> Perhaps, but will you be able to easily get a working bluray player in 25 years?
A working bluray player? Maybe, maybe not. A drive capable of reading BD-R discs, regardless of what its own native format might be? Almost certainly and beyond doubt. The last time I checked, CD-ROM is ALREADY somewhere between 20 and 25 years old , and my shiny new BD-R drive is perfectly capable of ripping redbook audio CDs from 1982, as well as a bunch of CD-ROM clipart discs I bought around 1993. In fact, the last time I checked, there are now even utilities to rip the raw data from an un-finalized or "coastered" disc from 10 years ago, and do offline recovery on the files.
I think it's safe to say that any new optical format MUST be formfactor-compatible with CD and DVD in order to succeed from now on. Some future format with larger-diameter discs might appear, but it's almost a given that its spindle hole will be the same size as CD/DVD/BD, and you'll be able to use old discs in it the same way you can use small-diameter CD/DVD discs in normal drives. At this point, there's so much invested in the form factor of CD/DVD/BD discs, it would take an epic shift of almost unprecedented change to overcome it at this point... and really, there's no compelling reason to deviate from it. Optical discs and their drives will never be thin enough to make the thin-at-all-costs crowd happy, and everyone else really doesn't *care* about a millimeter or two.
non-LTH BD-R has a HUGE advantage over any hard drive: you can throw it in a drawer, forget about it for the next 25 years, maybe even let it bake in a hot, humid Florida garage for 5-10 years, and end up with something that's likely to still be readable. There are so many things that can go wrong and break with a normal hard drive over the span of 25 years, the likelihood of ANY hard drive actually working even 10-15 years down the road after years of disuse and questionable storage is basically "nonexistent", and depressingly low even if you've kept it in a 70 degree room with low humidity the whole time.
DVD-R used organic dyes and isn't likely to be a reliable long-term storage medium, but the ORIGINAL (non-LTH) BD-R discs are about as close as you can get with modern media to "carving it in stone". There's even a company (Milleniata?) who makes discs that are basically BD-R media burned to DVD geometry (you need a supported BD-R drive to burn them), and a likely shelf life of a hundred years or more (especially if you burn 2 or 3 copies, and store them in different locations, so you can scrape the bits from all 3 and take advantage of error correction to reconstruct an intact copy decades from now).
VHS has been dead as a consumer format for more than a decade, but there are STILL companies selling new VCRs. What vanished were the cheap consumer models. What remains are heavy-duty pro models designed mainly for recovery and restoration work... and it's a market that's slowly growing as desperate consumers realize they no longer have the players for their old high school band/cheerleading/football tapes their parents made years ago, and they go looking for solutions (or people who can do it for them). Best of all, the patents have all basically expired, so now a smaller company with the ability to machine metal & plastic parts actually CAN step in to take over a market that companies like Sony & Matsushita lost interest in years ago.
For a hint of where the market for spinning drives is going, look at DLP. DLP never totally went away... it just walked away from the low end, then milked the high end for years.
SSDs are getting cheaper, but for raw bulk digital tonnage and petabytes of ripped Blu-ray pr0n, it's still hard to beat spinning hard drives. Manufacturers will just quit making small drives as SSDs catch up, add platters until they can't fit anymore into a 3.5" enclosure, then revisit the past and reintroduce 5.25" hard drives, just like Quantum did ~15 years ago. At some point (probably 10-20 years from now) SSDs might eclipse spinning hard drives, but I wouldn't write them out of the picture TOO soon. We'll be buying them LONG after Joe Sixpack and his kids have forgotten what they are.
Optical media will probably be around longer, as long as Hollywood doesn't manage to kill it off, because it has one concrete advantage: longevity (as long as it's not based on organic dyes). BD-R media is likely to be around (in single, 2, 3, 4, or more) layer forms for a really, really long time.
Prices won't necessarily go up per se, but drives will probably get more expensive over time because the low end will just cease to exist, and manufacturers will try to make the drives bigger, faster, more redundant, (god forbid) repairable, or some permutation of the above, while maintaining the same price points and gradually just eliminating the lower ones until the only spinning drive you can buy is a 5.25" 500TB Western Digital Diplodicus Max with 256GB flashcache for $299.
> Getting the drivers into mainline would be a far better solution.
Or at least stabilize Android's kernel ABI so that every new release doesn't break most/all of the loadable kernel modules that shipped with the phone. 99% of the time, the Linux kernel ABI breaks between versions for no particular reason besides "nobody even bothers to try and NOT break it".
Among some,.ko-breakage is considered a feature and punishment for not open-sourcing drivers. Unfortunately, Qualcomm, nVidia, and Broadcom own the chipset market (for American phones, at least), and no amount of market forces or consumer preferences are likely to make the slightest difference to them, because it's pretty much impossible to make a fully-working American LTE phone without their chips. The overwhelming majority of real changes to successive Android kernels are Google-made, anyway. If Google would just fork Android's kernel once and for all, ignore the mainline going forward (to avoid senseless ABI breakage with every new kernel, possibly attempting to incorporate Linus' latest improvements as faithfully as possible without breaking the ABI), we'd be better off.
The Nexus One rocked. It was groundbreaking, best of breed, and was unquestionably the best Android device money could buy at any price. Every Nexus since then has been... underwhelming, to put it politely.
The Nexus S was a joke, and literally WAS "last year's hardware". By the time Sprint got it, it was already obsolete.
The gNex was a step in the right direction, except they stupidly gave Verizon exclusivity for 6 months & shipped it with a locked bootloader. By the time the gNex hit other carriers, the Galaxy S3 was imminent (and a solid step up hardware-wise).
The Nexus 4 has a sealed battery, no microSD, steals screen real estate for the buttons, and has an insane glass back that cracks if you drop it more than a quarter inch.
Are the Nexi good values? Yeah. Are they competition-crushing best of breed devices that decisively take Android to the next level, and instantly render all Android phones that came before them functionally obsolete? Unfortunately, no. Google seems to have decided to settle for "cheap & throwaway" instead of "bully pulpit awesome".
As of this moment, there is no known "Java" VM that runs under Android. Dalvik begins its life as Java sourcecode, but it's actually "double-compiled" by the time it runs on Android: Java source to Java bytecode, then Java bytecode is compiled to Dalvik bytecode. Lots of Android developers eventually get bitten hard before grasping the true meaning of "Dalvik Isn't Java" (ie, there ARE things that work in Java, but don't work in Dalvik. Runtime dynamic compilation to achieve dependency injection is one example that comes to mind).
Getting Java to run in any kind of compliant manner, with or without Swing, would be a HUGE undertaking that's also prohibited by Oracle and Java's open-source license. Remember, Java is encumbered by patents now owned by Oracle, and a license to use those patents is granted ONLY to users running compliant implementations of Java on x86/AMD64-architecture hardware. You can have a license to use software, without necessarily having a license to make use of patents embodied within it.
If you spent months of your life getting OpenJDK to work under Android (with or without Swing) and published it, you'd be instantly sued by Oracle for patent infringement. Even pre-Oracle, Sun charged shitloads of money to license Java for embedded applications (remember, that's what it was invented for in the first place).
Either way, Java applets under Android aren't happening. Period. Even if you solved the software problems & fought off the lawyers, there's still the tiny problem that Android browsers have no concept of an "Applet".
Sprint is a bad example, because their network is SO dysfunctional right now, your data use on it is self-limiting by virtue of being 80kbps in most places on a GOOD day. In the months leading up to my switch from Sprint to AT&T, my Sprint use averaged 2.5gb/month (thanks mostly to wimax). Since switching to AT&T, I've gone over 3gb almost every single month. When you have 5mbps in a BAD area (usually 10-15mbps, I've seen 26mbps) and USED to get 1mbps in a GREAT area (10mbps when Sprint first launched 4g in south Florida), you can use a LOT more data without even trying.
