I'd be okay with "usage plans" if I got credited back for the data I didn't use. If I use 200 of my 250 plan this month, then use 300 next month, I shouldn't be charged an overage.
Be careful what you wish for. There's no free lunch here. These data caps aren't arbitrary. They're what you get from:
(amount of bandwidth ISP has to pay for) = (average customer bandwidth) * (number of customers)
(average cost per customer) = (cost of ISP's bandwidth) / (number of customers) + (profit margin)
(customer's monthly quota) = (whatever number keeps their average bandwidth at the above value)
There's a large psychological element in that last one. So if you increase the quota window to 2 months (let you carry over unused quota for one month) and it causes average use to increase from (say) 200 GB/mo to 220 GB/mo (because people don't feel as constrained by the cap), then the ISP has to respond by increasing the plan prices by 10%. If everyone gets too demanding about retaining unused quota, the plan price will have to be increased to the point where the only way to "get your money's worth" each month is to exactly use up your quota.
Of course the best way is with metered usage (pay per GB you use in a month). But for some reason everyone seems to be totally against that even though that's the method that makes the most sense. I don't really sympathize with ISPs and cellular carriers about it though. They brought this upon themselves by allowing their marketing department to sell "unlimited" plans to the public.
Despite being somewhat anti-Apple, I've always been on Apple's side when stories about "it's" human rights violations come up, specifically for the reasons OP cites. I would never mod it down simply because it favors Apple, and if it gets out of hand I've written posts defending Apple as OP did.
That said, I'm less inclined to go around correcting people when it's brought up in web forums because the general media is biased wildly in Apple's favor. e.g. Did you know one of the main patents in Apple's $1 billion award against Samsung was ruled invalid by the USPTO this summer? Nary a word about it in the general media (I only found out about it by accident a few weeks ago when the story came up while I was searching for who was suing whom for patent infringement). So I figure some biased bad PR is necessary to help balance out the biased good PR Apple normally gets.
Robocalls from political campaigns are exempted too. My parents made a largish political donation a few years back, which apparently put them on some sort of list. They were harassed by robocalls asking for more political donations for over a year. It was so bad that when I visited a few months before the election, they didn't even bother answering the phone anymore, letting it go straight to voicemail and answering only if it was someone they knew leaving a message. They got 5-10 of these calls every night while I was there. When they moved and I asked if they wanted to keep their old phone number, their answer was hell no.
10 years ago I thought light field sensors were the future of photography. But in that 10 years, processors (especially GPUs) and camera sensors have advanced so quickly. It's now a race between light field sensors (which are like recording the information a hologram records), versus simply mounting 2+ cameras which take pictures simultaneously and using image processing algorithms to extract the depth info from those pictures instead of recording it directly.
1. Log every website any of your customers visits and store it for a year.
2. We're not going to tell you how. That's your problem, but if you can't figure out a way we'll probably fine you. No, we're not excluding SSL.
3. You are paying for it too. Just pass the costs on to your customers or something.
All you need to prevent this type of idiocy is a law that requires:
4. The politicians who pass this law will be the first ones monitored as the law requires, and the results of said monitoring will be freely available for the public to examine.
If there was an explosion inside the cabin or luggage compartment, there will be internal paneling, structural members, etc., blackened and bent and peppered with explosive ejecta littering the deserts of the Sinai.
That's actually part of the difficulty. The parts which clearly indicate it was a bomb because of blackened and bent pieces are scattered around the crash site, impossible to tell where they originally came from. So there's usually no direct evidence of a bomb.
What you end up doing instead is looking for areas of missing pieces from other mundane pieces of debris. In Pan Am 103, they conducted tests using mock bombs stored in retired fuselages. The baggage containers adjacent to the bomb survived like this.. The suspect baggage container in Pan Am 103 was shredded like this. Which was their basis for concluding the bomb was inside that baggage container. A falling object only has so much energy once it hits terminal velocity, and that energy is only sufficient to deform or tear metal by a certain amount. So the pieces end up a certain size on average (that's why the debris from USAir 427 and American 77 were virtually unrecognizable as a plane - they struck head-on at such a high velocity their high kinetic energy went into tearing the metal into tiny pieces).
The evidence which led to the conclusion the bomb on Pan Am 103 was in a radio were these pieces of circuit board (overlayed on top of a pristine board in this first pic). An explosive decompression and free fall from altitude would not have fragmented the circuit board in that manner. An explosion very close to the board had to have done it. The police and investigators at Lockerbie did an impressive job of evidence collection to find these pieces. As Kogalymavia Flight 9268 crashed in a desert, they should be able to recover even small pieces like this and we'll eventually know if it was a bomb.
It's the fact that (according to what we've heard in the news) the plane broke up almost exactly when it reached a certain altitude. Why is this important? Because the single most effective way to trigger a bomb in an airliner is to tie it to an altimeter, so that the bomb will only go off once it's taken above the set height.
In this case, that's actually evidence against an altitude-triggered bomb. The flight was last tracked on radar at 31,000 ft. Far, far above the internal pressurization, which is usually equivalent to 7000-10,000 ft. So unless the bomb were somehow attached to the outside of the aircraft, it should have blown somewhere around 10,000 ft max if it were triggered by a pressure altimeter.
It's worth noting that structural failures also occur the exact moment a plane reaches a certain altitude. As the pressure differential between inside and outside increases, the stresses in the structure increase. If the structure is somehow weakened, it will give way once a certain pressure threshold is exceeded, which usually happens upon climbing in altitude.
