the bathwater here.
So in the network neutrality discussion, do we want to say that:
1. Caching should be off-limits forever and completely (as the article implies).
2. Caching should be available absolutely for free to all content providers who ask, at no charge and regardless of how much content the provider wants cached (forcing BT to basically cache the entire Internet if they choose to cache anything at all, and some things like NetFlix and encrypted data are un-cache-able without the content provider's cooperation and assistance).
3. Caching and charging for it is allowed but ISPs must maintain sufficient bandwidth to support non-cached activities as well, with the only difference in speed being the obvious speed boost you achieve by caching something locally.
I maintain that option #3 is the most logical choice. It maintains network neutrality while giving ISPs the opportunity to manage their network traffic.
Otherwise, BT's going to have to put in a shitload of backbone upgrades in order to accommodate every one of their customers doing streaming bandwidth 24/7 from anywhere on the planet. If they have to do that, your rates are going up.
I'm not saying they should be able to hold a high-bandwidth service hostage, but they might find it in their best interests to give that service a really good deal on caching services and prevent some of those upgrade costs they'll pass on to you.
A lot of ISP's have an Akamai presence. I'm frankly surprised that BT wants to even bother "rolling their own" when there are already third parties ready to do this for them. However, intelligent caching is good for everyone, and it does require a level of content provider cooperation (so the content provider can clearly identify what is cache-able and what is dynamic).
The problem is that the backbone necessary to do that costs serious money. A CDN is a more efficient use of the existing infrastructure, and costs a whole lot less than an extra few thousand miles of fiber.
What amazes me is that BT and other ISPs don't simply offer this for free, because it benefits them as much as it does their customers. I guess any revenue stream is a good revenue stream, but if they simply cached ALL static content and worked out deals with people like NetFlix to store encrypted copies of their movies for local serving rather than streaming every copy of every movie across the Internet backbone, it would make BT seem like the fastest ISP *EVER*, save them gobs and shitloads of money on backbone bandwidth, allow them to raise monthly caps and allotted speeds to very high levels without putting their network under significantly more strain, put off expensive backbone upgrades for years, and cost them very little
I think this is basically an Akamai-like caching server. And I can see where the controversy lies.
If BT implements this and doesn't intentionally throttle other services, I don't see this as a violation of neutrality - BT is not discriminating against anyone in using the Internet pipe, they are simply maintaining a cache service for those who want to cough up a little more dough for their web sites to be stored in a local cache. BP customers can still access anything they want on the Internet at Internet speeds, but certain things run at local speed which is faster.
However, I can see how this could be easily abused, if BT started speeding up the Internet packets (*) coming from their customers who paid for the caching service and slowing down everyone else's, or started blocking or throttling sites that refused to pay for it, or lowered their overall Internet connection to the point where only cached sites were useful.
I can also see how this could be interpreted as a "two tier" system, but such systems have been in use for quite some years in the US and have been very successful here. They do make some web pages faster than others, but I haven't seen many reports of ISPs intentionally throttling their regular Internet bandwidth to punish service providers who don't pay up. I've heard of ISPs who try to force high-bandwidth content providers to subscribe, and that's wrong, but that's a matter of abusing the technology, not a problem inherent to the technology.
Frankly, I don't understand why an ISP wouldn't want to simply start caching all static content. But, unfortunately, that means that most content they really want to cache is not going to be. Streaming video from someone like NetFlix is encrypted so the movie you watch is a different set of bits from the same exact movie your neighbor watches 5 minutes later. BitTorrent is not only comprised of a great deal of illegal content that the ISPs don't want in their cache servers, it's also frequently encrypted, and the BitTorrent protocol is going to tend to prefer "local" clients so it's already optimized to save backbone usage when possible anyway.
YouTube would be brilliant for this sort of thing, and YouTube actually uses Akamai if I recall correctly.
(*) By "Internet packets", I'm referring to caching customers who might only cache static content. This is how my company uses Akamai - we give Akamai a copy of all of our static content (ie. pictures of our product), they replicate it out to all of their edge servers around the world as needed, and we simply use an Akamai URL to access the image on the web site. Akamai automatically determines the closest server to the customer and serves them up the replacement image. All encrypted and dynamic content is directly between the customer and our web servers.
I've been known to pull out a notebook and take notes if someone is on a business call. One "can you repeat that, please?" usually is enough to remind them that talking about company business in a public place where everyone around them is pretty quiet is not only rude, it's terribly inadvisable.:)
For now. Until a few kids get killed by some asshole using his phone to send a text. Then it'll be a "think of the CHILDREN" law that no cell phone may operate at more than 5MPH.
