From what year are you counting? The Amiga 1000 came out in 1985, and its graphics capabilities were pretty close to dead-even with the best you could get from a SGI Iris 1000 series graphics terminal from the same era.
From what I remember, the most impressive mainframe graphics demos from that era actually had to cheat, because even the best graphics terminals weren't fast enough to update an entire raster frame during VBLANK... and unlike Amiga, they didn't have Jay Miner's magic working behind the scenes to let them output graphics that were more impressive than the raw chipset specs implied. Instead, they'd synchronize the outputs of two or more SGI terminals, use them to render alternating frames, and use custom hardware to switch a single monitor or projector between them -- often, using additional vector projectors to overlay additional detail on top. Had Amiga ethernet cards existed & somebody written a suitable host app to turn it into a terminal in its own right, it would have made a pretty impressive mainframe graphics terminal itself (especially an A2000 with a FlickerFixer).
OK, the NEXT generation of SGI's best terminals that arrived about a year or two later pretty much toasted everything, Amiga or otherwise, and it was 15 years before PCs even started to come close, but it IS kind of satisfying to know that there was a brief period when the graphics capabilities of at least one popular home computer & the best mainframe graphics terminals money could buy briefly converged.
The difference is, if DIY'ers do it, there will be relatively little public data that can be mined, the various formats will usually be mutually-incompatible, and it'll be self-limiting in its ability to cause harm to random strangers.
It's another matter entirely when you have a quarter-million people streaming everything in sight (literally) to Google's servers for eternal retention... geotagged, face-recognized, and there for anyone to search.
So, why,exactly, doesn't anybody who sells such hardware actually include the magic "802.11y" or "802.11y-2008" *anyplace* where Google might find it and associate it with their products? Is all 3.6Ghz Wimax 802.11y-2008 by definition and reality, or is it kind of like the situation with Zigbee-vs-802.15.4, where you have hardware that could technically be compatible if the firmware supported it... but doesn't?
^^^ The problem is, you really don't have any choice in the matter. Most MegaHuge Corporations won't reimburse you for purchases not made with the official corporate card, so if that card is Amex, they can pretty much rape your ass any way they feel like doing, and you really don't have any choice in the matter besides quitting your job or getting fired for being unable to travel and refusing to cooperate.
In a way, it's even worse than the usual form of corporate serfdom, because you can be required to run up literally thousands of dollars in travel expenses as a condition of remaining employed, but your employer can then turn around, declare bankruptcy without warning, and leave you personally on the hook for more money than you actually earned in salary going all the way back to day one. The media loves to talk about executives making $20,000 day trips by private jet to have a staff meeting at a vineyard in Napa... it never talks about staff members who were ordered to run up $2,000 in last-minute travel expenses the day before a mandatory "all hands" meeting on the other side of the country where they were told that they're getting laid off, and the company's lawyer is filing for bankruptcy at that very minute. It happens. (textbook example: Worldcom)
A *lot* of Worldcom employees got burned by that seeming technicality. They made business trips during the company's final days with their Worldcom Amex card, then when the company declared bankruptcy, all reimbursements were frozen, and there were rumors that employees might even have to pay back expenses that were previously reimbursed over the previous 5 months.
In the end, the court had mercy on the non-executive employees who were at risk of getting laid off with an unreimbursed $5-10k Amex bill to boot, but American Express made their lives absolutely *miserable* for months while the auditors scrutinized the bills to make sure the charges weren't "excessive" (and thus excluded from reimbursement) unless they had the cash to pay their bills out of pocket while waiting to get their expenses approved.
The real story here is that American Express, unlike Visa and Mastercard, rarely -- if ever -- sides with a cardholder over a merchant, even in situations where it's blatantly obvious that a charge is wrong. It's how they're able to get merchants to accept higher swipe fees and transaction charges.