You're missing the big picture. People are speculating the next big convergence in electronics will be between TV streaming/tuning hardware and game consoles. I wouldn't have called it a few years ago, but I'm starting to agree with them. The biggest capital expense for a gamer is purchasing the 3D hardware. With bandwidth increasing and costs coming down, streamed 3D games are becoming more feasible, where the hardware and its costs for many people are shared thus lowering overall costs.
Your Lamborghini analogy is completely off the mark because there's a huge difference between a car and a plane, and it would require a huge engineering investment to redesign a car to fly. Right now the only difference between a TV streaming unit and a game console is the 3D hardware, and that extra hardware becomes irrelevant if you can stream the game (as Sony's smart TVs and bluray players already let you do with PS3 games). The marginal cost to add gaming functionality to modern streaming video units is minuscule. Replace "Apple TV" with cell phones, and "games" with PDA functionality and think about it again. That was the last big convergence - phones and PDAs converged to create today's smartphones.
See, they don't drive the car, and measure the mileage you get. As I understand it, they hook it up to a test rig, do some tests, and then calculate the mileage.
The EPA MPG ratings are not intended to predict how many MPG you will get driving the car. There's just too much variability in driving styles, road conditions, and regional climate for any single number to be an accurate prediction.
The MPG rantings are intended to allow you to compare different cars to each other when shopping. If one car is rated at 25 MPG and another at 30 MPG, you may not get exactly 30 MPG with the second car. But you will most likely get about 20% better MPG with it than the first car.
To make it conceptually easier to grasp, the EPA tries to scale the figure so it's approximately equal to the MPG you'll get while driving. But this is so widely misunderstood by the public at this point that I think the EPA would be better off just changing to an abstracted scale where 100 represents the amount of fuel burned by the average sedan, higher numbers represent more fuel burned, lower numbers less.
MPG is already screwed up since it's the inverse of fuel consumption. Going from 25 MPG to 50 MPG saves only half as much fuel as going from 12.5 MPG to 25 MPG, not twice as much as the size of the numbers would suggest. This has led to all sorts of other harmful misconceptions among people, like ridiculing hybrid engines in SUVs when that's precisely the best place to put them if you want to maximize fuel savings. Putting hybrids in economy cars actually represents the smallest possible fuel savings, and quite frankly is a waste of money (if a hybrid gives 40% better MPG and the cost is an extra $4000, payback time is about 14 years on a Prius, but only 6 years on a SUV). It also means the average MPG for two vehicles is not (MPG1 + MPG2)/2. It's 1/MPGavg = 1/MPG1 + 1/MPG2. So if you're trying to save fuel by splitting your driving between a SUV and a hybrid, the SUV disproportionately affects your average MPG, and the hybrid makes little difference from a non-hybrid economy car. Changing the scale would be a good opportunity to correct this numerical distortion. The rest of the world uses liters per 100 km specifically to avoid this problem.
In other words they see that in their industry I am, and will remain, what is called a "deadbeat." This does not mean a non-payer, this means someone who PAYS and doesn't carry balances, thereby denying the banks the opportunity to collect interest at extortionate interest rates.
Is there something wrong with this whole system?
There is nothing wrong with the system. You're assuming the point of the bank (or credit card) is to offer you a service. It's not. The bank/credit card is lending money for the purpose of making more money (collecting interest). Nothing more. If they can't collect as much interest from you, in their eyes that makes you a worse investment than someone who carries a balance and pays interest.
If you have substantial savings and your financial habits are good enough to maintain a 800+ credit score, then what's the problem? You don't need to borrow money. Use your savings.
Or alternately, do what the banks and credit cards do - leverage your savings to generate income. Having zero debt is actually not the optimal fiscal situation. It's the safest, but it's not optimal. Basically by putting your money in a savings account, you're giving the bank a cheap loan, and the bank is leveraging your money to make more money (by doing things like lending it out to people who carry balances on their credit cards). If you have a savings, no debt, and excellent credit, you should be able to borrow money very cheaply, and invest it in something which gives a greater return than the interest you'll have to pay, and certainly a greater return than the interest on a savings account.
In order to get the latest news from Japan, several stations in the U.S. were carrying live video feeds from Japan. I was expecting awesome interactive 3D computer graphics using green screens to create a pseudo-holographic experience. Instead the weather report was a cloth map of Japan with felt cutouts of clouds, the sun, and numbers for the temperature velcro'ed on. The weatherman (woman) pointed to these using a pointing stick (hadn't seen one of those since the 1990s when they started being replaced by laser pointers). When covering the Fukushima accident, they'd gone to the trouble of recreating the entire facility in model scale using painted cardboard. It was damn good, would've made any model railroader proud, but was such a throwback to the era before computer graphics.
That's actually why I'm a bit skeptical of the EPA claims in this case. The EPA cites only the 2014 Touareg, 2015 Cayenne, and various 2016 Audi models for having the defeat device. But all three of those vehicle lines sharing the same engine all three years. If you found it on those specific vehicles, you'd expect to find it on all those vehicles for all three years because it's the same engine.
I'm inclined to believe VW on this one. The EPA letter narrows the problem down to an odd mix of vehicles. The 2014 Touareg, the 2015 Cayenne, and various 2016 Audi models. This despite all these vehicles sharing the same 3.0 liter engine each year. If the 2014 Touareg had the defeat device, you'd expect the 2014 Porsche and Audi vehicles to also have it. You'd also expect the 2015 Touareg to have it since it uses the same engine as the 2014 model, but for some reason only the Porsche is flagged as having it for 2015. Also, unlike the 2.0 liter TDI engine scandal, the 3.0 liter engines all use urea injection systems to reduce NOx emissions.
