The difference is, by eliminating "drive-by" attacks by requiring a little bit of real-world physical work by end users to enable rewrites, you can morally and legitimately blame stupid end users who do it so they can watch the dancing bunnies. I like the idea of printing the key on a sticker attached to the motherboard. I like the idea of a jumper that has to be shorted out to enable overwrites even better. It can even be a tiny switch on the rear (among all the ports) that has to be flipped while the computer is off, then flipped again in response to a prompt, to perform the overwrite. Just forcing users to shut down, flip the switch, launch flash mode by holding down 3 unlikely keys while powering up, then flip it again in response to some non-obvious prompt documented on page 739 of the pdf instruction manual, will weed out 99% of the aspiring dancing bunny victims because they won't have the attention span or motivation to go through with the hassle. The remaining 1% who do it to watch dancing bunnies are hopeless future Darwin Award recipients.
The solution, if Microsoft wants to be the non-Evil (if not actually "good") guys, would be to require UEFI secure boot AND require that the key be furnished to end users for logo compliance. If they're worried about social engineering, they can put it someplace where it won't stop anyone who's likely to care about Linux, but be a substantial barrier to clueless end users who'd be a danger to themselves and others if they had it. Say, a sticker on the motherboard (or, for laptops and factory-built PCs, under a factory-installed SoDIMM), with a second peel-off copy that can be removed by end users and pasted somewhere convenient.
LOL. Actually, I'm a bit of a geezer too (hint: I was a freshman in college the night I went to Incredible Universe at midnight to buy my copy of WIndows 95). One of my favorite laptops was the Zeos Meridian, which just so happened to have an IBM-style pointer stick in the location I'm talking about (and unfortunately, was utterly out of my price range at the time) -- http://cdn2.iofferphoto.com/img3/item/211/480/688/dIaQ.jpg
Actually, you could probably argue that there's a certain amount of de-facto herd immunity in places with antibiotic abuse, simply because people who get infected by syphilis are likely to come down with a sore throat or upper respiratory infection, and head home with a prescription for some powerful broad-spectrum antibiotic likely to blow the syphilis away with trivial ease long before they ever realized they HAD syphilis. Apparently, syphilis is pretty easy to cure, and even penicillin is good enough to do the job in most cases (let alone something like Levaquin, Zithromax, Velosef, Biaxin, or Keflex).
Ah... but you're forgetting one very, very big difference between personal debt and sovereign debt: US federal debt is dollar-denominated. It would be kind of a nuclear option, but if push came to shove, the US could truly print its way out of debt. Furthermore, unlike countries like Iceland and Greece, the US has some of the most abundant natural resources on planet earth. The US does, in fact, have options available that private citizens who owe money don't. The worst thing the US could possibly do is sell off its natural assets to foreign owners to pay down debt. It would be the financial equivalent of taking out a $200,000 home equity loan on a house that's mostly paid for to pay credit card debt that could be washed away by bankruptcy. Morally admirable, perhaps, but financially stupid.
The US also has an advantage few other countries enjoy besides China -- in addition to having sinfully abundant natural resources, the US has a gigantic domestic market with the electoral muscle to prevent those resources from being exported wholesale for a pittance instead of sold domestically. If Congress, the Senate, and the President decide that Exxon isn't going to be allowed to sell its Alaskan & CONUS oil to anybody besides Americans... it's not going to be sold to anyone besides Americans. If they try to take their ball & go home (as if "they" could go anywhere besides maybe Switzerland or Dubai) by shutting them down, they'd be nationalized in 3 hours, and their employees would be back at work the next morning as government employees. You can argue about whether it would be morally right, or a good idea... but the fact is, it would happen.
It's also important to remember that the US can get away with way more inflation than other countries, because the true value of the US dollar lies with its nearly infinite liquidity and network effects. Simply put, the dollar has value as currency because everyone else has vaults full of dollars to use as currency. Even in cases where you have a very strong currency, like Norway or Australia, by the time you factor in the bank fees to convert to any other currency, you're better off just leaving your (US) dollars as (US) dollars and spending them to buy online pr0n instead of handing 5% of your wealth to a bank somewhere just to exchange your dollars for some other currency.
Don't get me wrong. I'm not advocating the expenditure of money just for the sake of spending money. This is precisely the time when the government should be buying up all the freeway right of way (at fire sale prices) that it's ever wanted to own for future expansion projects, 16-laning every major freeway in America, running fiber to every halfway-urban neighborhood (not necessarily to every house, but at least getting it close enough to run cat6 cable for gigabit ethernet the last thousand feet from pod to home router), burying power lines in hurricane-prone areas, and generally investing in the kind of high-quality infrastructure that will still be useful 50-100 years from now, instead of doing stupid things like subsidizing half the purchase price of electric cars so environmentalists can feel smug and superior to their neighbors.
> Deflation is much better than inflation in case of recession, because it allows the poorest people to survive, > as they can buy more with whatever meager savings they have.
You're overlooking the fact that statistically, the poor (and for the most part, the middle class, and a fairly large chunk of the upper class as well) have no real savings to speak of once you factor in the negative value of their pre-existing debt (student loans, upside-down mortgages, and credit cards with 14% interest rates despite a nominal prime rate approaching 0%) that must now be repaid in deflated currency. In the entire United States, you'd be lucky to find 10,000 people who'd be genuinely better off under sustained real deflation. The remainder would be crucified by an impossibly crushing debt load hemmed in by falling wages for those who can even manage to get a job. On the other hand, you wouldn't have to look very hard at all to find 299 million Americans who'd benefit from sustained moderate inflation (and about 50-100 million towards the upper middle class end of the spectrum, whose debt is primarily fixed-interest long-term debt like student loans and mortgages, who'd effectively see their debt washed away within a couple of years).
> deflation would have allowed the bad debts to be liquidated and people would save and restart production.
Newsflash: nobody sane is going to invest a million dollars producing goods likely to have an ultimate resale value of $950,000 (ie, goods whose ultimate value is less than the cost of their production) when they can just put the money in a vault and wait for its absolute value to appreciate on its own. Especially when those same businesses can't get loans to finance the production of goods. Even under moderate deflation, it becomes almost impossible to raise investment funds because investors won't settle for anything less than zero risk with an impossibly high return.
Try this: name ONE ERA in modern history marked by deflation that's looked back upon fondly by the people who experienced it firsthand. I don't think there's anybody who has ever genuinely said, "You know, the Great Depression was a wonderful long-term boost for our nation's economy, and I'm glad it happened!"
>USA doesn't need more consumption, it needs real recession to kick in >and it needs underconsumption and savings so that real capital can be >once again accumulated to restart production
Unfortunately, the recession you're welcoming with open arms would end up being deflationary. During periods of deflation, the best thing a wealthy person can do is put his cash in a safe and forget about it, because just about anything he might try to invest in is likely to end up causing him to lose money.
Inflation isn't good, but deflation is deadly... especially to any industry with long supply chains and timeframes (ex: automakers and building construction), because just about anything you try to build is going to end up being worth less when you're finished than you spent to build it in the first place (with the possible execption of merchandise with extremely high markup, like a Rolls Royce, and construction work that's mandatory, like roof repair after a hurricane; neither case will generate enough economic activity to keep their respective industries alive, let alone prosperous).
