The ups and downs also depend on which party holds Congress. It tends to be whichever party doesn't hold the White House, more often than not.
The Constitution makes the division of power clear. Congress controls the purse strings, not the president. You can blame the president all you want for saying what the government's priorities should be, but ultimately it is Congress that is responsible for deciding how money is apportioned and how much tax is collected. The only thing the president can do is veto it, and only if Congress can't get a two-thirds majority to agree that it is a good idea.
In other words, President Trump's temper tantrum over the border wall can go away tomorrow if Congress would grow some balls and work across the aisles to present a budget approved by a veto-proof majority.
No, but pretty much every mobile home on the planet has a utility room, and those are pretty much all built after the 1950s. And lots of folks do things like add a screen room over the back deck or whatever. The same approach works with a wide range of styles.
I'm sure you where just trolling but we can't have you spreading bullshit can we. I fixed you comment for you after all it was Obama, a democrat, that doubled the national debit adding 10 trillion dollars to it. More than all other presidents before him combine.
Not trolling. Sarcastic, but not trolling. The President of the United States controls neither spending nor taxation. That's Congress's job. What you're actually saying is that the (almost always Republican-controlled) Congress ran up more national debt under Obama than under all other presidents before him combined.
As a rule, Democrats raise taxes to cover their spending, while Republicans borrow money to do so. There are exceptions, but they are rare. Thus, the national debt is mostly a Republican-caused problem. Tax cuts (the heart of the Republican platform) inherently increase the national debt. Anybody who says otherwise is kidding him/herself. It's like selling products at a loss, but making it up in volume.
What would replace Amazon? Do you want to order exclusively from Banggood, Alibaba, or Taobao? If you want to boost your Chinese social credit score, that is fine
When the Chinese foreclose on all that national debt the Republicans have been running up, you might wish you had done so.:-D
Alternatively, that's what utility rooms (or enclosed porches) are for. You put a lock between that room and the rest of the house, and if you know you're going to have a delivery when you can't be there, you provide them with a temporary code for the outer lock, and lock the door between the utility room (or porch) and the rest of the house.
If someone doesn't want to transmit files through Apple's servers, suggesting to use Dropbox instead is rather lame. Apart from fears that companies could intentionally cheat on you and your privacy, sharing files through iTunes is _intended_ not to be shared with others, while DropBox is _intended_ to be shared, so it is more likely that hackers might be able to "share" your data.
That's arguable. Dropbox has a system designed specifically to provide access to specific users authorized to access a given file, and requires the other party to know about the file, whereas iCloud is designed to automatically share files with any device tied to the account, complete with a mechanism for easily adding new devices to the account. So the question becomes one of which is more likely: Dropbox getting their ACL programming wrong or Apple having a bug that allows injecting a new device into an iCloud account.
I'm not saying you're wrong, because it's really anybody's guess which is more likely. Either way, I don't think it is necessarily safe to assume that a site designed for secure public sharing is more likely to leak data than a site designed for secure inter-device sharing within a single account. I actually suspect that the odds are about the same.
Tim Cook keeps blaming it on repairs, and in the case of China, tariffs, but that doesn't at all reflect what third party research data is showing: In nearly all cases, people say they're too expensive and don't deliver enough value to justify that.
Particularly now. I can't help but find it ironic that Apple is lecturing the rest of us about privacy just a few short months after they moved all their Chinese users onto servers in mainland China — one of the most dubiously privacy-risking moves in the history of the industry.
And now, their sales in China are weak. Imagine that. It's almost like after years of Apple using privacy as a selling point, their users actually care about privacy.
Kidney stones can be caused by lack of water, but once you have them, AFAIK, water won't do anything to dissolve them. And although water is preventative for gout, I don't think it is curative (but I could be wrong). Both conditions, however, can be improved with citrus juice.
Completely untrue. Many cancers are curable with extremely high success rates if caught early, including acute lymphoblastic leukemia (98%) and localized (non-metastatic) prostate cancer (~100%).
And even stage 4 metastatic cancer is sometimes cured, thanks to immunotherapy.
The only thing absolutely certain is that water has a 0% cure rate for any illness other than dehydration, and always will.
If a vehicle gets damaged between the assembly line and the customer, short of someone wrecking it, that's a design flaw. Vehicles should not be that fragile.
Ironically, this is a problem that automation is likely to make worse. If a part is defectively weak by design, the rate of early breakage during assembly will be much higher with human assembly than with automated assembly, because humans are less careful. This can make design flaws in parts obvious earlier, which increases the chances of the engineering team being called in to fix the design before a million of them get distributed to customers.
The grandparent is actually entirely correct. Comcast's internet advertised pricing is all-inclusive, with a few caveats: you have to own your own modem
Caveat emptor: If you go with Comcast Business and require static IPs, they won't let you own your own modem. You have to rent theirs for an extra fee (which is not included in the static IP fee).
Netflix isn't innocent in this. They throttled their own customers and blamed it on ISPs.
Citation needed.
As far as I'm aware, the only throttling Netflix has ever done has been in the form of providing lower-bitrate streams to pay-by-the-byte cellular customers, and they have never blamed the ISPs for this money-saving feature.
Yes, the lopsided peering agreements cost the ISPs far far more than the electricity costs for a couple of rack mounted boxes.
