Doubtful. This damn hurricane didn't even give us a waterspout or funnel cloud. Isaac was kind of a flop too, but at least it lit up the Miami and Fort Lauderdale TWDR like a Christmas tree with tornado vortex signatures (TVS) for a couple of hours as it approached. Sandy? I think I saw maybe a dozen TVS over the span of 3 hours between KAMX, TMIA, and TFLL, before I got bored and went to bed.
Actually, there WAS one thing about Isaac that was *very* cool. When I was driving to work, I noticed about 20 cars pulled over to the side of the ramp from eastbound I-595 to northbound I-95, and I quickly discovered the reason: for the first time I can remember, jets were making "Kai Tak" landings at Fort Lauderdale airport. there was a major rain band further west, so they couldn't use the normal "straight in" approach to runway 13 (31?). Instead, the jets made their final approach by flying south (directly above I-95), cleared the bridge over the New River and the tall bridge for State Road 84, then made a 45-degree hard left turn above the cheering crowd standing on the ramp from eastbound 595 to northbound 95. (aerial view: http://goo.gl/maps/kojPF )
I enjoyed the festivities for a few minutes (the jets were coming in about 1 every 2 minutes), then got back in my car and headed to the office, timed to see one jet approaching the bridge as I drove over the river, and another jet flying by overhead near Oakland Park. It was very cool. It's a shame they're getting rid of that runway next year. Personally, I think the airport's making a mistake, because without that runway, they would have been forced to shut down the airport completely for most of the day when Isaac was passing over.
Oh, for ${deity}'s sake, this is going to hit as a 2-day windy rainstorm, not fsck'ing Hurricane Andrew. It's most likely consequence is the largest power outage in American history, not the wholesale destruction of the northeastern US like a wet, soggy nuclear bomb.
Just remember this: you can buy generators from Amazon.com with $3.99 prime overnight shipping.
OK, a few more tips:
* Harbor Freight's latest sales flyer has $89 generators again. If you don't own one, and can't afford a better one, go buy one. Don't kid yourself -- it'll probably be in throw-away condition by next week, and won't run much more than your laptop, some lights, your phone chargers, and maybe your DSL modem (see the next point), but if you're reading Slashdot, those items are the essential items of life, without which you'll be unrelentingly miserable.
* If your DSL modem doesn't work, but you have dialtone, you might have to double-convert your power. In other words, plug a 12v adapter into the generator's 120v, and feed its output to a cigarette lighter socket. Plug a 12v-to-120v inverter into that socket, and plug the wall wart from your DSL modem into the inverter.
* Don't bother trying to use a UPS with your generator. it won't work. Seriously, it won't. Generator power is good enough for running almost anything you care about, but UPSes are picky about things like AC line frequency... and sadly, picky about it for no really good reason, besides the fact that 20 years ago, line frequency was something that was easy to measure and a good proxy for the electricity itself. Your laptop's PSU doesn't really care whether the line frequency is 60hz, 50hz, or actually just abuot anything between 42hz and 65hz. Unfortunately, your UPS will see the generator's frequency wobble, and will kick the UPS into battery mode. What? You have an expensive, huge, inverter-type generator? Great, but it's still not going to work. The moment the UPS is happy with the line power's quality and takes it off battery power, the surge load is going to make the generator stumble for a cycle or two... and the UPS will notice, and instantly switch back to battery power. Then a moment later, it'll decide the current's stable, and try to switch back. Stir, rinse, and repeat until the UPS's battery runs out as you stand there swearing at it. This is a common problem. Unless you're literally a company the size of a Google data center with your own private power plant that has a huge flywheel design, your likelihood of success with any generator+UPS combo is roughly nil, almost entirely due to the line-frequency-UPS-freakout problem. Please, for the love of god, will someone who works for APC please read this and let us have a UPS that's frequency-tolerant?!?
You need two ice chests and an igloo cooler. Fill the igloo cooler with ice. Ice for drinks comes off the top. Cold melted water comes out the tap. No need for crates of bottled water you'll never drink, because you don't actually drink water anyway;-) The first cooler is for drinks. The second cooler is for food. Have both full and ready at least 6 hours before you're likely to lose power, and DO NOT open the refrigerator or freezer until power comes back on unless you're planning to throw away everything inside. On the other hand, if the power's off for more than 2 days, clean out the refrigerator on day 3 or 4. If you don't, it'll turn into a real, honest to god biohazard before that first week is over, and you'll end up having to throw it out because you'll NEVER be able to safely decontaminate it once it turns into a mold colony. And it will, quickly. Oh, if possible, for 'drink' ice, buy bagged ice that's not random chippings -- they tend to melt together into a monolithic block that becomes useless for drinks. Cylindrical ice is the best.
Do all of your laundry NOW. You can run a washer off a generator, but unless you have gas, your dryer is gone until the power's back on. In a state like Florida where it's 99% humid outside, clotheslines don't wor
The most likely mode of failure for internet access during Sandy is likely to be "the storm knocked out commercial power, then persisted longer than the battery backup power at your service provider's facility or tower".
From the research I did, it looks like the best bet for datacard/hotspot #1 is Verizon. Apparently, they have 8-10 hours of battery backup at all of their cell sites, and 85% (in Florida, at least; not sure whether the statistic was specific to Florida or applies nationwide) have on-site generators that fire up automatically & have enough on-site fuel to run for a week. They also apparently allow you to buy an unsubsidized data card or hotspot on eBay, and activate it for $15 per day (250mb data per day) in a completely adhoc manner, with no strings, minimums, reactivation/inactivity fees, or other sneaky charges.
For some reason, they seem to explicitly NOT allow "day pass" use with PCMCIA/Cardbus/ExpressCard devices, and I'm still trying to find out whether you have to activate it before the storm (or at least have working phone/internet service by some other means at the time you activate it), or whether you can literally buy a $13 EVDO datacard on eBay, throw it in a drawer as a really cheap insurance policy against loss of internet access during a storm, then pull it out, plug it into your laptop, and do the whole process -- payment, activation, and all -- using only the connectivity provided by the Verizon datacard itself.
Apparently, AT&T has a similar "day pass" deal. I didn't bother to research it, because I already have an AT&T phone (Galaxy S3), and since my whole goal was to find cheap "backup plan" options for getting online if my AT&T cell phone lost data service during a storm, I didn't bother to look into them.
For a longer outage, especially if you have Cable internet (which tends to go out shortly after commercial power is lost, and stay that way until the day after it's restored... at least, going by everything I've ever seen from Comcast in Florida), you might want to look into something that's cheaper and less stingy with data, like maybe T-Mobile. I wasn't able to find anything specific about their backup power situation besides references to them having a fleet of portable generators, which suggests that they're worse than Verizon (who already has fixed generators on-site, in place, ready to go), no better than AT&T (call it a hunch, but I suspect that whatever Verizon does, AT&T probably pays lip service to doing as well), and probably at least a little bit worse. My assessment: T-Mobile probably won't stay up until the bitter end of the storm, but if your cable internet is going to be down for a few days or more, they're probably the best option for days #2 and beyond. I'd expect that even if they go down during the storm, they'll be up and running within a day afterwards.
One caveat about used T-Mobile devices... I'm not sure exactly why this is apparently a problem unique to T-Mobile (or at least a bigger problem with them), but apparently it's possible to buy a used T-mobile device after getting T-Mobile to verify that the ESN is 'clean', activate it with your own SIM, use it for months, then have it unceremoniously blacklisted by T-Mobile for something the seller did long after it was sold to you. For example, if someone buys a device on a 2-year contract, replaces it with another, sells the first one to you, then later defaults on the contract. Apparently, Sprint and Verizon keep track of transfers, but T-Mobile just indiscriminately blacklists whatever ESN was on file under the original contract without bothering to investigate further to avoid collateral damage).
Right now, I can't recommend Sprint under any circumstances. Their 3G network sucks so badly right now (with the possible exception of the 3 or 4 places they've semi-finished upgrading), power loss is almost the least of their problems. After Isaac strafed Miami (taking down Comcast and U-verse for about 6-8 hours), I ran speedtest on Sprint & got
The article's central argument is that data centers can be run at higher temperatures. I'm pointing out that if you run your data center at higher temperatures to save on your energy costs, much or all of those savings could end up getting neutralized by premature equipment failure, and the cost of mitigating it.
Hearing aids are unique among consumer electronic items, because they have almost zero tolerance for latency. If the media stream coming from your entertainment device is delayed by 12ms, you'll never notice the difference. If the sound coming out of your hearing aids is delayed by 12ms, your ability to locate items by sound and react to them is going to be completely borked. At best, you'll be stressed out and irritated. At worst, you'll feel disoriented and confused.