> The U.S. might be better served by associations of smaller companies that have cross-sharing deals
Oh Jesus/Xenu/FSM, please, no. That's what mobile phone service in the US *USED* to be like 10+ years ago. You couldn't drive from Miami to fucking Orlando or Tampa, or from Dallas to Houston, and use your phone in the other city, without getting assraped by both your provider AND the local company whose network you were roaming on. Southeast Florida was PrimeCo territory, Central Florida was Alltel. Southeast Florida was BellSouth mobility, Central Florida was GTE Mobilenet. AT&T had service in Miami and Orlando, but not Tampa, Naples, or most of the Turnpike between Fort Pierce and Kissimmee. Airtouch or Voicestream (the companies that, post-merger, became T-Mobile) had flickering flames of service in parts of Miami, Orlando, and Tampa, but you were pretty much SOL in the other 95% of the state. MetroPCS was dirt cheap in Miami, but your phone was LITERALLY a paperweight the moment you left Dade+Broward+Palm Beach Counties because they couldn't roam off-network AT ALL.
The day AT&T and Cingular (formerly BellSouth Mobility) merged, customers of both cheered. Ditto, for when Verizon (formerly PrimeCo) merged with Alltel.
Half the reason Sprint became so totally dominant in Florida in the early 2000s (a lead they held up until they completely squandered it over the past year or two) was because they were the first carrier that honestly and truly allowed you to use your phone anywhere in the state (and most of the remainder of the country, for that matter) without paying a cent in roaming fees.
As bad and fucked up as American mobile phone service is today, it was UNFATHOMABLY worse 10-20 years ago. Even back in LATE 2001, I remember going to Dallas with a friend who had AT&T who couldn't use his phone there to make or receive calls without paying something outrageous like $5 per day plus 89c/minute for airtime. The nearest place where he could use his phone without paying to roam was Houston (well, maybe Austin... I'm not sure).
The problem with Ting is that their per-megabyte fees make the actual usage of 5+ gigabytes of data cost-prohibitive. They're a lot cheaper if you honestly and truly use almost no data, but if you DO use a lot of data, Ting's break-even point (compared to T-Mo/Vz/AT&T) is around 250-500mb/month. Beyond that, and you'll be looking at charges of a hundred bucks or more for data that would cost less than half as much on any other network.
Ting's data charges are kind of like mileage overage charges for a leased car. As long as you don't drive much, the lease is cheap. But if you go over, they become EXPONENTIALLY more expensive than what you'd have paid to have just bought the car with a normal loan.
The problem is, AT&T is expensive, but if you don't really care about the cost, everyone else sucks in at least one major way:
Verizon's LTE coverage is "broad" (lots of cities with at least one spot of coverage), but "thin" (airport? Absolutely. downtown? Probably. Indoors in a fringe suburban area where AT&T has solid LTE coverage? Forget about it.) That said, Verizon has usable (if slow) coverage just about anywhere that's remotely populated, and most places that are really, really rural.
T-Mobile is awesome in the areas where they're awesome, and sucks horribly in the other 80-90% of the country. As a practical matter, if you live in a market where they're good, as long as you're in your hometown they'll be really good about 80% of the time, tolerable another 15-18%, and totally suck the remaining 2% or so. When they're fast, they're almost as fast as Verizon. When they're slow? Er... well... when roaming on AT&T, you're limited to EDGE speeds (though it appears they at least have the decency to let you connect via UMTS/HSPA, even if it's throttled, and don't LITERALLY force EDGE connections).
Sprint? OK, total epic suck just about everywhere right now, and likely to stay that way for at least another year or two. They were good up until ~3 years ago, and tolerable up until ~2, but the moment the iPhone hit, their network totally went down the shithole because most of their towers are serviced by literally two T-1 lines... one used for voice calls, control, and 1xRTT data, and one used for EVDO data. They've been putting band-aid fixes in place for the past year and adding another T-1 or two, but ask anybody with Sprint... until the day they switch on LTE and finish their NV deployment in your area, it just keeps getting worse and worse and worse every day. And just when you think it can't possibly get any worse (I was getting ~26kbps down, and ~40kbps up the day I finally walked away after ~14 years with them), you run speedtest one day and see SINGLE-DIGIT speedtest scores. And even in the cities where they've officially "turned on" LTE, their LTE coverage hasn't even gotten to where their wimax coverage WAS two years ago, so they still mostly suck.
AT&T isn't quite perfect in every way... there are areas (mostly fringe suburbia indoors, but NOT necessarily rural areas) where Verizon is slow-but-solid 1-2mbps where AT&T wheezes and struggles to do EDGE, but I've only seen a couple of them... and they're definitely the exception rather than the rule. In the western parts of urban Dade & Broward counties in Florida, AT&T totally spanks Verizon in most places. Verizon TOTALLY cherry-picks dense areas and almost completely ignores suburban areas when deploying LTE. In contrast, AT&T has LTE in fewer markets... but in those markets, pretty much the whole area is solidly blanketed with LTE.
That said, I firmly believe the government needs to forcibly break up AT&T again into at least 3 different companies: one that owns the wireless, one that owns the fiber/copper backhauls & rights of way, and U-verse. Then, the fiber/copper backhaul company would have every incentive to sell fiber to AT&T's competitors... and AT&T would be instantly stopped dead in its tracks from being able to launder surcharges and fees states agreed to 10-20 years ago to fund fiber deployments and use them to build out its wireless network instead.
AT&T's own execs have said publicly that they don't want to spend any money building or maintaining anything with wires anymore. It's time for the FCC to take them up on their offer, and force them to divest their entire wireline holdings to somebody who DOES. Not just the old copper POTS network they don't want, but the fiber runs and rights of way that go along with it as well. Right now, AT&T doesn't want to spend resources on fiber (unless it directly services one of their own towers), and they don't want anybody ELSE to be able to do it, either.
Oracle is really doing its best to kill Java. For them to even *THINK* auto-uninstalling 1.6 is a good idea at this point in time is like the Titanic's crew chopping holes in their lifeboats upon seeing the iceberg...
In terms of paradigm shifts, I'd say one of the more fundamental was the humble 555 IC. You can buy brand new products today that use more or less the exact same die design that was laid out by hand 40 years ago. All that's really changed is that it now comes in tiny SMD packaging with more or less the same silicon inside as always.
The 555 was fundamental in ways that the 74xx chips, and even FPGAs, aren't. 7400-series chips formed the foundation of modern computers, but they gradually (for commercial purposes, at least) evolved and consolidated into PALs, which evolved into CPLDs, which evolved into FPGAs.
The 555 was literally an overnight seismic shift in circuit design that's endured almost unchanged for nearly half a century -- it was useful the day it came out, and remains useful today. In contrast, 7400 logic chips were a vital part of a long chain of evolution that begin with chips that were mostly useless on their own until you combined them like electronic lego bricks, and continued their steady evolution (more or less faithfully demonstrating Moore's Law) up to the present.
The generic Op-amp was another. Over the years, we've refined them and made them cheaper, but to a large extent an op-amp is still an op-amp. And from op-amps, we got chips like the LM386 -- the infamous single-chip.5w amplifier that was pretty much a universal component in cheap consumer electronics gear until class D amplifiers with integrated simple DSP capabilities and I2C volume/effects control arrived like a tsunami in the early 2000s. (For anyone under ~40 or so, anything our parents called a "transistor radio" when we were kids was almost always a "LM386 radio").
> Under the plan, some of the $1.17 trillion that the U.S. owes China would be converted from debt to "equity".