Would those agreements still be made if streaming services, like Netflix, didn't exist? If so then it is a moot point.
The agreements would exist, but the total traffic (and thus, the total cost of those agreements) would be much, much less.
But that is hardly the entire issue. How many other streaming/information services does the ISP have to house? Why is Netflix able to dictate the business of a different company? A couple here a couple there add up quickly. Besides, just because it is cost effective doesn't mean that an ISP has to do whatever Netflix demands.
The reason the ISP has to do it is because they are a monopoly, and their only viable legal alternative is to provide enough bandwidth to their upstream providers so that their in-house video-on-demand services are not guilty of unfair competition against external video-on-demand services like Netflix et al. And the answer to the question of how many other services is "all of them" with the sole exceptions being external service that are so small that it is cheaper to provide enough bandwidth instead.
Or if they would prefer, Comcast could shut down all of their cable TV and video-on-demand services and become a pure ISP.
Which require rack-space and electricity. Do those costs offset the bandwidth savings?
For an ISP the size of Comcast and a content provide the size of Netflix, it is cheaper by nearly four orders of magnitude. Netflix actually publishes information about their caching servers online, though I can't find the power specifications, so I can only take a rough guess based on the typical power consumption of blade servers in that form factor.
I would not expect a 2U rack unit to draw much more than 1 kW of power, which at 8 cents per kWH costs about $700 per year to keep it powered. Each box can provide 10 GB per second of throughput, which is enough to handle 26,000 streams at 3 Mbps (this is an arbitrary example), or 2.6 cents per continuous stream per year. By contrast, a single upstream T3 trunk line costs about $36,000 per year, and can provide enough bandwidth for about ten streams at 3 Mbps, or $1,200 per continuous stream per year.
If an ISP gives such favorable treatment to Netflix do they have to give the same treatment to any streaming service/information service provider?
No, most streaming providers partner with companies like Akamai to do the same thing until they get big enough to warrant their own servers. And when they do, the ISP would be utterly stupid to not agree to a similar arrangement—utterly stupid almost beyond the limits of human imagination.
An ISP should pay Netflix's rack-space and electricity bill because ???
A few key points should explain it:
The nature of the Internet is that each party must pay for its own bandwidth to the nearest backbone. Netflix pays for the bandwidth it uses to get its data to the backbone by paying its upstream ISP, who pays its upstream ISP, and so on. Similarly, Comcast must do the same.
The caching equipment reduces the bandwidth, and thus the cost, of both of those connections. Therefore, it benefits Comcast and Netflix approximately equally.
The hardware costs a lot of money (racks of RAID arrays aren't cheap), and Netflix provides that hardware. Thus, given the equal benefits, it seems entirely reasonable for Comcast to foot the (funny-money) bill for a place to store it plus the trivial cost of electricity, which is dramatically less expensive than the upstream bandwidth they would otherwise have to pay for, and even over the long term, also costs dramatically less than the hardware.
In short, the caching approach represents huge savings for Comcast, and gives them a cost split that is very much in their favor compared with the 50-50 cost split that would occur if they had to pay for bandwidth to the backbone to handle that traffic. The only alternatives are for Comcast to either pay more by using bandwidth and pay a larger percentage of the total cost or to throttle Netflix in an anticompetitive manner and get sued.
Netflix should decide when and where an ISP upgrade its infrastructure because ???
Netflix does no such thing. They provide higher-quality stream options, and the ISP's customers either care enough to demand that the ISP improve its bandwidth or they don't.
Meanwhile, your solution seems to be the government force ISP's to eat those costs because Netflix and Comcast are fighting.
No, the government should force the ISP to eat those costs because Comcast is a monopoly that is behaving anticompetitively.
There has to be something very wrong with either your numbers or your approach. The damage per mile for an 18-wheeler is, assuming it is fully loaded down, ~10,000 times that of a passenger car, and you're advocating raising the gas tax to only a factor of 10 times what cars pay.
Diesel trucks use somewhere on the order of 4x as much fuel per mile. Divide that 10k by 10 and you have 1k, and divide it by 4, and you get 250. With your approach, trucks are paying on the order of 1/250th of their fair share.
For your numbers to be reasonable, diesel trucks would have to burn 1000x as much fuel per mile as passenger cars, which would require trucks with gas tanks approximately as big as a thirty-foot swimming pool. Clearly, this is not the case.
The biggest problem I see is adding the cost of equipment in, as though we should treat that as somehow paying for the cost of roadways. That causes you to be off by one entire order of magnitude compared with the fuel costs, which are only a small part of the cost of trucking. However, your 54B number is also problematic. The industry as a whole used 54B gallons of fuel, but only used 39B gallons of that were diesel fuel, which is to say that the 54B gallon number includes a very large number of vehicles that aren't even big enough to use diesel. If you only want to look at big rigs, the number is likely single-digit billions, and given that smaller trucks do orders of magnitude less damage, we really have to focus on big rigs here.
The only reasonable way to tax this would be to take (at least) the road repair portion of the gasoline tax, multiply it times the weight multiplier (9,600 or whatever), and divide it by some factor based on the vehicle's miles per gallon (4-ish) compared with a normal passenger vehicle.
