> Fiber has zero value as a "utility for society".
Bullshit. The value of fiber isn't just the value it brings to people willing to spend tens of thousands of dollars per month for an OC-12 today, it's the value it brings to people who'll find cool ways to enjoy having the equivalent of an OC-12 for slightly more than they're paying for shit DSL or cable today, amortized over the next 25-100 years (since buried fiber is basically a 50-100+ year capital investment).
If the federal government had any common sense, it would be taking advantage of its ability to borrow money at near-zero interest rates and making low-interest loans available to finance laying open-access fiber across America like there's no tomorrow. AT&T won't spend $10,000 laying fiber to a customer, because any horizon more than 5 years away represents too much capital risk, few customers could afford to cough up $10,000 (or even pay it off over some short period like 5 years that's a fraction of its useful life), and AT&T won't voluntarily share a fiber with anyone, because it likes keeping customers locked in. Even if the government burned $20,000 laying the same fiber and passed along the cost, it could be financed over 50 years for less than $60/month, interest included. And with inflation, by the time those last 10 years arrived, $60/month would seem as silly as a 12c/month luxury surtax on telephone service to finance the Spanish-American War.
Try using a USB interface with 5 pins to simultaneously output 720p60 and 1080i60 (let alone 1080p60, 3D, or 120fps) video while simultaneously charging the phone, controlling it with USB peripherals, and playing HD content from a ripped.iso image on an external flash drive.
The more aggressively you try to serialize lots of different high-speed buses into a single limited set of wires, the more expensive and complicated your external interface hardware inevitably becomes, and the more dependent you become upon explicit support for both the external hardware and your crossbar/interface by the OS itself.
> Try to put it in hub, will not charge. EVER HEARD OF TAKING WHAT YOU ARE GIVEN? > CHARGE DAMN IT, THE USB SPEC SAYS SO.
Actually, the USB spec says that a device is only allowed to draw 100mA unless the port it's connected to is capable of negotiating for 500mA. A de-facto extension allows devices to draw "up to" 1.7A (without necessarily guaranteeing that the source is physically capable of supplying that much without damage to itself) if the two data pins are shorted together.
As a practical matter, if you're a USB endpoint device, you detect that you're connected to something capable of supplying power... but NOT negotiating... it's almost guaranteed to be a powered USB hub that isn't connected to a computer, and can easily supply 500mA. But strictly speaking, it's not a guarantee.
Also, not all extensions to the USB standard are necessarily bad. Samsung's 11-pin micro-USB-compatible cable (downward-compatible with micro-USB, but with additional pads on the other side for devices capable of putting it to good use) was a definite improvement. The problem with trying to mega-serialize everything into two wires is that it makes it hard and expensive for different uses to coexist without either requiring some very expensive (at least initially) interface hardware along the way. The box that costs $200 today might very well cost $99 next year, and $12.99 at Wal Mart 5 years from now, but life is still going to suck mightily when you're spending almost as much as you paid for your phone just to connect anything to it. Not even APPLE can TOTALLY get away with driving up interface costs by THAT much.
Intel's "lightning bolt" interface is wonderful as an open-ended port that can ultimately be used to connect just about anything to anything in full bidirectional splendor. If I could have only one physical port on my phone or computer, that's definitely the one I'd want. BUT... if it's your only port on the device, you end up turning something that's supposed to be cheap & straightforward into an expensive abomination. Do you REALLY want to go out and spend $40-200 on an external box that deserializes a 10+gbps lightning bolt bus into a virtual PCI express bus, hardwired to a PCI Express to USB root hub chip, hardwired to a USB audio chip and USB-I2C/UART bridge, so that both can be hardwired to a class D amp and give you something to output actual audio to your headphones?
Oh... I forgot... the above interface will also require you to root or reflash your phone so you can install the kernel module it requires, or do without it until you buy your next phone, since the likelihood that any carrier in America will give two shits about having the manufacturer add support for it to an existing phone is somewhere between "slim" and "none". Then find out that if you want to charge at the same time, you need ANOTHER $50-200 interface to multiplex the "headphone audio interface" and "power interface" into your elegant (if expensive and demanding) minimalist serial bus...
Minimalism might be elegant, but it rarely ends up being cheap.
> I'm an old Usenet fan, but am perfectly aware the ability to nymshift led to a culture of spamming and verbal harrassment that are basically unacceptable
No, it was the AOL users. The darkest day in the history of Usenet was the day they started allowing people to post to it from AOL.
Actually, all the condom needs is a switch that when open, leaves D+ and D- connected to nothing at all, and when closed, shorts D+ and D- with each other. By definition, if the phone sees that the D+ and D- pins are shorted together, the device is entitled to draw 1.7A from the power supply.
Officially, if D+ and D- are neither shorted nor able to negotiate for higher current, the device is only supposed to draw 100mA. In reality, everything I've ever seen besides Motorola's annoying phones ignores that rule and draws 500mA if there's power, but no negotiation, because anything that's GENUINELY capable of supplying only 100mA while powered up is almost guaranteed to be either a laptop running in "BIOS mode" that's capable of saying, "No!" (or cutting the power if too much gets drawn), a 99c power supply from China whose lifespan will be measured in days *anyway*, or a powered USB hub that isn't connected to a computer and can supply 500mA per port without breaking a sweat.
> We would all be stuck on 1.5mbps DSL lines because there is no cost effective way to push data quickly over consumer grade circuits
To a large degree, that's true. You can have the greatest digital signal processing on earth, but if you have a copper pair is of a length and quality that's right at the edge of what could do 1.5mbps ADSL back in 1999, the harsh truth is that you aren't going to do a whole lot better with a copper pair of equal length and quality today.
What changed was:
* the length of the wire (phone companies deployed remote DSLAMs and VRADs to neighborhood edges to get the distance down to ~1,500 feet or less)
* the wire quality (some neighborhoods had old, decrepit copper pairs replaced... which actually was kind of insane, because the cost to have replaced it with fiber might have been double the cost once you factored in the cost of the backhoe, workers, etc, as well as the future value of fiber relative to the future value of copper).
* availability of copper pairs, and willingness of the phone company to use them for pair-bonding. Circa 2000, lots of neighborhoods were literally maxed out, and phone companies were using DSL as a way to multiplex several customers onto a single copper pair. In fact, that's WHY DSL was invented in the first place, before the internet motivated companies to repurpose it for bulk data. As customers dropped additional lines that used to be for fax machines and modems, it freed up copper pairs to use for DSL. As more customers dropped lines that used to be their home's voice line, a LOT more copper pairs began to get freed up.
* ability of DSL modems to adapt to line conditions, instead of rigidly forcing "one speed fits all" the way g.lite did. G.lite was fixated on 1.5mbps/256k (128k?) because it was a number consumers were demanding that was relatively do-able within most of the former Bell System, and allowed them to avoid truck rolls for the majority of installations... at least, in the majority of American suburbia. Send out a line tech to do some local cleanup work & one-off tweaking, and you can make DSL happen on wires that would have been regarded as "unusable" by the phone company 15 years ago, and you can even do it using technology that existed back then. Much of it was just bureaucracy & unwillingness to act like a customer-focused business instead of a Soviet-era bureaucracy that viewed customers as a cost sink.
That's not to say there haven't been ANY improvements... but really, the technology hasn't changed. It just became cheaper, and the network edge moved closer to the customers.
The POINT is that actual hardware failure of a SSD is unlikely, so they can offer long warranties, but despite their long nominal lives, they're more likely to fail early (and often, and repeatedly) at the one task you bought them for -- reliably and robustly storing data.