Inflation is Adam Smith's way of forcing the wealthy to invest their money and put it to work, by slowing depleting its value if they try to just sit on their capital for too long without putting it to productive use. High inflation (and wildly unpredictable inflation) is bad because it induces unproductive behavior (people end up spending lots of time managing their money instead of earning more), but a small, stable amount of annual inflation (let's say ~3%) is a good thing.
> Sprint and T-mobile use different technologies and have different market shares.
Believe it or not, the difference is more one of business policy than actual technology. As a practical matter, any phone built like the Motorola Photon could easily work on Sprint and T-Mobile (it can't now, because Sprint had the UMTS radio specifically created to block use of AT&T and T-Mobile). With the next version of the baseband chipset, it could probably even do HSPA+ ("4G on T-Mobile") as well.
If Sprint bought T-Mobile, the first thing they'd do is require all new high-end phones -- Sprint AND T-mobile -- to be like the Photon & able to do CDMA2000, UMTS, GSM, and wimax. After a year or two, they'd start repurposing some of their 1900MHz "Sprint" spectrum to UMTS uplinks, and pair it with chunks of T-Mobile's 2100MHz UMTS downlink spectrum. It wouldn't break a single T-Mobile phone, because they can all do 1900/2100MHz UMTS as easily as they can do 1700/2100MHz UMTS anyway.
Sprint would keep circuit-switched CDMA voice, but start phasing out EVDO in favor of UMTS for data, just like Telus did in Canada (except Telus skipped EVDO and went straight to UMTS). After 3-4 years, when all their high-end users were solidly migrated to phones capable of UMTS data, they'd probably shut down EVDO entirely in markets with tight 1900MHz spectrum (like San Diego) and only provide 1xRTT to the few remaining users who couldn't use UMTS (1xRTT can coexist alongside CDMA voice and dynamically share channel space with it, whereas UMTS needs dedicated spectrum). I believe this is what CDMA carriers in India have done as they've migrated users from EVDO to UMTS for data.
5-10 years down the line, the distinction between "Sprint" and "T-Mobile" would be academic. The towers would all be shared. Sprint would use a chunk of 1900MHz spectrum for legacy GSM voice & 2(.5)-G data (GPRS and EDGE), another chunk of 1900MHz spectrum for circuit-switched CDMA voice and 1xRTT data, and a third chunk of 1900MHz paired with T-Mobile's 2100MHz for 1900/2100MHz UMTS. They'd probably pair a fourth chunk of 1900MHz with their ~850MHz Nextel spectrum for premium UMTS (850 uplink, 1900 downlink), and use their 1700MHz spectrum for LTE. In the meantime, Sprint and T-Mobile customers with new phones would both get to suffer with Sprint's crap 4G service ("crap", because it doesn't work in moving vehicles due to the way Clear fucked up their tower-tower hand-offs... or more precisely, didn't bother to implement at all, so they all act like wifi access points with ~1km range instead of a real cellular network; instead of handing off gracefully, Sprint 4G just drops the connection and leaves you with no network connectivity for 10+ seconds while it handshakes with the next tower and gets a different IP address).
Anyway, the point is that there IS a graceful way to merge Sprint and T-Mobile's networks. That said, if Sprint and T-Mobile were to "merge", I'd only want it to happen if the Justice Department made them:
1. sell all their tower assets to TowerCo (the company Sprint created to own its towers), to maximize their availability for lease on open terms to competing networks
2. create a new company to handle everything RF-related. In other words, own the spectrum leases and run the CDMA2000, GSM, UMTS, and LTE radio services. Let's call this "RFco", and view them as the equivalent of an RBOC.
3. keep "Sprint" and "T-Mobile" as separate companies who provide their own data backhaul & provide actual PSTN connectivity, sell phones (but require Sprint to fully support USIMs, require both companies to allow the use of any hardware that's physically compatible, and publish their standards and protocols in open format to allow end users to implement them on their own hardware as desired).
In other words, Let's suppose you're a "Sprint" customer. You might (or might not) buy your phone from them, but it has a USIM (which works on everything from UMTS to CDMA), so Sprint can't lock you into their own propr
Please. There's no way in *hell* you could type arbitrary text that's not conversational English at 50wpm using Swype, if only because one out of every 20-40 words would end up getting mango into some other word by its prediction album, and everything you wrote would make you look like the 21st century reincarnation of http://en.wikipedia.org/wiki/William_Archibald_Spooner
I personally can do about 30-40wpm with Graffiti when I'm really into it and writing at full speed. It does admittedly take more motion than Swype, but has the advantage that once you're good at it, you really DON'T have to pay much attention to the screen while writing. Turn off prediction, disable character-by-character searching, use it on a 1GHz+ phone, and it really IS accurate enough to let you get away with blind-writing.
^^^ Exactly. People have forgotten that PCs were revolutionary BECAUSE they devolved control and power away from centralized IT departments, and put it directly in the hands of end users who could skirt bureaucracy and do cool, new useful things without having to wade through months of committee meetings first. Those who don't remember the past are doomed to repea...NO CARRIER
Fine. Let my desktop PC use UPnP to configure my router, let my router update my DDNS hostname, and let my Android phone & tablet sync directly with my desktop PC. Maybe add a server appliance running Samba into the equation. No need to screw with proprietary online services that either cost lots of money or can vanish tomorrow without warning. I guess we're the geezers now, but anyone old enough to remember dotcom services we depended upon disappearing overnight (or mid-afternoon), never to return, is unlikely to ever fully trust "the cloud", let alone any proprietary service provided by it that locks you in and gives you no way to escape to freedom with your data.
> Windows 7 works just fine. It's the new XP - didn't you know?
It's sad, but you're probably right. Microsoft today is kind of like a rock star who's made so much cash, he's just going to be weird and do whatever the fsck he feels like doing from now on. If Microsoft is hyping "Metro" in an effort to generate developer excitement, they're having the exact opposite effect. Everyone *I* know is like, "WTF, has Microsoft gone completely batshit insane?"
It's almost like Microsoft's entire developer elite just hit their mid-40s, had a midlife crisis, realized they have enough cash to spend the rest of their lives coding for fun, retired en masse, and handed over the company to a marketing department that thinks making Windows look like a tablet UI so it can run phone apps better is somehow a good idea.
> Unfortunately, there are no good implementations for modern phones
Not quite true. Access-nee-Palm (whatever name they go under now) semi-officially sells Graffiti for Android. There's a free ad-supported version that embeds ads in the graffiti area, and a $2.99 version without them. It's worth every penny. It's the ONLY Graffiti-1 implementation I know of for Android that properly recognizes all the alternate strokes hardcore PalmOS users have burned into our motor memory forever, like backwards-v for "V", up-over for "t", "6" on the "letter" side for "G", etc. I just wish there were some good way to do cursor-up and cursor-down, because Connectbot is quite painful without them.