To be fair, only about two-thirds of state gas taxes go to pay for road maintenance, but the rest are usually spent for things like public transit infrastructure, and it seems reasonable to pass that on to the trucking industry as well, to pay for rail infrastructure expansion and improvements that would dramatically improve long-haul transportation efficiency when compared with trucking.
Bearing in mind that light duty trucks do not fill up at the same gas stations as tractor trailers (they won't fit under the canopy at a normal gas station), it seems entirely plausible that we could tax their fuel differently.
An alternative approach would be to dramatically increase the heavy vehicle use tax, basing it on mileage. Currently, this tax is capped at only $550 per vehicle per year, and only for vehicles over 55,000 pounds. If we lowered the limit to 10,000 pounds (about twice the heaviest passenger vehicles on the road) and taxed it at a rate of m[*d*((a/5000)^4) - b] where m is the mileage, a is the axle weight, computed as the taxable weight divided by the number of axles, b is the road damage estimate for a two-axle vehicle that weighs 10,000 pounds, and d is twice the baseline damage caused by an average passenger vehicle, you'd be in the ballpark.
In a a state with 35.6 cents per gallon state tax plus 24.4 cents per gallon federal tax, with 60% of the state tax for road repair, we get 21.36 + 24.4 cents, or 45.76 cents per gallon going to road repairs, divided by 24.7 miles per gallon on average, or 1.85 cents per mile for an average passenger vehicle. That's b, and the average vehicle weight for those vehicles is about 4,000 pounds.
Road damage for a 10k pound vehicle is (2.5^4) times that, which is about 39x, so 72.15 cents per mile. That's d.
So for a five-axle vehicle that travels 5,000 miles and weighs 80,000 pounds (16,000 per axle), m[*d*((w/10000)^4) - b] is 5000[(.7215 *(3.2^4)) -.0185], or about $378k.
I'll let you ponder the degree of an externality so massive that a truck can do $378k in damage and pay only $550 in heavy vehicle use taxes plus about $370 in gas taxes to cover that damage.
Every time I see these 'net neutrality' things all I can think about is Idiocracy and the 'its got electrolytes' bit, and the NN version of the same is 'muh bandwidth wuz stranglified'.
Competition in a zero sum environment:
The total bandwidth available is not short time frame elastic, its completely static. So an example for sake of argument under the imposed 'bad' NN rules the 'muh bandwidth wuz stranglified' people keep pushing for:
Netflix in North America uses 50%(whatever the real number is doesn't matter) of available bandwidth and pays the same as everyone else under that rule, ISP/trunking/peering companies are unable to charge them more by the imposed rules.
Along comes SUPERNetflix with double goodness and 4X the bandwidth use which everyone starts using because double is mo gooderer, and now they use say 99% of the available bandwidth and the same ISP/trunking/peering companies (yes these numbers are exaggerated) are unable to charge more or negotiate a throttling plan due to the imposed rules.
The net result is all internet traffic is essentially throttled down to a max of 1% and the ISP is handcuffed, unable to charge for fair use, or throttle in the best interest of its customers.
Actually, you're wrong on pretty much every count here. For any sufficiently large ISP, Netflix will *give* the ISP a caching box for their data center that will take the vast majority of Netflix traffic entirely off of the upstream pipes. The only cost to them, other than the electricity to power the box and the cost of square footage to house it, is the cost of upgrading the infrastructure from the central office to the customer, which is usually a matter of upgrading the equipment at both ends. And given how often customer premises equipment fails and has to be replaced, it is usually just a matter of upgrading the equipment at the head end. In other words, it is highly short-term elastic.
And even if an ISP doesn't have that arrangement, nobody runs only a single fiber anywhere, so in practice, there is *always* extra fiber capacity available for upgrading the connection between two ISPs.
So the only way you would see the problems you're describing is if an ISP like Comcast gets too big for its britches and insists that Netflix pays a monthly fee to put that caching box in their data center, and then refuses to upgrade the peering point between the ISP's network and Netflix's ISP's network. Which it did. And their mutual customers got screwed, mainly because Comcast thought that throttling Netflix (and only Netflix) would cause more people to buy video-on-demand content from Comcast. And obviously it worked, or else they would have quickly discontinued that experiment.
Short of some sort of regulatory penalty against Comcast and similar ISPs when they pull these shenanigans, nothing will change. Net neutrality was an attempt to use the FCC's regulatory powers to do so. Unfortunately, because their interpretation of the law was purely a regulatory position rather than an explicit piece of legislation, it was subject to the whims of a given administration, and went away when the administration did.
The reason we have cable monopolies is because local governments awarded monopoly service contracts. Usually they granted the monopoly in exchange for coverage guarantees - to insure that low-income areas weren't excluded from cable and Internet service.
They did the same thing for telephone. But you're badly mistaken if you believe that removing monopolies will result in competition. Countless towns have tried that over the years, and essentially 100% of the time, they were back to a monopoly within about five years.
Why is this the case? Because wire infrastructure is something economists call a natural monopoly — a product or service in which the startup costs are so high and the payoff over such a long period of time that competition is infeasible.