Actually, the big problem with SSDs is the fact that unlike conventional hard drives, SSDs often DO have catastrophic data loss out of the blue long after the burn-in period, but long before the drive is even remotely "old". A conventional hard drive that's going to die tends to do it when the drive is new, before you've invested much in setting up Windows yet again. A Sandforce-based SSD is most likely to die 3-6 months after installation, then every 3-6 months after each "secure erase" resurrection, until the day you put it into panic mode trying desperately to recover a file you saved an hour ago that didn't make it to your latest backup.
You're right, because a shitty low-end no-name brand at least has the benefit of anonymity instead of a name that's tainted beyond repair or redemption.
The warranty period for a SSD is almost meaningless, because most SSD failures involve instant unrecoverable data loss by a drive that technically isn't "broken" in the legally-binding warranty sense, since you can "fix" it by erasing everything and reloading the virtual gun for another round of Russian Roulette.
Actually, there's an even bigger question... was the "failure" definition for SSDs carefully-crafted to exclude drives that experienced total data loss, but could technically be "repaired" and run for another 3-6 months before the next instance of total data loss?
At least when a traditional hard drive dies, it stays dead instead of coming back like an unstoppable undead zombie with an insatiable appetite eager to make a second attempt at destroying whatever data of yours it missed the first time around.
I can't think of a better gift for that special "frienemy" in your life than a Sandforce SSD. It's a gift that just keeps on giving... misery. Endless, eternal misery.
You forgot to mention the wonderful way Sandforce controllers encrypt the data with a key that you (the drive's owner) aren't allowed to have, so your ONLY data recovery option (on the rare occasions when it MIGHT be an option) is to pay an extortionate amount of money to one of Sandforce's "trusted partners" to decrypt it for you... and apparently, they actually charge more money to do what's now a 100% automated software-based recovery than the same companies USED to charge to remove the platters from a conventional hard drive in a cleanroom and mount them in a recovery unit.
I'm frankly surprised that some company like OCZ hasn't come up with an "innovative" new-economy (as in, "fuck you, consumer!") business model for SSDs, like selling 256-gig drives for $25 that lock themselves after some random period of time between 180 and 720 days after first use, and charge $2,000 for the key to unlock the drive (decreasing to $1,000 after a week, $500 after a month, $250 after 6 months, $125 after a year, and $64 after 2 years). Is there anybody who doubts that netbook manufacturers would pee their pants with glee if they could take advantage of that kind of "innovation"?
The craziest thing about Sandforce is the way their marketing department is somehow able to spin some of the worst design flaws in the history of computing into alleged sales points.
Their controller chips are the worst crock of steaming shit and snake oil to have disgraced the computer industry in *years*.
If The Onion's staff got together with a few sixpacks of beer, a pizza or two, and a mission to make up something funny for the "Tech" section of their next April Fool's Day edition, it would read like a Sandforce datasheet.
Let's be honest... if Wired Magazine got a press release proclaiming that Sandforce SSD controller chips maximized Windows performance by making users re-install it from scratch every 3-6 months, their editorial staff would have to call Sandforce's marketing department to confirm that it was real, and wasn't just a bored member of the Syrian Electronic Army playing practical jokes on them again...
Oops. Brain fart. It was the "Vista International" (at least, until Marriott bought it at some point after I was in Middle School). It's what happens when you grow up in a state like Florida with hotels literally everywhere, and your brain ends up using lossy compression to (somewhat) remember them. Occasionally, you'll remember a hotel whose name is partly right, and partly random fragments of other hotel names you've seen over the years.
Exactly. What good is a drive with a hundred million hours before physical failure if it commits data-suicide every 4-12 weeks? Maybe I'm just weird, but I don't *care* whether a drive (like the OCZ Vertex 2, for example) technically isn't "broken", and "only" has to be "securely erased" to temporarily make it usable again for a few weeks until its next episode of eternal amnesia. If it happens once, it's a random fluke. If it happens twice in six months, the drive is fucked and useless for its intended purpose.
> What makes you think you can't take FLASH devices and access them in a similar way to platters?
Sandforce controllers enforce mandatory AES encryption that can't be disabled, using a key that can't be recovered or set to a known value. So if your controller decides to quit allowing you to access your data, unsoldering the chips won't do you any good, because the values you read from them might as well be random noise.
The thing that really sucks about SSDs (at least, Sandforce-based drives) is the fact that 99% of their failures are due to firmware bugs that can be simultaneously triggered on an entire array at once (especially the sleep-related bugs). It's a mode of failure the creators of RAID 1, 5, and 10 never anticipated.
IMHO, the worst thing about SSDs (at least, those with Sandforce controllers) is the fact that they have mandatory full-drive encryption that can't be disabled, using a key you aren't allowed to set or recover, and gets blown away whenever you reflash the firmware. This means, among other things, if the drive's controller gets itself confused:
* You can't reflash data-recovery firmware onto the drive. The act flashing it would blow away the encryption key and render the data gone forever.
* If the drive decides you're trying "too hard" to systematically extract data from it while it's in a confused state, it'll go into "panic mode" by blowing away the encryption key. If this happens, your data is gone forever AND you have to send the drive back to OCZ or whomever you got it from in order to get it unlocked. For your protection, of course. And Hollywood's. Among other things, dd_rescue/ddrecover can trigger panic mode.
* You can't even do the equivalent of removing the platters from a conventional drive in a clean room and mount them to another drive for reading, because the data on the flash chips is all encrypted, and the key is unrecoverable.
This is BULLSHIT, and it's why I refuse to buy any more SSDs. I, as an end user, should be able to download a utility from somewhere, reflash the drive to firmware that includes an offline recovery mode that simply dumps the flash chip content from start to finish, and either disable the encryption or set it to a key *I* control, so the 99.99999% of the data on the drive that's good when the embedded firmware freaks out can be dumped and recovered offline.
If there's a God, Linus will go NUCLEAR over this, get a few seconds on CNN & other networks to rant about the unreliability of SSDs, and scare enough consumers to hit the industry HARD where it'll hurt the most... their bank accounts.
It might not be possible to make SSDs reliable, but DAMMIT, they should at least be RECOVERABLE. There were goddamn hard drives with recoverable data pulled out of laptops left in safes in the Vistamark hotel when a tower sheared it in half and buried it under flaming rubble, yet a SSD that dies if you so much as look at it the wrong way due to firmware bugs ends up being fundamentally unrecoverable for no hard technical reason.
And yes, I'm bitter about having my hard drive commit suicide for no reason besides Sandforce Business Policy. As long as they keep making controllers that cause drives to self-destruct at the drop of a hat, I'll keep doing my best to talk people out of buying drives tainted by their controller chips. Sandforce sucks.
Actually, people in China DO occasionally draw characters in the air with their fingers to disambiguate spoken homonyms. And it's clearly viewed as important by English-speakers, because it's the main reason why we still have silent letters and unusual spellings. Look for every instance of an "odd" English spelling, and there's almost always a homonym hiding behind it.
Interesting factoid: the only city in America where a Spanish-language TV station has EVER been the #1 station in its media market is Miami (Telemundo, I believe). In San Juan, where Spanish is both the de-facto and official language, the two top channels have always been Fox and NBC. And in Miami's case, it was kind of a weird situation where somebody at Univision said something non-condemnatory about Fidel Castro, Univision refused to denounce and fire them, and for a few years if you were Cuban in Miami, it was (*cough*) socially frowned upon to admit in polite company that you watched Univision as a restult. Meanwhile, Miami had 6 or 7 English-language networks prior to the big consolidation a few years ago, and no one other channel could match the popularity of "Betty la Fea" (just to give one example), so Univision ended up slightly beating out everyone else by a small plurality.