Part of the problem is the unlicensed bands need to be coordinated worldwide, since people will inevitably carry such devices from one country to another.
As with DOS, IBM contracted with Microsoft to develop OS/2, so it actually started off as Microsoft's code. Both partners were on board for a while and told the entire industry DOS (a CLI running programs in real mode) was going to be replaced by OS/2 (a GUI though it could run DOS in a window using the 80286's protected mode, so a crashed DOS app wouldn't hang the entire computer), so a lot of companies began porting their software over to OS/2.
Then there was some sort of falling out. Most people point the finger at Microsoft because Microsoft stopped talking about OS/2 and started talking about Windows (a GUI which ran on DOS). That's why early versions of Windows had a terrible reputation for crashing - because it ran on top of DOS, any app which crashed could hang the entire OS. You would typically have to reboot the computer 2-3 times a day, losing all your unsaved work each tine. This remained true all the way up to Windows ME. By Windows 3.0, Microsoft was completely pushing Windows and not OS/2. IBM wasn't happy but the contract called for Microsoft to do OS/2 development, so IBM couldn't take over or hire another company to do it. They negotiated to take over the project from Microsoft.
Because the then-current version of Windows was built with the same GUI elements IBM had established after a lot of usability R&D (IBM's Common User Access), IBM also got rights to Windows 3.x's code. That's what allowed them to include a copy of Windows in OS/2 3.0. They just modified Windows to run on the version of DOS running in OS/2's DOS box, instead of MS-DOS. It's also one of the reasons Microsoft really pushed to release Windows 95 in 1995 (it was code-named Chicago, and the joke was that Gates made the official name Windows 95 instead of Windows 4.0 to force his developers to release it in 1995). The terms of the split only gave IBM access to Windows 3.x code, so it was important for Microsoft to push the next version of Windows out there. Starting with Win95, Microsoft played a cat and mouse game making changes to prevent it from running in an OS/2 DOS box (or on DR-DOS).
NT was a separate project within Microsoft (built like OS/2 to not run programs in x86 real mode like DOS did), which had been on the back burner. Once they formalized their split with IBM, they began working on it in earnest, merging its API with Windows 9x. There were earlier enterprise-only releases, but the first real commercial release was Windows 2000 (in parallel with Windows ME). It's the foundation for the current versions of Windows.
Incidentally, the switcheroo between OS/2 and Windows ended up really helping Microsoft. Both IBM and Microsoft stressed that OS/2 was the future, so a lot of software companies which dominated DOS (e.g. WordPerfect, Lotus) invested a lot in porting their software to OS/2. When Microsoft switched gears and told them to develop for Windows version instead, these companies felt they'd been lied to and balked. They either made a half-hearted attempt to port to Windows, or put it off for a year or more. That time gap is what allowed Microsoft to swoop in and take over the office suite market on Windows with Microsoft Office. And thus began the early calls for an anti-trust investigation of Microsoft (which really swung into high gear when Microsoft tried to take over the web browser market by including it in Windows for free, essentially destroying the profit model for any competitor making a browser).
OS/2 1.3 used to run a lot of ATMs, until machines running Windows began replacing them in the 2000s. Coincidentally, that's when a lot of hacks to steal money from the ATMs started.
I know lots of state IT workers (from the university system) and the universal refrain is that they don't even have budgets for the basics. This is a big departure from the right wing meme of government being awash in tax dollars and lavishly spending, and these aren't the stereotypical lazy worker types either.
If your metric for whether or not something is underfunded is if it could be improved by spending more money, then you fail at business. Everything can be improved by spending more money on it. The trick to succeeding at business (running efficiently) is to spend the minimum amount necessary to consistently complete jobs to a certain standard.
I think that a lot of the reality is that the money goes to outsourcing giants like HP, IBM, Accenture, etc. and it's wasted in the inefficiencies that this brings to light. I've been in lots of outsourced IT departments and do work for outsourcers. The problem with outsourcing is this -- the company doing the outsourcing is paying $X to maintain their own environment. To win the contract, the outsourcer has to come in at $X - $Y for the bid to be low enough to accept. (X - Y) has to be greater than their cost to make $Z off the deal, where $Z is positive margin.
You hand-waved the last part to make the math work out in your favor. The fixed cost (equipment, wages for salaried workers, etc) to the bidding company is $F. These are costs the company would be paying regardless of whether or not they get the contract. The variable cost is $V and represents costs specific to that one contract. The total cost for the bidding company is then ($V + $F/n) where n is the number of contracts jobs it has. (This is a bit simplified since n would be weighted depending on the size of each job, but you get the idea.)
If the cost of doing the job to the company doing the outsourcing is $X, then assuming similar costs $X = $V + $F. And $Z = $F - $F/n.
Outsourcing then makes financial sense when a company specializing in the outsourcing task is able to lower costs by using the same equipment for multiple jobs (whether separated by space or by time). This is in fact how most business works, not just IT. You don't keep $10,000 of plumbing equipment in your garage. When you have a plumbing problem, you just call a plumber and pay him $200. He then uses his $10,000 of plumbing equipment to fix your problem. You have outsourced the plumbing job.