The problem is, all of the cheap ways to do digital signal processing add intolerable amounts of latency, so hearing aids are stuck with hybrid analog+digital designs that try to keep their filtering problems in the domian where they can be resolved the fastest. With digital designs, you can get away with sloppy designs that have corners cut and mostly get away with it if premature failure is OK as an option. With analog designs, every penny you shave off is going to have consequences, and those consequences add up quickly. Mixed-signal designs are the worst of both worlds -- you have to use premium-quality components and be aware of analog signal behavior every step of the way, then turn around and try to fix the noise and artifacts introduced by the digital part as well.
Yes, a hearing aid that simply amplifies sound through some cheap analog means, maybe with simple filtering, would be very cheap to make. However, for most users, that kind of hearing aid would be about as useful as a pair of drugstore reading glasses for somebody who has astigmatism. For profound hearing loss, making speech recognizable is about as hard as trying to fix botched laser surgery that's left somebody with higher-order optical aberrations that simply can't be fixed by a simple symmetric lens.
God/Nature/the Univrese has a cruel sense of humor, and here's an example that will make sense to people who had high-end car stereos at some point in the past. Remember what happened when you ran your stereo's line-level signal through a low-pass filter to separate out the bass channel? It flipped the phase, and made it lag. At the time, you probably dreamed of the day when you could use a DSP to implement an infinite-slope crossover that fixed both problems. Then, years later, you learned the cruel truth: in order to implement such a filter, you had to wait until you had a few thousand samples to analyze and work on... and the time you had to wait until you had a big enough window of samples to analyze ended up being almost exactly the same amount of time that the analog low-pass filter delayed the bass. The digital breakthrough is that if you don't have to do that analysis in realtime, and you have enough storage space to analyze the music offline, then re-sync everything up and store all the individual tracks separately, you can achieve the flawless perfection you always sought as a teenager with laggy bass. It's now cheap and easy to do, because you can take a whole CD, rip it to raw PCM, analyze it with your PC into separate 16-bit audio tracks for every single speaker element in your car, tweak their phase relationships to your heart's content, then write it all to a microSD card & have room to do the exact same thing to a few dozen more CDs.
The problem is, hearing aids don't have that luxury. They're one of the hardest-core realtime applications out there. You can't sample the sound, recursively process it, then go back and remix it at your leisure until it's *exactly* right, then play it over and over again thereafter. You have roughly half of a millisecond to do what you're going to do and send it to the transducer in the user's ear canal.
Of course, there's a big gray area of users whose hearing problems wouldn't be solved by cheap analog hearing aids, but like someone who's got a diopter of astigmatism and moderate far-sightedness, a pair of $12 reading glasses from the rack at the drug store would probably be better than nothing at all. But make no mistake... even if you could embrace the hacker/maker ethic, buy your own best-of-breed he
Heat is death to computer hardware. Maybe not instantly, but it definitely causes premature failure. Just look at electrolytic capacitors, to name one painfully obvious component that fails with horrifying regularity in modern hardware. Fifteen years ago, capacitors were made with bogus electrolyte and failed prematurely. Some apparently still do, but the bigger problem NOW is that lots of items are built with nominally-good electrolytic capacitors that fail within a few months, precisely when their official datasheet says they will. A given electrolytic capacitor might have a design half-life of 3-5 years at temperatures of X degrees, but be expected to have 50/50 odds of failing at any time after 6-9 months when used at temperates at or exceeding X+20 degrees. Guess what temperature modern hardware (especially cheap hardware with every possible component cost reduced by value engineering) operates at? X+Y, where Y >= 20.
Heat also does nasty things to semiconductors. A modern integrated circuit often has transistors whose junctions are literally just a few atoms wide (18 is the number I've seen tossed around a lot). In durability terms, ICs from the 1980s were metaphorically constructed from the paper used to make brown paper shopping bags, and 21st-century semiconductors are made from a single layer of 2-ply toilet paper that's also wet, has holes punched into it, and is held under tension. Heat stresses these already-stressed semiconductors out even more, and like electrolytic capacitors, it causes them to begin failing in months rather than years.
It serves HTC right. Hopefully they OneX taught them a lesson, and next year's models will have batteries that end users can swap/upgrade, microSD sockets, and real two-stage camera buttons.
Seriously. Name one single thing that makes the HTC OneX a better phone than the Galaxy S3. Nothing. Nothing whatsoever. If HTC had given it a two-stage camera button, or even any dedicated camera shutter button AT ALL, at least some people would have been left wringing their hands and agonizing between it and the S3. They didn't, so that's one opportunity to differentiate themselves for roughly 17 cents that HTC squandered.
The OneX has a sealed battery. Right there, they've instantly written off anyone who won't buy a phone that can't be used with a 2800mAH+ battery, and anybody who expects to be able to swap batteries at will. The Galaxy S3 allows you to do both. The OneX allows you to do neither. Strike two.
The OneX doesn't have a microSD card. The Galaxy S3 does. Once again, for the price of something that costs about 12 cents in HTC quantities, they blew it with a large segment of the Android market who won't even give a phone that lacks microSD expansion capabilities a second look.
Let's not forget HTC's nasty habit of releasing monolithic kernels that can't be built from source because the proprietary bits were just ripped out before they shat the source onto the curb and said "here it is". Samsung cleanly separates out their proprietary kernel code as proper loadable kernel modules, just like god and Linus intended. However, I'll only count this as a half-strike against HTC, because historically, they DO at least tend to release new kernels in half the time (or less) that it takes Samsung to release new loadable kernel modules for new kernels. This is a prime example of an area where HTC could spank Samsung... if they were to commit to separating out all of their proprietary bits as proper loadable kernel modules and released automated builds more or less immediately upon getting their hands on Google's new source (and in a "rapidly timely manner" if changes had to be made to fix problems with the automated builds), they'd have a HUGE competitive advantage over Samsung in this regard. They could just release them as unsupported early-access betas, and treat the users at XDA like a vast unpaid QA program.
It's not like HTC is uncreative. The Evo 3D had a very cool & compelling feature. It might not have been all that useful in daily life, but it was definitely a cool feature to have. I know lots of people who didn't really USE it, but I know of very few who genuinely wished their phone didn't have that feature at all. Most of the complaints about it were due to some of the hardware design compromises that were made to keep the cost down by limiting the resolution and bitrate at which you could capture in stereo.
Anyway, the point is that HTC decided to rest on its laurels and release a phone that doesn't suck, but doesn't really do anything BETTER than the Galaxy S3 does. It's basically the same price, targets the same market, and offers nothing to let its owners stand in front of a group of S3 owners and proudly say, "My phone does ______ better than yours does." In the Apple universe, annual incremental upgrades are doled out as the norm, and users applaud politely & line up to buy this year's refinement. In the Android universe, you have to either knock people's socks off and delight buyers every single year, or be content to sell phones that are basically 'free' no-name commodities.
Lest anybody accuse me of being a Samsung fanboy, I'll be the first to say that I *want* HTC to make phones that beat the crap out of Samsung's, because then Samsung will turn around and try harder to make phones that beat the crap out of HTC's. Then I want Google to use Motorola as its bully pulpit to pull the rug out from under both, and raise the hardware stakes even higher with phones that have unlocked bootloaders & make Samsung's and HTC's flagship models look like antiques, the same way the Nexus One did to the phones that came before it.
There's obvious proof that the NSA doesn't have the fastest computers on earth -- they aren't a Wall Street trading firm.
Nobody on EARTH spends more staggering amounts of cash on endless tiny incremental upgrades to ensure that their computers are always the fastest computers on earth. They develop their own FPGA-accelerated algorithm accelerators that are hand-tuned to execute their algorithms faster than even the fastest general-purpose computer hardware.
The NSA? They buy Wall Street's cast-offs, and have the SECOND-fastest computers on Earth.
Let's not forget the oh-so-environmentally-friendly RoHS lead-free solder that has almost zero tolerance for error, and has prematurely sent more consumer electronics products to the landfill than any other single component besides maybe bad electrolytic capacitors. It cracks, shrinks, forms whiskers, and fails in novel ways that continue to amaze people.
Between RoHS-compliant solder and bad capacitors, it's a miracle *anything* manages to last for more than a year or two anymore.
Oh, and there's another goodie on the horizon, though US and EU customs have been working overtime to keep them away as long as they can... transformerless power supplies. See, it's cheap and easy to make a 5v supply that can output a little more than 100mA, as long as you don't care about niceties we've taken for granted (more or less forever) in the west, like isolation from mains current. They have none.
The next time you see a 99c USB power supply that's impossibly small, crack it open and look for the transformer inside. The same goes for a HELL of a lot of the battery chargers currently for sale on eBay from China. High-quality USB chargers use switchmode power supplies. Cheap ones make do with a few capacitors and a zener diode. Let me put it this way... you don't EVER want to plug one of those chargers DIRECTLY into any Android phone you care about, because they offer NO protection from voltage surges besides their own destruction.
