Actually, Leviton has done some wonderful things within the constraints of the single-gang Decora formfactor. Half the light switches in my house have two or three mini paddle switches that flip left and right that are either 1/2 or 1/3 the size of a standard Decora-style switch. Pass & Seymour have similar switches that flip up and down (THEIR 3-switch variant has a half-height switch on top, and two half-height half-width switches on the bottom).
Although my original motivation was to avoid replacing the single-gang switchboxes with double- or triple-gang boxes, I've since come to actually prefer them. To me, my brother's newly-built house with 4-6 normal Decora-style paddle switches in a row just seems kind of tacky by comparison... even moreso when you consider that half the switches in each group hardly ever get used, and with that many switches in a row you literally have to stop and think which switch controls the main kitchen lights, as opposed to the undercabinet lights, the cove accent lights, the spotlights over the snackbar, etc.
There IS one lingering problem in America, though... what to do about prewired speaker connectors. A full-sized single-gang plate with two binding posts is WAY too big and looks horrible, but there really isn't any smaller variant that's just big enough for two binding posts. So... anyone who's prewiring for speakers ends up with two choices: put a (relatively) HUGE single-gang box with full-sized blank plate over it, or just bore a hole in the wall and pull the wires through when the day comes to install the speakers. I have more than a few friends who ended up just drilling a hole and pulling the wire pair through, and it was PRECISELY because there really isn't any good appropriately-sized alternative available in America.
> and its family members RJ11 and all the others > please give a price to the man or people who designed a plug so simple and effective!
Are we talking about the same plug with that horrid plastic tab that inevitably breaks off if it gets repeatedly connected and disconnected? I think THAT plug must have been invented by the same person who designed the inevitably-broken hinges for a CD jewelbox...
Don't even get me STARTED about RJ-45 and TIA-968A/B... same fragile plastic tab, with the added bonus of a wiring order that requires you somehow keep the wires tightly twisted to the last millimeter, yet somehow untwist enough of orange to flank blue & blue-white and get the whole thing to go in straight...
> Because it's not to code. A bathrooms outlets must be on a circuit unto themselves.
It depends. My house was built in 1982, and it had a single GFCI breaker that fed the outlets in both bathrooms, the garbage disposal, and the outdoor power outlets. However, when I gutted it to the studs & concrete, I redid all the bathroom's wiring, so they're now on a separate circuit. I even ran an extra neutral, so I could could make the two other second-floor circuits AFCI-protected. For the record, the biggest problem people run into when trying to retrofit AFCI breakers is the fact that every AFCI-protected 'hot' wire needs its own neutral, but most 20th century American homes run circuits with a single neutral wire shared by a pair of 'hot' wires between the circuit breaker panel and some outlet or switch box on the other side of the house where the two circuits diverge. It worked, because the two 'hot' wires come from opposite legs of the transformer, and it ironically decreases the current carried by the neutral wire to the panel (the two hot wires are always opposite in polarity, besides the brief moment every ~1/30 second when they're both at the zero crossing and equal). Unfortunately, if you share a neutral between two circuits, the AFCI breakers can't work.
Factoid: Microsoft BASIC for the TRS-80 Model 100/102 was the last version of Microsoft BASIC that was personally developed by Bill Gates.
Also, I believe it's not *quite* correct to say Bill Gates wrote Vic-20 and C64 BASIC... he wrote PET basic, licensed it to Commodore in Perpetuity, then Tramiel had someone hack at the raw machine language code to make it work on the Vic-20 and Commodore 64 so he wouldn't have to pay additional licensing fees. THAT was the REAL reason why nearly everything significant had to be done with PEEK and POKE... it was really a ~7 year old implementation of BASIC written for a computer with a fraction of the RAM & ROM, and generally devoid of anything resembling color graphics and sound. On the other hand, it was an incredibly tight, optimized implementation of BASIC. Anyone who grew up with a Vic-20/C64 remembers that Commodore BASIC's PRINT command was almost fast enough to use for full-screen animation. By comparison, PRINT on other popular 8-bit computers (and 16-bit, and early 32-bit, for that matter) was somewhere between "annoyingly slow" and "utterly glacial" (TI-99/4A)
From what I remember, the biggest problem with AmigaBasic wasn't so much the language itself as the fact that it encouraged you to do things that relied on slow system routines. For example, AmigaBASIC had a major fetish with BOBs, and didn't really like Sprites. The problem was, if you read the AmigaBasic manual, it gave you the impression you could do something utterly insane, like write a version of "Centipede" that used a Bob for every mushroom. In reality, it would collapse and die from the overhead somewhere around Bob #20. The other problem was the fact that in many ways, AmigaBasic was the pre-alpha version of Visual Basic, and it was the first language I can remember that was event-driven rather than procedural. For late-80s teenagers (and probably older users, too), event-driven programming was a *major* mindfuck, and few Amiga programmers *really* "got it" until they'd been using it for a year or more. At the end of the day, AmigaBasic still sucked (I ended up using Assembly and GFA Basic for everything, though I had copies of Lattice C and True Basic as well), but its suckiness wasn't *entirely* Microsoft's fault. Anyone who thinks AmigaBasic was the worst obviously never used ABasiC;-)
> "ICANN is to be congratulated for succeeding in expanding the Internet beyond the Latin alphabet.
The problem isn't that ICANN expanded "the Internet" beyond the Latin Alphabet (or at least the subset enshrined in ASCII's alphanumeric characters plus hyphen)... the problem is the collateral damage it caused, and continues to cause, because they did the equivalent of dumping a freeway interchange in the middle of an already-thriving residential neighborhood.
ICANN (possibly with IETF) needs to do three things:
1) Work with IETF to extend DNS so that TLD registrars can define a specific subset of UTF-8 that's valid for its subdomains. By definition,.com/.net/.org should be forever restricted to the historical [A-Za-z0-9\-] subset to put an end to homograph phishing. In other words, no TLD could indiscriminately include everything from legacy-ASCII to Klingon, Runes, and ancient Egyptian. They'd have to pick the characters used to write a single real language and stick with it.
2) Require that TLD character-validity rules be fully normalized against characters between 0x30 and 0x7f. In other words, if there's a letter in the language's unicode codepage that looks just like ASCII 'i', they can allow ASCII 'i', or the language's own version of 'i' with its own UTF-8 value, but NOT both. The choice of which 'a-zA-Z' to use would largely depend upon which value gets generated from a keypress by a keyboard in the target country.
3) Create new TLDs for writing systems used in more than one country, by at least 50 million people... preferably, short and understandable to anyone in a country that uses that writing system. So, Chinese might get.{zhong} or.{zhongwen} (but the PRC itself's country TLD might be.{zhongguo}), Cyrillic might sensibly get the characters that resemble.NHT (backwards 'N') which apparently is the abbreviation for "Int" in Russian, Ukranian, Serbian, and probably most other languages using Cyrillic, etc, and conveniently looks vaguely like ".NET" to everyone else (but the backwards-N would ensure only a complete idiot could think it really WAS.net). Ditto for Arabic. Languages almost synonymous with a single country (Hebrew, Greek, Japanese, Korean, etc) or spoken by fewer than 50 million real people in daily business wouldn't get their own TLDs... but their countries would get a new country TLD in the writing system (along with their old 2-letter TLD).