In fact the whole of modern civilization is built upon this very premise. Instead of each person growing his own crops, raising his own livestock, cutting his own lumber, building his own house, shoeing his own horse, etc., all of these jobs have become specialized. Instead of each person needing to learn a thousand different things, they just become extremely proficient at one thing. Goods and services they are unable to make or complete because they are untrained, they simply trade with other people who are extremely proficient at those one things. It's a more efficient arrangement since everyone is extremely proficient, which is why our average per capita productivity is so much higher than in the settler days when each person needed to be somewhat proficient at those thousand things.
Where outsourcing becomes unnecessary is when the job at the one organization is big enough that n can in fact be set to 1. If a job requires one company and one full-time IT staff, then $F has surpassed the threshold where there's no longer any advantage to breaking it down to $F/n. Arguably, a government should always be in this position. While a single government office's IT needs may be small enough not to require a full-time dedicated IT employee, the sum total of all those offices should require enough employees that it makes sense to have a dedicated IT department which provides part-time services to all those offices.
The only exception then is temporary projects which require additional staff and equipment, but that staff and equipment becomes superfluous after the project is completed.
There are good quality affordable journals, run by professional societies or universities, which are an excellent alternative to Elsevier and other expensive for-profit journals. For the health of science, it is important that people choose to submit there.
Actually I'd say the Journal model of publishing is outdated and backwards in today's interconnected society. They made sense back in the day when both publishing and buying (shipping) published copies cost a substantial amount of money. It was fiscally prudent to filter out the papers before publishing to a few judged to be "worthy" in the field.
Nowadays, publishing (on the web) and shipping (over the internet) is for all practical purposes free. There is no longer any need to filter prior to publishing - filtering can happen after. Researchers should just "publish" their papers on their own or school's website. "Journals" need to be replaced by some sort of search engine which reviews new papers and ranks them according to "worthiness". Something like an IMDB of scientific papers, with a running "top 10 papers in the last month" list for each field with rankings based on reviews by researchers in the field. That top 10 list for each month could be archived as the equivalent of a "Journal", while the entire database and ranking is accessible and searchable as a replacement for browsing through the journal section of the library.
For brownie points, you could even make the searches customizable. So if you think Prof. Henry Higgins is a quack, you can specifically tag him as such in your account, and the ranking algorithm would exclude all his votes in the paper rankings that are displayed to you.
so why can't Antarctica actually be losing massive amounts of ice and the resulting removal of mass cause uplifting of the underlying rock?
Ice has a density of approx 0.91 g/cm^3. The Earth's crust has a density of approx 2.7 g/cm^3. So the crust weighs about 3x more per unit volume than ice.
You are hypothesizing that (in terms of net potential energy) the removal of 1 ton of ice lowering the top of the ice by a net 1 meter results in the uplifting of a net 3 tons of rock by more than 1 meter. Basic physics says what you posit can only occur in the presence of a (massive) external energy source which replenishes that lost potential energy and then some. I suppose that might be possible if there's a huge pressurized magma bubble underneath Antarctica exerting a huge amount of upward pressure, but the burden of proof would be upon you.
You don't need to double-test everything. When a manufacturer produces 1 million units of a product, they don't test every single one. They pull out about one unit per thousand or one per ten thousand and test the sample. Yes that means some bad units get through. But the sampling will give you a statistical idea what percentage of the units are bad. That gives you a basis for improving the system if the flaw rate is higher than you deem acceptable. And perhaps most relevant to this case, it'll let you see if the flaw rate suddenly spikes. Then you can go hunt down the part or person responsible for the new flaws and fix or replace them.
"But these are court cases. We need to be 100% sure if the person is guilty, so everything should be double or triple tested!" Nothing in life is 100% sure. We only try criminal cases based on guilt beyond a reasonable doubt. All this would do is change the predominant probability bound from "there is only a 1 in 40 million chance that the accused's DNA would've matched that found at the crime scene by random chance" to "there is only a 1 in 10,000 chance the DNA testing lab made an error."
Actually, I suspect that's the reason police and prosecutors haven't widely implemented random redundancy testing to detect lab error rates. The statistics behind DNA matches (usually 1 per million to several hundred million) has created a popular belief that DNA tests are infallible, which makes the prosecutor's job a lot easier if they can get a DNA match. The error rate for anything involving people tends to be around 1 in a thousand. If the public learned about that, it would destroy juries' faith in DNA tests and take away one of the prosecutors' best modern tools.
it's just unbelievable, how often flash needs to be updated. [...] how many bugs/security holes can one poece of software have?
Flash was never supposed to be the universal scripting language of the web. It was designed as an artist's tool for transmitting animated graphics over low-bandwidth connections. Consequently, security was a very low priority in its original design.
If you want to blame someone, blame the folks behind the HTML standard who dragged their feet for a decade on adding features to the standard that web designers wanted to use. Websites resorted to using Flash because there wasn't a way to do it with HTML. It was so bad, for a while there I was thinking maybe it would've been better if Microsoft had completely succeeded at gaining control of HTML and "extended" it with a bunch of proprietary features.
Well, to be fair, it's arguably not very useful on a 1280x800 screen these days, especially not at 10". I can run two very primitive apps next to one another... It's only recently that Android devices commonly have enough pixels to actually warrant the feature. Now that it's relevant, it's becoming more common.
My first laptop had a 640x480 monochrome 9.5" screen. I was very productive running multiple windowed apps on it.