When using 802.11n in greenfield mode, does reducing the maximum Tx rate down to 6 or 13mbps just slow down the symbol rate, or does it also narrow the bandwidth as well/instead? Put another way, if you're surrounded by people using b/g to stream multicast video to TVs in their home (glares at neighbors with U-verse who can't be arsed to pull proper cat5e through the walls like *I* did), can you set your own AP to N-only "greenfield" mode, throttle it to 6mbps, then pick a channel like 3, 4, 8, or 9 with a relatively clean conscience knowing that you aren't *completely* stomping over BOTH neighbors? Or is reduced-rate wi-fi like USB1.1 with a cheap hub, where 1mbps of slow traffic uses as much resources as 10mbps of fast traffic?
The problem isn't that the US is more spread out... the problem is that Europe actually HAS open space to cheaply build new rail corridors in between cities, whereas in many parts of the eastern US, you can drive a hundred miles or more without seeing anything more rural than an occasional vacant lot next to the interstate. Nebraska and Kansas might have cities surrounded by cornfields, but east of the Mississippi, our cities tend to be surrounded by hundreds of miles of single-family homes, strip malls, and office parks.
The other problem in the US is our obsession with either keeping high-speed passenger trains 100% separate every last inch of the way, or forcing them to be capable of surviving a head-on collision at full speed with a mile-long coal train if they share tracks with a conventional train anywhere along the route... even if they'd only be running at low speed in the areas where they shared tracks (like the last mile or two into a big city station). In Europe (particularly in Germany), they built the first segment of the new high-speed tracks, and tied them in to the existing rail network at both ends... then extended them from there. In America, we piously plan to do stupid things, like build isolated segments of high-speed rail that don't directly connect to *anything*, and would force passengers to physically switch trains for years, or forever.
HSR between ONLY Bakersfield and Corcoran, or ONLY Tampa and Orlando, is insane. Brand new HSR tracks between Bakersfield and Corcoran that continue into LA and San Francisco along the existing tracks and immediately cut an hour or two off the time it would take to make the trip at low speed, then fill in the gaps to reduce the time even more, are a great start to what's going to be an awesome HSR network someday. Ditto, for new HSR tracks between Melbourne and Orlando (eventually Tampa) that connect to the existing FEC tracks between Jacksonville and Miami.
Engineering-wise, Acela-type trains aren't ideal... but they're actually pretty good. Their 150mph speed limit is due to Amtrak, not engineering -- Bombardier's engineers designed them to run at 186mph, and in a flat state like Florida, they could do 200mph without breaking a sweat given suitable tracks and administrative approval.
As far as subsidies go, EVERY transportation mode is subsidized from general tax revenues. Gas taxes haven't fully supported road construction and maintenance costs since the mid-1990s (they USED to, but as gas prices have increased, the federal and state governments have gradually reduced them to levels that no longer cover 100% of costs). In 2011, Amtrak's total subsidy came out to about $4.25 per American. Nothing to really be proud of, but far from the scandalous rape some would have you believe it is... and most of THAT is for fixed costs that are basically the same regardless of whether Amtrak runs one train or ten trains through any given station per day. Under the current status quo, Amtrak can't "win" regardless of what it does. If it raises fares, it gets decried for being expensive. If it lowers fares, it gets attacked for requiring subsidies. The point is, Amtrak is Amtrak. For better or worse, right now it's all we have. In a few years, we'll have the backbone of California HSR, and FEC Railroad's new passenger service in Florida running along with Amtrak.
You overlooked the #1 argument in *favor* of gifted class -- for many kids, it's the first time in their lives they get to spend extended amounts of time with "their people". I made my first real friend within a month of getting into the class. I was in middle school. It was literally the first time I'd met anybody who was equally smart, and was into the same things I was. I ultimately became friends with pretty much everyone in the class, and we all *might* have tripped over each other in high school (maybe in math club or later, in the Amiga users' group), but the point is that it meant I could finally have real friends after an utterly friendless elementary school experience, in a large school that nevertheless did an amazing job of keeping kids anonymously away from each other.
My brother's mother in law is a gifted teacher, and she agrees 100% that a major need for gifted class is to give kids who've always been untouchably-geeky outsiders a safe environment where they can make friends & avoid self-destructing before high school.
Run a client-side app on your phone that implements a little game of "Simon Says" for any phone number that isn't on your whitelist. It'll take out telemarketers, too, because they don't have 3 minutes to spend playing games to get to you (and most are robodialed anyway).
"Hi, your number isn't on my whitelist, so you need to play a little game of Simon Says in order to make my number ring or reach my voicemail. Press 7 to continue" [caller presses 7]
"I'm sorry, Simon didn't say to continue" [click]
[human caller tries again, presses nothing when told to press 3]
"Good, you understand the rules. SIMON SAYS, enter the first 3 digits of pi to continue, omitting the decimal point"
[caller enters 314]
"Congratulations, you aren't an elected official from the midwestern US! Press 1 to continue"
[caller enters 1, then swears violently a second later when he realizes he was tricked]
"I'm sorry, Simon didn't say to press 1" [click]
[Caller tries again, this time gets a new challenge, like...]
"There are two kinds of people in the world... those who understand binary, and those who don't. If you were starting a company to make X-10 compatible products and wanted a cute name that implied X-10++, but couldn't use X-11 because that's the name of Linux's graphics server, what single digit would arguably convey the same subtle inside joke without trademark infringement? Simon says, enter it now"
[Caller scratches his head, then has a flash of enlightenment and enters 3]
[System congratulates caller, tries to trick him into pressing 1 to continue, fails, throws another brainteaser at the caller, congratulates him for getting it right, tries to trick him into pressing a single digit again, has Simon tell him to press the digit, has simon tell him to press another, then tells him to press a third (without Simon), the caller isn't fooled]
"OK, you win. You clearly want to reach ${me} quite badly, and you're either a friend, a determined bill collector, or some pour soul in Bangalore who's getting paid almost nothing to endure hours and hours of brainteasers from others running the same software I am, so I'll put your call through."
(ring, ring... no answer, goes to voicemail)
"Oh dear, there's no answer. Look, since you've been such a good sport, Simon says to leave your message, then enter your callback number."
(for anybody who's wondering, I really did program something like once as a joke & programming exercise at the first company I worked for after college using Dialogic Visual Voice Pro. I've been meaning to try re-implementing it with Asterisk, and maybe even try to implement it as a client-side app for Android phones, if only to filter out calls from candidates during election years)
>Why don't you just register a single.com domain and run your stuff from there
Because that would make it easy for China, Iran, and other regimes to block users from using his services. That's what he *does* -- he enables people stuck behind oppressive (often government-run) firewalls to get to blocked sites & surf without frustration and/or fear.
The problem with SSD failures is that, unlike spinning drives, RAID won't save you. Several generations of OCZ's Sandforce-based drives had (and still have) weird firmware bugs that caused spontaneous data-suicide. Making matters worse, if you had multiple drives in RAID configuration, and something triggered the bug in one of them, there's a pretty good chance the same bug ended up getting triggered in ALL of them.
RAID was developed to protect data from progressive drive failure. It simply can't deal with drives that will kill themselves at the drop of a hat, for any reason, or no apparent reason at all.
The sad thing is, all this time people have been neurotic about SSDs getting write errors, while totally overlooking their REAL problem -- buggy firmware and insidious mandatory encryption that makes data recovery nearly impossible when your drive dies without warning. One minute, your computer is working fine. The next minute, your data is gone forever because the controller decided to commit suicide and take your data with it.
> Young people are cutting the cords faster than ever
The only "cord" young people have decisively cut are landline telephones, mostly because it's now an expensive service of minimal value compared to mobile phones. Some "young people" who don't view computers as important might try to use their wireless phones for internet access, but for most "cord cutters", it has more to do with avoidance of installation fees and stiff ETFs that would otherwise accompany frequent moves than any genuine desire to cut a metaphorical cord.
Also, let's not forget that "young people" who don't live with their parents or in a college dorm commonly share a house or apartment with at least one other housemate, which cuts the effective cost of the cable and internet bill in half. I don't know a male under 25 with a housemate who doesn't automatically subscribe to the most expensive channel package available, just because they can. Most young adults don't really *care* about monthly service costs until they get their own house or apartment, at which point the staggering cost of buying those services for one person becomes vividly apparent.
As far as cable companies dying, you're forgetting something -- even if content providers like ESPN and MTV were available over the internet directly to consumers, it's not cost-effective for them to individually bill millions of credit cards for each channel. Past a certain point, the transaction fees end up being more than the profit margin. So, we'll always have companies that will go out and negotiate bundle prices with multiple content providers in exchange for a monthly fee that's a fraction of what they'd cost individually. The catch is, it's a slippery slope for everyone involved, and there's tremendous motivation to increase "ARPU" by any means necessary... increasing channels from 2 to 3 (showing the same content, but repeating it more often), bundling 10 more channels for an extra 10 cents, whatever. It's no different than going to BK, and spending almost as much to buy a small hamburger, fries, and drink as you'd have spent on a large Whopper combo.