For the most part, if a cardholder files a chargeback with Mastercard or Visa that looks even slightly reasonable, they'll freeze the funds from the merchant account within a matter of minutes. They might not issue the refund to the cardholder immediately while they're investigating if there's a dispute, but the burden at that point is overwhelmingly on the merchant to prove the charge was legitimate and correct. And if there's still any doubt, Mastercard or Visa will issue the refund, furnish documentation to the merchant, and tell them to sue you in small claims court if they think payment is owed.
In stark contrast, American Express will demand copies of the receipt from YOU (the customer), demand nothing from the merchant until they're 100% satisfied, and will still probably side with the merchant absent overwhelming evidence to the contrary.
In my life, I've had two chargeback disputes with Amex... and lost both. One was careless stupidity on my part (though still blatantly wrong), the other was Amex showing just how anti-cardholder they really are.
The first time it happened (circa 2000), the clerk accidentally hit '0' twice when entering the amount into the credit card terminal, and turned my $13.90 purchase into a $139.00 purchase. I had a receipt showing that the price was ~$13.25 with ~$0.65 in sales tax, but because I didn't notice until a month later & signed the receipt, Amex refused to budge. When I challenged American Express to have the merchant produce anything resembling an itemized receipt showing PRECISELY $139.00 in purchases with a timestamp within an hour of my alleged $139.00 purchase, they refused. As far as they were concerned, all that mattered was the 2x4 inch nearly-illegible receipt printed by the credit card terminal with something that resembled a blue smudge of a signature on it that I admitted to having (in addition to the itemized receipt for 1/10 the amount). I cancelled my card over it in rage, and refused to do business with them for about 10 years.
Fast forward a decade. My employer required me to get an Amex corporate card as a condition of getting reimbursed on business trips. One morning, the cashier at a Waffle House ran the card through, then somehow screwed up the machine between the moment the charge went through and the moment the printer produced the receipt. She insisted that the charge didn't go through. I argued with her for 3 minutes (I heard the printer start, before the power cord short or whatever rebooted it), then gave in because I didn't have much of a choice. Sure enough, I got charged twice, about 3 minutes apart, for the same amount. This time, I was sure Amex would take my side, because the restaurant obviously didn't have a signed receipt from me for the first one, and there was no sane reason why somebody would have two ~$6.00 charges approximately 3 minutes apart. Goddamn if they didn't do it to me again, and refused to reverse the first charge. Their official excuse? I didn't have an unsigned copy of the first receipt. What. The. FUCK. That really, REALLY pissed me off. Yeah, my boss signed off on the override and told me it wasn't worth fighting with them over $6, but it was the moral principle of it.
So, if you're an AMEX cardholder, be warned: if anything goes wrong, American Express WILL NOT take your side. In fact, they won't necessarily even lift a finger to make the merchant defend the charge. They'll press YOU for receipts, and will disqualify your chargeback on the slightest technicality, but will let the merchant get away with almost anything short of blatant, systematic intentional fraud.
Or, if you can find any actual company with a real product to sell, buy some 802.11y gear, pay the (theoretical) hundred-buck fee for a 10-year license to use the 3.6GHz spectrum, and do it with complete unambiguous legality.
Now, finding actual 802.11y hardware might be a bit of a challenge... the last time I looked, it was *still* commercially-nonexistent circa 2-3 months ago.
Were they literally brand new isolated segments, or were they new lines that extended (or at least directly shared a station with) existing lines? I'm talking about the American situation where a city with no meaningful transit to begin with builds its first rail line, at least half of which passes through areas that are industrial wastelands, dangerous slums, and/or just plain stupid places to build it (generally, surrounded by an ocean of single-family homes whose owners will fight any attempts at skyscraper-density redevelopment next door).