Somehow, this guy named Jobs convinced a huge swath of the population that it was impossible to multitask on a small screen. Even Microsoft bought what he said, and it resulted in the widespread rebellion against Windows 8's Metro interface. It's not the screen size or resolution which makes it hard to multitask on phones and tablets. It's the clumsy navigation you get with a touch interface - too easy to tap the the wrong spot or to "click" when you meant to move. Jobs eliminated multitasking so that this drawback of touch interfaces wouldn't be as pronounced, in order to increase the chances of his devices (which only had a touch interface) being accepted.
It worked, but don't ever make the mistake of believing Job's marketing spin is the true explanation. Multitasking has nothing to do with screen size or resolution. Multitasking on a small low-res screen is absolutely not a problem if you have precise inputs like a keyboard, mouse, and even touchpoint/trackpad (the important part is that pointer movement is distinct from clicking).
Yeah, at my college, if the author of a textbook was at the school or a friend of someone at the school, the prof would just get permission to copy it (since the author still held the copyright, not the publishing company). They'd send a loose-leaf copy to the school copy center, and give the students the option of buying the textbook, or buying the spiral-bound copy for a few bucks (cost of making the copies). Great option for the students who were taking the course as a core requirement, not something related to their major.
Slashdot Mirror
Be careful what you wish for. There's no free lunch here. These data caps aren't arbitrary. They're what you get from:
(amount of bandwidth ISP has to pay for) = (average customer bandwidth) * (number of customers)
(average cost per customer) = (cost of ISP's bandwidth) / (number of customers) + (profit margin)
(customer's monthly quota) = (whatever number keeps their average bandwidth at the above value)
There's a large psychological element in that last one. So if you increase the quota window to 2 months (let you carry over unused quota for one month) and it causes average use to increase from (say) 200 GB/mo to 220 GB/mo (because people don't feel as constrained by the cap), then the ISP has to respond by increasing the plan prices by 10%. If everyone gets too demanding about retaining unused quota, the plan price will have to be increased to the point where the only way to "get your money's worth" each month is to exactly use up your quota.
Of course the best way is with metered usage (pay per GB you use in a month). But for some reason everyone seems to be totally against that even though that's the method that makes the most sense. I don't really sympathize with ISPs and cellular carriers about it though. They brought this upon themselves by allowing their marketing department to sell "unlimited" plans to the public.
Despite being somewhat anti-Apple, I've always been on Apple's side when stories about "it's" human rights violations come up, specifically for the reasons OP cites. I would never mod it down simply because it favors Apple, and if it gets out of hand I've written posts defending Apple as OP did.
That said, I'm less inclined to go around correcting people when it's brought up in web forums because the general media is biased wildly in Apple's favor. e.g. Did you know one of the main patents in Apple's $1 billion award against Samsung was ruled invalid by the USPTO this summer? Nary a word about it in the general media (I only found out about it by accident a few weeks ago when the story came up while I was searching for who was suing whom for patent infringement). So I figure some biased bad PR is necessary to help balance out the biased good PR Apple normally gets.
Robocalls from political campaigns are exempted too. My parents made a largish political donation a few years back, which apparently put them on some sort of list. They were harassed by robocalls asking for more political donations for over a year. It was so bad that when I visited a few months before the election, they didn't even bother answering the phone anymore, letting it go straight to voicemail and answering only if it was someone they knew leaving a message. They got 5-10 of these calls every night while I was there. When they moved and I asked if they wanted to keep their old phone number, their answer was hell no.
10 years ago I thought light field sensors were the future of photography. But in that 10 years, processors (especially GPUs) and camera sensors have advanced so quickly. It's now a race between light field sensors (which are like recording the information a hologram records), versus simply mounting 2+ cameras which take pictures simultaneously and using image processing algorithms to extract the depth info from those pictures instead of recording it directly.
All you need to prevent this type of idiocy is a law that requires:
4. The politicians who pass this law will be the first ones monitored as the law requires, and the results of said monitoring will be freely available for the public to examine.
That's actually part of the difficulty. The parts which clearly indicate it was a bomb because of blackened and bent pieces are scattered around the crash site, impossible to tell where they originally came from. So there's usually no direct evidence of a bomb.
What you end up doing instead is looking for areas of missing pieces from other mundane pieces of debris. In Pan Am 103, they conducted tests using mock bombs stored in retired fuselages. The baggage containers adjacent to the bomb survived like this.. The suspect baggage container in Pan Am 103 was shredded like this. Which was their basis for concluding the bomb was inside that baggage container. A falling object only has so much energy once it hits terminal velocity, and that energy is only sufficient to deform or tear metal by a certain amount. So the pieces end up a certain size on average (that's why the debris from USAir 427 and American 77 were virtually unrecognizable as a plane - they struck head-on at such a high velocity their high kinetic energy went into tearing the metal into tiny pieces).
The evidence which led to the conclusion the bomb on Pan Am 103 was in a radio were these pieces of circuit board (overlayed on top of a pristine board in this first pic). An explosive decompression and free fall from altitude would not have fragmented the circuit board in that manner. An explosion very close to the board had to have done it. The police and investigators at Lockerbie did an impressive job of evidence collection to find these pieces. As Kogalymavia Flight 9268 crashed in a desert, they should be able to recover even small pieces like this and we'll eventually know if it was a bomb.