When a new company comes into a territory with an existing cable company, the new company has to have some way of inducing people to switch — either by lowering their cost or by providing better service. Either way, it is very easy for the existing company to match their offer, because they have paid for their infrastructure, and almost all of their income is profit. So the existing company invariably either matches or undercuts the newcomer and provides enough improvements to their service so that the newcomer cannot steal enough subscribers to cover the payments on their construction costs plus operating expenses.
What happens then? Unless the local government decides to bail out the newcomer to keep competition going, the newcomer typically sells the infrastructure to the older cable company for little more than what they still owe on it, thus giving the incumbent a set of new replacement lines at significantly below cost. And you're back to a monopoly.
That's why, with the possible exception of extremely high-density areas where a new company can get by on a tiny percentage of residents/businesses, the only places where competition has ever really succeeded have been places that have had two or more competing companies since the dawn of cable TV, and even those usually break down to a monopoly eventually.
This is a problem that cannot be solved no matter what you do. Wire infrastructure is simply too expensive to allow for competition in practice. Heck, wireless infrastructure is almost too expensive in most places, forget wired.
The only way to get real competition in Internet service is to separate the wire provider from the actual Internet service provider. In practice, this can happen in one of two ways, both of which involve the government:
Governments can create regulations that require incumbent cable or fiber providers to lease access to their competitors at a reasonable rate.
Governments can build out a fiber infrastructure and then lease access to any ISP that wants to provide service over those fibers. Governments can optionally spin off the resulting fiber provider as a nonprofit corporation with a mandate to lease access to any ISP at a reasonable rate.
Either way, the result is the same: You have one company or organization providing wire service for multiple ISPs. When you do this, competition is possible at the ISP level. In practice, though, the second approach (government-built infrastructure) tends to work better in the long run, for two reasons:
Incumbent providers who own the lines tend to do only minimum maintenance on lines that are in use by other companies, resulting in two companies blaming each other for poor service, and the customer having no real recourse.
Those sorts of laws typically apply only to a given technology, and have to be constantly updated as the technology changes, or else they become useless. After all, we had such a law for ADSL, but the phone companies then ran fiber out to remote terminals to provide ADSL2 service, and wouldn't lease those fibers, so only the phone company could provide the faster ADSL2 service. And once you switched over, they would cut the existi
So - Internet speeds are increasing after NN is overturned. How is this bad for the consumer, again?
Speeds to speed test servers are increasing faster after NN is overturned, but those have little relation to actual Internet speeds. They represent best-case speed, rather than typical speed, because no ISP would be stupid enough to throttle connections to a speed test server. But actual average speeds may or may not be increasing any faster than they were before.
Also, by focusing on average speed, you're missing the whole point of net neutrality. It isn't about the average. It's about the worst case. It's about ensuring that ISPs aren't extorting companies who aren't their customers, about ensuring that ISPs don't artificially degrade performance on specific services like video-on-demand or VoIP to drive customers to their own competing services, and so on. That can't be measured using average bandwidth. At all.
To use a car analogy, saying that average Internet speeds increased after the repeal of NN is roughly like saying speed limits increased after overturning a law that prevents illegal speed traps.
States should really take the opportunity to rationalize the road construction and maintenance cost allocation. According to the US GAO, road damage is proportional to the fourth power of per-axle weight, times the number of axles. They estimate that an 18-wheeler does 9,600 times as much damage as a passenger sedan, and that's on a per-mile basis. When you also factor in that many trucks cover 100K miles per year, vs, say, 15K for a car, that means the trucker's share is 64,000 times that of the car owner.
This. One of the best things about driving a Tesla is realizing that I'm no longer helping subsidize the trucking industry's massive externality. We should be massively expanding our country's (electric) rail network, not using trucks (diesel or otherwise) to deliver everything.
If diesel fuel were charged at a fair tax (about $4,500 per gallon of diesel fuel), we would have the best road and rail system in the country. Heck, even if we charged trucking companies only 0.1% of their fair share (about 7x what they pay now), that alone would make a huge difference in improving the state of our country's failing infrastructure. Just saying.
"But Tesla has ELECTRIC cars!" Toyota unveiled three new electric cars last year and has seven more scheduled for release within 24 months. Not "pay us $40,000 today and maybe in three or four years we'll figure out how to build a car". And that's a relatively small project for Toyota - they aren't betting the company on the next model.
Did I miss something? A far as I can tell, Toyota has only built one electric car in their entire history — the Rav4 EV — and that is discontinued because almost nobody could deal with an EV that had only a 110-mile range.
Fuel cell vehicles are not real EVs. The biggest trick the automakers managed to pull on California's government was tricking them into giving carpool stickers to fuel cell vehicles. A fuel cell is not green. Most hydrogen comes from splitting natural gas, which is far from clean energy; it is dramatically less green than even the average for California electrical power. And even with huge subsidies, hydrogen still costs anywhere from 4x to 15x as much per mile as EVs. Fuel cell vehicle manufacturers are parasites on the clean energy bandwagon.
Similarly, plug-in hybrids are also not real EVs. Yes, you can at least use them as EVs, but only over a fairly short range, which is typically less than the average daily commute. So most drivers end up using at least some gasoline.
Basically, Toyota pays lip service to EVs while basically doing everything they can to avoid actually building one. I don't expect them to pull their heads out of their backsides any time soon. And Tesla told them that they needed to offer a model with a bigger battery, but Toyota execs thought nobody would buy it. And here we are, with people buying the Model X for four times as much money. And every single Model X sale was a lost Rav4 EV sale that Toyota could have easily had, if their management had not been completely and totally oblivious to the needs of their consumers.