I personally couldn't live without Graffiti. When I bought my Epic4G, I came within a day of taking it back and exchanging it for an Evo because Graffiti didn't work properly on the Epic (you had to press and hold for about a second to register a "dot" gesture), but after I filed the bug report, they literally had it fixed in 2 days. I can say that Android Graffiti is the best Graffiti-1 implementation that's existed since the Palm V (the Treo-era Graffiti-2 implementations sucked). For anybody who reads this, if you used to like Graffiti on Palm, try it on Android. You'll feel like ${deity} has personally heard your cries and answered your prayers. You'll feel like you can finally input text properly again for the first time in years.
IBM (now Lenovo) has the right pointer-stick mechanism, but the wrong location. The stick SHOULD go directly below the spacebar, aligned by a line perpendicular to the spacebar that would run through the center of the "B" key. Fujitsu got the right location, but patented a crap pointer-stick mechanism. Unfortunately, Fujitsu patented the stick's LOCATION too, so nobody besides Sony (with a sufficiently-large patent portfolio to bully Fujitsu into submission) has ever dared to put the stick there.
Don't believe me that it's the ideal location? Try this experiment: hold your hands so the thumb is over the Trackpoint the same way it would be if it were centered below the spacebar and try moving it. Ahhhh. Much nicer, isn't it? Your thumb is a lot stronger, so you can give the stick a good hard shove to hurl the pointer to the other side of the screen quickly instead of exerting your index finger in a hyperextended position.
Personally, I think keyboards should have TWO Trackpoint-style pointer sticks: one in the "GHB" triangle, and one centered below the spacebar, with 4 buttons flanking the lower stick (2 per side, placed so whichever thumb ISN'T on the stick can easily press one or the other, keeping in mind the ~45-degree axis of comfortable motion for the thumb. Then, make them all software-configurable, so you can assign the 4 buttons to be "left", "right", and/or some other function as you desire, and either use both pointer sticks as mice, or reassign one to be for scrolling/panning. I'd personally use the "GHB" one as a faux scrollwheel;-)
It's called two-way HD video chat with realtime collaborative desktop sharing. Incoming video from the other guy? Check. Video-capture and encoding to send him? Check. RDP, VNC, or some comparable protocol? Check. Gimp? Check.
It's kind of an extreme example by current phone standards, but not much beyond what you could do today with a fairly high-end laptop from Dell (or a fairly run-of-the-mill one, if you have dedicated hardware to handle the h.264 realtime capture and encoding, and a video chipset with hardware-assisted decoding).
The problem with predictions that future computers will be more power-efficient is the reality that users and programmers inevitably find creative ways to soak up 150% of any improvement to processor speed. On one hand, we all bitch about 3GHz quadcore PCs that feel slower than a 400 MHz Pentium 3 did running Windows 98... but none of us could stand using Windows 98 anymore, because it would feel almost like a toy compared to anything modern. And before anyone brings it up, fully tricked-out Linux is every bit as bad as Windows -- maybe even a little worse. Hell, look how long 1.6GHz Atom chips lasted before Intel had to boost the cores, clock speed, and cache enough to run Vista acceptably fast. Any long-term strategy that depends upon reducing speed for the sake of power savings is doomed, because mainstream software and consumer expectations will outstrip it and render it obsolete long before it hits the store shelves.
Perfect example: Pentile RGBW. On paper, it's awesome... good black & white text, great battery life. Put any Pentile RGBW display next to Super AmoLED+, and it's going to look like total crap and leave its owner feeling shortchanged because it just doesn't have the searing "snap" and saturation. It might not look as good for black & white text, and might not last as long battery-wise, but it's like the "Pepsi Challenge" from the 70s and 80s -- put the two side by side, and people will always make the impulse decision to grab the pretty one, regardless of intellectual arguments and pragmatism.
We can't get through an entire day on a single battery because every manufacturer feels compelled to make its new phone thinner than last year's iPhone. If companies like HTC/Samsung/Motorola would just bite the bullet and add a millimeter of extra thickness, we could have 4000mAH batteries that could make it through a day of active use & still be alive the next morning when somebody calls to wake you up.
~6 years ago, Samsung felt compelled to give every SPH-i500 owner two free batteries (one extended-life) because of its *scandalously poor* battery life -- fully-charged, it could barely make it through eighteen hours with the non-extended battery. Today's iPhone and Android owners would *kill* for that kind of battery life, but the fashionistas who dream of credit-card thickness phones won't let us have it.
> If it can play HD video (and it already can) then what else do I want performance for?
Play HD video on the screen while capturing and encoding realtime 720p60 from the camera, streaming it in realtime through a software firewall and VPN while using BGP to negotiate a multilink connection via EVDO/UMTS, LTE/Wimax, and/or Wi-Fi (depending on which one(s) are the best-available network link at that instant in time), and juggling your antivirus suite, anti-rootkit scanner, checking your IMAP mail server for updates every 3 seconds, keeping your Twitter timeline up to the second, and editing 3 ~4000x3000 HDR pics you took with the camera in its highest-resolution mode in Gimp. And somehow, manage to not crash or choke when somebody calls you, and it has to gracefully notify you about the incoming voice call and somehow allow you to answer it without seriously interrupting everything else you're trying to do with your phone at that moment.
> It's proven, it's developing and has no legacy dragging it back.
Yes. Until you try and turn it into a multi-core architecture with parallel branches, speculative & out-of-order execution, and all the other things x86 does in its sleep that are virgin territory with ARM. At that point, it's no longer proven and mature -- it's Internet Explorer 6 with lots of band-aids and a few upgrades pigtailed on & held in place with lots of duct tape.
If you're making the 2.0 version of "John's Phone" (a device intended to do absolutely nothing besides make phone calls as simply as possible, with long battery life and minimal scope creep), ARM is absolutely the best architecture to use for the job. On the other hand, if your phone has mutated into a de-facto pocket laptop driving an external display via wireless HDMI and running browser-based Flash applications, some future Franken-ARM architecture isn't necessarily the best choice.
None. It's a parallel evolutionary track. Start with Linus' kernel, throw out 99% of everything else that ends up in a modern Linux distro, then write your own windowing system that does the same job as x11, but does it in a way that's completely different and is mostly inseparable from the Java-like abstraction layer sitting between the kernel and userspace applications.
I'm struggling here to think of an Animal analogy,but I know there are at least a few cases where you have two species that superficially resemble one another, but actually got there by completely separate evolutionary tracks (possibly, after diverging from some common ancestor millions of years earlier). Zebras-vs-Horses, maybe? Eels vs water snakes?
ARM's one insurmountable advantage over x86 in the embedded realm is deterministic execution. x86 is now optimized for out-of-order and speculative execution, which makes multitasking applications "just work" better, at the cost of more complex silicon logic. The problem is, if you're doing something realtime ("realtime" != "responsive" or "fast" or "multitasking; it has a very specific meaning in embedded development) where you literally have to have an ironclad guarantee that a given block of code will execute in EXACTLY $n clock cycles, without interruptions, you can't use modern x86 architecture CPUs, because they're fundamentally incapable of making that specific guarantee. At least, in their Core2whatever form, and probably in most/all of their Atom forms as well. That's a realm that ARM owns, and x86 is unlikely to ever conquer because it's too fundamentally alien to everything modern x86 architecture regards as holy and sacred.