The point is, the way internationalization has been rolled out so far has created a worldwide party for fraud and phishing via homograph attacks. An end needs to be put to it NOW. If someone has an existing IDNS name that would be invalidated by the new rules (say, {nurren}.com), they'd get first chance at it in the new.{zhong} TLD (nurren}.{zhong}). If there were two or more existing.com|.net|.org domains that clashed (say, {nurren}.com and {nurren}.net), they'd have to share the TLD and settle for distinct subdomains of it, like {something}.{nurren}.{zhong} and {somethingdifferent}.{nurren}.{zhong}.
Why did it even HAVE to be represented as an exact multiple of 0.1 seconds? As opposed to something a bit more binary-friendly, like 1/8 or 1/16th of a second? Was this just a case of 1980s Waterfall Design Paradigm, where somebody who didn't necessarily understand the precise characteristics of the hardware involved pulled the 1/10 second specification out of a metaphorical hat because it sounded good, then everyone below that point was expected to just unquestioningly implement it instead of asking WHY it had to be an exact multiple of some error-prone value early in the design process?
IMHO, the worst decision ever made in the history of the internet was ICANN's decision to allow non-ASCII subdomains of.com,.net, and.org. Those three gTLDs, if not others as well, should be forever off-limits to any characters besides the original 26 letters a..z, the digits 0..9, and hyphen.
For other TLDs, and for national TLDs, DNS should be extended to allow a TLD's authoritative top-level registrar to authoritatively indicate which UTF8 characters (or range(s) of characters) beyond those historically-allowed for ASCII DNS are valid for its subdomains. The registrar for Spain might decide it needs accented vowels and tilde+n, but has no reason for Turkish vowels that conveniently (for phishers) look identical to ASCII characters.
The next step would have been the creation of a few brand new TLDs, like ".Zhong" (U4E2D) and ".Nihongo" (U65E5 + U672C) -- think ".(Chinese)" and ".(Japanese)", not to mention similar TLDs for Hindi, Korean, Russian, and other languages that use non-Roman alphabets.
The point is, ICANN could have done a much, much better job with this whole mess. I think everyone can agree that international domain names are something that needs to exist, but trying to staple them onto.com/.net/.org was an incredibly bad idea.
> So basically we're trading one evil (comcast) for another (government).
The big difference is, for the most part Comcast's remedies if you subvert them are largely civil in nature -- denying you future service, charging you penalty fees, suing you, or the like. The government can have you thrown in prison.
Of course, some companies and court jurisdictions have been hard at work finding creative ways to criminalize breaches of corporate policy (particularly through abuse of "theft of electricity" rationales), but for the most part there's still a line between things that are criminal vs merely civil. Comcast's mostly on the 'civil' side, but the government is almost exclusively on the 'criminal' side.
> The effort made in consumer electronics to save four cents (over 10 million units) would probably make your head spin.
Tell me about it. It's what condemned many very, very expensive first-generation CD-Rs to the coaster bin (no ram buffer to speak of, so if the CPU got distracted for even a fraction of a second, the CD was toast), and sent almost every Samsung SPH-i330 to an early grave (a few cents saved using mask rom instead of flash, coupled with a fatal bug that bricked them within a few weeks of use and couldn't be fixed). Not to mention the problems due to aggressive cost-cutting that have plagued videogame consoles for decades. Red Ring of Death, anyone?
A major complication is the fact that today's PDA phones are basically cellular winmodems. Ever wonder why a Samsung SPH-i300 with ~40MHz 680x0-ish Dragonfire didn't feel all that much slower than a 400MHz dual-core ARM running WinMo? Part of the problem is WinMo's bloat... but an even bigger part of the problem is the fact that cpu #2 spends basically 100% of its time being a software-based faux DSP anytime the radio is in use... and occasionally steals cycles from cpu #1 while it's at it. In contrast, the humble i300 was literally a cell phone radio bolted to a PalmOS PDA, connected by LITERALLY a serial port. In every meaningful way, the two were completely independent. Dialed a number on the LCD-rendered keypad? The Palm side sent what was basically an ATDT command to the phone side. Incoming call? The phone side did the actual ringtone (the beefy speaker was connected to the phone; the palm side just had a wimpy piezo), and sent ###RING xxxxxxxxxx down the serial port for the Palm to decide what to do with it. Pressed the green button to answer the call? The Palm side sent (literally) ATA down the serial port to the phone. And so on.
The good thing about Winmodem-like cellphones is... um... er... uh... well, I'm sure there's something good about it. But anyway, the bad thing is that it means that you're basically talking about a gigahertz-rate raw DAC and ADC creating a software-defined radio. The supreme irony is that the hardware in a G1 probably COULD be hacked to do UMTS at 850MHz on AT&T... but it would be utterly illegal, precisely because it would wreak havoc with AT&T's network and mess up your neighbors' service while you spent saturday afternoon debugging it.
As I understand it, a phone running Android (or Windows Mobile, for that matter) is kind of like a PC running Linux under VMware under Windows (or vice-versa). You have the hypervisor-like "Supreme Controller" that runs the software that makes it a radio, with Linux-nee-Android running under the hypervisor. So you really ARE root as far as Linux/Android is concerned, but behind the curtain, Root has a metaphorical deity of his own to serve. I'm not entirely sure how it's enforced, and how much of it is handled by actual hardware partitioning (ie, whether the CPU takes a cue from memory curtaining, so that the core running the radio has ram and registers that the core running the UI and apps can't touch), but that's a rough summary of what happens behind the scenes on an Android phone.
> So I'd guess a forked F/OSS version of MySQL would need to call themselves something else, losing the name.
Personally, I'm rather fond of "rSQL". It's short, it's catchy, and neatly emphasizes the difference between the open source version ("our" SQL) and the new Oracle-centric proprietary branch ("*my* **My** ***MY*** Sql").
> So why are those goodies disabled under Linux? > Come on NVIDIA, just release the specs and let us write our own damned drivers eh?
It's not so much that they're "disabled", as a case of being "not implemented". The problem is that the line between what's a hardware capability and what's implemented mainly at the driver level through software is increasingly blurred. Just to give a familiar example, look at a PCI Winmodem. At the end of the day, a Winmodem is basically a PCI soundcard that's hardwired to a phone jack and optimized for PSTN-level voltages & impedance. Someone like Conexant could flawlessly document how to use the chips on one of their Winmodem cards to generate and sample audio, and it wouldn't do a thing to help anyone actually make the card act like a 56k modem under Linux. Someone has probably done it by now, but back when it would have actually still mattered (circa 1999-2000 or so), there was no such thing as an open-source Linmodem driver for that precise reason. Documenting the hardware was necessary, but even fully-documented, it would have only gotten you ~2% of the way towards the ultimate goal of *being* a software-defined modem.
There's another problem with video drivers -- patents. As a practical matter, everyone in the industry violates at least one patent belonging to the other big players, and they're *all* sitting ducks for every patent troll who comes wandering along. If NVIDIA were 100% altruistic, fully implemented every Windows feature into their Linux drivers, and released the full documented source code to their proprietary Linux drivers, they'd essentially be painting a red target on their forehead and making the patent trolls' fishing expeditions that much easier. It's sad, but it's true.
So... does it have a proper SDK (well, ok, at least unlocked hardware that can be freely reflashed with guerrilla builds of Android), or maybe an API for UI customization (repurposing buttons, changing the way it works, etc)?