And right there, that's how you know the entire "plan" is pure bullshit. The US doesn't "owe" China anything besides some quantity of $100 (or larger, should the Fed choose to print them) bills -- hundred-dollar bills the Fed can print at will, in any quantity necessary -- once the bonds purchased by Chinese investors (or its government) reach maturity. Period.
US debt is dollar-denominated. US Treasury bonds say right on them that they're payable upon maturity for some quantity of dollars, and the Fed can print as many as it needs to. It's literally impossible for the US to default on its debt absent intentional sabotage by grandstanding elected officials.
The whole act of printing bonds & pretending they're a loan of some kind is theater whose purpose is to keep the amount of dollars growing, while simultaneously limiting the total dollars in active circulation, by encouraging investors to keep reinvesting matured bonds in new bonds.
So, what's to stop China from cashing in on them and collecting its mountain of $100 (or larger) bills? The logistics of trying to move around a hundred billion $100 bills (or 100 million $100,000 bills) and figure out what to DO with them, and the instant collapse in value they'd see that made 90% of their investment worthless if they tried to dump even a small fraction of them too quickly.
It's the same reason why guys who have millions of Bitcoins can't turn them into millions of dollars overnight -- there's a limit to the number of Bitcoins the market can absorb in any given day, and if the market were flooded with them, the amount people would pay for them would plummet quickly. Even if they tried to go on a real estate buying spree, spending even a hundred billion dollars on tangible goods & real estate is HARD.
Things like a 747 might be a few hundred million dollars, but there are only so many 747s available to buy in any given year. If you went to Boeing and said, "I'd like to buy a thousand 787 Dreamliners next year, with remote controls so I can have them take off, then crash them into the desert for my personal entertainment", they'd laugh at you. Well... ok, at this point, they'd probably be delighted to sell you a dozen or two, but their annual production capacity is finite. To get as many as you want, you'd literally have to build a dozen (or more) new factories for them.
Ditto for skyscrapers. To really invest hundreds of billions of dollars in skyscrapers, they'd have to start brand new companies to acquire land, pay people to build them, and find enough places that would even allow them to build them in the first place. And after a certain point, they're going to run out of small towns who'll say yes to almost anything, and the zoning department is going to start saying 'no, you can't build another 90 story building... there's already 37 buildings with 80 or more stories under construction in downtown [Macon/Cedar Rapids/Ocala/Modesto/Youngstown/Reno], and frankly, it's a good thing 89% of those buildings are going to be vacant, because we don't actually *have* the city infrastructure to handle even a fraction of the people who could theoretically live and/or work in the buildings we've already approved, anyway. Now, if you're interested in building our city a subway system, and rebuilding the interstate as a double-deck 24-lane highway with braided ramps at your expense to gain additional development rights...."
Investing billions of dollars in companies by buying stock is even harder than buying goods and skyscrapers. Warren Buffet has written about this problem quite often. When you have THAT MUCH money to invest, there really aren't very many companies that are big enough, with enough shares of sufficiently-expensive stock that's not over-valued, for you to buy enough stock to be worth your time and effort before the share price begins to skyrocket and make it too expensive to be a good investment
> the attacks were 'for the apparent purpose of monitoring the newspaper's China coverage'
Errrrrr... if the main activity of the "attacks" was to scrape articles about China and send them to servers in China (or somewhere else), did it occur to anybody that maybe... just MAYBE... the motive of the hacker(s) might have been to obtain copies of articles that they can't reliably read via normal means courtesy of the Great Firewall of China (or they don't feel like paying for, or they literally CAN'T pay for because the subscription system rejects them and won't allow them to subscribe or pay by some means readily available to them)?
Seriously. In case anybody hasn't noticed, there's been an upswing lately in seemingly-pointless (to Americans) content-aggregation sites that do little more than scrape content from American sites that have a habit of getting randomly blocked by Chinese ISPs (like Wikipedia, XDA-Developers.com, and StackOverflow.com) or sit behind paywalls (like WSJ), then regurgitate it... with ads stripped out, and more importantly, no NEW ads inserted)? Inevitably, from some domain whose name vaguely resembles some plausible combination of English words, but appears to be nonsensical (making up an example, something like "blueteapot.org")?
Remember, when you're ~12-20 years old, just about anything that someone in a position of authority says you aren't allowed to have/read/do instantly becomes your #1 goal in life to obtain. Chinese adults might be largely indifferent to the Great Firewall, but nerdy Chinese teenagers hate it with a passion the moment they find out the real reason why sites mysteriously quit working for no obvious reason... especially when they're trying to finish their Java programming homework, and discover that some idiot decided to block StackOverflow this week.
Of course, it *could* also just be that some powerful/wealthy Chinese official got pissed off about something the WSJ wrote, and paid a bunch of hackers to get revenge for him, but when I see an "attack" whose main consequence is the scraping and republishing of content likely to be of interest to people in China, my first thought is, "sigh... the things those poor guys have to do to get around the Great Firewall..."
> real-world Linux installation (where the mere presence of "root" nullifies any advantage of "secure boot").
You haven't had the misfortune of experiencing Linux on non-x86/AMD64 platforms, I see. Out in the ARM hinterlands, things aren't nearly as friendly. TI, in particular, has some UNBELIEVABLY nasty features baked into their SoC (used by Motorola for Android phones, among other things) that allow the manufacturer to dictate what memory, flash, and i/o ports you can read or modify based upon where the code is executing from... as well as enforce mandatory encryption for flash, so even if you physically desoldered the chips and programmed them independently of the device before soldering them back in, they wouldn't work. There are things that code executing from RAM (or some range of memory addresses) on a Motorola Photon/Electrify/Atrix2 is simply NOT ALLOWED TO DO, and those restrictions are enforced by the SoC and its integrated peripherals themselves.
In other words, on an ARM platform, even ROOT can be shackled and chained into obedience to our corporate overlord masters.
> An inbound port is open to the entire world, anyone can connect, and, (baring any on-device security), > they can do pretty much anything the device is capable of doing.
And 9 times out of 10, unless the homeowner couldn't figure out how to do it, any device that accepts incoming connections on a port probably has a port from the router's public IP address forwarded to its internal IP address *anyway*.
Yes, barring device security, they can do whatever they'd like. That's why the device HAS security. So they can't.
The biggest problem with internet cameras and DVRs isn't the fact that they can use UPnP to "punch holes" -- it's the fact that 99.9% of the damn cameras don't allow you to authenticate via SSL (valid certificate or not), and instead send your login credentials in the clear over the wi-fi network at Starbucks. I wish to ${deity} that routers had a "reverse https proxy" function that would accept inbound https connections, strip the ssl, and transparently forward the traffic to the same port of an internal IP address where there's a device that's too stupid to know how to do SSL.
I won't be losing sleep tonight worrying about my cameras' ability to coax the router into forwarding arbitrary ports to them. I'd lose quite a bit of sleep if I didn't have the internet-connected camera in my bedroom wired up to the burglar alarm through a relay that cuts the power to it whenever the alarm isn't in "away" mode, and a similar relay that cuts power to the switch connecting those cameras to the router. Technically, I could have gotten away with just one relay on the switch, but I couldn't sleep with the camera's red light blinking at me regardless of whether or not it was connected to the router at the time.
> Newegg just spent a lot of money fighting a lawsuit which has now established a precedent
This is actually an extraordinarily rare event, which is why it's being so rightfully celebrated. Patent trolls almost NEVER allow their cases to move all the way to final judgment... they either get the other party to settle for undisclosed terms, or they themselves withdraw the suit. If they settle for undisclosed terms, the court never actually rules upon whether or not the patent is valid, so they can keep using it against others. If they drop the case, the court won't allow them to pursue it again against the same defendants, but they can keep doing it to other people.