At the end of the day, under just about any likely scenario, somebody who tries to subscribe to exactly one high-value cable channel they care about (like ESPN, MTV, CNN, or Bravo) will probably end up paying $25. Two channels? $40. Three channels? $62. Twenty channels? $75. And so on, until you're right back to spending $120/month for the half-dozen or so channels you passionately care about, another 20 or so that you might watch once in a blue moon, and 6,000 you don't care about, but had to buy to get 3 of the other 20.
As a practical matter, most laws like this are only directly relevant to companies importing more than a single item for resale -- particularly large corporations like Asda importing items by the shipping crate. The EU doesn't have the resources to go out and inspect the inventory of every Chinese immigrant selling goods straight from Shenzhen from his crowded store in outer London or Amsterdam, let alone scrutinize every item purchased from the US. As a practical matter, it doesn't... it just requires companies like Asda to certify the compliance of the goods they sell, and knows that it can keep 99.9% of noncompliant goods out of most consumers' hands without lifting a finger or paying the salary of a single customs agent.
Here's an easy experiment: go find a small, independent store that sells imported computer parts purchased from Shenzhen. Say, ATX tower cases or USB hubs, owned by a guy who emigrated from China and has family members purchasing for him back in China. Buy the coolest-looking case with a brand name you've never heard of, and a random USB hub. Take them home, and scrutinize the legal compliance of each. If you see a FCC ID (or its European equivalent) etched into the circuit board, look it up... and feign surprise when you discover that it's either associated with some item that hasn't even been sold in 3 years, or was completely made up. Don't forget to check the compliance of things like the power supply, too. More likely than not, both items are technically illegal in the EU, US, or both... and nobody really cares, unless they're a retailer the size of Amazon or Walmart/Asda. Then, they care a lot.
That said, a law that sets standards that aren't already achievable on the assumption that consumers can just ignore it if necessary is a bad law.
> Simply use the net and get all the shows you want WHEN you want and with less commercials to boot.
Oh, really? And where, pray tell, is this magic IP-based virtual cable network that offers realtime IP feeds of the show that's literally running for the first time *right this second* -- as opposed to stuff that's either old (or at least non-realtime), low video quality, openly pirated, or some combination of the three?
There's no technological reason why CNN/MSNBC/Fox/TWC and the rest can't stream HD in realtime, and deliver encrypted copies of the shows you've already tagged for recording in advance & let you have the decryption key the moment the episode officially airs for the first time... but content providers just won't do it. And unfortunately, nothing can make them do it. Maybe the refusal of Dish and others to deal with AMC will prod them to do an end run around them and offer service directly to end users over the internet... but I doubt it.
As much as I'd like to believe that someday, someone will offer a boundary-free service over the internet that works like of like U-verse at the nuts & bolts level, but furnishes 19.2mbps content at original resolution in the native mode, and has a UI that's fully open & customizable by end users with programming backgrounds, I'm sufficiently nihilistic about the American content industry to know that when all is said and done, my monthly "TV-related" expenses would probably end up doubling compared to now anyway. Never, in the history of American TV, have prices genuinely gone down.
Competition from satellite HAS been nice, if only because it brought us Voom 5 years before Comcast even heard of HD, and brought us 200 HD channels on DirecTV when Comcast was still subjecting customers to "death of a thousand macroblocks" 10 new channels at a time, but don't kid yourself. One way or another, the industry is going to find ways to charge us a LOT more for a LITTLE more content, quality, and/or flexibility. And it'll probably compress the hell out of it, too.
Years ago, I laughed at my dad when he told me that color TV pictures were sharper in the 1960s than they were in the 80s and 90s. Then I saw my first DVD, and it sank in for the first time just how badly quality had been completely pwn3d by cost-cutting over the years. Now, the content providers are doing the same to us all over again. We have "HD", but the "HD" we have today via U-verse, Comcast, Dish, and even DirecTV is total and complete crap (minus 2 or 3 showcase channels) compared to the old days of Voom, where the SD channels had bitrates higher than U-verse uses NOW for its HD channels. MPEG-4 is good, but beyond a certain point, you can't fool mother nature. She's going to make sure you get blotches of red sparklies on areas that are supposed to be creamy white, just to remind you that someone is recompressing recompressed 12mbps video down to 6mbps. 10 years from now, we're going to be sold on "Super HD" with 3200x1800 resolution that, if we're lucky, will be indistinguishable from 60fps progressive 1920x1080 encoded at 25-40mbps, and basically just be oversampling with extra marketing and hype.
> Why not have kept the 400 and planted 1000 more?
Because then, the Shuttle would still be sitting in LAX's parking lot, unable to move anywhere due to the trees in the way. And the funding for those 1,000 trees exists only because people want to see the shuttle at the museum. Without a shuttle to motivate cutting down the trees and create a need to replace them with more than twice as many, the money would end up getting spent on beer & hookers, or buying a new tank for the LAPD to accidentally terrorize the neighbors of a drug dealer with.
It was smoldering long before Obama. In a rare showing of nonpartisan unity, both parties have actively beaten it up ever since the end of the Cold War.
I almost wish China would hurry up and announce plans to plant a red flag on the red planet, just to completely freak out everyone in Washington and instantly elevate NASA to the top of their funding priority list. Obviously nobody sane wants a return to the Cold War, but a little healthy rivalry between worthy adversaries can be a good alternative to the complacency we have right now.
On the other hand, NASA could try a little harder to learn how public relations is supposed to work. They need to pull the Puritanical stake out of their butt and visit Russia to see what a real launch party is supposed to be like. Russia's space program *knows* where their funding comes from, and they work hard to keep the funds flowing as freely as the vodka at their launch parties. NASA lost its way when it decided that its mission was to make space exploration boring and routine.
When there's a crisis at ISS, they should have live in-station cameras beaming the raw streaming footage to every news network 24/7 to make it their urgent story of the hour... and find ways to pitch for more funding along the way. When the crew is huddling in a refuge spot because of an impending collision with space junk, CNN should have America hanging on the drama and begging for more. 99 times out of 100, when something bad happens in space, the reaction of most people ends up being, "oh... I hadn't even though about NASA lately. I'd kind of forgotten they even exist." Then they wonder why their funding has dried up.
Over the past 30 years, NASA has managed to transform space from something that's cool and extreme, and make it seem about as exciting as working on an oil rig in the middle of the Gulf of Mexico. No, actually, they've made it seem less appealing and exciting. At least workers on the oil rig get to have uncensored broadband. NASA needs to hook up with the producers of shows like "Turbine Cowboys" and "Iron Men", and get a Journalist-Producer on board the ISS to shoot a season's worth of shows up there. Money can't *buy* the kind of PR a season of "Going Up?" on DiscoveryHD (with teaser that shows elevator running to the 100,000th floor) would get them.
You're preventing 'drive by' attacks, which are the ones likely to bite a *slashdot* user.
As long as Microsoft requires that secure boot keys be unconditionally furnished to end users, it's not a problem. The danger is if they allow vendors to charge extra for them or withhold them. A secure boot key *I* can use to sign my own bootloaders for my own PC is a good thing. It only becomes evil when Dell is allowed to sell $99 PCs w/ad-supported OS that can't be replaced by end users (because within 5 years, a non-locked PC would become an exotic niche item with a $1,800 price premium).
With a certificate bearing the same CN as the original? Low, as long as the bootloader realizes that it's never seen anything signed by s0m3hack3r@foo.to, and presents the user with a dialog that says something like, "You have never booted an OS signed by s0m3hack3r@foo.to, and foo.to is not recognized as a known OSS Organization. Click here to boot into your computer's mini-distro and perform an automated legitimacy lookup (internet access required), or (... options that include 'continue if you trust them' and 'cancel'...)
For a side trip, boot into a mini Linux burned into flash that can grab an ip via dhcp or connect to wifi with ssid/key stored in flash or entered now & wget a lookup of the CN from the UEFI bootloader's organization. Known malware CNs would be blacklisted & identified as such, others could be further researched using Lynx before either continuing the boot (optionally remembering the CN for future boots) or aborting.
Seriously. Learn ARM assembly, practice hitting the bare metal in an Android phone, and get a job working for someone like Nvidia, Qualcomm, Broadcom, Samsung, HTC, or someone comparable. You have a skill almost nobody does anymore, and you know how much more fun assembly is. Screw Java and boring corporate productivity apps. You can have more fun with assembly writing drivers, and make more money while you're at it.:-)
Slashdot Mirror
Doubtful. This damn hurricane didn't even give us a waterspout or funnel cloud. Isaac was kind of a flop too, but at least it lit up the Miami and Fort Lauderdale TWDR like a Christmas tree with tornado vortex signatures (TVS) for a couple of hours as it approached. Sandy? I think I saw maybe a dozen TVS over the span of 3 hours between KAMX, TMIA, and TFLL, before I got bored and went to bed.