That's why Tesla's stock is so valuable. It isn't an indication that the company is going to rapidly grow to be bigger than Toyota. Rather, it's a bet that over the long term, a company managed by people who actually understand technology and are looking forwards will do better than a company managed by people who don't understand technology and are always looking backwards.
Even without the huge subsidies, the cost savings are significant over the life of the vehicle, in part because gasoline is taxed so heavily. That's why places with a higher-than-average median income — places where a majority of people can afford new cars (the used EV market is basically nonexistent) — also have much higher EV sales than places with lower median income. Compare that EV sales in the U.S. as a whole (1.2%) to EV sales in San Jose (13%).
With Europe's exorbitant gas taxes, it is no surprise that the effect of these cost savings is magnified, as the savings themselves are magnified.
Conductor ampacity, resistance, and maximum voltage would be simple starting points.
Do not want. Putting additional intelligence in cables just increases the odds of the cable failing because some unnecessary chip decides to stop working. And a couple of those don't even make sense:
The maximum voltage for a USB cable is limited only by the distance between pins (arcing), which is defined by the shape and size of the connector itself, making that number entirely moot, barring something really bonkers.
The resistance should be approximately zero (at least too small to easily measure). Otherwise, there's something very wrong with the cable.
What you really care about is how much current the cable can carry before the voltage starts to sag, which depends mainly on the conductor size. There's a very easy way to find that out without the cable needing to be smart: ramp up the current until the voltage sags beyond reasonable tolerances, then back off the current. This small amount of extra intelligence on the device end, purchased once, eliminates the need for extra intelligence on cables, which you buy dozens of.
Furthermore, the cable's notion of how much current it can carry is, in practice, completely useless. When a cable starts to fail, it usually fails slowly, with individual wire strands breaking. This can create a hot spot in the cable or connector that, when driven at full amperage, could potentially cause a fire. The cable has no way of determining whether such damage has occurred. But with proper voltage drop detection at the endpoints, properly designed hardware actually *can* detect that type of failure and reduce the amperage to a safe level.
Slashdot Mirror
The ups and downs also depend on which party holds Congress. It tends to be whichever party doesn't hold the White House, more often than not.
The Constitution makes the division of power clear. Congress controls the purse strings, not the president. You can blame the president all you want for saying what the government's priorities should be, but ultimately it is Congress that is responsible for deciding how money is apportioned and how much tax is collected. The only thing the president can do is veto it, and only if Congress can't get a two-thirds majority to agree that it is a good idea.
In other words, President Trump's temper tantrum over the border wall can go away tomorrow if Congress would grow some balls and work across the aisles to present a budget approved by a veto-proof majority.
No, but pretty much every mobile home on the planet has a utility room, and those are pretty much all built after the 1950s. And lots of folks do things like add a screen room over the back deck or whatever. The same approach works with a wide range of styles.
Not trolling. Sarcastic, but not trolling. The President of the United States controls neither spending nor taxation. That's Congress's job. What you're actually saying is that the (almost always Republican-controlled) Congress ran up more national debt under Obama than under all other presidents before him combined.
As a rule, Democrats raise taxes to cover their spending, while Republicans borrow money to do so. There are exceptions, but they are rare. Thus, the national debt is mostly a Republican-caused problem. Tax cuts (the heart of the Republican platform) inherently increase the national debt. Anybody who says otherwise is kidding him/herself. It's like selling products at a loss, but making it up in volume.
When the Chinese foreclose on all that national debt the Republicans have been running up, you might wish you had done so. :-D
Alternatively, that's what utility rooms (or enclosed porches) are for. You put a lock between that room and the rest of the house, and if you know you're going to have a delivery when you can't be there, you provide them with a temporary code for the outer lock, and lock the door between the utility room (or porch) and the rest of the house.
That's arguable. Dropbox has a system designed specifically to provide access to specific users authorized to access a given file, and requires the other party to know about the file, whereas iCloud is designed to automatically share files with any device tied to the account, complete with a mechanism for easily adding new devices to the account. So the question becomes one of which is more likely: Dropbox getting their ACL programming wrong or Apple having a bug that allows injecting a new device into an iCloud account.
I'm not saying you're wrong, because it's really anybody's guess which is more likely. Either way, I don't think it is necessarily safe to assume that a site designed for secure public sharing is more likely to leak data than a site designed for secure inter-device sharing within a single account. I actually suspect that the odds are about the same.
Particularly now. I can't help but find it ironic that Apple is lecturing the rest of us about privacy just a few short months after they moved all their Chinese users onto servers in mainland China — one of the most dubiously privacy-risking moves in the history of the industry.
And now, their sales in China are weak. Imagine that. It's almost like after years of Apple using privacy as a selling point, their users actually care about privacy.
Just saying.
Kidney stones can be caused by lack of water, but once you have them, AFAIK, water won't do anything to dissolve them. And although water is preventative for gout, I don't think it is curative (but I could be wrong). Both conditions, however, can be improved with citrus juice.