As a practical matter, the shift of Android to x86 for high-end phones and tablets is as inevitable as Apple's was for the Macintosh. The real alternative would be to turn ARM into de-facto x86 (with speculative and out-of-order execution, parallel pipelines, etc) that does poorly and expensively what x86 does cheaply as a matter of commodity. ARM and x86 both have their place. ARM was appropriate for Android 1.x. Frankly, though, the idea of a 3GHz quadcore ARM CPU is just kind of ridiculous. It's like using a big screwdriver with a sledge hammer as a chisel. It's just not the right architecture for the job. You can kludge it and make it work, but at some point it makes sense to just go with the flow and transition to x86, where high performance is cheap and commodity-like.
Maybe... until Microsoft gets around to assessing a WebOS fee that's only 25% more than Android's, but includes Android rights as a free bonus.
Seriously, though. As much as it pains everyone to pay money to Microsoft, at least THEY'LL license to anybody with a checkbook. In contrast to, oh... say... Apple, who's said point blank that their goal is to get hardware alleged to be infringing taken off the market, period. It's not just about Apple's desire to ram iOS down everyone's throat... it's about their desire to take away Android owners' freedom to install whatever we want, be it pr0n or anything else that makes Steve Jobs Cry. Microsoft wants a few bucks. Apple wants your freedom and liberty. And your dedicated camera button. Steve Jobs *hates* dedicated camera buttons...
In historical terms, Android is Tesla, Microsoft is Westinghouse, and Apple is Edison. Edison was very successful at keeping competitors' superior products off the market, but in the long run consumers really didn't forgive him. There's a reason why Edison's remembered for the electric light, but almost nothing else -- his other products almost universally sucked, and were shunned by the marketplace the moment his patents expired and he could no longer keep them off the market.
> But the US is no longer an important part of the phone industry, y'know
That's a very dangerous assumption, and a large part of how Nokia got into its present mess. The US might be a tiny part of the GSM market, and account for almost no direct-to-consumer sales, but its mindshare influence is enormous everywhere that English is a major language of commerce and media. Nokia abandoned the US market, then wondered why its popularity at home slowly evaporated. It completely failed to grasp the significance of MeeGo & Symbian's complete nonexistence as far as sites like Gizmodo, Engadget, Wired, and US-based sites in general were concerned. They weren't marginalized -- as far as American authors were concerned, they might as well have gone out of business. And when readers in Europe saw nothing but stories about iPhones and Android, their mindshare at home dried up within a year or two as well.
Does Motorola even HAVE any meaningful non-Android phones that aren't 1-cent throwaways for TracFone and the "Jitterbug" crowd? Google or not, Android is what saved Motorola from irrelevance at the high end of the market.
Google's ownership of Moto probably means they aren't going to START making Windows phones, but it's hard to pull out of a market you were never in to begin with. It would be kind of like accusing Nokia of abandoning Android if Microsoft buys them outright.
I think they're more afraid that Google's going to use Motorola as its bully pulpit to push phone hardware beyond what US carriers will embrace, and leave them in a position of having to choose between gambling on cutting-edge designs with high R&D costs that US carriers might reject, or trying to run new versions of Android on hardware that's just not up to the task & having their phones suck compared to GoogleMoto's flagship models.
Google doesn't HAVE to give itself artifical exclusivity to put HTC & Samsung in a position of de-facto second-class status; all they have to do is make the latest version of Android demand substantially better hardware than the highest-end phones Samsung and HTC are willing to make at that instant in time. Google has the cash to play "chicken" with US carriers and push ahead with hardware its official customers (US carriers) aren't interested in buying.
If Sprint, Verizon, and AT&T aren't interested in selling a phone that would have a $399 subsidized cost and ship with a 3GHz quadcore CPU, a quarter-terrabyte of flash, 8 gigs of ram, a 4.5" 1280x720 display, and a GPU that would have brought tears of envy to the most hardcore PC gamer's eyes just 2 or 3 years ago, HTC & Samsung are unlikely to build them anyway in the hope that Sprint/Verizon/AT&T will back down. Google has the cash to build a warehouse full of them, ship out preview phones to reviewers, then pit customers against carriers for the phones they're dying to own. Worst-case, Google writes them off as a partial loss & makes lots of buyers in India & New Zealand happy. HTC & Samsung can't afford to take risks like that.
You're right about development costs for commercial products, but remember... the bar is much, much lower when you're talking about kits. This particular application of wideband FM might be new, but wideband FM itself is ancient technology. Frankly, if HD happened 20 years earlier, I think it's safe to say that there would be two dozen circuit designs floating around right now to do this same thing. The only thing making it hard NOW is the fact that (statistically) nobody under the age of 60 really understands wideband FM anymore (myself included). If someone could bring this idea to the attention of Wayne Green (the guy who used to publish 73 magazine), he could probably design it HIMSELF as an afternoon hobby project, even if he hasn't personally done anything FM-related in years. Or at least come up with a good, fairly authoritative explanation as to why it's not viable or would cost more to do than it's worth.
Remember, it's hard to design something like this that's power/spectrum-efficient, cheap, and tolerant of user error. Give it its own 75-ohm RG59, and you don't really have to care about efficiency. You CAN splatter one channel across the entire UHF band, because you've got the cable's entire bandwidth to do it. You don't have to make it user-friendly, because you're dealing with users who by definition MUST have RTFM. That basically leaves a two-sided 2 or 4 layer circuit board made for a buck or so in hundred quantities by a company in China, about $20 worth of parts, a ziploc bag, and a photocopied sheet with the URL of the pdf instruction manual online. You can get away with lots and lots of sins when your target market consists of DIY'ers who can help themselves anyway. This isn't an iPhone;-)
The rationale for doing something like this as a kit has less to do with saving labor costs than avoiding FCC regulation (which would add at least $100,000 to the production cost, assuming it even passed). As I understand it, you couldn't sell a finished product like this without FCC approval, but when you sell it as a bag of unsoldered parts and a circuit board, the official responsibility for FCC compliance falls on the buyer (who, of course, will never get it certified).
As a practical matter a component-video modulator like I described that operates with the typical power of an old RF modulator would be unlikely to have an impact beyond the user's house, even without FCC approval, unless he did something completely idiotic and/or illegal, like try to backfeed the signal into the cable company's cable or jack up the transmission power. If it can be done cheaply, a safety feature to sniff the cable for any hint of QAM (turning on a red error LED and refusing to transmit if it does) would be a nice addition to the design. Beyond that, the potential to interfere is really no better, worse, or different than the potential for someone to do it with an old RF modulator designed for channel 3/4, or even a more expensive "agile" modulator designed for UHF. Just make sure at least the transmitter gets sold with a metal case so the buyer can't try to save a few bucks and run with a bare circuit board hanging behind the TV.