Actually, there WAS one thing about Isaac that was *very* cool. When I was driving to work, I noticed about 20 cars pulled over to the side of the ramp from eastbound I-595 to northbound I-95, and I quickly discovered the reason: for the first time I can remember, jets were making "Kai Tak" landings at Fort Lauderdale airport. there was a major rain band further west, so they couldn't use the normal "straight in" approach to runway 13 (31?). Instead, the jets made their final approach by flying south (directly above I-95), cleared the bridge over the New River and the tall bridge for State Road 84, then made a 45-degree hard left turn above the cheering crowd standing on the ramp from eastbound 595 to northbound 95. (aerial view: http://goo.gl/maps/kojPF )
I enjoyed the festivities for a few minutes (the jets were coming in about 1 every 2 minutes), then got back in my car and headed to the office, timed to see one jet approaching the bridge as I drove over the river, and another jet flying by overhead near Oakland Park. It was very cool. It's a shame they're getting rid of that runway next year. Personally, I think the airport's making a mistake, because without that runway, they would have been forced to shut down the airport completely for most of the day when Isaac was passing over.
Oh, for ${deity}'s sake, this is going to hit as a 2-day windy rainstorm, not fsck'ing Hurricane Andrew. It's most likely consequence is the largest power outage in American history, not the wholesale destruction of the northeastern US like a wet, soggy nuclear bomb.
Just remember this: you can buy generators from Amazon.com with $3.99 prime overnight shipping.
OK, a few more tips:
* Harbor Freight's latest sales flyer has $89 generators again. If you don't own one, and can't afford a better one, go buy one. Don't kid yourself -- it'll probably be in throw-away condition by next week, and won't run much more than your laptop, some lights, your phone chargers, and maybe your DSL modem (see the next point), but if you're reading Slashdot, those items are the essential items of life, without which you'll be unrelentingly miserable.
* If your DSL modem doesn't work, but you have dialtone, you might have to double-convert your power. In other words, plug a 12v adapter into the generator's 120v, and feed its output to a cigarette lighter socket. Plug a 12v-to-120v inverter into that socket, and plug the wall wart from your DSL modem into the inverter.
* Don't bother trying to use a UPS with your generator. it won't work. Seriously, it won't. Generator power is good enough for running almost anything you care about, but UPSes are picky about things like AC line frequency... and sadly, picky about it for no really good reason, besides the fact that 20 years ago, line frequency was something that was easy to measure and a good proxy for the electricity itself. Your laptop's PSU doesn't really care whether the line frequency is 60hz, 50hz, or actually just abuot anything between 42hz and 65hz. Unfortunately, your UPS will see the generator's frequency wobble, and will kick the UPS into battery mode. What? You have an expensive, huge, inverter-type generator? Great, but it's still not going to work. The moment the UPS is happy with the line power's quality and takes it off battery power, the surge load is going to make the generator stumble for a cycle or two... and the UPS will notice, and instantly switch back to battery power. Then a moment later, it'll decide the current's stable, and try to switch back. Stir, rinse, and repeat until the UPS's battery runs out as you stand there swearing at it. This is a common problem. Unless you're literally a company the size of a Google data center with your own private power plant that has a huge flywheel design, your likelihood of success with any generator+UPS combo is roughly nil, almost entirely due to the line-frequency-UPS-freakout problem. Please, for the love of god, will someone who works for APC please read this and let us have a UPS that's frequency-tolerant?!?
You need two ice chests and an igloo cooler. Fill the igloo cooler with ice. Ice for drinks comes off the top. Cold melted water comes out the tap. No need for crates of bottled water you'll never drink, because you don't actually drink water anyway ;-) The first cooler is for drinks. The second cooler is for food. Have both full and ready at least 6 hours before you're likely to lose power, and DO NOT open the refrigerator or freezer until power comes back on unless you're planning to throw away everything inside. On the other hand, if the power's off for more than 2 days, clean out the refrigerator on day 3 or 4. If you don't, it'll turn into a real, honest to god biohazard before that first week is over, and you'll end up having to throw it out because you'll NEVER be able to safely decontaminate it once it turns into a mold colony. And it will, quickly. Oh, if possible, for 'drink' ice, buy bagged ice that's not random chippings -- they tend to melt together into a monolithic block that becomes useless for drinks. Cylindrical ice is the best.
Do all of your laundry NOW. You can run a washer off a generator, but unless you have gas, your dryer is gone until the power's back on. In a state like Florida where it's 99% humid outside, clotheslines don't wor
The most likely mode of failure for internet access during Sandy is likely to be "the storm knocked out commercial power, then persisted longer than the battery backup power at your service provider's facility or tower".
From the research I did, it looks like the best bet for datacard/hotspot #1 is Verizon. Apparently, they have 8-10 hours of battery backup at all of their cell sites, and 85% (in Florida, at least; not sure whether the statistic was specific to Florida or applies nationwide) have on-site generators that fire up automatically & have enough on-site fuel to run for a week. They also apparently allow you to buy an unsubsidized data card or hotspot on eBay, and activate it for $15 per day (250mb data per day) in a completely adhoc manner, with no strings, minimums, reactivation/inactivity fees, or other sneaky charges.
For some reason, they seem to explicitly NOT allow "day pass" use with PCMCIA/Cardbus/ExpressCard devices, and I'm still trying to find out whether you have to activate it before the storm (or at least have working phone/internet service by some other means at the time you activate it), or whether you can literally buy a $13 EVDO datacard on eBay, throw it in a drawer as a really cheap insurance policy against loss of internet access during a storm, then pull it out, plug it into your laptop, and do the whole process -- payment, activation, and all -- using only the connectivity provided by the Verizon datacard itself.
Apparently, AT&T has a similar "day pass" deal. I didn't bother to research it, because I already have an AT&T phone (Galaxy S3), and since my whole goal was to find cheap "backup plan" options for getting online if my AT&T cell phone lost data service during a storm, I didn't bother to look into them.
For a longer outage, especially if you have Cable internet (which tends to go out shortly after commercial power is lost, and stay that way until the day after it's restored... at least, going by everything I've ever seen from Comcast in Florida), you might want to look into something that's cheaper and less stingy with data, like maybe T-Mobile. I wasn't able to find anything specific about their backup power situation besides references to them having a fleet of portable generators, which suggests that they're worse than Verizon (who already has fixed generators on-site, in place, ready to go), no better than AT&T (call it a hunch, but I suspect that whatever Verizon does, AT&T probably pays lip service to doing as well), and probably at least a little bit worse. My assessment: T-Mobile probably won't stay up until the bitter end of the storm, but if your cable internet is going to be down for a few days or more, they're probably the best option for days #2 and beyond. I'd expect that even if they go down during the storm, they'll be up and running within a day afterwards.
One caveat about used T-Mobile devices... I'm not sure exactly why this is apparently a problem unique to T-Mobile (or at least a bigger problem with them), but apparently it's possible to buy a used T-mobile device after getting T-Mobile to verify that the ESN is 'clean', activate it with your own SIM, use it for months, then have it unceremoniously blacklisted by T-Mobile for something the seller did long after it was sold to you. For example, if someone buys a device on a 2-year contract, replaces it with another, sells the first one to you, then later defaults on the contract. Apparently, Sprint and Verizon keep track of transfers, but T-Mobile just indiscriminately blacklists whatever ESN was on file under the original contract without bothering to investigate further to avoid collateral damage).
Right now, I can't recommend Sprint under any circumstances. Their 3G network sucks so badly right now (with the possible exception of the 3 or 4 places they've semi-finished upgrading), power loss is almost the least of their problems. After Isaac strafed Miami (taking down Comcast and U-verse for about 6-8 hours), I ran speedtest on Sprint & got
The article's central argument is that data centers can be run at higher temperatures. I'm pointing out that if you run your data center at higher temperatures to save on your energy costs, much or all of those savings could end up getting neutralized by premature equipment failure, and the cost of mitigating it.
Hearing aids are unique among consumer electronic items, because they have almost zero tolerance for latency. If the media stream coming from your entertainment device is delayed by 12ms, you'll never notice the difference. If the sound coming out of your hearing aids is delayed by 12ms, your ability to locate items by sound and react to them is going to be completely borked. At best, you'll be stressed out and irritated. At worst, you'll feel disoriented and confused.
The problem is, all of the cheap ways to do digital signal processing add intolerable amounts of latency, so hearing aids are stuck with hybrid analog+digital designs that try to keep their filtering problems in the domian where they can be resolved the fastest. With digital designs, you can get away with sloppy designs that have corners cut and mostly get away with it if premature failure is OK as an option. With analog designs, every penny you shave off is going to have consequences, and those consequences add up quickly. Mixed-signal designs are the worst of both worlds -- you have to use premium-quality components and be aware of analog signal behavior every step of the way, then turn around and try to fix the noise and artifacts introduced by the digital part as well.