Now, for the harsh truth: it's impossible to live in the United States for 25+ years without achieving some degree of English competency, whether it's intentional or not. Even if you went out of your way to surround yourself only with people who speak Spanish, watched only Spanish TV shows, and read only Spanish publications, every time you leave the house you're going to be surrounded by English... even in the most hardcore Spanish enclaves in America. And regardless of how hard YOU might try to avoid English, your kids are going to watch English-language TV shows and movies with their friends, speak English at school, and around you whenever they don't want you to know what they're talking about with their friends.
Let me emphasize the last point, because I've had to cruelly disillusion many of my Cuban friends who were absolutely *convinced* that their parents, or at least their grandparents and elderly aunts/uncles "couldn't speak English", and would say things in front of them (in English) as though they weren't there. Without exception, they were wrong, and were horrified when the truth finally came out. I still remember when one of my friends told me, "My mom doesn't know English". I pointed out that she was a paralegal for a law firm. He countered with the excuse that it was in Hialeah. I replied that even if the law firm were in Hialeah, there was no... way... in... HELL any law firm would have hired her if she literally didn't speak a word of English. Six months later, he admitted I was right, and that he was shocked to discover that his mother was quite fluent in English. She just never spoke English around *him*, because she liked being able to eavesdrop on him when he was talking to his friends:-D
As a practical matter, anybody who's born in America is going to be fluent in English by the time he's a teenager... EVEN IF his parents try to send him to schools where Spanish is spoken. Anybody who's lived in America for more than 10 years is going to have a grasp of English, and by the time they've lived here for 25 years, they're going to know English at least as well as my 100% Ohio-bred pure-Gringo parents know Spanish after living in South Florida for 25 years. And if they're emigrating to the US via official means, they almost certainly have a college degree, which almost guarantees that they've studied English for at least a few years.
The fundamental problem with romanized Chinese is the fact that nearly every word in Chinese has multiple homonyms (to/too/two), even AFTER you take into account the various inflections called "tones" (which are really just ways of formally representing verbal inflections in writing).
English disambiguates homonyms with silent letters and alternate letter combinations. If Chinese followed the same strategy, the romanized spelling of Chinese words would be almost completely arbitrary, and Chinese kids would spend years memorizing the difference between "shi", "she", "shee", "shii", "shie", "schi", "sche", "schii", and "schie" (plus appropriate tone marks). In the end, it wouldn't be much of an improvement... assuming it were any improvement at all.
At one time, Chinese had a serious "keyboard problem", but it's been largely solved by keyboards like Wubizixing ( http://en.wikipedia.org/wiki/Wubizixing ) and Wubihua. At the simple end, Wubihua assigns 5 keys to the most fundamental strokes used to write Chinese: horizontal, vertical, left-falling, right-falling/dot, and hooked/complex. You press the keys corresponding to at least the first 4 strokes, then press the key corresponding to the last, and it presents you with a list of plausible characters that match. The more keys you press, the smaller the list gets, until you're left with either an unambiguous match or you've entered all the strokes.
Other methods, like Wubizixing, go a step further, and assign keys to the radicals themselves (if you think of characters as being like molecules, radicals are atoms, and strokes are quarks; in English terms, characters are words or stems, radicals are letters, and strokes are the way you'd write those letters... like "vertical, vertical, horizontal" for "uppercase H"). Somebody who's good at typing on a Wubizixing keyboard with the key-cadence of somebody who types English at ~100wpm can achieve an equivalent word-rate of about 120-150wpm (because Wubizixing makes more efficient use of the keys on the keyboard, and requires fewer keystrokes per communicated-word than English QWERTY).
The irony is that most people in China are amazed when they first encounter a Westerner who can type on a Wubi keyboard (-hua OR -zixing), because they think they're "too hard" to use. The reality is that stroke-based input is REALLY the only way somebody who doesn't know how to speak Chinese CAN enter characters on a keyboard. There's definitely room for algorithm-improvement in a "westerner-friendly" stroke-based input method, but I can guarantee that whatever we end up with ~10 years from now, it's going to look more like Wubi than anything else. It'll just be more forgiving of someone who enters "zhong" (level 'o' tone) as "vertical, horizontal, vertical, horizontal, vertical" (or some other permutation) instead of "vertical, hook, horizontal, vertical" (just to give one example).
As for "too hard", Wubizixing really isn't any harder for someone in China to master than QWERTY is for someone in the US. For geeks who type all day, every day, nonstop, it's a skill that pays HUGE personal dividends. For people who think computers in general are "hard to use", it doesn't really matter whether they're American or Chinese... they'll dick around with two-finger hunt & peck or Pinyin input, and endlessly predict the death of keyboards in favor of speech recognition. The rest of us, American and Chinese, will laugh at them and keep typing 120-150wpm while they struggle to send email and text messages with amusing autocorrect errors.
Anyway, getting back to romanization of Chinese... it's not going to happen. Chinese has romanized as much as it's ever going to romanize. Twenty years ago, keyboards and fonts were real problems. Now,
The problem with CDMA phones supporting Sprint and Verizon has nothing to do with *bands*, and everything to do with firmware licensing and Sprint's business policy.
* In Sprint's entire history, I think they've had maybe 3 or 4 phones that WEREN'T dual-band and capable of using Verizon's frequencies. Sprint phones ROAM on Verizon in many rural areas.
* The CDMA2000 equivalent of a SIM card's IMEI is the MEID. There's no technical reason why it can't be on a SIM-like card, just like GSM phones use. In fact, Qualcomm itself defined a standard called R-UIM that does precisely that, and was used just about everywhere in the world besides the United States back when CDMA phones were common in countries like Australia, New Zealand, Singapore, India, Brazil, and Chile. The problem is, Sprint refuses to activate any phone whose MEID isn't in their computer as a Sprint-branded phone sold by Sprint. Verizon, for various legal reasons, can't refuse to activate a non-Verizon phone, but unless you have an inside connection to help you get it working, your likelihood of getting a Sprint phone to ever work properly on Verizon is low ("properly" == "voicemail, MMS, and similar things work properly"). Moreover, without actual Verizon radio modem firmware, a non-Verizon phone will never be able to do EVDO on Verizon, and will be limited to 1xRTT. Basically, Verizon authenticates its phones differently to EVDO than Sprint, and AFAIK, nobody has ever independently gotten a non-Verizon phone to do EVDO on Verizon ("independently" == "doesn't work for Verizon or Qualcomm"). People HAVE gotten Sprint phones that are identical twins of Verizon phones to do EVDO on Verizon by reflashing them with a Verizon radio modem, but AFAIK, it's been *years* since anybody has gotten that trick to work.
But yeah, 99% of the bullshit and incompatibility we have in the US comes down to Qualcomm's licensing policies. As far as Qualcomm is concerned, their REAL customers are Verizon, Sprint, AT&T, and (to a lesser extent) T-Mobile, US Cellular, and a few others. Qualcomm will NEVER knowingly assist anybody who wants to do an end run around the carriers, and they're 100% cool with the status quo of carrier-locked phones. I'd go so far as to argue that if push came to shove, Qualcomm will probably be resisting manufacturer-licensed firmware long after the carriers themselves have gotten beaten into submission by the FCC, because carrier-licensed firmware goes to the deepest core of their existence.
If the FCC prohibits carriers from demanding vendor-licensed firmware, Qualcomm's lawyers will be busy finding ways to let carriers say "the devil made me do it" and continue to enforce the policies LONG after the carriers themselves are allowed to legally *demand* it (even if they ARE allowed to legally "favor" it by choosing to buy Qualcomm-based phones instead of other phones for sale to their own customers).
The only way we're ever going to have truly carrier-agnostic phones in the US is if the FCC starts refusing to certify any phone that can't pass the compatibility test suites of every major network in addition to the interference-related ones the FCC itself cares about now. It'll happen... someday... but you can BET Qualcomm's lobbyists in Washington won't be the ones encouraging it.