The app itself was kind of buggy & brittle, but Karsten Obarski is basically the father of modern game music that doesn't suck. OK, Chris Hülsbeck was breaking plenty of ground of his own & blowing us away with HIS work (and continues to do so today), but Obi was the one who gave us a programmer-friendly toolchain & notation for composition, editing, *and* incorporation of the music into working software. It fit the typical game-development workflow of its era like a glove, and almost overnight set the standard for what Amiga (and later, PC) owers *expected* from computer-generated music.
Jumpman: set the standard for 'playability' & 'fun'. I remember making fun of it when I saw the underwhelming graphics, but it had me hooked the first time I played it. Truly, one of the best games ever. Decades later, it's STILL playable
Archon: what can I say? It started where chess left off, hit the ground running, and just *oozed* "epic win" for concept & gameplay.
Barbarian: the game that INVENTED the concept of a "fatality" move
> A unionized pilot in USA for one of the big airlines makes $250k to $500k.
When, in the 1970s? My brother's a pilot. He's not poor, but the days of pilots making 250-500k/year (in the US, at least) ended with the bankruptcies of PanAm/TWA/Eastern, and the disappearance of navigators, flight engineers, and the 727. His hourly pay looks impressive, until you consider the fact that he literally gets paid only when the plane is in the air, and the day or two he spends commuting to and from his first/last airport, the time he spends overnighting, and the hours he has to BE at the airport, but isn't actually getting paid or doing anything. When you factor in the high unemployment rate for American pilots who entered the workforce after 2000 (first to be laid off, last to be called back), it's not exactly what you'd call a lucrative career anymore.
Worst of all, if he ever DID decide to pursue another career and declined to go back after being laid off, his career as a pilot would be fried and gone forever -- he'd lose his type ratings, and once you lose them, it's almost impossible to get hired by another airline without them. That's the dilemma he's forced into every couple of years -- take that job as a math professor, and forget about EVER flying a jet again after having gone through hell (and almost a hundred thousand dollars renting planes to get his flight time up, and years of poverty & taking real risks with his life daily as a flight instructor) getting there in the first place.
I don't know about 1990, but we definitely had some pretty nasty air pollution about 40-60 years ago. When I was < 10 and growing up in Ohio circa 1980, I remember that air pollution was pretty much everywhere, even in smaller cities, like the Warren-Youngstown-Sharon area roughly halfway between Cleveland and Pittsburgh. My first "omg" memory of Florida was looking up during recess one day about a month after we moved there, and freaking out because I could see the full moon in broad daylight. That was something you never, EVER saw in Ohio. Or at least something *I* had no memory of ever seeing.
Hell, I spent July 5, 1994 in New York, and remember BARELY being able to see the Twin Towers from Midtown. The whole city smelled like a burning log in a fireplace. Likewise, I spent a week in Los Angeles sometime in August 1996, and remember driving into L.A. on LaCienega drive... I made it over the mountain, and saw the famous vista with LA (well, OK, I guess it was actually Beverly Hills) spread out in front of me... except you couldn't actually see anything except faint rooftops a mile or two away, and a sea of opaque smog. In LA's defense, though, its smog didn't really have any particular odor. It was opaque to a degree I'd never seen in my life, but other than obscuring most of the views, it didn't really bother me.
> I've had fraudulent transactions reversed quite a few times over the years with my debit card.
There's a big, HUGE difference. The difference isn't that the charges won't EVENTUALLY get reversed... the difference is that one immediately empties your checking account, while the other is mostly an abstract paper loss.
If someone steals your credit card and makes fraudulent purchases, you submit the paperwork for the fraud claim, and that's pretty much the end of it until they finish their investigation.
If someone steals your DEBIT card, they can drain your checking account to zero, and possibly even overdraw it by a thousand dollars or more. If your finances are shaky to begin with, having your checking account emptied or technically-overdrawn for a week or longer can have financially devastating consequences & set off a chain reaction of late fees, penalties, credit-score lowering, and reactions to that credit-score lowering that's almost impossible to stop, let alone undo, once it begins. When your checking account is overdrawn by more than $2,000 in fraudulent charges, even your direct-deposited paycheck is going to partially or completely vaporize into a puff of antimoney smoke. If you have a savings account that's liked to the checking account for overdraft protection, they can drain your savings account to zero, too.
It's hard to over-emphasize just how completely you can be fucked by debit card fraud, and the fallout (in the form of fees and penalties from others, plus damage to your credit score) can persist and keep doing damage LONG after the fraudulent purchases themselves have been reversed.
You're forgetting about "textbook inertia" and fact that anything in a public school science textbook that remotely involves "evolution" has already been diluted, whitewashed, and minimized to make the book acceptable to the state's evangelical Christian gatekeepers. There are probably kids sitting in middle school classrooms *RIGHT NOW* reading more or less the same content their parents did, more for political and budgetary reasons than anything.
I kid you not... when I was in sixth grade (late 80s), my science teacher handed out brand new textbooks on the first day of school. After handing them out, he became visibly irate, then told us to open the book to page 234, tear it out, and throw it away after class without reading it, "because the school board decided it might make us question our religious beliefs, so they ordered teachers to have the offending page removed." He informed us that the trashcan was "over there" (pointing to a spot that was kind of by the door, but far enough out of the way that you'd have had to be really determined to use it) and made a point of looking the other way as we left at the end of class. It was a brilliant act of subversion that could have probably gotten him fired, because you can bet every single kid in the class read the page. Future classes never got to see the page (well, not counting the photocopies some of my friends and I made & passed out like contraband to some younger friends when we were in 8th grade), but he was able to make a small difference in the lives of at least one class. I know it was page 234, because I still have it;-)
> Except that your not technically buying it, your purchase is subsidised by the contract.
Legally irrelevant. You own your phone 100% the moment you walk out the door with it. The fact that you owe the carrier a hefty ETF if you don't satisfy the contract is why.
For various reasons, American mobile phone carriers have ALWAYS been prohibited from directly leasing phone hardware to consumers. BellSouth Mobility did its best to get the policy overturned around 1991 (and went into high gear after Hurricane Andrew, using it as their perfect pretense), and they got smacked down by the government every time.
I believe the rationale back then was that mobile phone manufacturing was an oligopoly (basically, Motorola owned the US market), and if carriers could lease phones to consumers, the manufacturers would jack up the retail cost to levels that made purchase by end users economically untenable, then rebate most of it back to their biggest customers (ie, carriers).
In AT&T's partial defense, they require real data plans because the experience of just about every wireless carrier on earth has been that an iPhone or Android phone in the hands of someone who "doesn't need a data plan, because they never use data" is a ticking timebomb waiting to explode into an ugly & expensive customer service nightmare.
The extortionately expensive per-kb/mb fees charged for adhoc data use don't help, of course, but the fact remains, it's almost IMPOSSIBLE to use an iPhone or Android phone without EVER using a single kilobyte of billable wireless data service. Very little data, sure. But ZERO data? Damn near impossible without crippling the phone.
That's not to say AT&T doesn't use this fact as an opportunity to make even more money... but honestly, using an iPhone or Android phone on a wireless plan with ZERO kb/mb per month of cheap/included data is a financial disaster waiting to happen, and you really can't really blame ANY carrier for not wanting to voluntarily put itself into the position of having to spend hours fighting with a livid customer over a $7.62 charge for data "they didn't use" (but their phone DID, in the background).
Slashdot Mirror
> Perhaps, but will you be able to easily get a working bluray player in 25 years?
A working bluray player? Maybe, maybe not. A drive capable of reading BD-R discs, regardless of what its own native format might be? Almost certainly and beyond doubt. The last time I checked, CD-ROM is ALREADY somewhere between 20 and 25 years old , and my shiny new BD-R drive is perfectly capable of ripping redbook audio CDs from 1982, as well as a bunch of CD-ROM clipart discs I bought around 1993. In fact, the last time I checked, there are now even utilities to rip the raw data from an un-finalized or "coastered" disc from 10 years ago, and do offline recovery on the files.
