In other words, you really drank the cool aid on the double dipping situation.
Mutual transit agreements are rightly predicated on a reasonable balance of traffic. It does indeed cost money to provide transit ( Allowing traffic from ISP A bound for ISP C to transit your network) and so free transit agreements need to be reciprocal.
Peering is an entirely different case. That's where traffic from ISP A bound for customers of ISP B is routed through a direct connection from A's router to B's router without 3rd party transit. It makes no ethical sense for peering connections to be metered in any way. After all, these are packets an already paying customer wants delivered as part of their already paid for service. If they don't route it to it's destination, they're already breaking a contract with a customer.In other words, ISPs get paid for sinking traffic by the customer that the traffic is bound for.
Going to the specific case of Comcast accepting traffic from Netflix, every packet handed to Comcast that originated from Netflix happens because aone of Comcast's paying customers requested it (as is their right by virtue of having paid Comcast a monthly service fee). Comcast already agreed to carry that packet, and is getting paid to do so by their residential customer. Any attempt to get Netflix or their ISP to pay for that packet as well is double dipping.
The traffic on that "everloaded gig-e link" was traffic that paying Comcast customers requested (because they wanted to watch Netflix). Comcast was actually providing crappy and sub-standard service to their paying customers by not doing anything about that overloaded link. Netflix kindly offered to provide Comcast with a caching server that would have reduced that traffic by orders of magnitude in exchange for a few U of rack space and a few KWh of power. It was a pretty good offer and a clear win all around. Comcast turned them down cold and countered with fuck you, pay me. Netflix said "no thanks, enjoy your transit bills" as was their right.
Had there been proper competition from a healthy market, none of this would have come up. If Comcast used an overloaded connection as a throttle, Comcast's Netflix watching customers would have switched to a competing ISP and Comcast would be kindly invited to pound sand. If we had decent consumer protection laws, Comcast would have been ordered to either deliver the traffic their customers already paid them for or start refunding money and paying fines.
The whole situation is best visualized as a group of people in business suits standing knee deep in a flow of cash fighting over a penny until it stretches into copper wire.
Both Netflix AND the ISP save tons of upstream bandwidth.
Or, without neutrality, ISP throttles the hell out of Netflix and zero rates CrapeeStreaming (a wholly owned subsidiary) and gives their customers the middle finger suggesting they go back to dial-up if they don't like it.
Even ignoring that, the problem is more fundamental. Let's imagine for a moment the world's one and only infallible drug sniffing dog. He alerts to the SMELL of drugs. That is not the same as alerting to drugs actually in the package. If someone else's package full of drugs got damaged, there may be drugs ON the innocent package. If it rode in the back of a hot truck next to an imperfectly sealed box of pot, it will smell like pot. In either case since it is rare for only 2 packages to be carried together in a truck, far more packages will smell like drugs than actually contain drugs.
Of course, in real life no dog is infallible. They all have a significant false positive rate above and beyond the innocent packages that came in contact with drugs.. When most of the packages checked do not contain drugs, it means MOST packages a dog alerts on will be innocent, which is more or less the opposite of reasonable suspicion.
You mean as opposed to now where they're strung out on illegal drugs and so have little to lose by adding one more criminal charge?
Try running the prisons properly and legalizing drugs. Then the prospects are: Avoid crime and use drugs or commit a crime and 'enjoy' a few years in a drug free prison.
With the significantly reduced prison population and taxes on the drugs, we will easily be able to afford to upgrade the prisons and still save money.
We already have a standard for specifying a time in absolute terms. Just use it when scheduling an event across time zones and all is well. No need for anyone to change their clocks or anything. If you're unsure if someone might be in a different time zone or observe summer time differently, use parentheses. For example, conference call is at 1:45 P.M. ( 17:45 UTC).
"So quit my job and hold out until an employer is willing to sign a contract that no employer of software developers has signed before? And they'll do it because it's more expensive and gives them less control because they love those things?
Actually, you need laws. Otherwise,all liability will land on the developers while they get no authority to withhold sign-off until they're actually satisfied.
At the same time, many of the security issues actually do exist in more conventional engineering fields. How many bridges do you suppose are resistant to multiple attempts to bring it down every day? If someone did get a key support to fail, by cutting it or blasting it, do you really think the designer would bear responsibility?
Bank vaults are't graded on breakable/unbreakable. They're rated in hours and minutes because they can ALL be broken in to given sufficient time. Makers of cheapie door locks DO routinely make wild boasts about security that don't hold water at all.
Major manufacturers of "security panels" produce units that all have the same key (on sale now on ebay) and are trivially hotwired with a paperclip.