And it's sad. Qualcomm has some of the world's best engineers, and makes some of the best chipsets on earth. I've seen demos of stuff they have in the pipeline, and it's really awesome... but every time I see their demos, I have to remind myself that we won't get to get anywhere near those cool features on American phones until at least 2-5 years after they're commonplace elsewhere, because American carriers only care about features that directly translate into higher monthly revenues for them, or features they can claim hardware exclusivity on & use as a way to enforce vendor lock-in with.
You'd be surprised how cost-effective it can be to manufacture pre-stressed concrete bridge modules at a factory in China, load them onto a barge, and ship them to California. By definition, pre-stressed concrete bridge modules have to be manufactured off-site somewhere, and 99% of the transportation costs involve loading and unloading. Whether the actual long-haul transportation occurs via superfreighter from China, freight railroad across the US, or post-Panamex freighter from the east/gulf coast of the US, doesn't really make all that much of a difference.
The main obstacle to actually DOING it isn't even "buy American" laws, so much as China's notorious "quality fade". God forbid, if you built a bridge with Chinese bridge modules, then ONE of them were discovered to be substandard, I can *guarantee* that the government entity paying for the bridge would end up replacing and demolishing the entire new bridge (or at best, leaving it in place for pedestrians and cyclists, if they thought it could safely handle the lesser dynamic load, alongside the new bridge) because they'd never be allowed to risk having a large-scale failure due to another substandard module somewhere in the bridge.
That's part of the reason why stuff from American, Japanese, and European companies costs several orders of magnitude more. We actually *do* have the ability to track every bolt, screw, piece of rebar, and pound of aggregate every inch of the way from mine to megastructure, and somebody with enough funding and determination can spend years digging through the documentation to verify it. China isn't quite "there" yet, and doesn't appear to even *want* to go there.
It's a fairly big paradigm shift, because it's the difference between "build two bridge spans that, when used within their explicit design parameters over their guaranteed design lives, must never, ever, EVER fail" and "build three spans for half the cost of building one American span, and roll the dice that when/if one fails before it's functionally obsolete and has to be replaced anyway, it won't kill or injure too many people", and just take for granted that you'll probably end up demolishing and rebuilding the whole thing to increase its capacity *anyway* long before it becomes a problem.
What China appears to have failed to appreciate, based on what's happened in cities like London and New York, is that when the future arrives, building the replacement won't necessarily be easy or cheap, because adjacent land use will have grown up around it, and taking the old infrastructure out of service to make room for its replacement just isn't a viable option. The Big Dig comes to mind as an obvious example of a project that would have been expensive in any case, but was made exponentially MORE expensive by the need to keep the old road in service throughout the construction.
The Washington Metro has a similar problem... everyone agrees that it has a serious capacity problem in stations between Rosslyn and L'Enfant Plaza that are only going to get worse once the Silver Line begins service, but there's no cheap or easy way to enlarge the existing stations. They were built 25-40 years ago in holes that were excavated, but now have expensive buildings sitting on top of them... and mining out new platforms from below would cost *almost* as much money as demolishing the buildings above and digging new holes. The London Underground has been dealing with the same problem for decades, and now spends almost as much money (adjusting for inflation) to mine out and rebuild a single station as it spent to build the entire original tube network ~150 years ago.
The worst thing about living anywhere near a "bad" neighborhood are the endless car break-ins that the authorities can't do jack shit to stop. The Coconut Grove area of Miami, and the adjacent neighborhoods in Coral Gables (where I used to live) are a perfect example. Thanks to both explosive gentrification and the enduring legacy of old-south segregation-era zoning laws, there are plenty of areas where you literally have expensive homes back to back with housing projects that will never go away.
In those areas, you can never have guests come over to see you unless they park elsewhere and take a cab, because YOUR BUILDING's parking garage might have 2 layers of gates & security, but for obvious logistical reasons, the guest parking sits unprotected out in the open. Let me tell you... the only thing that sucks worse than getting your own parked car broken into is having friends come to see you, and getting THEIR OWN car broken into. Or god forbid, your parents' car. If your parents' car gets broken into, you will NEVER be allowed to forget about it. My parents STILL bring it up at least two or three times at family gatherings on Thanksgiving, Christmas, and the Fourth of July, and it happened more than a DECADE ago.
Ask anybody who lives in an urban neighborhood what their #1 neurotic fear is, and they'll tell you -- "Friends coming to visit, and getting their car broken into". On the hierarchy of social shame, it pretty much tops the list. From that point forward, you no longer live in a nice, safe, gentrified urban neighborhood. As far as your friends and family are concerned, you live in the 'hood.
The beginning of the end for Nokia happened around 2004, when UMTS arrived in Europe, and Nokia made an intentional business decision to not support EDGE, and to basically walk away from CDMA as well (even though at that point, probably half the phones sold by Verizon and Sprint were Nokia, as well as the majority of high-end phones sold by AT&T and T-Mobile).
As a result, their phones became useless paperweights in the US as far as anybody who ever used data was concerned. EDGE wasn't exactly "high speed", but compared to GPRS, EDGE is just "annoyingly slow" compared to "uselessly slow". Circa 2005-2008, EDGE was the best that existed in most of the US anyway... T-Mobile hadn't even started deploying HSPA yet, and AT&T's HSPA data existed in maybe two dozen cities.
Nokia presumably wrote off the US market because, in terms of total unit sales, it was roughly equal to Portugal or Switzerland. What they overlooked was the importance of mindshare... half the world's tech blogs and web sites are American, and as far as anyone in America was concerned, by ~2007 Nokia had effectively ceased to exist. At best, they were a company that used to be popular, and now just made throw-away low-end phones sold to people in remote African villages.
Other companies learned their lesson. Today, companies like Sony-Ericsson are working as hard as they can to break their Qualcomm addiction(*), and make a point of getting their phones into the hands of American reviewers who live in cities where T-Mobile has good HSPA+ coverage.
(*) Qualcomm insists on licensing LTE radio firmware to carriers rather than manufacturers, which means it's basically impossible for a manufacturer to sell phones capable of using LTE on AT&T or T-Mobile without the active involvement of AT&T or T-Mobile, and de-facto impossible to sell a phone built with a Qualcomm LTE chipset that's carrier-agnostic and capable of doing LTE on both AT&T and T-Mobile.
It's technically possible to use a separate non-Qualcomm chipset (like Beceem's) for LTE, but the price premium is fairly stiff (about $100, by the time the phone gets to retail stores). That's why companies like Sony-Ericsson (who desperately want to break the stranglehold American carriers have over the American phone market as gatekeepers with economic -- or in the case of Verizon & Sprint, real -- veto power) have eagerly embraced chipsets like the Renesas MP5232 and MP6530, which will enable them to make phones capable of doing LTE on AT&T and T-Mobile, and break the "LTE Lock-In" AT&T in particular has been working overtime to exploit as a way of making their nominally-GSM network into one that's as de-facto proprietary as Verizon's.
Nobody outside the US, or who's vacationed in Florida (or at least the southern US) has any idea WTF "Key Lime Pie" is, especially people in China & India.
Hell, I *GREW UP* in Florida, and was in HIGH SCHOOL before I had any idea what Key Lime Pie is. I'd be shocked if a random developer in Bangalore or Hong Kong, let alone Mumbai or Beijing, would have had the slightest idea what it was without researching it on Google.
Kit Kat, in contrast, is a global brand. Not necessarily a #1 item like M&Ms, but it's known. And it sounds like a halfway-reasonable word to people whose native languages are Hindi or Mandarin.
Slashdot Mirror
> Fiber has zero value as a "utility for society".