In this case, that's actually evidence against an altitude-triggered bomb. The flight was last tracked on radar at 31,000 ft. Far, far above the internal pressurization, which is usually equivalent to 7000-10,000 ft. So unless the bomb were somehow attached to the outside of the aircraft, it should have blown somewhere around 10,000 ft max if it were triggered by a pressure altimeter.
It's worth noting that structural failures also occur the exact moment a plane reaches a certain altitude. As the pressure differential between inside and outside increases, the stresses in the structure increase. If the structure is somehow weakened, it will give way once a certain pressure threshold is exceeded, which usually happens upon climbing in altitude.
You're missing the big picture. People are speculating the next big convergence in electronics will be between TV streaming/tuning hardware and game consoles. I wouldn't have called it a few years ago, but I'm starting to agree with them. The biggest capital expense for a gamer is purchasing the 3D hardware. With bandwidth increasing and costs coming down, streamed 3D games are becoming more feasible, where the hardware and its costs for many people are shared thus lowering overall costs.
Your Lamborghini analogy is completely off the mark because there's a huge difference between a car and a plane, and it would require a huge engineering investment to redesign a car to fly. Right now the only difference between a TV streaming unit and a game console is the 3D hardware, and that extra hardware becomes irrelevant if you can stream the game (as Sony's smart TVs and bluray players already let you do with PS3 games). The marginal cost to add gaming functionality to modern streaming video units is minuscule. Replace "Apple TV" with cell phones, and "games" with PDA functionality and think about it again. That was the last big convergence - phones and PDAs converged to create today's smartphones.
The EPA MPG ratings are not intended to predict how many MPG you will get driving the car. There's just too much variability in driving styles, road conditions, and regional climate for any single number to be an accurate prediction.
The MPG rantings are intended to allow you to compare different cars to each other when shopping. If one car is rated at 25 MPG and another at 30 MPG, you may not get exactly 30 MPG with the second car. But you will most likely get about 20% better MPG with it than the first car.
To make it conceptually easier to grasp, the EPA tries to scale the figure so it's approximately equal to the MPG you'll get while driving. But this is so widely misunderstood by the public at this point that I think the EPA would be better off just changing to an abstracted scale where 100 represents the amount of fuel burned by the average sedan, higher numbers represent more fuel burned, lower numbers less.
MPG is already screwed up since it's the inverse of fuel consumption. Going from 25 MPG to 50 MPG saves only half as much fuel as going from 12.5 MPG to 25 MPG, not twice as much as the size of the numbers would suggest. This has led to all sorts of other harmful misconceptions among people, like ridiculing hybrid engines in SUVs when that's precisely the best place to put them if you want to maximize fuel savings. Putting hybrids in economy cars actually represents the smallest possible fuel savings, and quite frankly is a waste of money (if a hybrid gives 40% better MPG and the cost is an extra $4000, payback time is about 14 years on a Prius, but only 6 years on a SUV). It also means the average MPG for two vehicles is not (MPG1 + MPG2)/2. It's 1/MPGavg = 1/MPG1 + 1/MPG2. So if you're trying to save fuel by splitting your driving between a SUV and a hybrid, the SUV disproportionately affects your average MPG, and the hybrid makes little difference from a non-hybrid economy car. Changing the scale would be a good opportunity to correct this numerical distortion. The rest of the world uses liters per 100 km specifically to avoid this problem.
There is nothing wrong with the system. You're assuming the point of the bank (or credit card) is to offer you a service. It's not. The bank/credit card is lending money for the purpose of making more money (collecting interest). Nothing more. If they can't collect as much interest from you, in their eyes that makes you a worse investment than someone who carries a balance and pays interest.
If you have substantial savings and your financial habits are good enough to maintain a 800+ credit score, then what's the problem? You don't need to borrow money. Use your savings.
Or alternately, do what the banks and credit cards do - leverage your savings to generate income. Having zero debt is actually not the optimal fiscal situation. It's the safest, but it's not optimal. Basically by putting your money in a savings account, you're giving the bank a cheap loan, and the bank is leveraging your money to make more money (by doing things like lending it out to people who carry balances on their credit cards). If you have a savings, no debt, and excellent credit, you should be able to borrow money very cheaply, and invest it in something which gives a greater return than the interest you'll have to pay, and certainly a greater return than the interest on a savings account.
In order to get the latest news from Japan, several stations in the U.S. were carrying live video feeds from Japan. I was expecting awesome interactive 3D computer graphics using green screens to create a pseudo-holographic experience. Instead the weather report was a cloth map of Japan with felt cutouts of clouds, the sun, and numbers for the temperature velcro'ed on. The weatherman (woman) pointed to these using a pointing stick (hadn't seen one of those since the 1990s when they started being replaced by laser pointers). When covering the Fukushima accident, they'd gone to the trouble of recreating the entire facility in model scale using painted cardboard. It was damn good, would've made any model railroader proud, but was such a throwback to the era before computer graphics.
That's actually why I'm a bit skeptical of the EPA claims in this case. The EPA cites only the 2014 Touareg, 2015 Cayenne, and various 2016 Audi models for having the defeat device. But all three of those vehicle lines sharing the same engine all three years. If you found it on those specific vehicles, you'd expect to find it on all those vehicles for all three years because it's the same engine.
I'm inclined to believe VW on this one. The EPA letter narrows the problem down to an odd mix of vehicles. The 2014 Touareg, the 2015 Cayenne, and various 2016 Audi models. This despite all these vehicles sharing the same 3.0 liter engine each year. If the 2014 Touareg had the defeat device, you'd expect the 2014 Porsche and Audi vehicles to also have it. You'd also expect the 2015 Touareg to have it since it uses the same engine as the 2014 model, but for some reason only the Porsche is flagged as having it for 2015. Also, unlike the 2.0 liter TDI engine scandal, the 3.0 liter engines all use urea injection systems to reduce NOx emissions.