Completely untrue. Many cancers are curable with extremely high success rates if caught early, including acute lymphoblastic leukemia (98%) and localized (non-metastatic) prostate cancer (~100%).
And even stage 4 metastatic cancer is sometimes cured, thanks to immunotherapy.
The only thing absolutely certain is that water has a 0% cure rate for any illness other than dehydration, and always will.
FTFY.
If a vehicle gets damaged between the assembly line and the customer, short of someone wrecking it, that's a design flaw. Vehicles should not be that fragile.
Ironically, this is a problem that automation is likely to make worse. If a part is defectively weak by design, the rate of early breakage during assembly will be much higher with human assembly than with automated assembly, because humans are less careful. This can make design flaws in parts obvious earlier, which increases the chances of the engineering team being called in to fix the design before a million of them get distributed to customers.
Caveat emptor: If you go with Comcast Business and require static IPs, they won't let you own your own modem. You have to rent theirs for an extra fee (which is not included in the static IP fee).
Citation needed.
As far as I'm aware, the only throttling Netflix has ever done has been in the form of providing lower-bitrate streams to pay-by-the-byte cellular customers, and they have never blamed the ISPs for this money-saving feature.
The agreements would exist, but the total traffic (and thus, the total cost of those agreements) would be much, much less.
The reason the ISP has to do it is because they are a monopoly, and their only viable legal alternative is to provide enough bandwidth to their upstream providers so that their in-house video-on-demand services are not guilty of unfair competition against external video-on-demand services like Netflix et al. And the answer to the question of how many other services is "all of them" with the sole exceptions being external service that are so small that it is cheaper to provide enough bandwidth instead.
Or if they would prefer, Comcast could shut down all of their cable TV and video-on-demand services and become a pure ISP.
Those are the options.
Netflix offered free caching servers
Which require rack-space and electricity. Do those costs offset the bandwidth savings?
For an ISP the size of Comcast and a content provide the size of Netflix, it is cheaper by nearly four orders of magnitude. Netflix actually publishes information about their caching servers online, though I can't find the power specifications, so I can only take a rough guess based on the typical power consumption of blade servers in that form factor.
I would not expect a 2U rack unit to draw much more than 1 kW of power, which at 8 cents per kWH costs about $700 per year to keep it powered. Each box can provide 10 GB per second of throughput, which is enough to handle 26,000 streams at 3 Mbps (this is an arbitrary example), or 2.6 cents per continuous stream per year. By contrast, a single upstream T3 trunk line costs about $36,000 per year, and can provide enough bandwidth for about ten streams at 3 Mbps, or $1,200 per continuous stream per year.
If an ISP gives such favorable treatment to Netflix do they have to give the same treatment to any streaming service/information service provider?
No, most streaming providers partner with companies like Akamai to do the same thing until they get big enough to warrant their own servers. And when they do, the ISP would be utterly stupid to not agree to a similar arrangement—utterly stupid almost beyond the limits of human imagination.
A few key points should explain it:
In short, the caching approach represents huge savings for Comcast, and gives them a cost split that is very much in their favor compared with the 50-50 cost split that would occur if they had to pay for bandwidth to the backbone to handle that traffic. The only alternatives are for Comcast to either pay more by using bandwidth and pay a larger percentage of the total cost or to throttle Netflix in an anticompetitive manner and get sued.
Netflix does no such thing. They provide higher-quality stream options, and the ISP's customers either care enough to demand that the ISP improve its bandwidth or they don't.
No, the government should force the ISP to eat those costs because Comcast is a monopoly that is behaving anticompetitively.
There has to be something very wrong with either your numbers or your approach. The damage per mile for an 18-wheeler is, assuming it is fully loaded down, ~10,000 times that of a passenger car, and you're advocating raising the gas tax to only a factor of 10 times what cars pay.
Diesel trucks use somewhere on the order of 4x as much fuel per mile. Divide that 10k by 10 and you have 1k, and divide it by 4, and you get 250. With your approach, trucks are paying on the order of 1/250th of their fair share.
For your numbers to be reasonable, diesel trucks would have to burn 1000x as much fuel per mile as passenger cars, which would require trucks with gas tanks approximately as big as a thirty-foot swimming pool. Clearly, this is not the case.
The biggest problem I see is adding the cost of equipment in, as though we should treat that as somehow paying for the cost of roadways. That causes you to be off by one entire order of magnitude compared with the fuel costs, which are only a small part of the cost of trucking. However, your 54B number is also problematic. The industry as a whole used 54B gallons of fuel, but only used 39B gallons of that were diesel fuel, which is to say that the 54B gallon number includes a very large number of vehicles that aren't even big enough to use diesel. If you only want to look at big rigs, the number is likely single-digit billions, and given that smaller trucks do orders of magnitude less damage, we really have to focus on big rigs here.
The only reasonable way to tax this would be to take (at least) the road repair portion of the gasoline tax, multiply it times the weight multiplier (9,600 or whatever), and divide it by some factor based on the vehicle's miles per gallon (4-ish) compared with a normal passenger vehicle.
To be fair, only about two-thirds of state gas taxes go to pay for road maintenance, but the rest are usually spent for things like public transit infrastructure, and it seems reasonable to pass that on to the trucking industry as well, to pay for rail infrastructure expansion and improvements that would dramatically improve long-haul transportation efficiency when compared with trucking.