Bullshit. The value of fiber isn't just the value it brings to people willing to spend tens of thousands of dollars per month for an OC-12 today, it's the value it brings to people who'll find cool ways to enjoy having the equivalent of an OC-12 for slightly more than they're paying for shit DSL or cable today, amortized over the next 25-100 years (since buried fiber is basically a 50-100+ year capital investment).
If the federal government had any common sense, it would be taking advantage of its ability to borrow money at near-zero interest rates and making low-interest loans available to finance laying open-access fiber across America like there's no tomorrow. AT&T won't spend $10,000 laying fiber to a customer, because any horizon more than 5 years away represents too much capital risk, few customers could afford to cough up $10,000 (or even pay it off over some short period like 5 years that's a fraction of its useful life), and AT&T won't voluntarily share a fiber with anyone, because it likes keeping customers locked in. Even if the government burned $20,000 laying the same fiber and passed along the cost, it could be financed over 50 years for less than $60/month, interest included. And with inflation, by the time those last 10 years arrived, $60/month would seem as silly as a 12c/month luxury surtax on telephone service to finance the Spanish-American War.
> What exactly is crappy about USB?
Try using a USB interface with 5 pins to simultaneously output 720p60 and 1080i60 (let alone 1080p60, 3D, or 120fps) video while simultaneously charging the phone, controlling it with USB peripherals, and playing HD content from a ripped .iso image on an external flash drive.
The more aggressively you try to serialize lots of different high-speed buses into a single limited set of wires, the more expensive and complicated your external interface hardware inevitably becomes, and the more dependent you become upon explicit support for both the external hardware and your crossbar/interface by the OS itself.
> Try to put it in hub, will not charge. EVER HEARD OF TAKING WHAT YOU ARE GIVEN?
> CHARGE DAMN IT, THE USB SPEC SAYS SO.
Actually, the USB spec says that a device is only allowed to draw 100mA unless the port it's connected to is capable of negotiating for 500mA. A de-facto extension allows devices to draw "up to" 1.7A (without necessarily guaranteeing that the source is physically capable of supplying that much without damage to itself) if the two data pins are shorted together.
As a practical matter, if you're a USB endpoint device, you detect that you're connected to something capable of supplying power... but NOT negotiating... it's almost guaranteed to be a powered USB hub that isn't connected to a computer, and can easily supply 500mA. But strictly speaking, it's not a guarantee.
Also, not all extensions to the USB standard are necessarily bad. Samsung's 11-pin micro-USB-compatible cable (downward-compatible with micro-USB, but with additional pads on the other side for devices capable of putting it to good use) was a definite improvement. The problem with trying to mega-serialize everything into two wires is that it makes it hard and expensive for different uses to coexist without either requiring some very expensive (at least initially) interface hardware along the way. The box that costs $200 today might very well cost $99 next year, and $12.99 at Wal Mart 5 years from now, but life is still going to suck mightily when you're spending almost as much as you paid for your phone just to connect anything to it. Not even APPLE can TOTALLY get away with driving up interface costs by THAT much.
Intel's "lightning bolt" interface is wonderful as an open-ended port that can ultimately be used to connect just about anything to anything in full bidirectional splendor. If I could have only one physical port on my phone or computer, that's definitely the one I'd want. BUT... if it's your only port on the device, you end up turning something that's supposed to be cheap & straightforward into an expensive abomination. Do you REALLY want to go out and spend $40-200 on an external box that deserializes a 10+gbps lightning bolt bus into a virtual PCI express bus, hardwired to a PCI Express to USB root hub chip, hardwired to a USB audio chip and USB-I2C/UART bridge, so that both can be hardwired to a class D amp and give you something to output actual audio to your headphones?
Oh... I forgot... the above interface will also require you to root or reflash your phone so you can install the kernel module it requires, or do without it until you buy your next phone, since the likelihood that any carrier in America will give two shits about having the manufacturer add support for it to an existing phone is somewhere between "slim" and "none". Then find out that if you want to charge at the same time, you need ANOTHER $50-200 interface to multiplex the "headphone audio interface" and "power interface" into your elegant (if expensive and demanding) minimalist serial bus...
Minimalism might be elegant, but it rarely ends up being cheap.
> I'm an old Usenet fan, but am perfectly aware the ability to nymshift led to a culture of spamming and verbal harrassment that are basically unacceptable
No, it was the AOL users. The darkest day in the history of Usenet was the day they started allowing people to post to it from AOL.
Actually, all the condom needs is a switch that when open, leaves D+ and D- connected to nothing at all, and when closed, shorts D+ and D- with each other. By definition, if the phone sees that the D+ and D- pins are shorted together, the device is entitled to draw 1.7A from the power supply.
Officially, if D+ and D- are neither shorted nor able to negotiate for higher current, the device is only supposed to draw 100mA. In reality, everything I've ever seen besides Motorola's annoying phones ignores that rule and draws 500mA if there's power, but no negotiation, because anything that's GENUINELY capable of supplying only 100mA while powered up is almost guaranteed to be either a laptop running in "BIOS mode" that's capable of saying, "No!" (or cutting the power if too much gets drawn), a 99c power supply from China whose lifespan will be measured in days *anyway*, or a powered USB hub that isn't connected to a computer and can supply 500mA per port without breaking a sweat.
> We would all be stuck on 1.5mbps DSL lines because there is no cost effective way to push data quickly over consumer grade circuits
To a large degree, that's true. You can have the greatest digital signal processing on earth, but if you have a copper pair is of a length and quality that's right at the edge of what could do 1.5mbps ADSL back in 1999, the harsh truth is that you aren't going to do a whole lot better with a copper pair of equal length and quality today.
What changed was:
* the length of the wire (phone companies deployed remote DSLAMs and VRADs to neighborhood edges to get the distance down to ~1,500 feet or less)
* the wire quality (some neighborhoods had old, decrepit copper pairs replaced... which actually was kind of insane, because the cost to have replaced it with fiber might have been double the cost once you factored in the cost of the backhoe, workers, etc, as well as the future value of fiber relative to the future value of copper).
* availability of copper pairs, and willingness of the phone company to use them for pair-bonding. Circa 2000, lots of neighborhoods were literally maxed out, and phone companies were using DSL as a way to multiplex several customers onto a single copper pair. In fact, that's WHY DSL was invented in the first place, before the internet motivated companies to repurpose it for bulk data. As customers dropped additional lines that used to be for fax machines and modems, it freed up copper pairs to use for DSL. As more customers dropped lines that used to be their home's voice line, a LOT more copper pairs began to get freed up.
* ability of DSL modems to adapt to line conditions, instead of rigidly forcing "one speed fits all" the way g.lite did. G.lite was fixated on 1.5mbps/256k (128k?) because it was a number consumers were demanding that was relatively do-able within most of the former Bell System, and allowed them to avoid truck rolls for the majority of installations... at least, in the majority of American suburbia. Send out a line tech to do some local cleanup work & one-off tweaking, and you can make DSL happen on wires that would have been regarded as "unusable" by the phone company 15 years ago, and you can even do it using technology that existed back then. Much of it was just bureaucracy & unwillingness to act like a customer-focused business instead of a Soviet-era bureaucracy that viewed customers as a cost sink.
That's not to say there haven't been ANY improvements... but really, the technology hasn't changed. It just became cheaper, and the network edge moved closer to the customers.
The POINT is that actual hardware failure of a SSD is unlikely, so they can offer long warranties, but despite their long nominal lives, they're more likely to fail early (and often, and repeatedly) at the one task you bought them for -- reliably and robustly storing data.
Actually, the big problem with SSDs is the fact that unlike conventional hard drives, SSDs often DO have catastrophic data loss out of the blue long after the burn-in period, but long before the drive is even remotely "old". A conventional hard drive that's going to die tends to do it when the drive is new, before you've invested much in setting up Windows yet again. A Sandforce-based SSD is most likely to die 3-6 months after installation, then every 3-6 months after each "secure erase" resurrection, until the day you put it into panic mode trying desperately to recover a file you saved an hour ago that didn't make it to your latest backup.