Yes, a hearing aid that simply amplifies sound through some cheap analog means, maybe with simple filtering, would be very cheap to make. However, for most users, that kind of hearing aid would be about as useful as a pair of drugstore reading glasses for somebody who has astigmatism. For profound hearing loss, making speech recognizable is about as hard as trying to fix botched laser surgery that's left somebody with higher-order optical aberrations that simply can't be fixed by a simple symmetric lens.
God/Nature/the Univrese has a cruel sense of humor, and here's an example that will make sense to people who had high-end car stereos at some point in the past. Remember what happened when you ran your stereo's line-level signal through a low-pass filter to separate out the bass channel? It flipped the phase, and made it lag. At the time, you probably dreamed of the day when you could use a DSP to implement an infinite-slope crossover that fixed both problems. Then, years later, you learned the cruel truth: in order to implement such a filter, you had to wait until you had a few thousand samples to analyze and work on... and the time you had to wait until you had a big enough window of samples to analyze ended up being almost exactly the same amount of time that the analog low-pass filter delayed the bass. The digital breakthrough is that if you don't have to do that analysis in realtime, and you have enough storage space to analyze the music offline, then re-sync everything up and store all the individual tracks separately, you can achieve the flawless perfection you always sought as a teenager with laggy bass. It's now cheap and easy to do, because you can take a whole CD, rip it to raw PCM, analyze it with your PC into separate 16-bit audio tracks for every single speaker element in your car, tweak their phase relationships to your heart's content, then write it all to a microSD card & have room to do the exact same thing to a few dozen more CDs.
The problem is, hearing aids don't have that luxury. They're one of the hardest-core realtime applications out there. You can't sample the sound, recursively process it, then go back and remix it at your leisure until it's *exactly* right, then play it over and over again thereafter. You have roughly half of a millisecond to do what you're going to do and send it to the transducer in the user's ear canal.
Of course, there's a big gray area of users whose hearing problems wouldn't be solved by cheap analog hearing aids, but like someone who's got a diopter of astigmatism and moderate far-sightedness, a pair of $12 reading glasses from the rack at the drug store would probably be better than nothing at all. But make no mistake... even if you could embrace the hacker/maker ethic, buy your own best-of-breed he
Heat is death to computer hardware. Maybe not instantly, but it definitely causes premature failure. Just look at electrolytic capacitors, to name one painfully obvious component that fails with horrifying regularity in modern hardware. Fifteen years ago, capacitors were made with bogus electrolyte and failed prematurely. Some apparently still do, but the bigger problem NOW is that lots of items are built with nominally-good electrolytic capacitors that fail within a few months, precisely when their official datasheet says they will. A given electrolytic capacitor might have a design half-life of 3-5 years at temperatures of X degrees, but be expected to have 50/50 odds of failing at any time after 6-9 months when used at temperates at or exceeding X+20 degrees. Guess what temperature modern hardware (especially cheap hardware with every possible component cost reduced by value engineering) operates at? X+Y, where Y >= 20.
Heat also does nasty things to semiconductors. A modern integrated circuit often has transistors whose junctions are literally just a few atoms wide (18 is the number I've seen tossed around a lot). In durability terms, ICs from the 1980s were metaphorically constructed from the paper used to make brown paper shopping bags, and 21st-century semiconductors are made from a single layer of 2-ply toilet paper that's also wet, has holes punched into it, and is held under tension. Heat stresses these already-stressed semiconductors out even more, and like electrolytic capacitors, it causes them to begin failing in months rather than years.
It serves HTC right. Hopefully they OneX taught them a lesson, and next year's models will have batteries that end users can swap/upgrade, microSD sockets, and real two-stage camera buttons.
Seriously. Name one single thing that makes the HTC OneX a better phone than the Galaxy S3. Nothing. Nothing whatsoever. If HTC had given it a two-stage camera button, or even any dedicated camera shutter button AT ALL, at least some people would have been left wringing their hands and agonizing between it and the S3. They didn't, so that's one opportunity to differentiate themselves for roughly 17 cents that HTC squandered.
The OneX has a sealed battery. Right there, they've instantly written off anyone who won't buy a phone that can't be used with a 2800mAH+ battery, and anybody who expects to be able to swap batteries at will. The Galaxy S3 allows you to do both. The OneX allows you to do neither. Strike two.
The OneX doesn't have a microSD card. The Galaxy S3 does. Once again, for the price of something that costs about 12 cents in HTC quantities, they blew it with a large segment of the Android market who won't even give a phone that lacks microSD expansion capabilities a second look.
Let's not forget HTC's nasty habit of releasing monolithic kernels that can't be built from source because the proprietary bits were just ripped out before they shat the source onto the curb and said "here it is". Samsung cleanly separates out their proprietary kernel code as proper loadable kernel modules, just like god and Linus intended. However, I'll only count this as a half-strike against HTC, because historically, they DO at least tend to release new kernels in half the time (or less) that it takes Samsung to release new loadable kernel modules for new kernels. This is a prime example of an area where HTC could spank Samsung... if they were to commit to separating out all of their proprietary bits as proper loadable kernel modules and released automated builds more or less immediately upon getting their hands on Google's new source (and in a "rapidly timely manner" if changes had to be made to fix problems with the automated builds), they'd have a HUGE competitive advantage over Samsung in this regard. They could just release them as unsupported early-access betas, and treat the users at XDA like a vast unpaid QA program.
It's not like HTC is uncreative. The Evo 3D had a very cool & compelling feature. It might not have been all that useful in daily life, but it was definitely a cool feature to have. I know lots of people who didn't really USE it, but I know of very few who genuinely wished their phone didn't have that feature at all. Most of the complaints about it were due to some of the hardware design compromises that were made to keep the cost down by limiting the resolution and bitrate at which you could capture in stereo.
Anyway, the point is that HTC decided to rest on its laurels and release a phone that doesn't suck, but doesn't really do anything BETTER than the Galaxy S3 does. It's basically the same price, targets the same market, and offers nothing to let its owners stand in front of a group of S3 owners and proudly say, "My phone does ______ better than yours does." In the Apple universe, annual incremental upgrades are doled out as the norm, and users applaud politely & line up to buy this year's refinement. In the Android universe, you have to either knock people's socks off and delight buyers every single year, or be content to sell phones that are basically 'free' no-name commodities.
Lest anybody accuse me of being a Samsung fanboy, I'll be the first to say that I *want* HTC to make phones that beat the crap out of Samsung's, because then Samsung will turn around and try harder to make phones that beat the crap out of HTC's. Then I want Google to use Motorola as its bully pulpit to pull the rug out from under both, and raise the hardware stakes even higher with phones that have unlocked bootloaders & make Samsung's and HTC's flagship models look like antiques, the same way the Nexus One did to the phones that came before it.
There's obvious proof that the NSA doesn't have the fastest computers on earth -- they aren't a Wall Street trading firm.
Nobody on EARTH spends more staggering amounts of cash on endless tiny incremental upgrades to ensure that their computers are always the fastest computers on earth. They develop their own FPGA-accelerated algorithm accelerators that are hand-tuned to execute their algorithms faster than even the fastest general-purpose computer hardware.
The NSA? They buy Wall Street's cast-offs, and have the SECOND-fastest computers on Earth.
WinNT => WNT == VMS++
Let's not forget the oh-so-environmentally-friendly RoHS lead-free solder that has almost zero tolerance for error, and has prematurely sent more consumer electronics products to the landfill than any other single component besides maybe bad electrolytic capacitors. It cracks, shrinks, forms whiskers, and fails in novel ways that continue to amaze people.
Between RoHS-compliant solder and bad capacitors, it's a miracle *anything* manages to last for more than a year or two anymore.
Oh, and there's another goodie on the horizon, though US and EU customs have been working overtime to keep them away as long as they can... transformerless power supplies. See, it's cheap and easy to make a 5v supply that can output a little more than 100mA, as long as you don't care about niceties we've taken for granted (more or less forever) in the west, like isolation from mains current. They have none.
The next time you see a 99c USB power supply that's impossibly small, crack it open and look for the transformer inside. The same goes for a HELL of a lot of the battery chargers currently for sale on eBay from China. High-quality USB chargers use switchmode power supplies. Cheap ones make do with a few capacitors and a zener diode. Let me put it this way... you don't EVER want to plug one of those chargers DIRECTLY into any Android phone you care about, because they offer NO protection from voltage surges besides their own destruction.