I think it's safe to say that any new optical format MUST be formfactor-compatible with CD and DVD in order to succeed from now on. Some future format with larger-diameter discs might appear, but it's almost a given that its spindle hole will be the same size as CD/DVD/BD, and you'll be able to use old discs in it the same way you can use small-diameter CD/DVD discs in normal drives. At this point, there's so much invested in the form factor of CD/DVD/BD discs, it would take an epic shift of almost unprecedented change to overcome it at this point... and really, there's no compelling reason to deviate from it. Optical discs and their drives will never be thin enough to make the thin-at-all-costs crowd happy, and everyone else really doesn't *care* about a millimeter or two.
non-LTH BD-R has a HUGE advantage over any hard drive: you can throw it in a drawer, forget about it for the next 25 years, maybe even let it bake in a hot, humid Florida garage for 5-10 years, and end up with something that's likely to still be readable. There are so many things that can go wrong and break with a normal hard drive over the span of 25 years, the likelihood of ANY hard drive actually working even 10-15 years down the road after years of disuse and questionable storage is basically "nonexistent", and depressingly low even if you've kept it in a 70 degree room with low humidity the whole time.
DVD-R used organic dyes and isn't likely to be a reliable long-term storage medium, but the ORIGINAL (non-LTH) BD-R discs are about as close as you can get with modern media to "carving it in stone". There's even a company (Milleniata?) who makes discs that are basically BD-R media burned to DVD geometry (you need a supported BD-R drive to burn them), and a likely shelf life of a hundred years or more (especially if you burn 2 or 3 copies, and store them in different locations, so you can scrape the bits from all 3 and take advantage of error correction to reconstruct an intact copy decades from now).
VHS has been dead as a consumer format for more than a decade, but there are STILL companies selling new VCRs. What vanished were the cheap consumer models. What remains are heavy-duty pro models designed mainly for recovery and restoration work... and it's a market that's slowly growing as desperate consumers realize they no longer have the players for their old high school band/cheerleading/football tapes their parents made years ago, and they go looking for solutions (or people who can do it for them). Best of all, the patents have all basically expired, so now a smaller company with the ability to machine metal & plastic parts actually CAN step in to take over a market that companies like Sony & Matsushita lost interest in years ago.
For a hint of where the market for spinning drives is going, look at DLP. DLP never totally went away... it just walked away from the low end, then milked the high end for years.
SSDs are getting cheaper, but for raw bulk digital tonnage and petabytes of ripped Blu-ray pr0n, it's still hard to beat spinning hard drives. Manufacturers will just quit making small drives as SSDs catch up, add platters until they can't fit anymore into a 3.5" enclosure, then revisit the past and reintroduce 5.25" hard drives, just like Quantum did ~15 years ago. At some point (probably 10-20 years from now) SSDs might eclipse spinning hard drives, but I wouldn't write them out of the picture TOO soon. We'll be buying them LONG after Joe Sixpack and his kids have forgotten what they are.
Optical media will probably be around longer, as long as Hollywood doesn't manage to kill it off, because it has one concrete advantage: longevity (as long as it's not based on organic dyes). BD-R media is likely to be around (in single, 2, 3, 4, or more) layer forms for a really, really long time.
Prices won't necessarily go up per se, but drives will probably get more expensive over time because the low end will just cease to exist, and manufacturers will try to make the drives bigger, faster, more redundant, (god forbid) repairable, or some permutation of the above, while maintaining the same price points and gradually just eliminating the lower ones until the only spinning drive you can buy is a 5.25" 500TB Western Digital Diplodicus Max with 256GB flashcache for $299.
> Getting the drivers into mainline would be a far better solution.
Or at least stabilize Android's kernel ABI so that every new release doesn't break most/all of the loadable kernel modules that shipped with the phone. 99% of the time, the Linux kernel ABI breaks between versions for no particular reason besides "nobody even bothers to try and NOT break it".
Among some, .ko-breakage is considered a feature and punishment for not open-sourcing drivers. Unfortunately, Qualcomm, nVidia, and Broadcom own the chipset market (for American phones, at least), and no amount of market forces or consumer preferences are likely to make the slightest difference to them, because it's pretty much impossible to make a fully-working American LTE phone without their chips. The overwhelming majority of real changes to successive Android kernels are Google-made, anyway. If Google would just fork Android's kernel once and for all, ignore the mainline going forward (to avoid senseless ABI breakage with every new kernel, possibly attempting to incorporate Linus' latest improvements as faithfully as possible without breaking the ABI), we'd be better off.
The Nexus One rocked. It was groundbreaking, best of breed, and was unquestionably the best Android device money could buy at any price. Every Nexus since then has been... underwhelming, to put it politely.
The Nexus S was a joke, and literally WAS "last year's hardware". By the time Sprint got it, it was already obsolete.
The gNex was a step in the right direction, except they stupidly gave Verizon exclusivity for 6 months & shipped it with a locked bootloader. By the time the gNex hit other carriers, the Galaxy S3 was imminent (and a solid step up hardware-wise).
The Nexus 4 has a sealed battery, no microSD, steals screen real estate for the buttons, and has an insane glass back that cracks if you drop it more than a quarter inch.
Are the Nexi good values? Yeah. Are they competition-crushing best of breed devices that decisively take Android to the next level, and instantly render all Android phones that came before them functionally obsolete? Unfortunately, no. Google seems to have decided to settle for "cheap & throwaway" instead of "bully pulpit awesome".
As of this moment, there is no known "Java" VM that runs under Android. Dalvik begins its life as Java sourcecode, but it's actually "double-compiled" by the time it runs on Android: Java source to Java bytecode, then Java bytecode is compiled to Dalvik bytecode. Lots of Android developers eventually get bitten hard before grasping the true meaning of "Dalvik Isn't Java" (ie, there ARE things that work in Java, but don't work in Dalvik. Runtime dynamic compilation to achieve dependency injection is one example that comes to mind).
Getting Java to run in any kind of compliant manner, with or without Swing, would be a HUGE undertaking that's also prohibited by Oracle and Java's open-source license. Remember, Java is encumbered by patents now owned by Oracle, and a license to use those patents is granted ONLY to users running compliant implementations of Java on x86/AMD64-architecture hardware. You can have a license to use software, without necessarily having a license to make use of patents embodied within it.
If you spent months of your life getting OpenJDK to work under Android (with or without Swing) and published it, you'd be instantly sued by Oracle for patent infringement. Even pre-Oracle, Sun charged shitloads of money to license Java for embedded applications (remember, that's what it was invented for in the first place).
Either way, Java applets under Android aren't happening. Period. Even if you solved the software problems & fought off the lawyers, there's still the tiny problem that Android browsers have no concept of an "Applet".
Sprint is a bad example, because their network is SO dysfunctional right now, your data use on it is self-limiting by virtue of being 80kbps in most places on a GOOD day. In the months leading up to my switch from Sprint to AT&T, my Sprint use averaged 2.5gb/month (thanks mostly to wimax). Since switching to AT&T, I've gone over 3gb almost every single month. When you have 5mbps in a BAD area (usually 10-15mbps, I've seen 26mbps) and USED to get 1mbps in a GREAT area (10mbps when Sprint first launched 4g in south Florida), you can use a LOT more data without even trying.