> but OCZ isn't some shitty low-end no-name brand
You're right, because a shitty low-end no-name brand at least has the benefit of anonymity instead of a name that's tainted beyond repair or redemption.
The warranty period for a SSD is almost meaningless, because most SSD failures involve instant unrecoverable data loss by a drive that technically isn't "broken" in the legally-binding warranty sense, since you can "fix" it by erasing everything and reloading the virtual gun for another round of Russian Roulette.
Actually, there's an even bigger question... was the "failure" definition for SSDs carefully-crafted to exclude drives that experienced total data loss, but could technically be "repaired" and run for another 3-6 months before the next instance of total data loss?
At least when a traditional hard drive dies, it stays dead instead of coming back like an unstoppable undead zombie with an insatiable appetite eager to make a second attempt at destroying whatever data of yours it missed the first time around.
I can't think of a better gift for that special "frienemy" in your life than a Sandforce SSD. It's a gift that just keeps on giving... misery. Endless, eternal misery.
You forgot to mention the wonderful way Sandforce controllers encrypt the data with a key that you (the drive's owner) aren't allowed to have, so your ONLY data recovery option (on the rare occasions when it MIGHT be an option) is to pay an extortionate amount of money to one of Sandforce's "trusted partners" to decrypt it for you... and apparently, they actually charge more money to do what's now a 100% automated software-based recovery than the same companies USED to charge to remove the platters from a conventional hard drive in a cleanroom and mount them in a recovery unit.
I'm frankly surprised that some company like OCZ hasn't come up with an "innovative" new-economy (as in, "fuck you, consumer!") business model for SSDs, like selling 256-gig drives for $25 that lock themselves after some random period of time between 180 and 720 days after first use, and charge $2,000 for the key to unlock the drive (decreasing to $1,000 after a week, $500 after a month, $250 after 6 months, $125 after a year, and $64 after 2 years). Is there anybody who doubts that netbook manufacturers would pee their pants with glee if they could take advantage of that kind of "innovation"?
The craziest thing about Sandforce is the way their marketing department is somehow able to spin some of the worst design flaws in the history of computing into alleged sales points.
Their controller chips are the worst crock of steaming shit and snake oil to have disgraced the computer industry in *years*.
If The Onion's staff got together with a few sixpacks of beer, a pizza or two, and a mission to make up something funny for the "Tech" section of their next April Fool's Day edition, it would read like a Sandforce datasheet.
Let's be honest... if Wired Magazine got a press release proclaiming that Sandforce SSD controller chips maximized Windows performance by making users re-install it from scratch every 3-6 months, their editorial staff would have to call Sandforce's marketing department to confirm that it was real, and wasn't just a bored member of the Syrian Electronic Army playing practical jokes on them again...
Oops. Brain fart. It was the "Vista International" (at least, until Marriott bought it at some point after I was in Middle School). It's what happens when you grow up in a state like Florida with hotels literally everywhere, and your brain ends up using lossy compression to (somewhat) remember them. Occasionally, you'll remember a hotel whose name is partly right, and partly random fragments of other hotel names you've seen over the years.
Exactly. What good is a drive with a hundred million hours before physical failure if it commits data-suicide every 4-12 weeks? Maybe I'm just weird, but I don't *care* whether a drive (like the OCZ Vertex 2, for example) technically isn't "broken", and "only" has to be "securely erased" to temporarily make it usable again for a few weeks until its next episode of eternal amnesia. If it happens once, it's a random fluke. If it happens twice in six months, the drive is fucked and useless for its intended purpose.
> What makes you think you can't take FLASH devices and access them in a similar way to platters?
Sandforce controllers enforce mandatory AES encryption that can't be disabled, using a key that can't be recovered or set to a known value. So if your controller decides to quit allowing you to access your data, unsoldering the chips won't do you any good, because the values you read from them might as well be random noise.
The thing that really sucks about SSDs (at least, Sandforce-based drives) is the fact that 99% of their failures are due to firmware bugs that can be simultaneously triggered on an entire array at once (especially the sleep-related bugs). It's a mode of failure the creators of RAID 1, 5, and 10 never anticipated.
IMHO, the worst thing about SSDs (at least, those with Sandforce controllers) is the fact that they have mandatory full-drive encryption that can't be disabled, using a key you aren't allowed to set or recover, and gets blown away whenever you reflash the firmware. This means, among other things, if the drive's controller gets itself confused:
* You can't reflash data-recovery firmware onto the drive. The act flashing it would blow away the encryption key and render the data gone forever.
* If the drive decides you're trying "too hard" to systematically extract data from it while it's in a confused state, it'll go into "panic mode" by blowing away the encryption key. If this happens, your data is gone forever AND you have to send the drive back to OCZ or whomever you got it from in order to get it unlocked. For your protection, of course. And Hollywood's. Among other things, dd_rescue/ddrecover can trigger panic mode.
* You can't even do the equivalent of removing the platters from a conventional drive in a clean room and mount them to another drive for reading, because the data on the flash chips is all encrypted, and the key is unrecoverable.
This is BULLSHIT, and it's why I refuse to buy any more SSDs. I, as an end user, should be able to download a utility from somewhere, reflash the drive to firmware that includes an offline recovery mode that simply dumps the flash chip content from start to finish, and either disable the encryption or set it to a key *I* control, so the 99.99999% of the data on the drive that's good when the embedded firmware freaks out can be dumped and recovered offline.
If there's a God, Linus will go NUCLEAR over this, get a few seconds on CNN & other networks to rant about the unreliability of SSDs, and scare enough consumers to hit the industry HARD where it'll hurt the most... their bank accounts.
It might not be possible to make SSDs reliable, but DAMMIT, they should at least be RECOVERABLE. There were goddamn hard drives with recoverable data pulled out of laptops left in safes in the Vistamark hotel when a tower sheared it in half and buried it under flaming rubble, yet a SSD that dies if you so much as look at it the wrong way due to firmware bugs ends up being fundamentally unrecoverable for no hard technical reason.
And yes, I'm bitter about having my hard drive commit suicide for no reason besides Sandforce Business Policy. As long as they keep making controllers that cause drives to self-destruct at the drop of a hat, I'll keep doing my best to talk people out of buying drives tainted by their controller chips. Sandforce sucks.
Actually, people in China DO occasionally draw characters in the air with their fingers to disambiguate spoken homonyms. And it's clearly viewed as important by English-speakers, because it's the main reason why we still have silent letters and unusual spellings. Look for every instance of an "odd" English spelling, and there's almost always a homonym hiding behind it.
Way fewer than you think.
Interesting factoid: the only city in America where a Spanish-language TV station has EVER been the #1 station in its media market is Miami (Telemundo, I believe). In San Juan, where Spanish is both the de-facto and official language, the two top channels have always been Fox and NBC. And in Miami's case, it was kind of a weird situation where somebody at Univision said something non-condemnatory about Fidel Castro, Univision refused to denounce and fire them, and for a few years if you were Cuban in Miami, it was (*cough*) socially frowned upon to admit in polite company that you watched Univision as a restult. Meanwhile, Miami had 6 or 7 English-language networks prior to the big consolidation a few years ago, and no one other channel could match the popularity of "Betty la Fea" (just to give one example), so Univision ended up slightly beating out everyone else by a small plurality.