If you don't allow your users to have Admin access, the differences probably aren't that huge. There is Windows Defender and BitLocker, so it might save you a few bucks on drive encryption software and possibly antivirus, and Libraries might make their documents a little easier to find.
If you do allow Admin access, the security popups can be a significant improvement in security, as they warn you when system areas are being futzed with.
The company I work for is still on XP and quite happy with it, for now. We're in the final throes of getting rid of IE6, though.
It also surprises me sometimes that the overages cost so much more per gigabyte than the base plan does.
Verizon's cell data plan is 2GB for $20. That's a decent price, and not a completely unacceptable cap. But their overage is $20 per gigabyte. That's their one and only smartphone/featurephone plan. So if I use 3GB a month on one phone, that's the same as 4GB would cost me if I had two phones at 2GB each. If I used 4GB a month, I'm paying $60 a month, enough to have three data plans.
AT&T's DataPro 2GB plan is $25 a month, but overages are at $10/GB, which means it's a little better, but their laptop data plan overages are five and ten cents a megabyte, which works out to $50 or $100 a gigabyte. If you buy their 5GB $60 plan and go a gigabyte over, it's $110. If you buy their 200MB $35 plan and use 1GB over, it's $135! It's far cheaper to buy a basic phone and tether it if you need over 5GB a month...
Slashdot Mirror
No, a Microsoft solution is perfectly acceptable. That's not the problem. You can have central password administration in Linux, too. The problem is having that central administration under the control of a single individual.
You're simply stating that you have the same problem that the OP is trying to solve, but on a different platform. You have a single person holding the keys to the kingdom, who doles out access as he/she sees fit. If that sysadmin were to get honked off at you, he could cripple you and you'd be facing significant pain to recover from it. If he wanted to get access to secrets, he could easily do so. He might lose his job or spend quality time with Bubba in prison as a result of either action later, but the damage would be done. Your company would have lost access to your critical systems and/or be facing reporters asking how your customer database was sold to the highest bidder in East Bumfukistan.
The question is about spreading around the rights and responsibilities so your entire infrastructure is not under the absolute control of a single (hopefully trustworthy) sysadmin, just in case that person turns out not to be so trustworthy. You should pay attention to this topic, because the system you've described has the same vulnerabilities.
If, by "quasi", you are referring to "queasy", which means "ill", yes....
Nah. Because then cell companies would have to compete based on features rather than supposed technical advantages of their underlying technology. That's CRAZY talk. Next thing you know, people can use (gasp) UNLOCKED cell phones on any carrier they choose, then it's utter chaos when customers aren't locked into their comfortable multi-year agreements with multi-hundred-dollar early termination fees! Why, that'd be unAmerican!
Here in 'Merika, we believe in this thing called the "free market" which means companies are "free" to screw us over.
Not that I'm a big iPhone fan, but you did hear that the iPhone will be on Verizon, supposedly this month, right? So if you're happy with your appley goodness, you can still get it in a Verizon crust, almost certainly by August of this year. Yet to be seen is if Verizon locks the GPS or pulls other stunts like they did with the Blackberry...
you can be sure that while the Car of Tomorrow may not fly
It all depends. Do the brief airborne moments of the rollover count as "flying" when the driver's front wheel comes off at 75MPH?
[Napoleon Dynamite: "You caught some air on that one, Pedro!"]
Chicken-and-egg.
At the moment, even if every car manufacturer made all of their cars with such a system and every consumer were forced to buy it on any new car, it would be over half a decade before there was a 50/50 chance the car who just tried to pull out in front of you had a computer yours could honk at.
The best systems are independently defensive and merely informative to those around you (they don't depend on the people around you to act).
Then, of course, there's the notion of hacking.
Of course, if you want to really screw something like this up, an RF jammer would do the job quite nicely and with a lot less fuss about encryption and all that nonsense. You don't need to commandeer it, just make it fail.
Not making any suggestions, because of course that would be illegal, but such a thing would be relatively trivial to build. It doesn't need to broadcast anything specific, just a crapload of static at a shifting frequency until the altitude of the drone is observed to be decreasing at a rate approximating terminal velocity.
The sort of person who finds themselves under sudden drone scrutiny is statistically less likely to be concerned about the legality of their methods to elude capture or avoid observation, or who the drone might hurt in the resulting impact.
In fact, the resulting impact might slow the pursuers down, what with the possible road damage and possible injured civilians to care for.