> The U.S. might be better served by associations of smaller companies that have cross-sharing deals
Oh Jesus/Xenu/FSM, please, no. That's what mobile phone service in the US *USED* to be like 10+ years ago. You couldn't drive from Miami to fucking Orlando or Tampa, or from Dallas to Houston, and use your phone in the other city, without getting assraped by both your provider AND the local company whose network you were roaming on. Southeast Florida was PrimeCo territory, Central Florida was Alltel. Southeast Florida was BellSouth mobility, Central Florida was GTE Mobilenet. AT&T had service in Miami and Orlando, but not Tampa, Naples, or most of the Turnpike between Fort Pierce and Kissimmee. Airtouch or Voicestream (the companies that, post-merger, became T-Mobile) had flickering flames of service in parts of Miami, Orlando, and Tampa, but you were pretty much SOL in the other 95% of the state. MetroPCS was dirt cheap in Miami, but your phone was LITERALLY a paperweight the moment you left Dade+Broward+Palm Beach Counties because they couldn't roam off-network AT ALL.
The day AT&T and Cingular (formerly BellSouth Mobility) merged, customers of both cheered. Ditto, for when Verizon (formerly PrimeCo) merged with Alltel.
Half the reason Sprint became so totally dominant in Florida in the early 2000s (a lead they held up until they completely squandered it over the past year or two) was because they were the first carrier that honestly and truly allowed you to use your phone anywhere in the state (and most of the remainder of the country, for that matter) without paying a cent in roaming fees.
As bad and fucked up as American mobile phone service is today, it was UNFATHOMABLY worse 10-20 years ago. Even back in LATE 2001, I remember going to Dallas with a friend who had AT&T who couldn't use his phone there to make or receive calls without paying something outrageous like $5 per day plus 89c/minute for airtime. The nearest place where he could use his phone without paying to roam was Houston (well, maybe Austin... I'm not sure).
The problem with Ting is that their per-megabyte fees make the actual usage of 5+ gigabytes of data cost-prohibitive. They're a lot cheaper if you honestly and truly use almost no data, but if you DO use a lot of data, Ting's break-even point (compared to T-Mo/Vz/AT&T) is around 250-500mb/month. Beyond that, and you'll be looking at charges of a hundred bucks or more for data that would cost less than half as much on any other network.
Ting's data charges are kind of like mileage overage charges for a leased car. As long as you don't drive much, the lease is cheap. But if you go over, they become EXPONENTIALLY more expensive than what you'd have paid to have just bought the car with a normal loan.
The problem is, AT&T is expensive, but if you don't really care about the cost, everyone else sucks in at least one major way:
Verizon's LTE coverage is "broad" (lots of cities with at least one spot of coverage), but "thin" (airport? Absolutely. downtown? Probably. Indoors in a fringe suburban area where AT&T has solid LTE coverage? Forget about it.) That said, Verizon has usable (if slow) coverage just about anywhere that's remotely populated, and most places that are really, really rural.
T-Mobile is awesome in the areas where they're awesome, and sucks horribly in the other 80-90% of the country. As a practical matter, if you live in a market where they're good, as long as you're in your hometown they'll be really good about 80% of the time, tolerable another 15-18%, and totally suck the remaining 2% or so. When they're fast, they're almost as fast as Verizon. When they're slow? Er... well... when roaming on AT&T, you're limited to EDGE speeds (though it appears they at least have the decency to let you connect via UMTS/HSPA, even if it's throttled, and don't LITERALLY force EDGE connections).
Sprint? OK, total epic suck just about everywhere right now, and likely to stay that way for at least another year or two. They were good up until ~3 years ago, and tolerable up until ~2, but the moment the iPhone hit, their network totally went down the shithole because most of their towers are serviced by literally two T-1 lines... one used for voice calls, control, and 1xRTT data, and one used for EVDO data. They've been putting band-aid fixes in place for the past year and adding another T-1 or two, but ask anybody with Sprint... until the day they switch on LTE and finish their NV deployment in your area, it just keeps getting worse and worse and worse every day. And just when you think it can't possibly get any worse (I was getting ~26kbps down, and ~40kbps up the day I finally walked away after ~14 years with them), you run speedtest one day and see SINGLE-DIGIT speedtest scores. And even in the cities where they've officially "turned on" LTE, their LTE coverage hasn't even gotten to where their wimax coverage WAS two years ago, so they still mostly suck.
AT&T isn't quite perfect in every way... there are areas (mostly fringe suburbia indoors, but NOT necessarily rural areas) where Verizon is slow-but-solid 1-2mbps where AT&T wheezes and struggles to do EDGE, but I've only seen a couple of them... and they're definitely the exception rather than the rule. In the western parts of urban Dade & Broward counties in Florida, AT&T totally spanks Verizon in most places. Verizon TOTALLY cherry-picks dense areas and almost completely ignores suburban areas when deploying LTE. In contrast, AT&T has LTE in fewer markets... but in those markets, pretty much the whole area is solidly blanketed with LTE.
That said, I firmly believe the government needs to forcibly break up AT&T again into at least 3 different companies: one that owns the wireless, one that owns the fiber/copper backhauls & rights of way, and U-verse. Then, the fiber/copper backhaul company would have every incentive to sell fiber to AT&T's competitors... and AT&T would be instantly stopped dead in its tracks from being able to launder surcharges and fees states agreed to 10-20 years ago to fund fiber deployments and use them to build out its wireless network instead.
AT&T's own execs have said publicly that they don't want to spend any money building or maintaining anything with wires anymore. It's time for the FCC to take them up on their offer, and force them to divest their entire wireline holdings to somebody who DOES. Not just the old copper POTS network they don't want, but the fiber runs and rights of way that go along with it as well. Right now, AT&T doesn't want to spend resources on fiber (unless it directly services one of their own towers), and they don't want anybody ELSE to be able to do it, either.
Oracle is really doing its best to kill Java. For them to even *THINK* auto-uninstalling 1.6 is a good idea at this point in time is like the Titanic's crew chopping holes in their lifeboats upon seeing the iceberg...
In terms of paradigm shifts, I'd say one of the more fundamental was the humble 555 IC. You can buy brand new products today that use more or less the exact same die design that was laid out by hand 40 years ago. All that's really changed is that it now comes in tiny SMD packaging with more or less the same silicon inside as always.
The 555 was fundamental in ways that the 74xx chips, and even FPGAs, aren't. 7400-series chips formed the foundation of modern computers, but they gradually (for commercial purposes, at least) evolved and consolidated into PALs, which evolved into CPLDs, which evolved into FPGAs.
The 555 was literally an overnight seismic shift in circuit design that's endured almost unchanged for nearly half a century -- it was useful the day it came out, and remains useful today. In contrast, 7400 logic chips were a vital part of a long chain of evolution that begin with chips that were mostly useless on their own until you combined them like electronic lego bricks, and continued their steady evolution (more or less faithfully demonstrating Moore's Law) up to the present.
The generic Op-amp was another. Over the years, we've refined them and made them cheaper, but to a large extent an op-amp is still an op-amp. And from op-amps, we got chips like the LM386 -- the infamous single-chip .5w amplifier that was pretty much a universal component in cheap consumer electronics gear until class D amplifiers with integrated simple DSP capabilities and I2C volume/effects control arrived like a tsunami in the early 2000s. (For anyone under ~40 or so, anything our parents called a "transistor radio" when we were kids was almost always a "LM386 radio").
> Under the plan, some of the $1.17 trillion that the U.S. owes China would be converted from debt to "equity".
And right there, that's how you know the entire "plan" is pure bullshit. The US doesn't "owe" China anything besides some quantity of $100 (or larger, should the Fed choose to print them) bills -- hundred-dollar bills the Fed can print at will, in any quantity necessary -- once the bonds purchased by Chinese investors (or its government) reach maturity. Period.
US debt is dollar-denominated. US Treasury bonds say right on them that they're payable upon maturity for some quantity of dollars, and the Fed can print as many as it needs to. It's literally impossible for the US to default on its debt absent intentional sabotage by grandstanding elected officials.