Now, for the harsh truth: it's impossible to live in the United States for 25+ years without achieving some degree of English competency, whether it's intentional or not. Even if you went out of your way to surround yourself only with people who speak Spanish, watched only Spanish TV shows, and read only Spanish publications, every time you leave the house you're going to be surrounded by English... even in the most hardcore Spanish enclaves in America. And regardless of how hard YOU might try to avoid English, your kids are going to watch English-language TV shows and movies with their friends, speak English at school, and around you whenever they don't want you to know what they're talking about with their friends.
Let me emphasize the last point, because I've had to cruelly disillusion many of my Cuban friends who were absolutely *convinced* that their parents, or at least their grandparents and elderly aunts/uncles "couldn't speak English", and would say things in front of them (in English) as though they weren't there. Without exception, they were wrong, and were horrified when the truth finally came out. I still remember when one of my friends told me, "My mom doesn't know English". I pointed out that she was a paralegal for a law firm. He countered with the excuse that it was in Hialeah. I replied that even if the law firm were in Hialeah, there was no... way... in... HELL any law firm would have hired her if she literally didn't speak a word of English. Six months later, he admitted I was right, and that he was shocked to discover that his mother was quite fluent in English. She just never spoke English around *him*, because she liked being able to eavesdrop on him when he was talking to his friends :-D
As a practical matter, anybody who's born in America is going to be fluent in English by the time he's a teenager... EVEN IF his parents try to send him to schools where Spanish is spoken. Anybody who's lived in America for more than 10 years is going to have a grasp of English, and by the time they've lived here for 25 years, they're going to know English at least as well as my 100% Ohio-bred pure-Gringo parents know Spanish after living in South Florida for 25 years. And if they're emigrating to the US via official means, they almost certainly have a college degree, which almost guarantees that they've studied English for at least a few years.
> honestly believe that the Chinese should switch to some sort of romanization
http://en.wikipedia.org/wiki/Lion-Eating_Poet_in_the_Stone_Den
The fundamental problem with romanized Chinese is the fact that nearly every word in Chinese has multiple homonyms (to/too/two), even AFTER you take into account the various inflections called "tones" (which are really just ways of formally representing verbal inflections in writing).
English disambiguates homonyms with silent letters and alternate letter combinations. If Chinese followed the same strategy, the romanized spelling of Chinese words would be almost completely arbitrary, and Chinese kids would spend years memorizing the difference between "shi", "she", "shee", "shii", "shie", "schi", "sche", "schii", and "schie" (plus appropriate tone marks). In the end, it wouldn't be much of an improvement... assuming it were any improvement at all.
At one time, Chinese had a serious "keyboard problem", but it's been largely solved by keyboards like Wubizixing ( http://en.wikipedia.org/wiki/Wubizixing ) and Wubihua. At the simple end, Wubihua assigns 5 keys to the most fundamental strokes used to write Chinese: horizontal, vertical, left-falling, right-falling/dot, and hooked/complex. You press the keys corresponding to at least the first 4 strokes, then press the key corresponding to the last, and it presents you with a list of plausible characters that match. The more keys you press, the smaller the list gets, until you're left with either an unambiguous match or you've entered all the strokes.
Other methods, like Wubizixing, go a step further, and assign keys to the radicals themselves (if you think of characters as being like molecules, radicals are atoms, and strokes are quarks; in English terms, characters are words or stems, radicals are letters, and strokes are the way you'd write those letters... like "vertical, vertical, horizontal" for "uppercase H"). Somebody who's good at typing on a Wubizixing keyboard with the key-cadence of somebody who types English at ~100wpm can achieve an equivalent word-rate of about 120-150wpm (because Wubizixing makes more efficient use of the keys on the keyboard, and requires fewer keystrokes per communicated-word than English QWERTY).
The irony is that most people in China are amazed when they first encounter a Westerner who can type on a Wubi keyboard (-hua OR -zixing), because they think they're "too hard" to use. The reality is that stroke-based input is REALLY the only way somebody who doesn't know how to speak Chinese CAN enter characters on a keyboard. There's definitely room for algorithm-improvement in a "westerner-friendly" stroke-based input method, but I can guarantee that whatever we end up with ~10 years from now, it's going to look more like Wubi than anything else. It'll just be more forgiving of someone who enters "zhong" (level 'o' tone) as "vertical, horizontal, vertical, horizontal, vertical" (or some other permutation) instead of "vertical, hook, horizontal, vertical" (just to give one example).
As for "too hard", Wubizixing really isn't any harder for someone in China to master than QWERTY is for someone in the US. For geeks who type all day, every day, nonstop, it's a skill that pays HUGE personal dividends. For people who think computers in general are "hard to use", it doesn't really matter whether they're American or Chinese... they'll dick around with two-finger hunt & peck or Pinyin input, and endlessly predict the death of keyboards in favor of speech recognition. The rest of us, American and Chinese, will laugh at them and keep typing 120-150wpm while they struggle to send email and text messages with amusing autocorrect errors.
Anyway, getting back to romanization of Chinese... it's not going to happen. Chinese has romanized as much as it's ever going to romanize. Twenty years ago, keyboards and fonts were real problems. Now,
The problem with CDMA phones supporting Sprint and Verizon has nothing to do with *bands*, and everything to do with firmware licensing and Sprint's business policy.
* In Sprint's entire history, I think they've had maybe 3 or 4 phones that WEREN'T dual-band and capable of using Verizon's frequencies. Sprint phones ROAM on Verizon in many rural areas.
* The CDMA2000 equivalent of a SIM card's IMEI is the MEID. There's no technical reason why it can't be on a SIM-like card, just like GSM phones use. In fact, Qualcomm itself defined a standard called R-UIM that does precisely that, and was used just about everywhere in the world besides the United States back when CDMA phones were common in countries like Australia, New Zealand, Singapore, India, Brazil, and Chile. The problem is, Sprint refuses to activate any phone whose MEID isn't in their computer as a Sprint-branded phone sold by Sprint. Verizon, for various legal reasons, can't refuse to activate a non-Verizon phone, but unless you have an inside connection to help you get it working, your likelihood of getting a Sprint phone to ever work properly on Verizon is low ("properly" == "voicemail, MMS, and similar things work properly"). Moreover, without actual Verizon radio modem firmware, a non-Verizon phone will never be able to do EVDO on Verizon, and will be limited to 1xRTT. Basically, Verizon authenticates its phones differently to EVDO than Sprint, and AFAIK, nobody has ever independently gotten a non-Verizon phone to do EVDO on Verizon ("independently" == "doesn't work for Verizon or Qualcomm"). People HAVE gotten Sprint phones that are identical twins of Verizon phones to do EVDO on Verizon by reflashing them with a Verizon radio modem, but AFAIK, it's been *years* since anybody has gotten that trick to work.
But yeah, 99% of the bullshit and incompatibility we have in the US comes down to Qualcomm's licensing policies. As far as Qualcomm is concerned, their REAL customers are Verizon, Sprint, AT&T, and (to a lesser extent) T-Mobile, US Cellular, and a few others. Qualcomm will NEVER knowingly assist anybody who wants to do an end run around the carriers, and they're 100% cool with the status quo of carrier-locked phones. I'd go so far as to argue that if push came to shove, Qualcomm will probably be resisting manufacturer-licensed firmware long after the carriers themselves have gotten beaten into submission by the FCC, because carrier-licensed firmware goes to the deepest core of their existence.
If the FCC prohibits carriers from demanding vendor-licensed firmware, Qualcomm's lawyers will be busy finding ways to let carriers say "the devil made me do it" and continue to enforce the policies LONG after the carriers themselves are allowed to legally *demand* it (even if they ARE allowed to legally "favor" it by choosing to buy Qualcomm-based phones instead of other phones for sale to their own customers).