Slashdot Mirror
The difference is, by eliminating "drive-by" attacks by requiring a little bit of real-world physical work by end users to enable rewrites, you can morally and legitimately blame stupid end users who do it so they can watch the dancing bunnies. I like the idea of printing the key on a sticker attached to the motherboard. I like the idea of a jumper that has to be shorted out to enable overwrites even better. It can even be a tiny switch on the rear (among all the ports) that has to be flipped while the computer is off, then flipped again in response to a prompt, to perform the overwrite. Just forcing users to shut down, flip the switch, launch flash mode by holding down 3 unlikely keys while powering up, then flip it again in response to some non-obvious prompt documented on page 739 of the pdf instruction manual, will weed out 99% of the aspiring dancing bunny victims because they won't have the attention span or motivation to go through with the hassle. The remaining 1% who do it to watch dancing bunnies are hopeless future Darwin Award recipients.
The solution, if Microsoft wants to be the non-Evil (if not actually "good") guys, would be to require UEFI secure boot AND require that the key be furnished to end users for logo compliance. If they're worried about social engineering, they can put it someplace where it won't stop anyone who's likely to care about Linux, but be a substantial barrier to clueless end users who'd be a danger to themselves and others if they had it. Say, a sticker on the motherboard (or, for laptops and factory-built PCs, under a factory-installed SoDIMM), with a second peel-off copy that can be removed by end users and pasted somewhere convenient.
LOL. Actually, I'm a bit of a geezer too (hint: I was a freshman in college the night I went to Incredible Universe at midnight to buy my copy of WIndows 95). One of my favorite laptops was the Zeos Meridian, which just so happened to have an IBM-style pointer stick in the location I'm talking about (and unfortunately, was utterly out of my price range at the time) -- http://cdn2.iofferphoto.com/img3/item/211/480/688/dIaQ.jpg
Actually, you could probably argue that there's a certain amount of de-facto herd immunity in places with antibiotic abuse, simply because people who get infected by syphilis are likely to come down with a sore throat or upper respiratory infection, and head home with a prescription for some powerful broad-spectrum antibiotic likely to blow the syphilis away with trivial ease long before they ever realized they HAD syphilis. Apparently, syphilis is pretty easy to cure, and even penicillin is good enough to do the job in most cases (let alone something like Levaquin, Zithromax, Velosef, Biaxin, or Keflex).
Ah... but you're forgetting one very, very big difference between personal debt and sovereign debt: US federal debt is dollar-denominated. It would be kind of a nuclear option, but if push came to shove, the US could truly print its way out of debt. Furthermore, unlike countries like Iceland and Greece, the US has some of the most abundant natural resources on planet earth. The US does, in fact, have options available that private citizens who owe money don't. The worst thing the US could possibly do is sell off its natural assets to foreign owners to pay down debt. It would be the financial equivalent of taking out a $200,000 home equity loan on a house that's mostly paid for to pay credit card debt that could be washed away by bankruptcy. Morally admirable, perhaps, but financially stupid.
The US also has an advantage few other countries enjoy besides China -- in addition to having sinfully abundant natural resources, the US has a gigantic domestic market with the electoral muscle to prevent those resources from being exported wholesale for a pittance instead of sold domestically. If Congress, the Senate, and the President decide that Exxon isn't going to be allowed to sell its Alaskan & CONUS oil to anybody besides Americans... it's not going to be sold to anyone besides Americans. If they try to take their ball & go home (as if "they" could go anywhere besides maybe Switzerland or Dubai) by shutting them down, they'd be nationalized in 3 hours, and their employees would be back at work the next morning as government employees. You can argue about whether it would be morally right, or a good idea... but the fact is, it would happen.
It's also important to remember that the US can get away with way more inflation than other countries, because the true value of the US dollar lies with its nearly infinite liquidity and network effects. Simply put, the dollar has value as currency because everyone else has vaults full of dollars to use as currency. Even in cases where you have a very strong currency, like Norway or Australia, by the time you factor in the bank fees to convert to any other currency, you're better off just leaving your (US) dollars as (US) dollars and spending them to buy online pr0n instead of handing 5% of your wealth to a bank somewhere just to exchange your dollars for some other currency.
Don't get me wrong. I'm not advocating the expenditure of money just for the sake of spending money. This is precisely the time when the government should be buying up all the freeway right of way (at fire sale prices) that it's ever wanted to own for future expansion projects, 16-laning every major freeway in America, running fiber to every halfway-urban neighborhood (not necessarily to every house, but at least getting it close enough to run cat6 cable for gigabit ethernet the last thousand feet from pod to home router), burying power lines in hurricane-prone areas, and generally investing in the kind of high-quality infrastructure that will still be useful 50-100 years from now, instead of doing stupid things like subsidizing half the purchase price of electric cars so environmentalists can feel smug and superior to their neighbors.
> Deflation is much better than inflation in case of recession, because it allows the poorest people to survive,
> as they can buy more with whatever meager savings they have.
You're overlooking the fact that statistically, the poor (and for the most part, the middle class, and a fairly large chunk of the upper class as well) have no real savings to speak of once you factor in the negative value of their pre-existing debt (student loans, upside-down mortgages, and credit cards with 14% interest rates despite a nominal prime rate approaching 0%) that must now be repaid in deflated currency. In the entire United States, you'd be lucky to find 10,000 people who'd be genuinely better off under sustained real deflation. The remainder would be crucified by an impossibly crushing debt load hemmed in by falling wages for those who can even manage to get a job. On the other hand, you wouldn't have to look very hard at all to find 299 million Americans who'd benefit from sustained moderate inflation (and about 50-100 million towards the upper middle class end of the spectrum, whose debt is primarily fixed-interest long-term debt like student loans and mortgages, who'd effectively see their debt washed away within a couple of years).
> deflation would have allowed the bad debts to be liquidated and people would save and restart production.
Newsflash: nobody sane is going to invest a million dollars producing goods likely to have an ultimate resale value of $950,000 (ie, goods whose ultimate value is less than the cost of their production) when they can just put the money in a vault and wait for its absolute value to appreciate on its own. Especially when those same businesses can't get loans to finance the production of goods. Even under moderate deflation, it becomes almost impossible to raise investment funds because investors won't settle for anything less than zero risk with an impossibly high return.
Try this: name ONE ERA in modern history marked by deflation that's looked back upon fondly by the people who experienced it firsthand. I don't think there's anybody who has ever genuinely said, "You know, the Great Depression was a wonderful long-term boost for our nation's economy, and I'm glad it happened!"
>USA doesn't need more consumption, it needs real recession to kick in
>and it needs underconsumption and savings so that real capital can be
>once again accumulated to restart production
Unfortunately, the recession you're welcoming with open arms would end up being deflationary. During periods of deflation, the best thing a wealthy person can do is put his cash in a safe and forget about it, because just about anything he might try to invest in is likely to end up causing him to lose money.