When using 802.11n in greenfield mode, does reducing the maximum Tx rate down to 6 or 13mbps just slow down the symbol rate, or does it also narrow the bandwidth as well/instead? Put another way, if you're surrounded by people using b/g to stream multicast video to TVs in their home (glares at neighbors with U-verse who can't be arsed to pull proper cat5e through the walls like *I* did), can you set your own AP to N-only "greenfield" mode, throttle it to 6mbps, then pick a channel like 3, 4, 8, or 9 with a relatively clean conscience knowing that you aren't *completely* stomping over BOTH neighbors? Or is reduced-rate wi-fi like USB1.1 with a cheap hub, where 1mbps of slow traffic uses as much resources as 10mbps of fast traffic?
The problem isn't that the US is more spread out... the problem is that Europe actually HAS open space to cheaply build new rail corridors in between cities, whereas in many parts of the eastern US, you can drive a hundred miles or more without seeing anything more rural than an occasional vacant lot next to the interstate. Nebraska and Kansas might have cities surrounded by cornfields, but east of the Mississippi, our cities tend to be surrounded by hundreds of miles of single-family homes, strip malls, and office parks.
The other problem in the US is our obsession with either keeping high-speed passenger trains 100% separate every last inch of the way, or forcing them to be capable of surviving a head-on collision at full speed with a mile-long coal train if they share tracks with a conventional train anywhere along the route... even if they'd only be running at low speed in the areas where they shared tracks (like the last mile or two into a big city station). In Europe (particularly in Germany), they built the first segment of the new high-speed tracks, and tied them in to the existing rail network at both ends... then extended them from there. In America, we piously plan to do stupid things, like build isolated segments of high-speed rail that don't directly connect to *anything*, and would force passengers to physically switch trains for years, or forever.
HSR between ONLY Bakersfield and Corcoran, or ONLY Tampa and Orlando, is insane. Brand new HSR tracks between Bakersfield and Corcoran that continue into LA and San Francisco along the existing tracks and immediately cut an hour or two off the time it would take to make the trip at low speed, then fill in the gaps to reduce the time even more, are a great start to what's going to be an awesome HSR network someday. Ditto, for new HSR tracks between Melbourne and Orlando (eventually Tampa) that connect to the existing FEC tracks between Jacksonville and Miami.
Engineering-wise, Acela-type trains aren't ideal... but they're actually pretty good. Their 150mph speed limit is due to Amtrak, not engineering -- Bombardier's engineers designed them to run at 186mph, and in a flat state like Florida, they could do 200mph without breaking a sweat given suitable tracks and administrative approval.
As far as subsidies go, EVERY transportation mode is subsidized from general tax revenues. Gas taxes haven't fully supported road construction and maintenance costs since the mid-1990s (they USED to, but as gas prices have increased, the federal and state governments have gradually reduced them to levels that no longer cover 100% of costs). In 2011, Amtrak's total subsidy came out to about $4.25 per American. Nothing to really be proud of, but far from the scandalous rape some would have you believe it is... and most of THAT is for fixed costs that are basically the same regardless of whether Amtrak runs one train or ten trains through any given station per day. Under the current status quo, Amtrak can't "win" regardless of what it does. If it raises fares, it gets decried for being expensive. If it lowers fares, it gets attacked for requiring subsidies. The point is, Amtrak is Amtrak. For better or worse, right now it's all we have. In a few years, we'll have the backbone of California HSR, and FEC Railroad's new passenger service in Florida running along with Amtrak.
You overlooked the #1 argument in *favor* of gifted class -- for many kids, it's the first time in their lives they get to spend extended amounts of time with "their people". I made my first real friend within a month of getting into the class. I was in middle school. It was literally the first time I'd met anybody who was equally smart, and was into the same things I was. I ultimately became friends with pretty much everyone in the class, and we all *might* have tripped over each other in high school (maybe in math club or later, in the Amiga users' group), but the point is that it meant I could finally have real friends after an utterly friendless elementary school experience, in a large school that nevertheless did an amazing job of keeping kids anonymously away from each other.
My brother's mother in law is a gifted teacher, and she agrees 100% that a major need for gifted class is to give kids who've always been untouchably-geeky outsiders a safe environment where they can make friends & avoid self-destructing before high school.
Run a client-side app on your phone that implements a little game of "Simon Says" for any phone number that isn't on your whitelist. It'll take out telemarketers, too, because they don't have 3 minutes to spend playing games to get to you (and most are robodialed anyway).
"Hi, your number isn't on my whitelist, so you need to play a little game of Simon Says in order to make my number ring or reach my voicemail. Press 7 to continue" [caller presses 7]
"I'm sorry, Simon didn't say to continue" [click]
[human caller tries again, presses nothing when told to press 3]
"Good, you understand the rules. SIMON SAYS, enter the first 3 digits of pi to continue, omitting the decimal point"
[caller enters 314]
"Congratulations, you aren't an elected official from the midwestern US! Press 1 to continue"
[caller enters 1, then swears violently a second later when he realizes he was tricked]
"I'm sorry, Simon didn't say to press 1" [click]
[Caller tries again, this time gets a new challenge, like...]
"There are two kinds of people in the world... those who understand binary, and those who don't. If you were starting a company to make X-10 compatible products and wanted a cute name that implied X-10++, but couldn't use X-11 because that's the name of Linux's graphics server, what single digit would arguably convey the same subtle inside joke without trademark infringement? Simon says, enter it now"
[Caller scratches his head, then has a flash of enlightenment and enters 3]
[System congratulates caller, tries to trick him into pressing 1 to continue, fails, throws another brainteaser at the caller, congratulates him for getting it right, tries to trick him into pressing a single digit again, has Simon tell him to press the digit, has simon tell him to press another, then tells him to press a third (without Simon), the caller isn't fooled]
"OK, you win. You clearly want to reach ${me} quite badly, and you're either a friend, a determined bill collector, or some pour soul in Bangalore who's getting paid almost nothing to endure hours and hours of brainteasers from others running the same software I am, so I'll put your call through."
(ring, ring... no answer, goes to voicemail)
"Oh dear, there's no answer. Look, since you've been such a good sport, Simon says to leave your message, then enter your callback number."
(for anybody who's wondering, I really did program something like once as a joke & programming exercise at the first company I worked for after college using Dialogic Visual Voice Pro. I've been meaning to try re-implementing it with Asterisk, and maybe even try to implement it as a client-side app for Android phones, if only to filter out calls from candidates during election years)
>Why don't you just register a single .com domain and run your stuff from there
Because that would make it easy for China, Iran, and other regimes to block users from using his services. That's what he *does* -- he enables people stuck behind oppressive (often government-run) firewalls to get to blocked sites & surf without frustration and/or fear.
The problem with SSD failures is that, unlike spinning drives, RAID won't save you. Several generations of OCZ's Sandforce-based drives had (and still have) weird firmware bugs that caused spontaneous data-suicide. Making matters worse, if you had multiple drives in RAID configuration, and something triggered the bug in one of them, there's a pretty good chance the same bug ended up getting triggered in ALL of them.
RAID was developed to protect data from progressive drive failure. It simply can't deal with drives that will kill themselves at the drop of a hat, for any reason, or no apparent reason at all.
The sad thing is, all this time people have been neurotic about SSDs getting write errors, while totally overlooking their REAL problem -- buggy firmware and insidious mandatory encryption that makes data recovery nearly impossible when your drive dies without warning. One minute, your computer is working fine. The next minute, your data is gone forever because the controller decided to commit suicide and take your data with it.
> Young people are cutting the cords faster than ever
The only "cord" young people have decisively cut are landline telephones, mostly because it's now an expensive service of minimal value compared to mobile phones. Some "young people" who don't view computers as important might try to use their wireless phones for internet access, but for most "cord cutters", it has more to do with avoidance of installation fees and stiff ETFs that would otherwise accompany frequent moves than any genuine desire to cut a metaphorical cord.
Also, let's not forget that "young people" who don't live with their parents or in a college dorm commonly share a house or apartment with at least one other housemate, which cuts the effective cost of the cable and internet bill in half. I don't know a male under 25 with a housemate who doesn't automatically subscribe to the most expensive channel package available, just because they can. Most young adults don't really *care* about monthly service costs until they get their own house or apartment, at which point the staggering cost of buying those services for one person becomes vividly apparent.
As far as cable companies dying, you're forgetting something -- even if content providers like ESPN and MTV were available over the internet directly to consumers, it's not cost-effective for them to individually bill millions of credit cards for each channel. Past a certain point, the transaction fees end up being more than the profit margin. So, we'll always have companies that will go out and negotiate bundle prices with multiple content providers in exchange for a monthly fee that's a fraction of what they'd cost individually. The catch is, it's a slippery slope for everyone involved, and there's tremendous motivation to increase "ARPU" by any means necessary... increasing channels from 2 to 3 (showing the same content, but repeating it more often), bundling 10 more channels for an extra 10 cents, whatever. It's no different than going to BK, and spending almost as much to buy a small hamburger, fries, and drink as you'd have spent on a large Whopper combo.