There is something else going on here.
Part of the problem is the unlicensed bands need to be coordinated worldwide, since people will inevitably carry such devices from one country to another.
As with DOS, IBM contracted with Microsoft to develop OS/2, so it actually started off as Microsoft's code. Both partners were on board for a while and told the entire industry DOS (a CLI running programs in real mode) was going to be replaced by OS/2 (a GUI though it could run DOS in a window using the 80286's protected mode, so a crashed DOS app wouldn't hang the entire computer), so a lot of companies began porting their software over to OS/2.
Then there was some sort of falling out. Most people point the finger at Microsoft because Microsoft stopped talking about OS/2 and started talking about Windows (a GUI which ran on DOS). That's why early versions of Windows had a terrible reputation for crashing - because it ran on top of DOS, any app which crashed could hang the entire OS. You would typically have to reboot the computer 2-3 times a day, losing all your unsaved work each tine. This remained true all the way up to Windows ME. By Windows 3.0, Microsoft was completely pushing Windows and not OS/2. IBM wasn't happy but the contract called for Microsoft to do OS/2 development, so IBM couldn't take over or hire another company to do it. They negotiated to take over the project from Microsoft.
Because the then-current version of Windows was built with the same GUI elements IBM had established after a lot of usability R&D (IBM's Common User Access), IBM also got rights to Windows 3.x's code. That's what allowed them to include a copy of Windows in OS/2 3.0. They just modified Windows to run on the version of DOS running in OS/2's DOS box, instead of MS-DOS. It's also one of the reasons Microsoft really pushed to release Windows 95 in 1995 (it was code-named Chicago, and the joke was that Gates made the official name Windows 95 instead of Windows 4.0 to force his developers to release it in 1995). The terms of the split only gave IBM access to Windows 3.x code, so it was important for Microsoft to push the next version of Windows out there. Starting with Win95, Microsoft played a cat and mouse game making changes to prevent it from running in an OS/2 DOS box (or on DR-DOS).
NT was a separate project within Microsoft (built like OS/2 to not run programs in x86 real mode like DOS did), which had been on the back burner. Once they formalized their split with IBM, they began working on it in earnest, merging its API with Windows 9x. There were earlier enterprise-only releases, but the first real commercial release was Windows 2000 (in parallel with Windows ME). It's the foundation for the current versions of Windows.
Incidentally, the switcheroo between OS/2 and Windows ended up really helping Microsoft. Both IBM and Microsoft stressed that OS/2 was the future, so a lot of software companies which dominated DOS (e.g. WordPerfect, Lotus) invested a lot in porting their software to OS/2. When Microsoft switched gears and told them to develop for Windows version instead, these companies felt they'd been lied to and balked. They either made a half-hearted attempt to port to Windows, or put it off for a year or more. That time gap is what allowed Microsoft to swoop in and take over the office suite market on Windows with Microsoft Office. And thus began the early calls for an anti-trust investigation of Microsoft (which really swung into high gear when Microsoft tried to take over the web browser market by including it in Windows for free, essentially destroying the profit model for any competitor making a browser).
OS/2 1.3 used to run a lot of ATMs, until machines running Windows began replacing them in the 2000s. Coincidentally, that's when a lot of hacks to steal money from the ATMs started.
If your metric for whether or not something is underfunded is if it could be improved by spending more money, then you fail at business. Everything can be improved by spending more money on it. The trick to succeeding at business (running efficiently) is to spend the minimum amount necessary to consistently complete jobs to a certain standard.
You hand-waved the last part to make the math work out in your favor. The fixed cost (equipment, wages for salaried workers, etc) to the bidding company is $F. These are costs the company would be paying regardless of whether or not they get the contract. The variable cost is $V and represents costs specific to that one contract. The total cost for the bidding company is then ($V + $F/n) where n is the number of contracts jobs it has. (This is a bit simplified since n would be weighted depending on the size of each job, but you get the idea.)
If the cost of doing the job to the company doing the outsourcing is $X, then assuming similar costs $X = $V + $F. And $Z = $F - $F/n.
Outsourcing then makes financial sense when a company specializing in the outsourcing task is able to lower costs by using the same equipment for multiple jobs (whether separated by space or by time). This is in fact how most business works, not just IT. You don't keep $10,000 of plumbing equipment in your garage. When you have a plumbing problem, you just call a plumber and pay him $200. He then uses his $10,000 of plumbing equipment to fix your problem. You have outsourced the plumbing job.
In fact the whole of modern civilization is built upon this very premise. Instead of each person growing his own crops, raising his own livestock, cutting his own lumber, building his own house, shoeing his own horse, etc., all of these jobs have become specialized. Instead of each person needing to learn a thousand different things, they just become extremely proficient at one thing. Goods and services they are unable to make or complete because they are untrained, they simply trade with other people who are extremely proficient at those one things. It's a more efficient arrangement since everyone is extremely proficient, which is why our average per capita productivity is so much higher than in the settler days when each person needed to be somewhat proficient at those thousand things.