Inflation isn't good, but deflation is deadly... especially to any industry with long supply chains and timeframes (ex: automakers and building construction), because just about anything you try to build is going to end up being worth less when you're finished than you spent to build it in the first place (with the possible execption of merchandise with extremely high markup, like a Rolls Royce, and construction work that's mandatory, like roof repair after a hurricane; neither case will generate enough economic activity to keep their respective industries alive, let alone prosperous).
Inflation is Adam Smith's way of forcing the wealthy to invest their money and put it to work, by slowing depleting its value if they try to just sit on their capital for too long without putting it to productive use. High inflation (and wildly unpredictable inflation) is bad because it induces unproductive behavior (people end up spending lots of time managing their money instead of earning more), but a small, stable amount of annual inflation (let's say ~3%) is a good thing.
> Sprint and T-mobile use different technologies and have different market shares.
Believe it or not, the difference is more one of business policy than actual technology. As a practical matter, any phone built like the Motorola Photon could easily work on Sprint and T-Mobile (it can't now, because Sprint had the UMTS radio specifically created to block use of AT&T and T-Mobile). With the next version of the baseband chipset, it could probably even do HSPA+ ("4G on T-Mobile") as well.
If Sprint bought T-Mobile, the first thing they'd do is require all new high-end phones -- Sprint AND T-mobile -- to be like the Photon & able to do CDMA2000, UMTS, GSM, and wimax. After a year or two, they'd start repurposing some of their 1900MHz "Sprint" spectrum to UMTS uplinks, and pair it with chunks of T-Mobile's 2100MHz UMTS downlink spectrum. It wouldn't break a single T-Mobile phone, because they can all do 1900/2100MHz UMTS as easily as they can do 1700/2100MHz UMTS anyway.
Sprint would keep circuit-switched CDMA voice, but start phasing out EVDO in favor of UMTS for data, just like Telus did in Canada (except Telus skipped EVDO and went straight to UMTS). After 3-4 years, when all their high-end users were solidly migrated to phones capable of UMTS data, they'd probably shut down EVDO entirely in markets with tight 1900MHz spectrum (like San Diego) and only provide 1xRTT to the few remaining users who couldn't use UMTS (1xRTT can coexist alongside CDMA voice and dynamically share channel space with it, whereas UMTS needs dedicated spectrum). I believe this is what CDMA carriers in India have done as they've migrated users from EVDO to UMTS for data.
5-10 years down the line, the distinction between "Sprint" and "T-Mobile" would be academic. The towers would all be shared. Sprint would use a chunk of 1900MHz spectrum for legacy GSM voice & 2(.5)-G data (GPRS and EDGE), another chunk of 1900MHz spectrum for circuit-switched CDMA voice and 1xRTT data, and a third chunk of 1900MHz paired with T-Mobile's 2100MHz for 1900/2100MHz UMTS. They'd probably pair a fourth chunk of 1900MHz with their ~850MHz Nextel spectrum for premium UMTS (850 uplink, 1900 downlink), and use their 1700MHz spectrum for LTE. In the meantime, Sprint and T-Mobile customers with new phones would both get to suffer with Sprint's crap 4G service ("crap", because it doesn't work in moving vehicles due to the way Clear fucked up their tower-tower hand-offs... or more precisely, didn't bother to implement at all, so they all act like wifi access points with ~1km range instead of a real cellular network; instead of handing off gracefully, Sprint 4G just drops the connection and leaves you with no network connectivity for 10+ seconds while it handshakes with the next tower and gets a different IP address).
Anyway, the point is that there IS a graceful way to merge Sprint and T-Mobile's networks. That said, if Sprint and T-Mobile were to "merge", I'd only want it to happen if the Justice Department made them:
1. sell all their tower assets to TowerCo (the company Sprint created to own its towers), to maximize their availability for lease on open terms to competing networks
2. create a new company to handle everything RF-related. In other words, own the spectrum leases and run the CDMA2000, GSM, UMTS, and LTE radio services. Let's call this "RFco", and view them as the equivalent of an RBOC.
3. keep "Sprint" and "T-Mobile" as separate companies who provide their own data backhaul & provide actual PSTN connectivity, sell phones (but require Sprint to fully support USIMs, require both companies to allow the use of any hardware that's physically compatible, and publish their standards and protocols in open format to allow end users to implement them on their own hardware as desired).
In other words, Let's suppose you're a "Sprint" customer. You might (or might not) buy your phone from them, but it has a USIM (which works on everything from UMTS to CDMA), so Sprint can't lock you into their own propr
Please. There's no way in *hell* you could type arbitrary text that's not conversational English at 50wpm using Swype, if only because one out of every 20-40 words would end up getting mango into some other word by its prediction album, and everything you wrote would make you look like the 21st century reincarnation of http://en.wikipedia.org/wiki/William_Archibald_Spooner
I personally can do about 30-40wpm with Graffiti when I'm really into it and writing at full speed. It does admittedly take more motion than Swype, but has the advantage that once you're good at it, you really DON'T have to pay much attention to the screen while writing. Turn off prediction, disable character-by-character searching, use it on a 1GHz+ phone, and it really IS accurate enough to let you get away with blind-writing.
^^^ Exactly. People have forgotten that PCs were revolutionary BECAUSE they devolved control and power away from centralized IT departments, and put it directly in the hands of end users who could skirt bureaucracy and do cool, new useful things without having to wade through months of committee meetings first. Those who don't remember the past are doomed to repea...NO CARRIER
Fine. Let my desktop PC use UPnP to configure my router, let my router update my DDNS hostname, and let my Android phone & tablet sync directly with my desktop PC. Maybe add a server appliance running Samba into the equation. No need to screw with proprietary online services that either cost lots of money or can vanish tomorrow without warning. I guess we're the geezers now, but anyone old enough to remember dotcom services we depended upon disappearing overnight (or mid-afternoon), never to return, is unlikely to ever fully trust "the cloud", let alone any proprietary service provided by it that locks you in and gives you no way to escape to freedom with your data.
> Windows 7 works just fine. It's the new XP - didn't you know?
It's sad, but you're probably right. Microsoft today is kind of like a rock star who's made so much cash, he's just going to be weird and do whatever the fsck he feels like doing from now on. If Microsoft is hyping "Metro" in an effort to generate developer excitement, they're having the exact opposite effect. Everyone *I* know is like, "WTF, has Microsoft gone completely batshit insane?"
It's almost like Microsoft's entire developer elite just hit their mid-40s, had a midlife crisis, realized they have enough cash to spend the rest of their lives coding for fun, retired en masse, and handed over the company to a marketing department that thinks making Windows look like a tablet UI so it can run phone apps better is somehow a good idea.
> Unfortunately, there are no good implementations for modern phones
Not quite true. Access-nee-Palm (whatever name they go under now) semi-officially sells Graffiti for Android. There's a free ad-supported version that embeds ads in the graffiti area, and a $2.99 version without them. It's worth every penny. It's the ONLY Graffiti-1 implementation I know of for Android that properly recognizes all the alternate strokes hardcore PalmOS users have burned into our motor memory forever, like backwards-v for "V", up-over for "t", "6" on the "letter" side for "G", etc. I just wish there were some good way to do cursor-up and cursor-down, because Connectbot is quite painful without them.