10,500 is just the maximum operating altitude. Given its 40-minute operating limit, I expect they'll hover it up to about 500-1000 feet and use it there, probably launched from a vehicle because 40 minutes isn't a whole lot of range. 500-1000 is below minimums for flying over most populated areas anyway, so there won't be a lot of fixed-wing or whirlybird traffic down there. Once they enter the upside down wedding cake, though, they'll HAVE to be talking to a controller, it HAS to have a transponder, and the controller will be slicing them a piece of wedding cake to fly in.
I'm actually more concerned with something like this in "C" and higher airspaces, where there's generally no RADAR approach and avoidance is more often than not VFR-only. Unless it has a lot of side-view cameras and several operators, something like this is going to be pretty much incapable of see-and-avoid.
I suspect something like this will be more likely to be seen outside a skyscraper window (where planes are not allowed) than over a rural house.
[Glasses_Man (Cusco? something like that) takes control of the drone and solves a crime]
[Local officer comes up and starts yelling at Glasses_Man while G_M is trying to land the drone].
Cusco: "What was that, I couldn't hear you over the..."
[puts on glasses]
"...droning."
YEAAAAAAHHHH!!!!
(sorry)
Assuming, of course, that the police model (unlike the military model) is actually encrypted.
Now why do you think he made a character called "Greed"-o and made him green? I mean, c'mon, it's a self-portrait!
LOL! Good point.
A magnetic compass is only accurate, as you say, if you are flying straight and level. For navigation, it's more convenient to use a gyro (which is a very reliable piece of kit, but also has gyroscopic precession issues so you need to recalibrate it to the mag compass occasionally).
Since the gyro has to be calibrated, and the mag compass is a cheap simple and reliable backup nav unit anyway, pilots use magnetic north.
You can get as complex as you want with gizmos, but at some point in your flying career there's a chance you're going to rely solely on a mag compass for navigation. If you end up in that mode, trust me, you DO NOT want to make your calculations any more complex than you need.
Aviation calibrates everything around your simplest instruments. For directions, this is the mag compass. The GPS can calculate mag north easily, the compass cannot calculate true north on its own. It's too simple a device, which is of course what makes it so reliable.
When you're under complete instrument failure and things are going pear-shaped on you, you really don't need any more work on your plate. Get to straight and level for a second, now you have an accurate compass reading.
Until it dies, then you either use a compass and have to adjust for declination, or use a compass and use mag north.
A GPS can read in mag north easily. They're complex little gewgaws with computers. Why make a pilot keep one more bit of data in his head that could potentially be inconsistent with what he's observing and make him adjust for it when the computers can calculate mag north easily?
Maybe only a dozen or so pilots have to navigate solely off a mag compass because their glass cockpit or directional gyro went kapoof, but that's a dozen people a year who already have a lot on their plates when trying to land an aircraft after a major instrument failure. Why make that any harder than it has to be? Because we don't want to write declination adjustment software (stuff that's been in GPS receivers since the very early days of GPS) in the GPS units and fancy gewgaws?
Which is why you have a directional gyro on board. The DG, which runs off very simple gizmos and is very reliable, gives you a convenient reading of direction most of the time, the compass is there as the base frame of reference (because DGs have to be reset occasionally), and just in case the DG goes kerbonk. You may only know your direction with precision when flying straight and level if the DG goes bye-bye, but you can always level off for a few seconds and get a directional reading.
Even cockpits with all the fancy glass and such usually have a basic compass, a simple electrically-driven gyro, a vacuum-driven airspeed indicator, and a pressure-sensitive altimeter. That way, if you forgot to load software version 4.334.84.779.83.56.23885.2341a revision 12b, and the screen goes black taking all your radios and GPS and weather reporting and $18,000 hiney wipers, you know how high you are, what direction you are going, and how fast you are traveling.
We use mag north because the simplest systems can accommodate it easily. And, guess what? The most complicated systems can accommodate it easily, too. Name me one model of GPS that does not have a selection between "mag north" and "true north".
When flying, no matter where you are in the world, you use Zulu (GMT) because everyone always knows exactly what time it is.
When flying, you use mag north because the simplest instruments always know where "North" is.
When flying, you use English because every pilot can talk to every other pilot and every controller across the planet.
When flying, you keep the simplest instruments available on board as backups to the complex stuff that breaks.
Flyin's complicated enough. Keep the basics simple.
Both.