The whole act of printing bonds & pretending they're a loan of some kind is theater whose purpose is to keep the amount of dollars growing, while simultaneously limiting the total dollars in active circulation, by encouraging investors to keep reinvesting matured bonds in new bonds.
So, what's to stop China from cashing in on them and collecting its mountain of $100 (or larger) bills? The logistics of trying to move around a hundred billion $100 bills (or 100 million $100,000 bills) and figure out what to DO with them, and the instant collapse in value they'd see that made 90% of their investment worthless if they tried to dump even a small fraction of them too quickly.
It's the same reason why guys who have millions of Bitcoins can't turn them into millions of dollars overnight -- there's a limit to the number of Bitcoins the market can absorb in any given day, and if the market were flooded with them, the amount people would pay for them would plummet quickly. Even if they tried to go on a real estate buying spree, spending even a hundred billion dollars on tangible goods & real estate is HARD.
Things like a 747 might be a few hundred million dollars, but there are only so many 747s available to buy in any given year. If you went to Boeing and said, "I'd like to buy a thousand 787 Dreamliners next year, with remote controls so I can have them take off, then crash them into the desert for my personal entertainment", they'd laugh at you. Well... ok, at this point, they'd probably be delighted to sell you a dozen or two, but their annual production capacity is finite. To get as many as you want, you'd literally have to build a dozen (or more) new factories for them.
Ditto for skyscrapers. To really invest hundreds of billions of dollars in skyscrapers, they'd have to start brand new companies to acquire land, pay people to build them, and find enough places that would even allow them to build them in the first place. And after a certain point, they're going to run out of small towns who'll say yes to almost anything, and the zoning department is going to start saying 'no, you can't build another 90 story building... there's already 37 buildings with 80 or more stories under construction in downtown [Macon/Cedar Rapids/Ocala/Modesto/Youngstown/Reno], and frankly, it's a good thing 89% of those buildings are going to be vacant, because we don't actually *have* the city infrastructure to handle even a fraction of the people who could theoretically live and/or work in the buildings we've already approved, anyway. Now, if you're interested in building our city a subway system, and rebuilding the interstate as a double-deck 24-lane highway with braided ramps at your expense to gain additional development rights...."
Investing billions of dollars in companies by buying stock is even harder than buying goods and skyscrapers. Warren Buffet has written about this problem quite often. When you have THAT MUCH money to invest, there really aren't very many companies that are big enough, with enough shares of sufficiently-expensive stock that's not over-valued, for you to buy enough stock to be worth your time and effort before the share price begins to skyrocket and make it too expensive to be a good investment
> the attacks were 'for the apparent purpose of monitoring the newspaper's China coverage'
Errrrrr... if the main activity of the "attacks" was to scrape articles about China and send them to servers in China (or somewhere else), did it occur to anybody that maybe... just MAYBE... the motive of the hacker(s) might have been to obtain copies of articles that they can't reliably read via normal means courtesy of the Great Firewall of China (or they don't feel like paying for, or they literally CAN'T pay for because the subscription system rejects them and won't allow them to subscribe or pay by some means readily available to them)?
Seriously. In case anybody hasn't noticed, there's been an upswing lately in seemingly-pointless (to Americans) content-aggregation sites that do little more than scrape content from American sites that have a habit of getting randomly blocked by Chinese ISPs (like Wikipedia, XDA-Developers.com, and StackOverflow.com) or sit behind paywalls (like WSJ), then regurgitate it... with ads stripped out, and more importantly, no NEW ads inserted)? Inevitably, from some domain whose name vaguely resembles some plausible combination of English words, but appears to be nonsensical (making up an example, something like "blueteapot.org")?
Remember, when you're ~12-20 years old, just about anything that someone in a position of authority says you aren't allowed to have/read/do instantly becomes your #1 goal in life to obtain. Chinese adults might be largely indifferent to the Great Firewall, but nerdy Chinese teenagers hate it with a passion the moment they find out the real reason why sites mysteriously quit working for no obvious reason... especially when they're trying to finish their Java programming homework, and discover that some idiot decided to block StackOverflow this week.
Of course, it *could* also just be that some powerful/wealthy Chinese official got pissed off about something the WSJ wrote, and paid a bunch of hackers to get revenge for him, but when I see an "attack" whose main consequence is the scraping and republishing of content likely to be of interest to people in China, my first thought is, "sigh... the things those poor guys have to do to get around the Great Firewall..."
> real-world Linux installation (where the mere presence of "root" nullifies any advantage of "secure boot").
You haven't had the misfortune of experiencing Linux on non-x86/AMD64 platforms, I see. Out in the ARM hinterlands, things aren't nearly as friendly. TI, in particular, has some UNBELIEVABLY nasty features baked into their SoC (used by Motorola for Android phones, among other things) that allow the manufacturer to dictate what memory, flash, and i/o ports you can read or modify based upon where the code is executing from... as well as enforce mandatory encryption for flash, so even if you physically desoldered the chips and programmed them independently of the device before soldering them back in, they wouldn't work. There are things that code executing from RAM (or some range of memory addresses) on a Motorola Photon/Electrify/Atrix2 is simply NOT ALLOWED TO DO, and those restrictions are enforced by the SoC and its integrated peripherals themselves.
In other words, on an ARM platform, even ROOT can be shackled and chained into obedience to our corporate overlord masters.
> An inbound port is open to the entire world, anyone can connect, and, (baring any on-device security),
> they can do pretty much anything the device is capable of doing.
And 9 times out of 10, unless the homeowner couldn't figure out how to do it, any device that accepts incoming connections on a port probably has a port from the router's public IP address forwarded to its internal IP address *anyway*.
Yes, barring device security, they can do whatever they'd like. That's why the device HAS security. So they can't.
The biggest problem with internet cameras and DVRs isn't the fact that they can use UPnP to "punch holes" -- it's the fact that 99.9% of the damn cameras don't allow you to authenticate via SSL (valid certificate or not), and instead send your login credentials in the clear over the wi-fi network at Starbucks. I wish to ${deity} that routers had a "reverse https proxy" function that would accept inbound https connections, strip the ssl, and transparently forward the traffic to the same port of an internal IP address where there's a device that's too stupid to know how to do SSL.
I won't be losing sleep tonight worrying about my cameras' ability to coax the router into forwarding arbitrary ports to them. I'd lose quite a bit of sleep if I didn't have the internet-connected camera in my bedroom wired up to the burglar alarm through a relay that cuts the power to it whenever the alarm isn't in "away" mode, and a similar relay that cuts power to the switch connecting those cameras to the router. Technically, I could have gotten away with just one relay on the switch, but I couldn't sleep with the camera's red light blinking at me regardless of whether or not it was connected to the router at the time.
> Newegg just spent a lot of money fighting a lawsuit which has now established a precedent
This is actually an extraordinarily rare event, which is why it's being so rightfully celebrated. Patent trolls almost NEVER allow their cases to move all the way to final judgment... they either get the other party to settle for undisclosed terms, or they themselves withdraw the suit. If they settle for undisclosed terms, the court never actually rules upon whether or not the patent is valid, so they can keep using it against others. If they drop the case, the court won't allow them to pursue it again against the same defendants, but they can keep doing it to other people.
The app itself was kind of buggy & brittle, but Karsten Obarski is basically the father of modern game music that doesn't suck. OK, Chris Hülsbeck was breaking plenty of ground of his own & blowing us away with HIS work (and continues to do so today), but Obi was the one who gave us a programmer-friendly toolchain & notation for composition, editing, *and* incorporation of the music into working software. It fit the typical game-development workflow of its era like a glove, and almost overnight set the standard for what Amiga (and later, PC) owers *expected* from computer-generated music.