Up to now, there's one eBook market that should have publishers absolutely salivating: computer and technical books. It's the one niche where non-crippled eBooks really DO have a compelling value proposition. Think about it... what's the biggest single problem with most computer books? They're already a version behind 6 weeks before they even arrive at Amazon, let alone Borders or Barnes & Noble. Computer-related eBooks have the potential to change that, by letting you auto-update the books regularly to incorporate fixes and revisions long after the book first became available. Plus, they have the potential for instant gratification. I'd probably buy twice as many books if I could have them *right now*, *this instant*, instead of having to wait until at least tomorrow afternoon to get them from Amazon (I've pretty much given up entirely on Borders -- once the gold standard of local computer book availability -- and Barnes & Noble has maybe 3% of the books I actually want available for immediate purchase at a store within 50 miles of my house whenever I go searching).
What I *really* want is an ebook that starts with the button configuration of an Ectaco Jetbook ( http://tinyurl.com/ctaq3q ) that keeps the transflective LCD, but bumps the resolution up to 1280x960 or 1600x1200, has approximately the width and height of a Manning book, and adds a capacitive digitizer for finger-friendly Graffiti-like handwriting recognition. And of course, a published, programmer-friendly API that lets you more or less reinvent its user interface without having to actually reinvent Acrobat reader *itself*.
Why LCD instead of e-ink? Latency. Novels are usually read page by page, sequentially. Computer books are ready by jumping all over the place, and rapidly flipping through pages to find something nearby. Frankly, 500-800ms is just plain unacceptable in that context. It would drive me nuts. I want to hit the 'next page' or 'previous page' button, and have the page visible before I even finish releasing the button (using onboard ram to cache the entire active pdf file since flash is painfully slow). Hell, if it becomes affordable, give me a second display while you're at it, so I can either see adjacent pages, or have two books open at once side by side. Maybe make one display e-ink, and one LCD (with a UI that juggles them around, so the page of interest during active page-flipping always ends up on the LCD instead of e-ink).
Would I pay $40-50 for an ebook that costs the same in print form, that can only be read on a Kindle, and is basically a warmed-over scan of the print edition? Hell no. At best, current ebooks are basically a crutch to get me over today's crisis until the paper copy arrives tomorrow. Half the time, my workflow can be described as: 1) find book(s) at Amazon; 2) order before the FedEx deadline; 3) download it 20 minutes later from eMule, since I *really* need it *right now* (admittedly, skipping step #2 most of the time if it's already past the evening deadline for next-day delivery and too late to get the real book by tomorrow anyway. It's amazing how not being able to get it tomorrow under any circumstances has a way of diminishing the perceived urgency of buying it). Yeah, I DO get my money's worth from my Amazon Prime membership. The FedEx man personally knows my name;-)
Would I do it for a book on something rapidly-changing, like Android programming, with constant revisions for at least 2 or 3 years as Android evolves from 1.5 to 1.6, 2.0, and beyond? Hell yeah. In fact, even limited to reading it on my PC, I paid the author of the Busy Coder Android books for a 12-month subscription that gives me permanent pdf copies of his Android books (3 so far), plus constant revisions. Google 'warescripti
> The patent is on using dedicated, non-host accessible RAM on the network adapter for the ring buffers and packets, together with memory-mapped IO for setting up DMA transfers.
Well... if that's the case, it shouldn't get too far in court since just about every ethernet implementation from the past ~5 years has been crap integrated ethernet based on host signal processing (think: winmodem on crack), I don't see how this patent would apply.
I mean, how many years has it been since MaximumPC even REVIEWED a high-performance PCI ethernet card with a real, onboard DSP and controller, let alone dedicated onboard RAM of its own? Unless the slimy trolls have sunken so far that they're claiming that an onboard function sucked into the mainboard chipset using system ram with HSP is tantamount to a PCI card with ram of its own...
> then in six months and one year, you query those people, see if they're still alive. After all, we don't care > if you died from influenza or the marthambles - if the vaccine keeps you out of the grave, then it's a win.
Er... death isn't the only consequence worthy of avoidance. That's the same logic health insurance companies (and government agencies in countries with socialized medicine) use in non-pandemic years when trying to argue that Tamiflu and Relenza aren't cost effective, on the grounds that it only reduces the "acute phase" by a day or so. What they conveniently omit is the fact that they might "only" reduce the phase when you're laying in bed in the fetal position hating life by a day or so, but they ALSO reduce both the duration and severity of the "misery" phase afterwards... the point when you're well enough to go back to work, but still feel and look like shit, with a runny nose that just won't quit, sore throat, and the rest. Both drugs can dramatically improve the overall experience, and make it a lot less awful than it would be if you did nothing to stop it.
As far as efficacy goes, I've religiously caught the flu at least once every year when I didn't get vaccinated. I've only gotten sick once when I was vaccinated, and I'm not sure whether THAT was a mild flu, or just a really bad cold.
There IS one problem with the "pick 3 strains" strategy -- it works a lot better in places like Iowa, Arizona, and South Carolina than it does in places like Miami, Orlando, and New York City, where there are literally millions of visitors every year from every single continent, and whose local flu strains weren't part of the vaccine mix.
I believe most CRTs actually consumed less power in active use than comparable *LCD* sets circa 2007 did. From what I know about electrical engineering, the main culprit was the power supply. CRTs had a fairly small AC-DC power supply, and drew most of their power from a transformer that converted 110VAC/220VAC into a few thousand volts... with relatively high efficiency. In contrast, LCD panels need lots of DC, much of which needs to be tightly regulated. To save money, they often use linear power supplies. Generalizing a bit, a linear supply draws about as many amperes of source voltage as it outputs in target voltage.
Here's an example. Suppose your LCD needs to draw 10 amperes at 5v, and you're feeding it 12vdc that came from a transformer and ran through a diode bridge. It's going to draw at least 10 amperes of 12v at the input end. In contrast, if you're using a more efficient (but slightly more expensive to build) active power supply, you might need ~6-8amperes of 12v input power to supply 5v @ 5A on the output side. CRTs had linear power supplies that were inefficient too, but the power needs of the tuner and control circuitry were pretty minimal, so it didn't matter much. Simply replacing the linear power supply with an active one can dramatically cut the power consumption. It would be insane to spend $25 more on the power supply for a wireless access point that draws ~150mA and sells for $20 on sale right now, but makes quite a bit of sense to do it for a $600 TV that draws hundreds of times more power.
As for the regulation, I think it's heavy handed. California could achieve most of the same benefit, with lower compliance cost, by simply requiring that stores clearly indicate the total amperes the TV draws from 4 hours of continuous use, and the estimated annual cost of that use based on the moving 5-year average price of power in California (with the state publishing the official value to be used for the next 12 months' of calculations each year). Let consumers see that TV A (on sale for $499) will cost them $10/month more to use than TV B (not on sale, $699), enable them to easily determine that TV A will ultimately cost a LOT more than TV B if they keep it more than 20 months, and Adam Smith's invisible hand will do its job just fine.
As I understand it, in Las Vegas particularly, it's actually a legal requirement that casinos try to identify and prevent card counting. Why? Because by law, all casinos have to have the exact same payout rate. For things like video slot machines, that's trivially easy to enforce through software alone. The problem is, if a casino "off the strip" openly allowed card counting to boost its popularity with tourists (imposing a max bet, and counting on the fact that most people who claim to know how to count cards really can't), the state of Nevada would have its agents promptly there to shut down the casino.