Where outsourcing becomes unnecessary is when the job at the one organization is big enough that n can in fact be set to 1. If a job requires one company and one full-time IT staff, then $F has surpassed the threshold where there's no longer any advantage to breaking it down to $F/n. Arguably, a government should always be in this position. While a single government office's IT needs may be small enough not to require a full-time dedicated IT employee, the sum total of all those offices should require enough employees that it makes sense to have a dedicated IT department which provides part-time services to all those offices.
The only exception then is temporary projects which require additional staff and equipment, but that staff and equipment becomes superfluous after the project is completed.
Actually I'd say the Journal model of publishing is outdated and backwards in today's interconnected society. They made sense back in the day when both publishing and buying (shipping) published copies cost a substantial amount of money. It was fiscally prudent to filter out the papers before publishing to a few judged to be "worthy" in the field.
Nowadays, publishing (on the web) and shipping (over the internet) is for all practical purposes free. There is no longer any need to filter prior to publishing - filtering can happen after. Researchers should just "publish" their papers on their own or school's website. "Journals" need to be replaced by some sort of search engine which reviews new papers and ranks them according to "worthiness". Something like an IMDB of scientific papers, with a running "top 10 papers in the last month" list for each field with rankings based on reviews by researchers in the field. That top 10 list for each month could be archived as the equivalent of a "Journal", while the entire database and ranking is accessible and searchable as a replacement for browsing through the journal section of the library.
For brownie points, you could even make the searches customizable. So if you think Prof. Henry Higgins is a quack, you can specifically tag him as such in your account, and the ranking algorithm would exclude all his votes in the paper rankings that are displayed to you.
I like how you made it a trifecta by posting that as anonymous coward.
Ice has a density of approx 0.91 g/cm^3. The Earth's crust has a density of approx 2.7 g/cm^3. So the crust weighs about 3x more per unit volume than ice.
You are hypothesizing that (in terms of net potential energy) the removal of 1 ton of ice lowering the top of the ice by a net 1 meter results in the uplifting of a net 3 tons of rock by more than 1 meter. Basic physics says what you posit can only occur in the presence of a (massive) external energy source which replenishes that lost potential energy and then some. I suppose that might be possible if there's a huge pressurized magma bubble underneath Antarctica exerting a huge amount of upward pressure, but the burden of proof would be upon you.
You don't need to double-test everything. When a manufacturer produces 1 million units of a product, they don't test every single one. They pull out about one unit per thousand or one per ten thousand and test the sample. Yes that means some bad units get through. But the sampling will give you a statistical idea what percentage of the units are bad. That gives you a basis for improving the system if the flaw rate is higher than you deem acceptable. And perhaps most relevant to this case, it'll let you see if the flaw rate suddenly spikes. Then you can go hunt down the part or person responsible for the new flaws and fix or replace them.
"But these are court cases. We need to be 100% sure if the person is guilty, so everything should be double or triple tested!" Nothing in life is 100% sure. We only try criminal cases based on guilt beyond a reasonable doubt. All this would do is change the predominant probability bound from "there is only a 1 in 40 million chance that the accused's DNA would've matched that found at the crime scene by random chance" to "there is only a 1 in 10,000 chance the DNA testing lab made an error."
Actually, I suspect that's the reason police and prosecutors haven't widely implemented random redundancy testing to detect lab error rates. The statistics behind DNA matches (usually 1 per million to several hundred million) has created a popular belief that DNA tests are infallible, which makes the prosecutor's job a lot easier if they can get a DNA match. The error rate for anything involving people tends to be around 1 in a thousand. If the public learned about that, it would destroy juries' faith in DNA tests and take away one of the prosecutors' best modern tools.
Good judgment comes from experience.
Experience comes from bad judgment.
Flash was never supposed to be the universal scripting language of the web. It was designed as an artist's tool for transmitting animated graphics over low-bandwidth connections. Consequently, security was a very low priority in its original design.
If you want to blame someone, blame the folks behind the HTML standard who dragged their feet for a decade on adding features to the standard that web designers wanted to use. Websites resorted to using Flash because there wasn't a way to do it with HTML. It was so bad, for a while there I was thinking maybe it would've been better if Microsoft had completely succeeded at gaining control of HTML and "extended" it with a bunch of proprietary features.
My first laptop had a 640x480 monochrome 9.5" screen. I was very productive running multiple windowed apps on it.
Somehow, this guy named Jobs convinced a huge swath of the population that it was impossible to multitask on a small screen. Even Microsoft bought what he said, and it resulted in the widespread rebellion against Windows 8's Metro interface. It's not the screen size or resolution which makes it hard to multitask on phones and tablets. It's the clumsy navigation you get with a touch interface - too easy to tap the the wrong spot or to "click" when you meant to move. Jobs eliminated multitasking so that this drawback of touch interfaces wouldn't be as pronounced, in order to increase the chances of his devices (which only had a touch interface) being accepted.
It worked, but don't ever make the mistake of believing Job's marketing spin is the true explanation. Multitasking has nothing to do with screen size or resolution. Multitasking on a small low-res screen is absolutely not a problem if you have precise inputs like a keyboard, mouse, and even touchpoint/trackpad (the important part is that pointer movement is distinct from clicking).
Yeah, at my college, if the author of a textbook was at the school or a friend of someone at the school, the prof would just get permission to copy it (since the author still held the copyright, not the publishing company). They'd send a loose-leaf copy to the school copy center, and give the students the option of buying the textbook, or buying the spiral-bound copy for a few bucks (cost of making the copies). Great option for the students who were taking the course as a core requirement, not something related to their major.