The only way we're ever going to have truly carrier-agnostic phones in the US is if the FCC starts refusing to certify any phone that can't pass the compatibility test suites of every major network in addition to the interference-related ones the FCC itself cares about now. It'll happen... someday... but you can BET Qualcomm's lobbyists in Washington won't be the ones encouraging it.
And it's sad. Qualcomm has some of the world's best engineers, and makes some of the best chipsets on earth. I've seen demos of stuff they have in the pipeline, and it's really awesome... but every time I see their demos, I have to remind myself that we won't get to get anywhere near those cool features on American phones until at least 2-5 years after they're commonplace elsewhere, because American carriers only care about features that directly translate into higher monthly revenues for them, or features they can claim hardware exclusivity on & use as a way to enforce vendor lock-in with.
> That work cannot be "off shored"
You'd be surprised how cost-effective it can be to manufacture pre-stressed concrete bridge modules at a factory in China, load them onto a barge, and ship them to California. By definition, pre-stressed concrete bridge modules have to be manufactured off-site somewhere, and 99% of the transportation costs involve loading and unloading. Whether the actual long-haul transportation occurs via superfreighter from China, freight railroad across the US, or post-Panamex freighter from the east/gulf coast of the US, doesn't really make all that much of a difference.
The main obstacle to actually DOING it isn't even "buy American" laws, so much as China's notorious "quality fade". God forbid, if you built a bridge with Chinese bridge modules, then ONE of them were discovered to be substandard, I can *guarantee* that the government entity paying for the bridge would end up replacing and demolishing the entire new bridge (or at best, leaving it in place for pedestrians and cyclists, if they thought it could safely handle the lesser dynamic load, alongside the new bridge) because they'd never be allowed to risk having a large-scale failure due to another substandard module somewhere in the bridge.
That's part of the reason why stuff from American, Japanese, and European companies costs several orders of magnitude more. We actually *do* have the ability to track every bolt, screw, piece of rebar, and pound of aggregate every inch of the way from mine to megastructure, and somebody with enough funding and determination can spend years digging through the documentation to verify it. China isn't quite "there" yet, and doesn't appear to even *want* to go there.
It's a fairly big paradigm shift, because it's the difference between "build two bridge spans that, when used within their explicit design parameters over their guaranteed design lives, must never, ever, EVER fail" and "build three spans for half the cost of building one American span, and roll the dice that when/if one fails before it's functionally obsolete and has to be replaced anyway, it won't kill or injure too many people", and just take for granted that you'll probably end up demolishing and rebuilding the whole thing to increase its capacity *anyway* long before it becomes a problem.
What China appears to have failed to appreciate, based on what's happened in cities like London and New York, is that when the future arrives, building the replacement won't necessarily be easy or cheap, because adjacent land use will have grown up around it, and taking the old infrastructure out of service to make room for its replacement just isn't a viable option. The Big Dig comes to mind as an obvious example of a project that would have been expensive in any case, but was made exponentially MORE expensive by the need to keep the old road in service throughout the construction.
The Washington Metro has a similar problem... everyone agrees that it has a serious capacity problem in stations between Rosslyn and L'Enfant Plaza that are only going to get worse once the Silver Line begins service, but there's no cheap or easy way to enlarge the existing stations. They were built 25-40 years ago in holes that were excavated, but now have expensive buildings sitting on top of them... and mining out new platforms from below would cost *almost* as much money as demolishing the buildings above and digging new holes. The London Underground has been dealing with the same problem for decades, and now spends almost as much money (adjusting for inflation) to mine out and rebuild a single station as it spent to build the entire original tube network ~150 years ago.
The worst thing about living anywhere near a "bad" neighborhood are the endless car break-ins that the authorities can't do jack shit to stop. The Coconut Grove area of Miami, and the adjacent neighborhoods in Coral Gables (where I used to live) are a perfect example. Thanks to both explosive gentrification and the enduring legacy of old-south segregation-era zoning laws, there are plenty of areas where you literally have expensive homes back to back with housing projects that will never go away.
In those areas, you can never have guests come over to see you unless they park elsewhere and take a cab, because YOUR BUILDING's parking garage might have 2 layers of gates & security, but for obvious logistical reasons, the guest parking sits unprotected out in the open. Let me tell you... the only thing that sucks worse than getting your own parked car broken into is having friends come to see you, and getting THEIR OWN car broken into. Or god forbid, your parents' car. If your parents' car gets broken into, you will NEVER be allowed to forget about it. My parents STILL bring it up at least two or three times at family gatherings on Thanksgiving, Christmas, and the Fourth of July, and it happened more than a DECADE ago.
Ask anybody who lives in an urban neighborhood what their #1 neurotic fear is, and they'll tell you -- "Friends coming to visit, and getting their car broken into". On the hierarchy of social shame, it pretty much tops the list. From that point forward, you no longer live in a nice, safe, gentrified urban neighborhood. As far as your friends and family are concerned, you live in the 'hood.
The beginning of the end for Nokia happened around 2004, when UMTS arrived in Europe, and Nokia made an intentional business decision to not support EDGE, and to basically walk away from CDMA as well (even though at that point, probably half the phones sold by Verizon and Sprint were Nokia, as well as the majority of high-end phones sold by AT&T and T-Mobile).
As a result, their phones became useless paperweights in the US as far as anybody who ever used data was concerned. EDGE wasn't exactly "high speed", but compared to GPRS, EDGE is just "annoyingly slow" compared to "uselessly slow". Circa 2005-2008, EDGE was the best that existed in most of the US anyway... T-Mobile hadn't even started deploying HSPA yet, and AT&T's HSPA data existed in maybe two dozen cities.
Nokia presumably wrote off the US market because, in terms of total unit sales, it was roughly equal to Portugal or Switzerland. What they overlooked was the importance of mindshare... half the world's tech blogs and web sites are American, and as far as anyone in America was concerned, by ~2007 Nokia had effectively ceased to exist. At best, they were a company that used to be popular, and now just made throw-away low-end phones sold to people in remote African villages.
Other companies learned their lesson. Today, companies like Sony-Ericsson are working as hard as they can to break their Qualcomm addiction(*), and make a point of getting their phones into the hands of American reviewers who live in cities where T-Mobile has good HSPA+ coverage.
(*) Qualcomm insists on licensing LTE radio firmware to carriers rather than manufacturers, which means it's basically impossible for a manufacturer to sell phones capable of using LTE on AT&T or T-Mobile without the active involvement of AT&T or T-Mobile, and de-facto impossible to sell a phone built with a Qualcomm LTE chipset that's carrier-agnostic and capable of doing LTE on both AT&T and T-Mobile.
It's technically possible to use a separate non-Qualcomm chipset (like Beceem's) for LTE, but the price premium is fairly stiff (about $100, by the time the phone gets to retail stores). That's why companies like Sony-Ericsson (who desperately want to break the stranglehold American carriers have over the American phone market as gatekeepers with economic -- or in the case of Verizon & Sprint, real -- veto power) have eagerly embraced chipsets like the Renesas MP5232 and MP6530, which will enable them to make phones capable of doing LTE on AT&T and T-Mobile, and break the "LTE Lock-In" AT&T in particular has been working overtime to exploit as a way of making their nominally-GSM network into one that's as de-facto proprietary as Verizon's.
Nobody outside the US, or who's vacationed in Florida (or at least the southern US) has any idea WTF "Key Lime Pie" is, especially people in China & India.
Hell, I *GREW UP* in Florida, and was in HIGH SCHOOL before I had any idea what Key Lime Pie is. I'd be shocked if a random developer in Bangalore or Hong Kong, let alone Mumbai or Beijing, would have had the slightest idea what it was without researching it on Google.
Kit Kat, in contrast, is a global brand. Not necessarily a #1 item like M&Ms, but it's known. And it sounds like a halfway-reasonable word to people whose native languages are Hindi or Mandarin.