At the end of the day, under just about any likely scenario, somebody who tries to subscribe to exactly one high-value cable channel they care about (like ESPN, MTV, CNN, or Bravo) will probably end up paying $25. Two channels? $40. Three channels? $62. Twenty channels? $75. And so on, until you're right back to spending $120/month for the half-dozen or so channels you passionately care about, another 20 or so that you might watch once in a blue moon, and 6,000 you don't care about, but had to buy to get 3 of the other 20.
As a practical matter, most laws like this are only directly relevant to companies importing more than a single item for resale -- particularly large corporations like Asda importing items by the shipping crate. The EU doesn't have the resources to go out and inspect the inventory of every Chinese immigrant selling goods straight from Shenzhen from his crowded store in outer London or Amsterdam, let alone scrutinize every item purchased from the US. As a practical matter, it doesn't... it just requires companies like Asda to certify the compliance of the goods they sell, and knows that it can keep 99.9% of noncompliant goods out of most consumers' hands without lifting a finger or paying the salary of a single customs agent.
Here's an easy experiment: go find a small, independent store that sells imported computer parts purchased from Shenzhen. Say, ATX tower cases or USB hubs, owned by a guy who emigrated from China and has family members purchasing for him back in China. Buy the coolest-looking case with a brand name you've never heard of, and a random USB hub. Take them home, and scrutinize the legal compliance of each. If you see a FCC ID (or its European equivalent) etched into the circuit board, look it up... and feign surprise when you discover that it's either associated with some item that hasn't even been sold in 3 years, or was completely made up. Don't forget to check the compliance of things like the power supply, too. More likely than not, both items are technically illegal in the EU, US, or both... and nobody really cares, unless they're a retailer the size of Amazon or Walmart/Asda. Then, they care a lot.
That said, a law that sets standards that aren't already achievable on the assumption that consumers can just ignore it if necessary is a bad law.
> Simply use the net and get all the shows you want WHEN you want and with less commercials to boot.
Oh, really? And where, pray tell, is this magic IP-based virtual cable network that offers realtime IP feeds of the show that's literally running for the first time *right this second* -- as opposed to stuff that's either old (or at least non-realtime), low video quality, openly pirated, or some combination of the three?
There's no technological reason why CNN/MSNBC/Fox/TWC and the rest can't stream HD in realtime, and deliver encrypted copies of the shows you've already tagged for recording in advance & let you have the decryption key the moment the episode officially airs for the first time... but content providers just won't do it. And unfortunately, nothing can make them do it. Maybe the refusal of Dish and others to deal with AMC will prod them to do an end run around them and offer service directly to end users over the internet... but I doubt it.
As much as I'd like to believe that someday, someone will offer a boundary-free service over the internet that works like of like U-verse at the nuts & bolts level, but furnishes 19.2mbps content at original resolution in the native mode, and has a UI that's fully open & customizable by end users with programming backgrounds, I'm sufficiently nihilistic about the American content industry to know that when all is said and done, my monthly "TV-related" expenses would probably end up doubling compared to now anyway. Never, in the history of American TV, have prices genuinely gone down.
Competition from satellite HAS been nice, if only because it brought us Voom 5 years before Comcast even heard of HD, and brought us 200 HD channels on DirecTV when Comcast was still subjecting customers to "death of a thousand macroblocks" 10 new channels at a time, but don't kid yourself. One way or another, the industry is going to find ways to charge us a LOT more for a LITTLE more content, quality, and/or flexibility. And it'll probably compress the hell out of it, too.
Years ago, I laughed at my dad when he told me that color TV pictures were sharper in the 1960s than they were in the 80s and 90s. Then I saw my first DVD, and it sank in for the first time just how badly quality had been completely pwn3d by cost-cutting over the years. Now, the content providers are doing the same to us all over again. We have "HD", but the "HD" we have today via U-verse, Comcast, Dish, and even DirecTV is total and complete crap (minus 2 or 3 showcase channels) compared to the old days of Voom, where the SD channels had bitrates higher than U-verse uses NOW for its HD channels. MPEG-4 is good, but beyond a certain point, you can't fool mother nature. She's going to make sure you get blotches of red sparklies on areas that are supposed to be creamy white, just to remind you that someone is recompressing recompressed 12mbps video down to 6mbps. 10 years from now, we're going to be sold on "Super HD" with 3200x1800 resolution that, if we're lucky, will be indistinguishable from 60fps progressive 1920x1080 encoded at 25-40mbps, and basically just be oversampling with extra marketing and hype.
> Why not have kept the 400 and planted 1000 more?
Because then, the Shuttle would still be sitting in LAX's parking lot, unable to move anywhere due to the trees in the way. And the funding for those 1,000 trees exists only because people want to see the shuttle at the museum. Without a shuttle to motivate cutting down the trees and create a need to replace them with more than twice as many, the money would end up getting spent on beer & hookers, or buying a new tank for the LAPD to accidentally terrorize the neighbors of a drug dealer with.
It was smoldering long before Obama. In a rare showing of nonpartisan unity, both parties have actively beaten it up ever since the end of the Cold War.
I almost wish China would hurry up and announce plans to plant a red flag on the red planet, just to completely freak out everyone in Washington and instantly elevate NASA to the top of their funding priority list. Obviously nobody sane wants a return to the Cold War, but a little healthy rivalry between worthy adversaries can be a good alternative to the complacency we have right now.
On the other hand, NASA could try a little harder to learn how public relations is supposed to work. They need to pull the Puritanical stake out of their butt and visit Russia to see what a real launch party is supposed to be like. Russia's space program *knows* where their funding comes from, and they work hard to keep the funds flowing as freely as the vodka at their launch parties. NASA lost its way when it decided that its mission was to make space exploration boring and routine.
When there's a crisis at ISS, they should have live in-station cameras beaming the raw streaming footage to every news network 24/7 to make it their urgent story of the hour... and find ways to pitch for more funding along the way. When the crew is huddling in a refuge spot because of an impending collision with space junk, CNN should have America hanging on the drama and begging for more. 99 times out of 100, when something bad happens in space, the reaction of most people ends up being, "oh... I hadn't even though about NASA lately. I'd kind of forgotten they even exist." Then they wonder why their funding has dried up.
Over the past 30 years, NASA has managed to transform space from something that's cool and extreme, and make it seem about as exciting as working on an oil rig in the middle of the Gulf of Mexico. No, actually, they've made it seem less appealing and exciting. At least workers on the oil rig get to have uncensored broadband. NASA needs to hook up with the producers of shows like "Turbine Cowboys" and "Iron Men", and get a Journalist-Producer on board the ISS to shoot a season's worth of shows up there. Money can't *buy* the kind of PR a season of "Going Up?" on DiscoveryHD (with teaser that shows elevator running to the 100,000th floor) would get them.
You're preventing 'drive by' attacks, which are the ones likely to bite a *slashdot* user.
As long as Microsoft requires that secure boot keys be unconditionally furnished to end users, it's not a problem. The danger is if they allow vendors to charge extra for them or withhold them. A secure boot key *I* can use to sign my own bootloaders for my own PC is a good thing. It only becomes evil when Dell is allowed to sell $99 PCs w/ad-supported OS that can't be replaced by end users (because within 5 years, a non-locked PC would become an exotic niche item with a $1,800 price premium).
>and still find a way to keep the code signed?
With a certificate bearing the same CN as the original? Low, as long as the bootloader realizes that it's never seen anything signed by s0m3hack3r@foo.to, and presents the user with a dialog that says something like, "You have never booted an OS signed by s0m3hack3r@foo.to, and foo.to is not recognized as a known OSS Organization. Click here to boot into your computer's mini-distro and perform an automated legitimacy lookup (internet access required), or (... options that include 'continue if you trust them' and 'cancel'...)
For a side trip, boot into a mini Linux burned into flash that can grab an ip via dhcp or connect to wifi with ssid/key stored in flash or entered now & wget a lookup of the CN from the UEFI bootloader's organization. Known malware CNs would be blacklisted & identified as such, others could be further researched using Lynx before either continuing the boot (optionally remembering the CN for future boots) or aborting.
Seriously. Learn ARM assembly, practice hitting the bare metal in an Android phone, and get a job working for someone like Nvidia, Qualcomm, Broadcom, Samsung, HTC, or someone comparable. You have a skill almost nobody does anymore, and you know how much more fun assembly is. Screw Java and boring corporate productivity apps. You can have more fun with assembly writing drivers, and make more money while you're at it. :-)
Try opening a second terminal window.
Just click the icon... oh, wait a second. Damn, it just re-selected the one that was already open.
Right-click the icon... crap, no option to do it there, either.
Er... um... (swear violently)
sudo apt-get install ({KDE, Gnome, LXDE, it doesn't matter... anything's better than the steaming mass of poo called 'Unity'})