The requirement itself is a blatant anticompetitive strategy used by the mega-casinos. It ensures that the big, new, shiny casinos on the strip have the exact same payout rate as the old, small casinos downtown and elsewhere, so the others can't compete with them by offering better odds to players.
> Mouse/keyboard activity timeout works nicely for that.
You know, I'm sure it's just my imagination or a coincidence, but it really DOES seem like Vista on my desktop PC doesn't go into power-saving mode UNTIL it perceives a tiny, tiny mouse wiggle after a long period of inactivity. It seems like I can leave my computer on, walk away, and see it running in full-power mode for hours... then the nanosecond I sit down and bump the desk, it decides to power down the panel. I'm halfway tempted to put a camcorder on a tripod and just leave it capturing for a while, then the next time I think it happens, go back and watch the last few minutes of video beforehand to see what really happened to make me think it did something crazy like that.
> Do you think all those netbooks and cheap desktops are being sold for graphic designers or gamers?
No, they're being sold to Slashdot users who'd never be content to endure the limits of a laptop as their "real" computer (last I checked, there's no quadcore 3GHz i7 available for notebooks), but will happily throw down a few hundred bucks for a decent semi-throwaway netbook to use on weekend trips and other occasional situations where it's nice to have.
> 2. FCC mandates all wireless carriers switch to GSM.
Er, actually GSM is an officially-deprecated dinosaur. Even in Europe. 3G "GSM" is actually "W-CDMA".
Now, that doesn't stop everyone from CALLING W-CDMA on a 3G GSM phone "GSM", but it doesn't MAKE it "GSM". GSM is a variant of TDMA (3 virtual channels that share a single frequency channel by sequentially using their own exclusive timeslices within it). TDMA is the reason why GSM-based devices cause a bursty buzzing sound if they're placed too close to an amplifier... the buzzing noise is caused by the click when the carrier rapidly cuts off for a fraction of a second, and the next phone's carrier cuts. Think: the click made by a Commodore 64 when you changed the volume register (which was really a wavetable DAC hardwired to a circuit emulating a single byte of RAM, with the various SID effects produced by directly manipulating the apparent value of that virtual byte).
Need more proof? Read up about picocells. Picocells that handle CDMA and *ONLY* 3G GSM are fairly cheap, and becoming commodities. Picocells that handle oldschool GSM *and* CDMA/WCDMA("3G GSM") are quite a bit more expensive, because THEY require implementation of two completely different technologies. Adding the ability to roam on a European 3G network to an American CDMA phone (a-la Verizon's Touch HD, I believe) is more of a political and business decision than a technological problem. Adding true GSM capabilities to a CDMA phone requires a second radio module & baseband processor, because the technologies are completely different.
> But Toyota's invention is only an improvement upon the other invention. Toyota can patent the improvement, > but they and all licensees must also obtain a license for the original patent. Just because you add a redial > function to a phone doesn't mean you then can patent the phone.
And a major part of what's broken with the current patent system is the fact that it gives nearly unlimited veto power to whomever had that first "idea", without regard to other, equally fundamental patents that make the "original" patent commercially viable in the first place. Given enough cash, you could sit around with a bunch of patent attorneys, brainstorm ideas all day with no real idea of how you might actually make them *work* (or be manufacturable, or work more than.3 seconds before blowing up, or...), send the attorneys off to patent what's patentable, and sit on them for the next 20 years in case someone invents some real product that in any way, shape, or form bears some similarity to one of your patents.
Of course, the devil's in the details. A partial start might be a law that says that if you're challenged for patent infringement, and you can establish that YOU have a viable patent that meaningfully improves upon a patented item, improves its manufacturability, or uses it in some novel way, you can pay 50-80% of the net profits from the sales of the allegedly infringing item into a fund that gets split up among everyone who can establish that they have a patent being infringed. If you make no money, you owe nothing. If you make a hundred million dollars, you owe $50-80 million.
OK, there's a million details to work out, not the least of being the definitions of "novel" and "improvement", and a strategy for determining how it should be divided up, but the point is that it would ensure that the patent owners make at least as much money as the alleged infringers, while simultaneously maximizing the odds of someone finding a cool way to make use of patents that ultimately serves the constitutional foundation of American IP law -- advancement of the sciences and useful arts.
> But what I want to know is, why hasn't anyone mass produced a Thin CRT yet?
They've been prototyped -- 10 years ago, I was convinced that the future of television was the Field Emission Display (FED) after I saw a demo at CES. Absolutely *beautiful*. The best of all worlds. Bright, saturated, distortion-free, and viewable from angles just like a regular CRT.
Basically, coat a sheet of glass with colored phosphors, and put individually-addressable solid-state electron sources behind them. To light up a particular phosphor group, turn on the emitters behind it to make it glow. Unfortunately, the technology went nowhere.:(
> First of all it's going to be a high-profile target for terrorists. > So, expect check-in situations exactly like the airport.
One big difference, though. It doesn't take much explosion-wise to kill lots of people on a plane. American trains, by comparison, are rolling bank vaults. A bomb detonated in a passenger car would kill ~50 people, max. Passengers in adjacent cars would barely notice the muffled thud. Sure, lots of terrorists could try to pull off a synchronized event... but then the complexity goes way up, and the "bang per terrorist buck" is still pretty low compared to much, MUCH softer targets... like a McDonalds across the street from an elementary school, or a high school/college football stadium at homecoming.
I *would* like to know, though, how in god's name a trainset consisting of two locomotives and 3-5 passenger cars ends up costing more than a f***ing brand new 777. I tend to support rail projects, but someone first needs to figure out why anything involving a train in America ends up costing more than the goddamn Apollo space program did. It somehow cost more money to double-track Tri-Rail from Miami to West Palm Beach -- through an existing rail corridor that USED TO *BE* double-track at one point in the past -- than it cost to progressively tear down and rebuild I-95 in Broward County, taking it from a 6-lane antique to a fairly impressive 12+ lane modern freeway in the process. Yeah, they built *a* new bridge for Tri-Rail. Big deal... they built two more, each 3-4x as wide as the new rail bridge, right next to it as part of I-95's reconstruction.
I really think that half the problem is that American states send people to study trains in Europe and Japan. Really, they should be going to INDIA to learn how to build and run a huge train network that's relatively fast, dirt cheap, at least as reliable & on-time as American air travel is, and genuinely useful to millions of people every day. Personally, I'd kill for hourly 100mph service from Miami to Orlando, with the ability to do rental car paperwork on the train & walk directly to the parking garage -- key in hand -- once the train arrives. It's a nearly ideal intermediate-speed rail route... farther than anyone really wants to drive, but not *quite* far enough to be worth the cost, grief, and "hurry up and wait" stress of air travel. The problem HERE is that every time FDOT gets ready to build something useful and sane, the HSR monster rears its ugly head, and derails the whole thing for another decade or two.
Slashdot Mirror
Actually, Leviton has done some wonderful things within the constraints of the single-gang Decora formfactor. Half the light switches in my house have two or three mini paddle switches that flip left and right that are either 1/2 or 1/3 the size of a standard Decora-style switch. Pass & Seymour have similar switches that flip up and down (THEIR 3-switch variant has a half-height switch on top, and two half-height half-width switches on the bottom).