Jumpman: set the standard for 'playability' & 'fun'. I remember making fun of it when I saw the underwhelming graphics, but it had me hooked the first time I played it. Truly, one of the best games ever. Decades later, it's STILL playable
Archon: what can I say? It started where chess left off, hit the ground running, and just *oozed* "epic win" for concept & gameplay.
Barbarian: the game that INVENTED the concept of a "fatality" move
https://www.youtube.com/watch?v=C4Ii_YfJNvw&feature=youtube_gdata_player
> A unionized pilot in USA for one of the big airlines makes $250k to $500k.
When, in the 1970s? My brother's a pilot. He's not poor, but the days of pilots making 250-500k/year (in the US, at least) ended with the bankruptcies of PanAm/TWA/Eastern, and the disappearance of navigators, flight engineers, and the 727. His hourly pay looks impressive, until you consider the fact that he literally gets paid only when the plane is in the air, and the day or two he spends commuting to and from his first/last airport, the time he spends overnighting, and the hours he has to BE at the airport, but isn't actually getting paid or doing anything. When you factor in the high unemployment rate for American pilots who entered the workforce after 2000 (first to be laid off, last to be called back), it's not exactly what you'd call a lucrative career anymore.
Worst of all, if he ever DID decide to pursue another career and declined to go back after being laid off, his career as a pilot would be fried and gone forever -- he'd lose his type ratings, and once you lose them, it's almost impossible to get hired by another airline without them. That's the dilemma he's forced into every couple of years -- take that job as a math professor, and forget about EVER flying a jet again after having gone through hell (and almost a hundred thousand dollars renting planes to get his flight time up, and years of poverty & taking real risks with his life daily as a flight instructor) getting there in the first place.
I don't know about 1990, but we definitely had some pretty nasty air pollution about 40-60 years ago. When I was < 10 and growing up in Ohio circa 1980, I remember that air pollution was pretty much everywhere, even in smaller cities, like the Warren-Youngstown-Sharon area roughly halfway between Cleveland and Pittsburgh. My first "omg" memory of Florida was looking up during recess one day about a month after we moved there, and freaking out because I could see the full moon in broad daylight. That was something you never, EVER saw in Ohio. Or at least something *I* had no memory of ever seeing.
Hell, I spent July 5, 1994 in New York, and remember BARELY being able to see the Twin Towers from Midtown. The whole city smelled like a burning log in a fireplace. Likewise, I spent a week in Los Angeles sometime in August 1996, and remember driving into L.A. on LaCienega drive... I made it over the mountain, and saw the famous vista with LA (well, OK, I guess it was actually Beverly Hills) spread out in front of me... except you couldn't actually see anything except faint rooftops a mile or two away, and a sea of opaque smog. In LA's defense, though, its smog didn't really have any particular odor. It was opaque to a degree I'd never seen in my life, but other than obscuring most of the views, it didn't really bother me.
Anyway, onto the pics:
Pittsburgh, 1948... during the DAY: http://bike-pgh.org/wp-content/uploads/2009/10/smog1.jpg
Cleveland, 1973: http://upload.wikimedia.org/wikipedia/commons/8/83/CLEVELAND_SKYLINE_IN_THE_SMOG_OF_JULY_20%2C_1973%2C_DAY_OF_POLLUTION_ALERT_-_NARA_-_550190.jpg
New York, 1972: http://earth911.com/wp-content/uploads/2012/04/Smog-1970s.jpg
Los Angeles, 1948: http://cdn.theatlantic.com/static/mt/assets/jamesfallows/los-angeles-smog_53499058.jpg
Manhattan, 1966: http://www.flickr.com/photos/wavz13/4083896787/
> I've had fraudulent transactions reversed quite a few times over the years with my debit card.
There's a big, HUGE difference. The difference isn't that the charges won't EVENTUALLY get reversed... the difference is that one immediately empties your checking account, while the other is mostly an abstract paper loss.
If someone steals your credit card and makes fraudulent purchases, you submit the paperwork for the fraud claim, and that's pretty much the end of it until they finish their investigation.
If someone steals your DEBIT card, they can drain your checking account to zero, and possibly even overdraw it by a thousand dollars or more. If your finances are shaky to begin with, having your checking account emptied or technically-overdrawn for a week or longer can have financially devastating consequences & set off a chain reaction of late fees, penalties, credit-score lowering, and reactions to that credit-score lowering that's almost impossible to stop, let alone undo, once it begins. When your checking account is overdrawn by more than $2,000 in fraudulent charges, even your direct-deposited paycheck is going to partially or completely vaporize into a puff of antimoney smoke. If you have a savings account that's liked to the checking account for overdraft protection, they can drain your savings account to zero, too.
It's hard to over-emphasize just how completely you can be fucked by debit card fraud, and the fallout (in the form of fees and penalties from others, plus damage to your credit score) can persist and keep doing damage LONG after the fraudulent purchases themselves have been reversed.
You're forgetting about "textbook inertia" and fact that anything in a public school science textbook that remotely involves "evolution" has already been diluted, whitewashed, and minimized to make the book acceptable to the state's evangelical Christian gatekeepers. There are probably kids sitting in middle school classrooms *RIGHT NOW* reading more or less the same content their parents did, more for political and budgetary reasons than anything.
I kid you not... when I was in sixth grade (late 80s), my science teacher handed out brand new textbooks on the first day of school. After handing them out, he became visibly irate, then told us to open the book to page 234, tear it out, and throw it away after class without reading it, "because the school board decided it might make us question our religious beliefs, so they ordered teachers to have the offending page removed." He informed us that the trashcan was "over there" (pointing to a spot that was kind of by the door, but far enough out of the way that you'd have had to be really determined to use it) and made a point of looking the other way as we left at the end of class. It was a brilliant act of subversion that could have probably gotten him fired, because you can bet every single kid in the class read the page. Future classes never got to see the page (well, not counting the photocopies some of my friends and I made & passed out like contraband to some younger friends when we were in 8th grade), but he was able to make a small difference in the lives of at least one class. I know it was page 234, because I still have it ;-)
> Except that your not technically buying it, your purchase is subsidised by the contract.
Legally irrelevant. You own your phone 100% the moment you walk out the door with it. The fact that you owe the carrier a hefty ETF if you don't satisfy the contract is why.
For various reasons, American mobile phone carriers have ALWAYS been prohibited from directly leasing phone hardware to consumers. BellSouth Mobility did its best to get the policy overturned around 1991 (and went into high gear after Hurricane Andrew, using it as their perfect pretense), and they got smacked down by the government every time.
I believe the rationale back then was that mobile phone manufacturing was an oligopoly (basically, Motorola owned the US market), and if carriers could lease phones to consumers, the manufacturers would jack up the retail cost to levels that made purchase by end users economically untenable, then rebate most of it back to their biggest customers (ie, carriers).
In AT&T's partial defense, they require real data plans because the experience of just about every wireless carrier on earth has been that an iPhone or Android phone in the hands of someone who "doesn't need a data plan, because they never use data" is a ticking timebomb waiting to explode into an ugly & expensive customer service nightmare.
The extortionately expensive per-kb/mb fees charged for adhoc data use don't help, of course, but the fact remains, it's almost IMPOSSIBLE to use an iPhone or Android phone without EVER using a single kilobyte of billable wireless data service. Very little data, sure. But ZERO data? Damn near impossible without crippling the phone.
That's not to say AT&T doesn't use this fact as an opportunity to make even more money... but honestly, using an iPhone or Android phone on a wireless plan with ZERO kb/mb per month of cheap/included data is a financial disaster waiting to happen, and you really can't really blame ANY carrier for not wanting to voluntarily put itself into the position of having to spend hours fighting with a livid customer over a $7.62 charge for data "they didn't use" (but their phone DID, in the background).