ALL security sucks donkey balls. It's just more apparent in software because the relatively rare non-dumb crooks don't have to actually travel to attack it and because they're safely tucked away in some other jurisdiction, they're rarely tracked down.
It's funny how silly your arguments are getting, especially in light of the news that the disagree indicator light will now be made standard.
By silly, I mean, how can MCAS possibly operate correctly when it has only two identical sensors to work with and they disagree. The system has no way to determine which one, if either, is right.
Currently the various expert systems and automation have zero morals and ethics. Their only criteria are maximize profit, minimize risk. If someone ends up dying in screaming agony, meh.
This is just another extension of the principle that the only people you can get on the phone are people with no power to say yes. Their job isn't to make an ethical decision, their job is to make sure the people with authority to make a decision don't have to personally feel the consequences of tossing ethics out the window.
Same with the software. The programmers are just following orders, it's not like they're using the software on actual people. The people using it are just following orders, it's not like they wrote the software to make those decisions. The people at the top just specified software and ordered it's use. It's not actual people, just boring statistical data on a quarterly report.
Of course, in reality the software is an extension of those who give the orders. They just want people to blame "the computer" for as long as possible, just like in the '70s when, according to the CSR on the phone, the computer was infallible.
True, but a square going "BIP! Boop!" doesn't present much moral dilemma. It took a bit over a decade and a half before video games presented anything like realistic imagery.
Slashdot Mirror
In other words, you really drank the cool aid on the double dipping situation.
Mutual transit agreements are rightly predicated on a reasonable balance of traffic. It does indeed cost money to provide transit ( Allowing traffic from ISP A bound for ISP C to transit your network) and so free transit agreements need to be reciprocal.
Peering is an entirely different case. That's where traffic from ISP A bound for customers of ISP B is routed through a direct connection from A's router to B's router without 3rd party transit. It makes no ethical sense for peering connections to be metered in any way. After all, these are packets an already paying customer wants delivered as part of their already paid for service. If they don't route it to it's destination, they're already breaking a contract with a customer.In other words, ISPs get paid for sinking traffic by the customer that the traffic is bound for.
Going to the specific case of Comcast accepting traffic from Netflix, every packet handed to Comcast that originated from Netflix happens because aone of Comcast's paying customers requested it (as is their right by virtue of having paid Comcast a monthly service fee). Comcast already agreed to carry that packet, and is getting paid to do so by their residential customer. Any attempt to get Netflix or their ISP to pay for that packet as well is double dipping.
The traffic on that "everloaded gig-e link" was traffic that paying Comcast customers requested (because they wanted to watch Netflix). Comcast was actually providing crappy and sub-standard service to their paying customers by not doing anything about that overloaded link. Netflix kindly offered to provide Comcast with a caching server that would have reduced that traffic by orders of magnitude in exchange for a few U of rack space and a few KWh of power. It was a pretty good offer and a clear win all around. Comcast turned them down cold and countered with fuck you, pay me. Netflix said "no thanks, enjoy your transit bills" as was their right.
Had there been proper competition from a healthy market, none of this would have come up. If Comcast used an overloaded connection as a throttle, Comcast's Netflix watching customers would have switched to a competing ISP and Comcast would be kindly invited to pound sand. If we had decent consumer protection laws, Comcast would have been ordered to either deliver the traffic their customers already paid them for or start refunding money and paying fines.
The whole situation is best visualized as a group of people in business suits standing knee deep in a flow of cash fighting over a penny until it stretches into copper wire.
NASA director reads the memo again:
by any means necessary
Hrmmm, clickety clickety clickety...Wall funds diverted to NASA!
Both Netflix AND the ISP save tons of upstream bandwidth.
Or, without neutrality, ISP throttles the hell out of Netflix and zero rates CrapeeStreaming (a wholly owned subsidiary) and gives their customers the middle finger suggesting they go back to dial-up if they don't like it.
You might be surprised. How much trace do you suppose you leave in a footstep? We know dogs can smell that and differentiate between two people.
The AC's link is good. Here's another
Even ignoring that, the problem is more fundamental. Let's imagine for a moment the world's one and only infallible drug sniffing dog. He alerts to the SMELL of drugs. That is not the same as alerting to drugs actually in the package. If someone else's package full of drugs got damaged, there may be drugs ON the innocent package. If it rode in the back of a hot truck next to an imperfectly sealed box of pot, it will smell like pot. In either case since it is rare for only 2 packages to be carried together in a truck, far more packages will smell like drugs than actually contain drugs.
Of course, in real life no dog is infallible. They all have a significant false positive rate above and beyond the innocent packages that came in contact with drugs.. When most of the packages checked do not contain drugs, it means MOST packages a dog alerts on will be innocent, which is more or less the opposite of reasonable suspicion.