2-switch -- http://www.westsidewholesale.com/media/catalog/product/cache/2/image/5e06319eda06f020e43594a9c230972d/8/6/861517-1_11.jpg
3-switch -- http://ak.buy.com/db_assets/large_images/253/90140253.jpg
P&S 3-switch -- http://www.homeandbeyond.com/prod-0085785-zoom.html
Although my original motivation was to avoid replacing the single-gang switchboxes with double- or triple-gang boxes, I've since come to actually prefer them. To me, my brother's newly-built house with 4-6 normal Decora-style paddle switches in a row just seems kind of tacky by comparison... even moreso when you consider that half the switches in each group hardly ever get used, and with that many switches in a row you literally have to stop and think which switch controls the main kitchen lights, as opposed to the undercabinet lights, the cove accent lights, the spotlights over the snackbar, etc.
There IS one lingering problem in America, though... what to do about prewired speaker connectors. A full-sized single-gang plate with two binding posts is WAY too big and looks horrible, but there really isn't any smaller variant that's just big enough for two binding posts. So... anyone who's prewiring for speakers ends up with two choices: put a (relatively) HUGE single-gang box with full-sized blank plate over it, or just bore a hole in the wall and pull the wires through when the day comes to install the speakers. I have more than a few friends who ended up just drilling a hole and pulling the wire pair through, and it was PRECISELY because there really isn't any good appropriately-sized alternative available in America.
> and its family members RJ11 and all the others
> please give a price to the man or people who designed a plug so simple and effective!
Are we talking about the same plug with that horrid plastic tab that inevitably breaks off if it gets repeatedly connected and disconnected? I think THAT plug must have been invented by the same person who designed the inevitably-broken hinges for a CD jewelbox...
Don't even get me STARTED about RJ-45 and TIA-968A/B... same fragile plastic tab, with the added bonus of a wiring order that requires you somehow keep the wires tightly twisted to the last millimeter, yet somehow untwist enough of orange to flank blue & blue-white and get the whole thing to go in straight...
> Because it's not to code. A bathrooms outlets must be on a circuit unto themselves.
It depends. My house was built in 1982, and it had a single GFCI breaker that fed the outlets in both bathrooms, the garbage disposal, and the outdoor power outlets. However, when I gutted it to the studs & concrete, I redid all the bathroom's wiring, so they're now on a separate circuit. I even ran an extra neutral, so I could could make the two other second-floor circuits AFCI-protected. For the record, the biggest problem people run into when trying to retrofit AFCI breakers is the fact that every AFCI-protected 'hot' wire needs its own neutral, but most 20th century American homes run circuits with a single neutral wire shared by a pair of 'hot' wires between the circuit breaker panel and some outlet or switch box on the other side of the house where the two circuits diverge. It worked, because the two 'hot' wires come from opposite legs of the transformer, and it ironically decreases the current carried by the neutral wire to the panel (the two hot wires are always opposite in polarity, besides the brief moment every ~1/30 second when they're both at the zero crossing and equal). Unfortunately, if you share a neutral between two circuits, the AFCI breakers can't work.
Factoid: Microsoft BASIC for the TRS-80 Model 100/102 was the last version of Microsoft BASIC that was personally developed by Bill Gates.
Also, I believe it's not *quite* correct to say Bill Gates wrote Vic-20 and C64 BASIC... he wrote PET basic, licensed it to Commodore in Perpetuity, then Tramiel had someone hack at the raw machine language code to make it work on the Vic-20 and Commodore 64 so he wouldn't have to pay additional licensing fees. THAT was the REAL reason why nearly everything significant had to be done with PEEK and POKE... it was really a ~7 year old implementation of BASIC written for a computer with a fraction of the RAM & ROM, and generally devoid of anything resembling color graphics and sound. On the other hand, it was an incredibly tight, optimized implementation of BASIC. Anyone who grew up with a Vic-20/C64 remembers that Commodore BASIC's PRINT command was almost fast enough to use for full-screen animation. By comparison, PRINT on other popular 8-bit computers (and 16-bit, and early 32-bit, for that matter) was somewhere between "annoyingly slow" and "utterly glacial" (TI-99/4A)
From what I remember, the biggest problem with AmigaBasic wasn't so much the language itself as the fact that it encouraged you to do things that relied on slow system routines. For example, AmigaBASIC had a major fetish with BOBs, and didn't really like Sprites. The problem was, if you read the AmigaBasic manual, it gave you the impression you could do something utterly insane, like write a version of "Centipede" that used a Bob for every mushroom. In reality, it would collapse and die from the overhead somewhere around Bob #20. The other problem was the fact that in many ways, AmigaBasic was the pre-alpha version of Visual Basic, and it was the first language I can remember that was event-driven rather than procedural. For late-80s teenagers (and probably older users, too), event-driven programming was a *major* mindfuck, and few Amiga programmers *really* "got it" until they'd been using it for a year or more. At the end of the day, AmigaBasic still sucked (I ended up using Assembly and GFA Basic for everything, though I had copies of Lattice C and True Basic as well), but its suckiness wasn't *entirely* Microsoft's fault. Anyone who thinks AmigaBasic was the worst obviously never used ABasiC ;-)
> "ICANN is to be congratulated for succeeding in expanding the Internet beyond the Latin alphabet.
The problem isn't that ICANN expanded "the Internet" beyond the Latin Alphabet (or at least the subset enshrined in ASCII's alphanumeric characters plus hyphen)... the problem is the collateral damage it caused, and continues to cause, because they did the equivalent of dumping a freeway interchange in the middle of an already-thriving residential neighborhood.
ICANN (possibly with IETF) needs to do three things:
1) Work with IETF to extend DNS so that TLD registrars can define a specific subset of UTF-8 that's valid for its subdomains. By definition, .com/.net/.org should be forever restricted to the historical [A-Za-z0-9\-] subset to put an end to homograph phishing. In other words, no TLD could indiscriminately include everything from legacy-ASCII to Klingon, Runes, and ancient Egyptian. They'd have to pick the characters used to write a single real language and stick with it.
2) Require that TLD character-validity rules be fully normalized against characters between 0x30 and 0x7f. In other words, if there's a letter in the language's unicode codepage that looks just like ASCII 'i', they can allow ASCII 'i', or the language's own version of 'i' with its own UTF-8 value, but NOT both. The choice of which 'a-zA-Z' to use would largely depend upon which value gets generated from a keypress by a keyboard in the target country.
3) Create new TLDs for writing systems used in more than one country, by at least 50 million people... preferably, short and understandable to anyone in a country that uses that writing system. So, Chinese might get .{zhong} or .{zhongwen} (but the PRC itself's country TLD might be .{zhongguo}), Cyrillic might sensibly get the characters that resemble .NHT (backwards 'N') which apparently is the abbreviation for "Int" in Russian, Ukranian, Serbian, and probably most other languages using Cyrillic, etc, and conveniently looks vaguely like ".NET" to everyone else (but the backwards-N would ensure only a complete idiot could think it really WAS .net). Ditto for Arabic. Languages almost synonymous with a single country (Hebrew, Greek, Japanese, Korean, etc) or spoken by fewer than 50 million real people in daily business wouldn't get their own TLDs... but their countries would get a new country TLD in the writing system (along with their old 2-letter TLD).