The regulations are there for a very good reason. Every airplane needs a magnetic compass on board, because in all the years of using magnetic compasses, they aren't known for running out of batteries or needing to be rebooted at a critical moment or failing because a programmer fucked up and you just crossed time zones. Electronics can easily accommodate using mag north, magnetics cannot as easily accommodate using true north. So we use the system the simplest system can accommodate easily, and the more complex systems that are MORE than capable of adjusting for it do so.
Hobbyist planes are where professional pilots learn to fly, and most of them start out using a magnetic compass as their primary directional instrument (or at least use that as input to set a a directional gyroscope that's easier to read). That way, when the pro pilot is up there and the instrument panel suddenly goes dead in a puff of smoke, there's no reason to write off the lives of the passengers on board. The pilot knows where he's going, and this is due in part to the simple instruments that are on board, and in part to the fact that he's prepared in case this happens.
You learn to fly using shit that don't break, then you get to play with the fancy doodads later, but you never forget how to use the shit that don't break. People's lives depend on that.
I think it's rather the opposite. It's the freedom from copyright that allows some people to censor it as they like, but also allows the original, uncensored work to continue to exist and be copied by people who prefer that version.
If the copyright holder decided to censor the work, they could simply stop distribution of any previous versions and the new censored version would be the only new one available. Your only chance of getting a copy of the uncensored work would be to purchase a used copy (which would quickly become desirable and therefore expensive) or pirate it.
Switching to an equally (among certain geeks) controversial but far more recent censorship issue, when Star Wars IV was re-released with the "Greedo shot first" bit, it was before the DVD player existed. Lucas stopped distribution of the original work, which was only available on VCR tape that could wear out. Within legal copyright, there is no way to possess anything but an original VHS tape of the movie, which are a rare and vanishing breed. If copyright was anything approaching sane, that movie would have fallen out of copyright several years ago and would be freely available for distribution and copying in its original form.
If a work is still under copyright, no one else can mess with it. But if the copyright owner decides to mess with it, no one else can stop them.
Lack of copyright means everyone owns it. They can do what they want with it, copy it, distribute it, and censor their own version for re-distribution.
Copyright means only one person can do anything with it. If they decide not to censor it, then it cannot be censored. If they decide to censor it, then it must be censored.
Copyright is only beneficial in this way if everyone agrees that there should be only one version of a work, and everyone agrees with the copyright holder as to which version that is (in which case the work protects itself).
In the same way, dictatorship is the most efficient form of government if you have a truly benevolent dictator, and having an unregulated monopoly is always the cheapest and best way of making a product if you assume the monopolistic company would have a natural inclination to serve their community and employees in the most efficient and mutually beneficial manner possible.
For those keeping score at home, none of the three will ever work out that way.
I think you may have missed my point, or I made it poorly (which is likely). Let me try again.
The issue at hand is network neutrality.
A caching service like this can easily benefit both the ISP and the customer by reducing unnecessary bandwidth, as long as it isn't abused. The gist of the article is that "caching servers bad", and I disagree. I agree that caching servers can be abused, but let's not throw the baby out with the bathwater here.
So in the network neutrality discussion, do we want to say that:
1. Caching should be off-limits forever and completely (as the article implies).
2. Caching should be available absolutely for free to all content providers who ask, at no charge and regardless of how much content the provider wants cached (forcing BT to basically cache the entire Internet if they choose to cache anything at all, and some things like NetFlix and encrypted data are un-cache-able without the content provider's cooperation and assistance).
3. Caching and charging for it is allowed but ISPs must maintain sufficient bandwidth to support non-cached activities as well, with the only difference in speed being the obvious speed boost you achieve by caching something locally.
I maintain that option #3 is the most logical choice. It maintains network neutrality while giving ISPs the opportunity to manage their network traffic.
Otherwise, BT's going to have to put in a shitload of backbone upgrades in order to accommodate every one of their customers doing streaming bandwidth 24/7 from anywhere on the planet. If they have to do that, your rates are going up.
I'm not saying they should be able to hold a high-bandwidth service hostage, but they might find it in their best interests to give that service a really good deal on caching services and prevent some of those upgrade costs they'll pass on to you.
A lot of ISP's have an Akamai presence. I'm frankly surprised that BT wants to even bother "rolling their own" when there are already third parties ready to do this for them. However, intelligent caching is good for everyone, and it does require a level of content provider cooperation (so the content provider can clearly identify what is cache-able and what is dynamic).
The problem is that the backbone necessary to do that costs serious money. A CDN is a more efficient use of the existing infrastructure, and costs a whole lot less than an extra few thousand miles of fiber.