I personally couldn't live without Graffiti. When I bought my Epic4G, I came within a day of taking it back and exchanging it for an Evo because Graffiti didn't work properly on the Epic (you had to press and hold for about a second to register a "dot" gesture), but after I filed the bug report, they literally had it fixed in 2 days. I can say that Android Graffiti is the best Graffiti-1 implementation that's existed since the Palm V (the Treo-era Graffiti-2 implementations sucked). For anybody who reads this, if you used to like Graffiti on Palm, try it on Android. You'll feel like ${deity} has personally heard your cries and answered your prayers. You'll feel like you can finally input text properly again for the first time in years.
IBM (now Lenovo) has the right pointer-stick mechanism, but the wrong location. The stick SHOULD go directly below the spacebar, aligned by a line perpendicular to the spacebar that would run through the center of the "B" key. Fujitsu got the right location, but patented a crap pointer-stick mechanism. Unfortunately, Fujitsu patented the stick's LOCATION too, so nobody besides Sony (with a sufficiently-large patent portfolio to bully Fujitsu into submission) has ever dared to put the stick there.
Don't believe me that it's the ideal location? Try this experiment: hold your hands so the thumb is over the Trackpoint the same way it would be if it were centered below the spacebar and try moving it. Ahhhh. Much nicer, isn't it? Your thumb is a lot stronger, so you can give the stick a good hard shove to hurl the pointer to the other side of the screen quickly instead of exerting your index finger in a hyperextended position.
Personally, I think keyboards should have TWO Trackpoint-style pointer sticks: one in the "GHB" triangle, and one centered below the spacebar, with 4 buttons flanking the lower stick (2 per side, placed so whichever thumb ISN'T on the stick can easily press one or the other, keeping in mind the ~45-degree axis of comfortable motion for the thumb. Then, make them all software-configurable, so you can assign the 4 buttons to be "left", "right", and/or some other function as you desire, and either use both pointer sticks as mice, or reassign one to be for scrolling/panning. I'd personally use the "GHB" one as a faux scrollwheel ;-)
It's called two-way HD video chat with realtime collaborative desktop sharing. Incoming video from the other guy? Check. Video-capture and encoding to send him? Check. RDP, VNC, or some comparable protocol? Check. Gimp? Check.
It's kind of an extreme example by current phone standards, but not much beyond what you could do today with a fairly high-end laptop from Dell (or a fairly run-of-the-mill one, if you have dedicated hardware to handle the h.264 realtime capture and encoding, and a video chipset with hardware-assisted decoding).
The problem with predictions that future computers will be more power-efficient is the reality that users and programmers inevitably find creative ways to soak up 150% of any improvement to processor speed. On one hand, we all bitch about 3GHz quadcore PCs that feel slower than a 400 MHz Pentium 3 did running Windows 98... but none of us could stand using Windows 98 anymore, because it would feel almost like a toy compared to anything modern. And before anyone brings it up, fully tricked-out Linux is every bit as bad as Windows -- maybe even a little worse. Hell, look how long 1.6GHz Atom chips lasted before Intel had to boost the cores, clock speed, and cache enough to run Vista acceptably fast. Any long-term strategy that depends upon reducing speed for the sake of power savings is doomed, because mainstream software and consumer expectations will outstrip it and render it obsolete long before it hits the store shelves.
Perfect example: Pentile RGBW. On paper, it's awesome... good black & white text, great battery life. Put any Pentile RGBW display next to Super AmoLED+, and it's going to look like total crap and leave its owner feeling shortchanged because it just doesn't have the searing "snap" and saturation. It might not look as good for black & white text, and might not last as long battery-wise, but it's like the "Pepsi Challenge" from the 70s and 80s -- put the two side by side, and people will always make the impulse decision to grab the pretty one, regardless of intellectual arguments and pragmatism.
We can't get through an entire day on a single battery because every manufacturer feels compelled to make its new phone thinner than last year's iPhone. If companies like HTC/Samsung/Motorola would just bite the bullet and add a millimeter of extra thickness, we could have 4000mAH batteries that could make it through a day of active use & still be alive the next morning when somebody calls to wake you up.
~6 years ago, Samsung felt compelled to give every SPH-i500 owner two free batteries (one extended-life) because of its *scandalously poor* battery life -- fully-charged, it could barely make it through eighteen hours with the non-extended battery. Today's iPhone and Android owners would *kill* for that kind of battery life, but the fashionistas who dream of credit-card thickness phones won't let us have it.
> If it can play HD video (and it already can) then what else do I want performance for?
Play HD video on the screen while capturing and encoding realtime 720p60 from the camera, streaming it in realtime through a software firewall and VPN while using BGP to negotiate a multilink connection via EVDO/UMTS, LTE/Wimax, and/or Wi-Fi (depending on which one(s) are the best-available network link at that instant in time), and juggling your antivirus suite, anti-rootkit scanner, checking your IMAP mail server for updates every 3 seconds, keeping your Twitter timeline up to the second, and editing 3 ~4000x3000 HDR pics you took with the camera in its highest-resolution mode in Gimp. And somehow, manage to not crash or choke when somebody calls you, and it has to gracefully notify you about the incoming voice call and somehow allow you to answer it without seriously interrupting everything else you're trying to do with your phone at that moment.
> It's proven, it's developing and has no legacy dragging it back.
Yes. Until you try and turn it into a multi-core architecture with parallel branches, speculative & out-of-order execution, and all the other things x86 does in its sleep that are virgin territory with ARM. At that point, it's no longer proven and mature -- it's Internet Explorer 6 with lots of band-aids and a few upgrades pigtailed on & held in place with lots of duct tape.
If you're making the 2.0 version of "John's Phone" (a device intended to do absolutely nothing besides make phone calls as simply as possible, with long battery life and minimal scope creep), ARM is absolutely the best architecture to use for the job. On the other hand, if your phone has mutated into a de-facto pocket laptop driving an external display via wireless HDMI and running browser-based Flash applications, some future Franken-ARM architecture isn't necessarily the best choice.
None. It's a parallel evolutionary track. Start with Linus' kernel, throw out 99% of everything else that ends up in a modern Linux distro, then write your own windowing system that does the same job as x11, but does it in a way that's completely different and is mostly inseparable from the Java-like abstraction layer sitting between the kernel and userspace applications.
I'm struggling here to think of an Animal analogy,but I know there are at least a few cases where you have two species that superficially resemble one another, but actually got there by completely separate evolutionary tracks (possibly, after diverging from some common ancestor millions of years earlier). Zebras-vs-Horses, maybe? Eels vs water snakes?
ARM's one insurmountable advantage over x86 in the embedded realm is deterministic execution. x86 is now optimized for out-of-order and speculative execution, which makes multitasking applications "just work" better, at the cost of more complex silicon logic. The problem is, if you're doing something realtime ("realtime" != "responsive" or "fast" or "multitasking; it has a very specific meaning in embedded development) where you literally have to have an ironclad guarantee that a given block of code will execute in EXACTLY $n clock cycles, without interruptions, you can't use modern x86 architecture CPUs, because they're fundamentally incapable of making that specific guarantee. At least, in their Core2whatever form, and probably in most/all of their Atom forms as well. That's a realm that ARM owns, and x86 is unlikely to ever conquer because it's too fundamentally alien to everything modern x86 architecture regards as holy and sacred.