The point is, the way internationalization has been rolled out so far has created a worldwide party for fraud and phishing via homograph attacks. An end needs to be put to it NOW. If someone has an existing IDNS name that would be invalidated by the new rules (say, {nurren}.com), they'd get first chance at it in the new .{zhong} TLD (nurren}.{zhong}). If there were two or more existing .com|.net|.org domains that clashed (say, {nurren}.com and {nurren}.net), they'd have to share the TLD and settle for distinct subdomains of it, like {something}.{nurren}.{zhong} and {somethingdifferent}.{nurren}.{zhong}.
Why did it even HAVE to be represented as an exact multiple of 0.1 seconds? As opposed to something a bit more binary-friendly, like 1/8 or 1/16th of a second? Was this just a case of 1980s Waterfall Design Paradigm, where somebody who didn't necessarily understand the precise characteristics of the hardware involved pulled the 1/10 second specification out of a metaphorical hat because it sounded good, then everyone below that point was expected to just unquestioningly implement it instead of asking WHY it had to be an exact multiple of some error-prone value early in the design process?
IMHO, the worst decision ever made in the history of the internet was ICANN's decision to allow non-ASCII subdomains of .com, .net, and .org. Those three gTLDs, if not others as well, should be forever off-limits to any characters besides the original 26 letters a..z, the digits 0..9, and hyphen.
For other TLDs, and for national TLDs, DNS should be extended to allow a TLD's authoritative top-level registrar to authoritatively indicate which UTF8 characters (or range(s) of characters) beyond those historically-allowed for ASCII DNS are valid for its subdomains. The registrar for Spain might decide it needs accented vowels and tilde+n, but has no reason for Turkish vowels that conveniently (for phishers) look identical to ASCII characters.
The next step would have been the creation of a few brand new TLDs, like ".Zhong" (U4E2D) and ".Nihongo" (U65E5 + U672C) -- think ".(Chinese)" and ".(Japanese)", not to mention similar TLDs for Hindi, Korean, Russian, and other languages that use non-Roman alphabets.
The point is, ICANN could have done a much, much better job with this whole mess. I think everyone can agree that international domain names are something that needs to exist, but trying to staple them onto .com/.net/.org was an incredibly bad idea.
> So basically we're trading one evil (comcast) for another (government).
The big difference is, for the most part Comcast's remedies if you subvert them are largely civil in nature -- denying you future service, charging you penalty fees, suing you, or the like. The government can have you thrown in prison.
Of course, some companies and court jurisdictions have been hard at work finding creative ways to criminalize breaches of corporate policy (particularly through abuse of "theft of electricity" rationales), but for the most part there's still a line between things that are criminal vs merely civil. Comcast's mostly on the 'civil' side, but the government is almost exclusively on the 'criminal' side.
> What do you mean fucked? There's plenty of wireless carriers that sell intenet access.
As long as your monthly broadband needs don't exceed 5 gigabytes total per month...
> The effort made in consumer electronics to save four cents (over 10 million units) would probably make your head spin.
Tell me about it. It's what condemned many very, very expensive first-generation CD-Rs to the coaster bin (no ram buffer to speak of, so if the CPU got distracted for even a fraction of a second, the CD was toast), and sent almost every Samsung SPH-i330 to an early grave (a few cents saved using mask rom instead of flash, coupled with a fatal bug that bricked them within a few weeks of use and couldn't be fixed). Not to mention the problems due to aggressive cost-cutting that have plagued videogame consoles for decades. Red Ring of Death, anyone?
A major complication is the fact that today's PDA phones are basically cellular winmodems. Ever wonder why a Samsung SPH-i300 with ~40MHz 680x0-ish Dragonfire didn't feel all that much slower than a 400MHz dual-core ARM running WinMo? Part of the problem is WinMo's bloat... but an even bigger part of the problem is the fact that cpu #2 spends basically 100% of its time being a software-based faux DSP anytime the radio is in use... and occasionally steals cycles from cpu #1 while it's at it. In contrast, the humble i300 was literally a cell phone radio bolted to a PalmOS PDA, connected by LITERALLY a serial port. In every meaningful way, the two were completely independent. Dialed a number on the LCD-rendered keypad? The Palm side sent what was basically an ATDT command to the phone side. Incoming call? The phone side did the actual ringtone (the beefy speaker was connected to the phone; the palm side just had a wimpy piezo), and sent ###RING xxxxxxxxxx down the serial port for the Palm to decide what to do with it. Pressed the green button to answer the call? The Palm side sent (literally) ATA down the serial port to the phone. And so on.
The good thing about Winmodem-like cellphones is... um... er... uh... well, I'm sure there's something good about it. But anyway, the bad thing is that it means that you're basically talking about a gigahertz-rate raw DAC and ADC creating a software-defined radio. The supreme irony is that the hardware in a G1 probably COULD be hacked to do UMTS at 850MHz on AT&T... but it would be utterly illegal, precisely because it would wreak havoc with AT&T's network and mess up your neighbors' service while you spent saturday afternoon debugging it.
As I understand it, a phone running Android (or Windows Mobile, for that matter) is kind of like a PC running Linux under VMware under Windows (or vice-versa). You have the hypervisor-like "Supreme Controller" that runs the software that makes it a radio, with Linux-nee-Android running under the hypervisor. So you really ARE root as far as Linux/Android is concerned, but behind the curtain, Root has a metaphorical deity of his own to serve. I'm not entirely sure how it's enforced, and how much of it is handled by actual hardware partitioning (ie, whether the CPU takes a cue from memory curtaining, so that the core running the radio has ram and registers that the core running the UI and apps can't touch), but that's a rough summary of what happens behind the scenes on an Android phone.
> So I'd guess a forked F/OSS version of MySQL would need to call themselves something else, losing the name.
Personally, I'm rather fond of "rSQL". It's short, it's catchy, and neatly emphasizes the difference between the open source version ("our" SQL) and the new Oracle-centric proprietary branch ("*my* **My** ***MY*** Sql").
> So why are those goodies disabled under Linux?
> Come on NVIDIA, just release the specs and let us write our own damned drivers eh?
It's not so much that they're "disabled", as a case of being "not implemented". The problem is that the line between what's a hardware capability and what's implemented mainly at the driver level through software is increasingly blurred. Just to give a familiar example, look at a PCI Winmodem. At the end of the day, a Winmodem is basically a PCI soundcard that's hardwired to a phone jack and optimized for PSTN-level voltages & impedance. Someone like Conexant could flawlessly document how to use the chips on one of their Winmodem cards to generate and sample audio, and it wouldn't do a thing to help anyone actually make the card act like a 56k modem under Linux. Someone has probably done it by now, but back when it would have actually still mattered (circa 1999-2000 or so), there was no such thing as an open-source Linmodem driver for that precise reason. Documenting the hardware was necessary, but even fully-documented, it would have only gotten you ~2% of the way towards the ultimate goal of *being* a software-defined modem.
There's another problem with video drivers -- patents. As a practical matter, everyone in the industry violates at least one patent belonging to the other big players, and they're *all* sitting ducks for every patent troll who comes wandering along. If NVIDIA were 100% altruistic, fully implemented every Windows feature into their Linux drivers, and released the full documented source code to their proprietary Linux drivers, they'd essentially be painting a red target on their forehead and making the patent trolls' fishing expeditions that much easier. It's sad, but it's true.