If they did that, they would be forced to admit that there's enough residue on common items that the dog would alert to everything.
For example, money. Pretty much all of it will test positive for at least cocaine.
The dirty secret: Most of the time the dog alerts due to subtle cues from it's handler, not from something it smells.
You mean as opposed to now where they're strung out on illegal drugs and so have little to lose by adding one more criminal charge?
Try running the prisons properly and legalizing drugs. Then the prospects are: Avoid crime and use drugs or commit a crime and 'enjoy' a few years in a drug free prison.
With the significantly reduced prison population and taxes on the drugs, we will easily be able to afford to upgrade the prisons and still save money.
Nobody cares. Everyone, obviously including you, understood what was meant. Haven't you picked up on that yet?
And they'll argue that what we do now ALREADY seems normal and nobody even has to list a finger!
We already have a standard for specifying a time in absolute terms. Just use it when scheduling an event across time zones and all is well. No need for anyone to change their clocks or anything. If you're unsure if someone might be in a different time zone or observe summer time differently, use parentheses. For example, conference call is at 1:45 P.M. ( 17:45 UTC).
And nobody had to change a clock or anything.
Sorry, no. It's turtles all the way down.
Consider, your rent-a-court Says you're right (SURPRISE!) and I should pay you $100. I say no. rent-a-court does what?
The law is the only reason PE's have final authority to sign off now and it's the only reason firms must hire them in spite of the expense.
You DO realize that the only reason the civil courts have any authority is those same men with guns, don't you?
Only if they expect me to bear responsibility as a PE would for a bridge (for example).
If they want to keep the authority, they get to keep the responsibility.
Who said anything about entitlement. With responsibility must come authority or the responsible party is merely a scapegoat.
What's so entitled about not accepting the role of scapegoat?
The music industry profits by selling bank robbers the music they use to get themselves psyched up for the big hit.
"So quit my job and hold out until an employer is willing to sign a contract that no employer of software developers has signed before? And they'll do it because it's more expensive and gives them less control because they love those things?
SUUUUUUUUUre.
I'll bet more people would recognize a stiletto as a dangerous weapon (especially if it's held to someone's throat) than a zip gun.
If the gun blows up in your hand, mission fails. I'm pretty sure they care about that.
Actually, you need laws. Otherwise,all liability will land on the developers while they get no authority to withhold sign-off until they're actually satisfied.
At the same time, many of the security issues actually do exist in more conventional engineering fields. How many bridges do you suppose are resistant to multiple attempts to bring it down every day? If someone did get a key support to fail, by cutting it or blasting it, do you really think the designer would bear responsibility?
Bank vaults are't graded on breakable/unbreakable. They're rated in hours and minutes because they can ALL be broken in to given sufficient time. Makers of cheapie door locks DO routinely make wild boasts about security that don't hold water at all.
Major manufacturers of "security panels" produce units that all have the same key (on sale now on ebay) and are trivially hotwired with a paperclip.
ALL security sucks donkey balls. It's just more apparent in software because the relatively rare non-dumb crooks don't have to actually travel to attack it and because they're safely tucked away in some other jurisdiction, they're rarely tracked down.
The problem here is not a live and let live attitude, the problem is when we provide material support.
Who said anything about CGA? SVGA was the key evolution there (though EGA looked good compared to CGA, it's not that great in retrospect).
Even with that, it took a really immersive game to get the player to forget they were looking at CGI.
It's funny how silly your arguments are getting, especially in light of the news that the disagree indicator light will now be made standard.
By silly, I mean, how can MCAS possibly operate correctly when it has only two identical sensors to work with and they disagree. The system has no way to determine which one, if either, is right.
Currently the various expert systems and automation have zero morals and ethics. Their only criteria are maximize profit, minimize risk. If someone ends up dying in screaming agony, meh.
This is just another extension of the principle that the only people you can get on the phone are people with no power to say yes. Their job isn't to make an ethical decision, their job is to make sure the people with authority to make a decision don't have to personally feel the consequences of tossing ethics out the window.
Same with the software. The programmers are just following orders, it's not like they're using the software on actual people. The people using it are just following orders, it's not like they wrote the software to make those decisions. The people at the top just specified software and ordered it's use. It's not actual people, just boring statistical data on a quarterly report.
Of course, in reality the software is an extension of those who give the orders. They just want people to blame "the computer" for as long as possible, just like in the '70s when, according to the CSR on the phone, the computer was infallible.
True, but a square going "BIP! Boop!" doesn't present much moral dilemma. It took a bit over a decade and a half before video games presented anything like realistic imagery.
Easier still to make an ABS stiletto and hide it inside a hair brush.
Less likely to blow up in your hand as well.
The answer is to try not to screw up 2 or 3 times within a critical window.