As a practical matter, the shift of Android to x86 for high-end phones and tablets is as inevitable as Apple's was for the Macintosh. The real alternative would be to turn ARM into de-facto x86 (with speculative and out-of-order execution, parallel pipelines, etc) that does poorly and expensively what x86 does cheaply as a matter of commodity. ARM and x86 both have their place. ARM was appropriate for Android 1.x. Frankly, though, the idea of a 3GHz quadcore ARM CPU is just kind of ridiculous. It's like using a big screwdriver with a sledge hammer as a chisel. It's just not the right architecture for the job. You can kludge it and make it work, but at some point it makes sense to just go with the flow and transition to x86, where high performance is cheap and commodity-like.
Maybe... until Microsoft gets around to assessing a WebOS fee that's only 25% more than Android's, but includes Android rights as a free bonus.
Seriously, though. As much as it pains everyone to pay money to Microsoft, at least THEY'LL license to anybody with a checkbook. In contrast to, oh... say... Apple, who's said point blank that their goal is to get hardware alleged to be infringing taken off the market, period. It's not just about Apple's desire to ram iOS down everyone's throat... it's about their desire to take away Android owners' freedom to install whatever we want, be it pr0n or anything else that makes Steve Jobs Cry. Microsoft wants a few bucks. Apple wants your freedom and liberty. And your dedicated camera button. Steve Jobs *hates* dedicated camera buttons...
In historical terms, Android is Tesla, Microsoft is Westinghouse, and Apple is Edison. Edison was very successful at keeping competitors' superior products off the market, but in the long run consumers really didn't forgive him. There's a reason why Edison's remembered for the electric light, but almost nothing else -- his other products almost universally sucked, and were shunned by the marketplace the moment his patents expired and he could no longer keep them off the market.
> But the US is no longer an important part of the phone industry, y'know
That's a very dangerous assumption, and a large part of how Nokia got into its present mess. The US might be a tiny part of the GSM market, and account for almost no direct-to-consumer sales, but its mindshare influence is enormous everywhere that English is a major language of commerce and media. Nokia abandoned the US market, then wondered why its popularity at home slowly evaporated. It completely failed to grasp the significance of MeeGo & Symbian's complete nonexistence as far as sites like Gizmodo, Engadget, Wired, and US-based sites in general were concerned. They weren't marginalized -- as far as American authors were concerned, they might as well have gone out of business. And when readers in Europe saw nothing but stories about iPhones and Android, their mindshare at home dried up within a year or two as well.
Does Motorola even HAVE any meaningful non-Android phones that aren't 1-cent throwaways for TracFone and the "Jitterbug" crowd? Google or not, Android is what saved Motorola from irrelevance at the high end of the market.
Google's ownership of Moto probably means they aren't going to START making Windows phones, but it's hard to pull out of a market you were never in to begin with. It would be kind of like accusing Nokia of abandoning Android if Microsoft buys them outright.
I think they're more afraid that Google's going to use Motorola as its bully pulpit to push phone hardware beyond what US carriers will embrace, and leave them in a position of having to choose between gambling on cutting-edge designs with high R&D costs that US carriers might reject, or trying to run new versions of Android on hardware that's just not up to the task & having their phones suck compared to GoogleMoto's flagship models.
Google doesn't HAVE to give itself artifical exclusivity to put HTC & Samsung in a position of de-facto second-class status; all they have to do is make the latest version of Android demand substantially better hardware than the highest-end phones Samsung and HTC are willing to make at that instant in time. Google has the cash to play "chicken" with US carriers and push ahead with hardware its official customers (US carriers) aren't interested in buying.
If Sprint, Verizon, and AT&T aren't interested in selling a phone that would have a $399 subsidized cost and ship with a 3GHz quadcore CPU, a quarter-terrabyte of flash, 8 gigs of ram, a 4.5" 1280x720 display, and a GPU that would have brought tears of envy to the most hardcore PC gamer's eyes just 2 or 3 years ago, HTC & Samsung are unlikely to build them anyway in the hope that Sprint/Verizon/AT&T will back down. Google has the cash to build a warehouse full of them, ship out preview phones to reviewers, then pit customers against carriers for the phones they're dying to own. Worst-case, Google writes them off as a partial loss & makes lots of buyers in India & New Zealand happy. HTC & Samsung can't afford to take risks like that.
You're right about development costs for commercial products, but remember... the bar is much, much lower when you're talking about kits. This particular application of wideband FM might be new, but wideband FM itself is ancient technology. Frankly, if HD happened 20 years earlier, I think it's safe to say that there would be two dozen circuit designs floating around right now to do this same thing. The only thing making it hard NOW is the fact that (statistically) nobody under the age of 60 really understands wideband FM anymore (myself included). If someone could bring this idea to the attention of Wayne Green (the guy who used to publish 73 magazine), he could probably design it HIMSELF as an afternoon hobby project, even if he hasn't personally done anything FM-related in years. Or at least come up with a good, fairly authoritative explanation as to why it's not viable or would cost more to do than it's worth.
Remember, it's hard to design something like this that's power/spectrum-efficient, cheap, and tolerant of user error. Give it its own 75-ohm RG59, and you don't really have to care about efficiency. You CAN splatter one channel across the entire UHF band, because you've got the cable's entire bandwidth to do it. You don't have to make it user-friendly, because you're dealing with users who by definition MUST have RTFM. That basically leaves a two-sided 2 or 4 layer circuit board made for a buck or so in hundred quantities by a company in China, about $20 worth of parts, a ziploc bag, and a photocopied sheet with the URL of the pdf instruction manual online. You can get away with lots and lots of sins when your target market consists of DIY'ers who can help themselves anyway. This isn't an iPhone ;-)
The rationale for doing something like this as a kit has less to do with saving labor costs than avoiding FCC regulation (which would add at least $100,000 to the production cost, assuming it even passed). As I understand it, you couldn't sell a finished product like this without FCC approval, but when you sell it as a bag of unsoldered parts and a circuit board, the official responsibility for FCC compliance falls on the buyer (who, of course, will never get it certified).
As a practical matter a component-video modulator like I described that operates with the typical power of an old RF modulator would be unlikely to have an impact beyond the user's house, even without FCC approval, unless he did something completely idiotic and/or illegal, like try to backfeed the signal into the cable company's cable or jack up the transmission power. If it can be done cheaply, a safety feature to sniff the cable for any hint of QAM (turning on a red error LED and refusing to transmit if it does) would be a nice addition to the design. Beyond that, the potential to interfere is really no better, worse, or different than the potential for someone to do it with an old RF modulator designed for channel 3/4, or even a more expensive "agile" modulator designed for UHF. Just make sure at least the transmitter gets sold with a metal case so the buyer can't try to save a few bucks and run with a bare circuit board hanging behind the TV.