So... does it have a proper SDK (well, ok, at least unlocked hardware that can be freely reflashed with guerrilla builds of Android), or maybe an API for UI customization (repurposing buttons, changing the way it works, etc)?
Up to now, there's one eBook market that should have publishers absolutely salivating: computer and technical books. It's the one niche where non-crippled eBooks really DO have a compelling value proposition. Think about it... what's the biggest single problem with most computer books? They're already a version behind 6 weeks before they even arrive at Amazon, let alone Borders or Barnes & Noble. Computer-related eBooks have the potential to change that, by letting you auto-update the books regularly to incorporate fixes and revisions long after the book first became available. Plus, they have the potential for instant gratification. I'd probably buy twice as many books if I could have them *right now*, *this instant*, instead of having to wait until at least tomorrow afternoon to get them from Amazon (I've pretty much given up entirely on Borders -- once the gold standard of local computer book availability -- and Barnes & Noble has maybe 3% of the books I actually want available for immediate purchase at a store within 50 miles of my house whenever I go searching).
What I *really* want is an ebook that starts with the button configuration of an Ectaco Jetbook ( http://tinyurl.com/ctaq3q ) that keeps the transflective LCD, but bumps the resolution up to 1280x960 or 1600x1200, has approximately the width and height of a Manning book, and adds a capacitive digitizer for finger-friendly Graffiti-like handwriting recognition. And of course, a published, programmer-friendly API that lets you more or less reinvent its user interface without having to actually reinvent Acrobat reader *itself*.
Why LCD instead of e-ink? Latency. Novels are usually read page by page, sequentially. Computer books are ready by jumping all over the place, and rapidly flipping through pages to find something nearby. Frankly, 500-800ms is just plain unacceptable in that context. It would drive me nuts. I want to hit the 'next page' or 'previous page' button, and have the page visible before I even finish releasing the button (using onboard ram to cache the entire active pdf file since flash is painfully slow). Hell, if it becomes affordable, give me a second display while you're at it, so I can either see adjacent pages, or have two books open at once side by side. Maybe make one display e-ink, and one LCD (with a UI that juggles them around, so the page of interest during active page-flipping always ends up on the LCD instead of e-ink).
Would I pay $40-50 for an ebook that costs the same in print form, that can only be read on a Kindle, and is basically a warmed-over scan of the print edition? Hell no. At best, current ebooks are basically a crutch to get me over today's crisis until the paper copy arrives tomorrow. Half the time, my workflow can be described as: 1) find book(s) at Amazon; 2) order before the FedEx deadline; 3) download it 20 minutes later from eMule, since I *really* need it *right now* (admittedly, skipping step #2 most of the time if it's already past the evening deadline for next-day delivery and too late to get the real book by tomorrow anyway. It's amazing how not being able to get it tomorrow under any circumstances has a way of diminishing the perceived urgency of buying it). Yeah, I DO get my money's worth from my Amazon Prime membership. The FedEx man personally knows my name ;-)
Would I do it for a book on something rapidly-changing, like Android programming, with constant revisions for at least 2 or 3 years as Android evolves from 1.5 to 1.6, 2.0, and beyond? Hell yeah. In fact, even limited to reading it on my PC, I paid the author of the Busy Coder Android books for a 12-month subscription that gives me permanent pdf copies of his Android books (3 so far), plus constant revisions. Google 'warescripti
> The patent is on using dedicated, non-host accessible RAM on the network adapter for the ring buffers and packets, together with memory-mapped IO for setting up DMA transfers.
Well... if that's the case, it shouldn't get too far in court since just about every ethernet implementation from the past ~5 years has been crap integrated ethernet based on host signal processing (think: winmodem on crack), I don't see how this patent would apply.
I mean, how many years has it been since MaximumPC even REVIEWED a high-performance PCI ethernet card with a real, onboard DSP and controller, let alone dedicated onboard RAM of its own? Unless the slimy trolls have sunken so far that they're claiming that an onboard function sucked into the mainboard chipset using system ram with HSP is tantamount to a PCI card with ram of its own...
> then in six months and one year, you query those people, see if they're still alive. After all, we don't care
> if you died from influenza or the marthambles - if the vaccine keeps you out of the grave, then it's a win.
Er... death isn't the only consequence worthy of avoidance. That's the same logic health insurance companies (and government agencies in countries with socialized medicine) use in non-pandemic years when trying to argue that Tamiflu and Relenza aren't cost effective, on the grounds that it only reduces the "acute phase" by a day or so. What they conveniently omit is the fact that they might "only" reduce the phase when you're laying in bed in the fetal position hating life by a day or so, but they ALSO reduce both the duration and severity of the "misery" phase afterwards... the point when you're well enough to go back to work, but still feel and look like shit, with a runny nose that just won't quit, sore throat, and the rest. Both drugs can dramatically improve the overall experience, and make it a lot less awful than it would be if you did nothing to stop it.
As far as efficacy goes, I've religiously caught the flu at least once every year when I didn't get vaccinated. I've only gotten sick once when I was vaccinated, and I'm not sure whether THAT was a mild flu, or just a really bad cold.
There IS one problem with the "pick 3 strains" strategy -- it works a lot better in places like Iowa, Arizona, and South Carolina than it does in places like Miami, Orlando, and New York City, where there are literally millions of visitors every year from every single continent, and whose local flu strains weren't part of the vaccine mix.
I believe most CRTs actually consumed less power in active use than comparable *LCD* sets circa 2007 did. From what I know about electrical engineering, the main culprit was the power supply. CRTs had a fairly small AC-DC power supply, and drew most of their power from a transformer that converted 110VAC/220VAC into a few thousand volts... with relatively high efficiency. In contrast, LCD panels need lots of DC, much of which needs to be tightly regulated. To save money, they often use linear power supplies. Generalizing a bit, a linear supply draws about as many amperes of source voltage as it outputs in target voltage.
Here's an example. Suppose your LCD needs to draw 10 amperes at 5v, and you're feeding it 12vdc that came from a transformer and ran through a diode bridge. It's going to draw at least 10 amperes of 12v at the input end. In contrast, if you're using a more efficient (but slightly more expensive to build) active power supply, you might need ~6-8amperes of 12v input power to supply 5v @ 5A on the output side. CRTs had linear power supplies that were inefficient too, but the power needs of the tuner and control circuitry were pretty minimal, so it didn't matter much. Simply replacing the linear power supply with an active one can dramatically cut the power consumption. It would be insane to spend $25 more on the power supply for a wireless access point that draws ~150mA and sells for $20 on sale right now, but makes quite a bit of sense to do it for a $600 TV that draws hundreds of times more power.
As for the regulation, I think it's heavy handed. California could achieve most of the same benefit, with lower compliance cost, by simply requiring that stores clearly indicate the total amperes the TV draws from 4 hours of continuous use, and the estimated annual cost of that use based on the moving 5-year average price of power in California (with the state publishing the official value to be used for the next 12 months' of calculations each year). Let consumers see that TV A (on sale for $499) will cost them $10/month more to use than TV B (not on sale, $699), enable them to easily determine that TV A will ultimately cost a LOT more than TV B if they keep it more than 20 months, and Adam Smith's invisible hand will do its job just fine.