Bearing in mind that light duty trucks do not fill up at the same gas stations as tractor trailers (they won't fit under the canopy at a normal gas station), it seems entirely plausible that we could tax their fuel differently.
An alternative approach would be to dramatically increase the heavy vehicle use tax, basing it on mileage. Currently, this tax is capped at only $550 per vehicle per year, and only for vehicles over 55,000 pounds. If we lowered the limit to 10,000 pounds (about twice the heaviest passenger vehicles on the road) and taxed it at a rate of m[*d*((a/5000)^4) - b] where m is the mileage, a is the axle weight, computed as the taxable weight divided by the number of axles, b is the road damage estimate for a two-axle vehicle that weighs 10,000 pounds, and d is twice the baseline damage caused by an average passenger vehicle, you'd be in the ballpark.
In a a state with 35.6 cents per gallon state tax plus 24.4 cents per gallon federal tax, with 60% of the state tax for road repair, we get 21.36 + 24.4 cents, or 45.76 cents per gallon going to road repairs, divided by 24.7 miles per gallon on average, or 1.85 cents per mile for an average passenger vehicle. That's b, and the average vehicle weight for those vehicles is about 4,000 pounds.
Road damage for a 10k pound vehicle is (2.5^4) times that, which is about 39x, so 72.15 cents per mile. That's d.
So for a five-axle vehicle that travels 5,000 miles and weighs 80,000 pounds (16,000 per axle), m[*d*((w/10000)^4) - b] is 5000[(.7215 *(3.2^4)) - .0185], or about $378k.
I'll let you ponder the degree of an externality so massive that a truck can do $378k in damage and pay only $550 in heavy vehicle use taxes plus about $370 in gas taxes to cover that damage.
Every time I see these 'net neutrality' things all I can think about is Idiocracy and the 'its got electrolytes' bit, and the NN version of the same is 'muh bandwidth wuz stranglified'.
Competition in a zero sum environment:
The total bandwidth available is not short time frame elastic, its completely static. So an example for sake of argument under the imposed 'bad' NN rules the 'muh bandwidth wuz stranglified' people keep pushing for: Netflix in North America uses 50%(whatever the real number is doesn't matter) of available bandwidth and pays the same as everyone else under that rule, ISP/trunking/peering companies are unable to charge them more by the imposed rules. Along comes SUPERNetflix with double goodness and 4X the bandwidth use which everyone starts using because double is mo gooderer, and now they use say 99% of the available bandwidth and the same ISP/trunking/peering companies (yes these numbers are exaggerated) are unable to charge more or negotiate a throttling plan due to the imposed rules.
The net result is all internet traffic is essentially throttled down to a max of 1% and the ISP is handcuffed, unable to charge for fair use, or throttle in the best interest of its customers.
Actually, you're wrong on pretty much every count here. For any sufficiently large ISP, Netflix will *give* the ISP a caching box for their data center that will take the vast majority of Netflix traffic entirely off of the upstream pipes. The only cost to them, other than the electricity to power the box and the cost of square footage to house it, is the cost of upgrading the infrastructure from the central office to the customer, which is usually a matter of upgrading the equipment at both ends. And given how often customer premises equipment fails and has to be replaced, it is usually just a matter of upgrading the equipment at the head end. In other words, it is highly short-term elastic.
And even if an ISP doesn't have that arrangement, nobody runs only a single fiber anywhere, so in practice, there is *always* extra fiber capacity available for upgrading the connection between two ISPs.
So the only way you would see the problems you're describing is if an ISP like Comcast gets too big for its britches and insists that Netflix pays a monthly fee to put that caching box in their data center, and then refuses to upgrade the peering point between the ISP's network and Netflix's ISP's network. Which it did. And their mutual customers got screwed, mainly because Comcast thought that throttling Netflix (and only Netflix) would cause more people to buy video-on-demand content from Comcast. And obviously it worked, or else they would have quickly discontinued that experiment.
Short of some sort of regulatory penalty against Comcast and similar ISPs when they pull these shenanigans, nothing will change. Net neutrality was an attempt to use the FCC's regulatory powers to do so. Unfortunately, because their interpretation of the law was purely a regulatory position rather than an explicit piece of legislation, it was subject to the whims of a given administration, and went away when the administration did.
They did the same thing for telephone. But you're badly mistaken if you believe that removing monopolies will result in competition. Countless towns have tried that over the years, and essentially 100% of the time, they were back to a monopoly within about five years.
Why is this the case? Because wire infrastructure is something economists call a natural monopoly — a product or service in which the startup costs are so high and the payoff over such a long period of time that competition is infeasible.
When a new company comes into a territory with an existing cable company, the new company has to have some way of inducing people to switch — either by lowering their cost or by providing better service. Either way, it is very easy for the existing company to match their offer, because they have paid for their infrastructure, and almost all of their income is profit. So the existing company invariably either matches or undercuts the newcomer and provides enough improvements to their service so that the newcomer cannot steal enough subscribers to cover the payments on their construction costs plus operating expenses.
What happens then? Unless the local government decides to bail out the newcomer to keep competition going, the newcomer typically sells the infrastructure to the older cable company for little more than what they still owe on it, thus giving the incumbent a set of new replacement lines at significantly below cost. And you're back to a monopoly.