What amazes me is that BT and other ISPs don't simply offer this for free, because it benefits them as much as it does their customers. I guess any revenue stream is a good revenue stream, but if they simply cached ALL static content and worked out deals with people like NetFlix to store encrypted copies of their movies for local serving rather than streaming every copy of every movie across the Internet backbone, it would make BT seem like the fastest ISP *EVER*, save them gobs and shitloads of money on backbone bandwidth, allow them to raise monthly caps and allotted speeds to very high levels without putting their network under significantly more strain, put off expensive backbone upgrades for years, and cost them very little
I think this is basically an Akamai-like caching server. And I can see where the controversy lies.
If BT implements this and doesn't intentionally throttle other services, I don't see this as a violation of neutrality - BT is not discriminating against anyone in using the Internet pipe, they are simply maintaining a cache service for those who want to cough up a little more dough for their web sites to be stored in a local cache. BP customers can still access anything they want on the Internet at Internet speeds, but certain things run at local speed which is faster.
However, I can see how this could be easily abused, if BT started speeding up the Internet packets (*) coming from their customers who paid for the caching service and slowing down everyone else's, or started blocking or throttling sites that refused to pay for it, or lowered their overall Internet connection to the point where only cached sites were useful.
I can also see how this could be interpreted as a "two tier" system, but such systems have been in use for quite some years in the US and have been very successful here. They do make some web pages faster than others, but I haven't seen many reports of ISPs intentionally throttling their regular Internet bandwidth to punish service providers who don't pay up. I've heard of ISPs who try to force high-bandwidth content providers to subscribe, and that's wrong, but that's a matter of abusing the technology, not a problem inherent to the technology.
Frankly, I don't understand why an ISP wouldn't want to simply start caching all static content. But, unfortunately, that means that most content they really want to cache is not going to be. Streaming video from someone like NetFlix is encrypted so the movie you watch is a different set of bits from the same exact movie your neighbor watches 5 minutes later. BitTorrent is not only comprised of a great deal of illegal content that the ISPs don't want in their cache servers, it's also frequently encrypted, and the BitTorrent protocol is going to tend to prefer "local" clients so it's already optimized to save backbone usage when possible anyway.
YouTube would be brilliant for this sort of thing, and YouTube actually uses Akamai if I recall correctly.
(*) By "Internet packets", I'm referring to caching customers who might only cache static content. This is how my company uses Akamai - we give Akamai a copy of all of our static content (ie. pictures of our product), they replicate it out to all of their edge servers around the world as needed, and we simply use an Akamai URL to access the image on the web site. Akamai automatically determines the closest server to the customer and serves them up the replacement image. All encrypted and dynamic content is directly between the customer and our web servers.
I've been known to pull out a notebook and take notes if someone is on a business call. One "can you repeat that, please?" usually is enough to remind them that talking about company business in a public place where everyone around them is pretty quiet is not only rude, it's terribly inadvisable. :)
Yeah, having to click through a bunch of "you shouldn't do this while you're driving" bullshit is very distracting when I'm driving. :)
are all voluntary.
For now. Until a few kids get killed by some asshole using his phone to send a text. Then it'll be a "think of the CHILDREN" law that no cell phone may operate at more than 5MPH.
If you don't allow your users to have Admin access, the differences probably aren't that huge. There is Windows Defender and BitLocker, so it might save you a few bucks on drive encryption software and possibly antivirus, and Libraries might make their documents a little easier to find.
If you do allow Admin access, the security popups can be a significant improvement in security, as they warn you when system areas are being futzed with.
The company I work for is still on XP and quite happy with it, for now. We're in the final throes of getting rid of IE6, though.
It also surprises me sometimes that the overages cost so much more per gigabyte than the base plan does.
Verizon's cell data plan is 2GB for $20. That's a decent price, and not a completely unacceptable cap. But their overage is $20 per gigabyte. That's their one and only smartphone/featurephone plan. So if I use 3GB a month on one phone, that's the same as 4GB would cost me if I had two phones at 2GB each. If I used 4GB a month, I'm paying $60 a month, enough to have three data plans.
AT&T's DataPro 2GB plan is $25 a month, but overages are at $10/GB, which means it's a little better, but their laptop data plan overages are five and ten cents a megabyte, which works out to $50 or $100 a gigabyte. If you buy their 5GB $60 plan and go a gigabyte over, it's $110. If you buy their 200MB $35 plan and use 1GB over, it's $135! It's far cheaper to buy a basic phone and tether it if you need over 5GB a month...