As I understand it, in Las Vegas particularly, it's actually a legal requirement that casinos try to identify and prevent card counting. Why? Because by law, all casinos have to have the exact same payout rate. For things like video slot machines, that's trivially easy to enforce through software alone. The problem is, if a casino "off the strip" openly allowed card counting to boost its popularity with tourists (imposing a max bet, and counting on the fact that most people who claim to know how to count cards really can't), the state of Nevada would have its agents promptly there to shut down the casino.
The requirement itself is a blatant anticompetitive strategy used by the mega-casinos. It ensures that the big, new, shiny casinos on the strip have the exact same payout rate as the old, small casinos downtown and elsewhere, so the others can't compete with them by offering better odds to players.
> Mouse/keyboard activity timeout works nicely for that.
You know, I'm sure it's just my imagination or a coincidence, but it really DOES seem like Vista on my desktop PC doesn't go into power-saving mode UNTIL it perceives a tiny, tiny mouse wiggle after a long period of inactivity. It seems like I can leave my computer on, walk away, and see it running in full-power mode for hours... then the nanosecond I sit down and bump the desk, it decides to power down the panel. I'm halfway tempted to put a camcorder on a tripod and just leave it capturing for a while, then the next time I think it happens, go back and watch the last few minutes of video beforehand to see what really happened to make me think it did something crazy like that.
> Do you think all those netbooks and cheap desktops are being sold for graphic designers or gamers?
No, they're being sold to Slashdot users who'd never be content to endure the limits of a laptop as their "real" computer (last I checked, there's no quadcore 3GHz i7 available for notebooks), but will happily throw down a few hundred bucks for a decent semi-throwaway netbook to use on weekend trips and other occasional situations where it's nice to have.
> 2. FCC mandates all wireless carriers switch to GSM.
Er, actually GSM is an officially-deprecated dinosaur. Even in Europe. 3G "GSM" is actually "W-CDMA".
Now, that doesn't stop everyone from CALLING W-CDMA on a 3G GSM phone "GSM", but it doesn't MAKE it "GSM". GSM is a variant of TDMA (3 virtual channels that share a single frequency channel by sequentially using their own exclusive timeslices within it). TDMA is the reason why GSM-based devices cause a bursty buzzing sound if they're placed too close to an amplifier... the buzzing noise is caused by the click when the carrier rapidly cuts off for a fraction of a second, and the next phone's carrier cuts. Think: the click made by a Commodore 64 when you changed the volume register (which was really a wavetable DAC hardwired to a circuit emulating a single byte of RAM, with the various SID effects produced by directly manipulating the apparent value of that virtual byte).
Need more proof? Read up about picocells. Picocells that handle CDMA and *ONLY* 3G GSM are fairly cheap, and becoming commodities. Picocells that handle oldschool GSM *and* CDMA/WCDMA("3G GSM") are quite a bit more expensive, because THEY require implementation of two completely different technologies. Adding the ability to roam on a European 3G network to an American CDMA phone (a-la Verizon's Touch HD, I believe) is more of a political and business decision than a technological problem. Adding true GSM capabilities to a CDMA phone requires a second radio module & baseband processor, because the technologies are completely different.
> Wow, you sure like to set the bar low...
load "$",8
list
> But Toyota's invention is only an improvement upon the other invention. Toyota can patent the improvement,
> but they and all licensees must also obtain a license for the original patent. Just because you add a redial
> function to a phone doesn't mean you then can patent the phone.
And a major part of what's broken with the current patent system is the fact that it gives nearly unlimited veto power to whomever had that first "idea", without regard to other, equally fundamental patents that make the "original" patent commercially viable in the first place. Given enough cash, you could sit around with a bunch of patent attorneys, brainstorm ideas all day with no real idea of how you might actually make them *work* (or be manufacturable, or work more than .3 seconds before blowing up, or...), send the attorneys off to patent what's patentable, and sit on them for the next 20 years in case someone invents some real product that in any way, shape, or form bears some similarity to one of your patents.
Of course, the devil's in the details. A partial start might be a law that says that if you're challenged for patent infringement, and you can establish that YOU have a viable patent that meaningfully improves upon a patented item, improves its manufacturability, or uses it in some novel way, you can pay 50-80% of the net profits from the sales of the allegedly infringing item into a fund that gets split up among everyone who can establish that they have a patent being infringed. If you make no money, you owe nothing. If you make a hundred million dollars, you owe $50-80 million.
OK, there's a million details to work out, not the least of being the definitions of "novel" and "improvement", and a strategy for determining how it should be divided up, but the point is that it would ensure that the patent owners make at least as much money as the alleged infringers, while simultaneously maximizing the odds of someone finding a cool way to make use of patents that ultimately serves the constitutional foundation of American IP law -- advancement of the sciences and useful arts.
> But what I want to know is, why hasn't anyone mass produced a Thin CRT yet?
They've been prototyped -- 10 years ago, I was convinced that the future of television was the Field Emission Display (FED) after I saw a demo at CES. Absolutely *beautiful*. The best of all worlds. Bright, saturated, distortion-free, and viewable from angles just like a regular CRT.
Basically, coat a sheet of glass with colored phosphors, and put individually-addressable solid-state electron sources behind them. To light up a particular phosphor group, turn on the emitters behind it to make it glow. Unfortunately, the technology went nowhere. :(
http://en.wikipedia.org/wiki/Field_emission_display
> First of all it's going to be a high-profile target for terrorists.
> So, expect check-in situations exactly like the airport.
One big difference, though. It doesn't take much explosion-wise to kill lots of people on a plane. American trains, by comparison, are rolling bank vaults. A bomb detonated in a passenger car would kill ~50 people, max. Passengers in adjacent cars would barely notice the muffled thud. Sure, lots of terrorists could try to pull off a synchronized event... but then the complexity goes way up, and the "bang per terrorist buck" is still pretty low compared to much, MUCH softer targets... like a McDonalds across the street from an elementary school, or a high school/college football stadium at homecoming.
I *would* like to know, though, how in god's name a trainset consisting of two locomotives and 3-5 passenger cars ends up costing more than a f***ing brand new 777. I tend to support rail projects, but someone first needs to figure out why anything involving a train in America ends up costing more than the goddamn Apollo space program did. It somehow cost more money to double-track Tri-Rail from Miami to West Palm Beach -- through an existing rail corridor that USED TO *BE* double-track at one point in the past -- than it cost to progressively tear down and rebuild I-95 in Broward County, taking it from a 6-lane antique to a fairly impressive 12+ lane modern freeway in the process. Yeah, they built *a* new bridge for Tri-Rail. Big deal... they built two more, each 3-4x as wide as the new rail bridge, right next to it as part of I-95's reconstruction.
I really think that half the problem is that American states send people to study trains in Europe and Japan. Really, they should be going to INDIA to learn how to build and run a huge train network that's relatively fast, dirt cheap, at least as reliable & on-time as American air travel is, and genuinely useful to millions of people every day. Personally, I'd kill for hourly 100mph service from Miami to Orlando, with the ability to do rental car paperwork on the train & walk directly to the parking garage -- key in hand -- once the train arrives. It's a nearly ideal intermediate-speed rail route... farther than anyone really wants to drive, but not *quite* far enough to be worth the cost, grief, and "hurry up and wait" stress of air travel. The problem HERE is that every time FDOT gets ready to build something useful and sane, the HSR monster rears its ugly head, and derails the whole thing for another decade or two.