That's why, with the possible exception of extremely high-density areas where a new company can get by on a tiny percentage of residents/businesses, the only places where competition has ever really succeeded have been places that have had two or more competing companies since the dawn of cable TV, and even those usually break down to a monopoly eventually.
This is a problem that cannot be solved no matter what you do. Wire infrastructure is simply too expensive to allow for competition in practice. Heck, wireless infrastructure is almost too expensive in most places, forget wired.
The only way to get real competition in Internet service is to separate the wire provider from the actual Internet service provider. In practice, this can happen in one of two ways, both of which involve the government:
Either way, the result is the same: You have one company or organization providing wire service for multiple ISPs. When you do this, competition is possible at the ISP level. In practice, though, the second approach (government-built infrastructure) tends to work better in the long run, for two reasons:
Speeds to speed test servers are increasing faster after NN is overturned, but those have little relation to actual Internet speeds. They represent best-case speed, rather than typical speed, because no ISP would be stupid enough to throttle connections to a speed test server. But actual average speeds may or may not be increasing any faster than they were before.
Also, by focusing on average speed, you're missing the whole point of net neutrality. It isn't about the average. It's about the worst case. It's about ensuring that ISPs aren't extorting companies who aren't their customers, about ensuring that ISPs don't artificially degrade performance on specific services like video-on-demand or VoIP to drive customers to their own competing services, and so on. That can't be measured using average bandwidth. At all.
To use a car analogy, saying that average Internet speeds increased after the repeal of NN is roughly like saying speed limits increased after overturning a law that prevents illegal speed traps.
This. One of the best things about driving a Tesla is realizing that I'm no longer helping subsidize the trucking industry's massive externality. We should be massively expanding our country's (electric) rail network, not using trucks (diesel or otherwise) to deliver everything.
If diesel fuel were charged at a fair tax (about $4,500 per gallon of diesel fuel), we would have the best road and rail system in the country. Heck, even if we charged trucking companies only 0.1% of their fair share (about 7x what they pay now), that alone would make a huge difference in improving the state of our country's failing infrastructure. Just saying.
Did I miss something? A far as I can tell, Toyota has only built one electric car in their entire history — the Rav4 EV — and that is discontinued because almost nobody could deal with an EV that had only a 110-mile range.
Fuel cell vehicles are not real EVs. The biggest trick the automakers managed to pull on California's government was tricking them into giving carpool stickers to fuel cell vehicles. A fuel cell is not green. Most hydrogen comes from splitting natural gas, which is far from clean energy; it is dramatically less green than even the average for California electrical power. And even with huge subsidies, hydrogen still costs anywhere from 4x to 15x as much per mile as EVs. Fuel cell vehicle manufacturers are parasites on the clean energy bandwagon.
Similarly, plug-in hybrids are also not real EVs. Yes, you can at least use them as EVs, but only over a fairly short range, which is typically less than the average daily commute. So most drivers end up using at least some gasoline.
Basically, Toyota pays lip service to EVs while basically doing everything they can to avoid actually building one. I don't expect them to pull their heads out of their backsides any time soon. And Tesla told them that they needed to offer a model with a bigger battery, but Toyota execs thought nobody would buy it. And here we are, with people buying the Model X for four times as much money. And every single Model X sale was a lost Rav4 EV sale that Toyota could have easily had, if their management had not been completely and totally oblivious to the needs of their consumers.
That's why Tesla's stock is so valuable. It isn't an indication that the company is going to rapidly grow to be bigger than Toyota. Rather, it's a bet that over the long term, a company managed by people who actually understand technology and are looking forwards will do better than a company managed by people who don't understand technology and are always looking backwards.
Even without the huge subsidies, the cost savings are significant over the life of the vehicle, in part because gasoline is taxed so heavily. That's why places with a higher-than-average median income — places where a majority of people can afford new cars (the used EV market is basically nonexistent) — also have much higher EV sales than places with lower median income. Compare that EV sales in the U.S. as a whole (1.2%) to EV sales in San Jose (13%).
With Europe's exorbitant gas taxes, it is no surprise that the effect of these cost savings is magnified, as the savings themselves are magnified.
Do not want. Putting additional intelligence in cables just increases the odds of the cable failing because some unnecessary chip decides to stop working. And a couple of those don't even make sense:
What you really care about is how much current the cable can carry before the voltage starts to sag, which depends mainly on the conductor size. There's a very easy way to find that out without the cable needing to be smart: ramp up the current until the voltage sags beyond reasonable tolerances, then back off the current. This small amount of extra intelligence on the device end, purchased once, eliminates the need for extra intelligence on cables, which you buy dozens of.
Furthermore, the cable's notion of how much current it can carry is, in practice, completely useless. When a cable starts to fail, it usually fails slowly, with individual wire strands breaking. This can create a hot spot in the cable or connector that, when driven at full amperage, could potentially cause a fire. The cable has no way of determining whether such damage has occurred. But with proper voltage drop detection at the endpoints, properly designed hardware actually *can* detect that type of failure and reduce the amperage to a safe level.
Nope. I could spend zero for food. I'd have to learn to spear fish, but it is possible.
Heh. They also want $49 for the "free" repair. F*** Apple.