In America, skyscrapers next to the station are almost mandatory for viability, because Americans have very little tolerance for extended outdoor walks. If their destination isn't next door or across the street from the station, Americans generally won't take the train (however, transit systems can cheat a bit, and build underground air-conditioned corridors that psychologically extend the station's useful footprint, because "next door" is measured from the point where the rider has to actually go outside. The underground corridors at Crystal City (Arlington, VA) can get kind of creepy (it almost feels like you stumbled onto a movie set or into some area where you aren't really allowed, and you're going to get run down by security guards at any moment), but they DO serve the useful purpose of making a fairly large area feel like it's all adjacent to the station.
All were mostly useless the day they opened, and were seriously limited in their usefulness for at least a decade... but now, roughly 30-40 years after the first segment opened, all four cities have at least gotten to the point where somebody who lives within walking distance of a station can find just about everything he or she needs within walking distance of another station, and depending upon the local job market and one's career field, has roughly 50-50 odds of being able to work within "viable" distance of a station.
More importantly, all four rail lines now have at least a few segments that are of sufficient importance that none of the cities would *dare* to take up an evil genie on his offer to refund the lines' original construction cost (not adjusted for inflation) in return for making them disappear forever. That's not to say that all four don't have examples of glaring mistakes that are painfully obvious today... but they each offer examples of the circumstances under which they can prosper and flourish in their respective markets (most of which can be summed up as, "embrace sprawl, proudly take ownership of it, and direct it into new suburban skyscraper clusters around outlying stations").
The problem isn't that LA needs wider freeways OR an expanded rail/subway network... LA needs wider freeways *AND* an expanded rail/subway network. People who argue for one to the exclusion of the other are missing the point.
A new rail line won't do a damn thing for gridlock *today* -- it takes a minimum of 25-40 years before a new rail line really starts to pull its own weight (40 when it's the first segment in a city, 25 when you're extending one that's already established). That doesn't mean the rail line shouldn't get built... it means the rail line should get built today, along with the freeway's reconstruction, so that 25-40 years from now, the rail line will have matured enough to start absorbing travel loads from the freeway, and it won't be necessary to demolish and rebuild the whole freeway yet again.
Verizon's bigger problem is that their implementation of EVDO is different from everyone else, so a phone from any other EVDO network on earth that's unlocked and made to work on Verizon will never be able to do better than 1xRTT. That's why Sprint/Metro/other CDMA phones roaming on Verizon can only do 1xRTT.
So, how many gigabytes per month can you use with an average data speed of 23kbps down (the actual real-world speeds I was getting from Sprint the day I finally left them in disgust), and 500kbps being cause for a standing ovation? OK, there might be 2 or 3 LTE towers somewhere within 10 miles, but right now, their LTE coverage is thinly-spread to a degree that makes T-Mobile's 2007 HSPA+ coverage look downright *saturated*. Sprint screwed up badly by doing a clean break from wimax, instead of using dualmode wimax-LTE chipsets for 2-3 years, deploying LTE into places without wimax, and THEN migrating the wimax to LTE. Instead, they took a relatively-functional wimax network, and pretty much threw it away the moment everyone bought a new phone last year. Just about everyone I know left Sprint after they got their new phone, and went from "ok" wimax that compensated for dysfunctional 3G/EVDO, to having no meaningful mobile data service at all because their 3G network was (and from what I've heard, pretty much still IS) in a state of dysfunctional meltdown.
I'm afraid to look at the rest of the specs to see which non-negotiable spec ended up getting gimped this time, since it seems to be a grand tradition among manufacturers that the closer they get to perfection, the more horribly they have to ruin at least one specific thing.
Please, sweet baby Jesus, fierce Xenu, and most noodly FSM, let this phone have a removable battery whose stock capacity exceeds 2500mAH, microSD, and a 5" display having a minimum of 1920x1080 resolution. With abundant ram and DVI (possibly via MLH), of course.
> Unless they really right the ship with Windows 9
(...) So far, their effort has been mostly a flop. Unless they really right the ship with Windows 9 by giving us back Aero Glass, or raise the bar several notches and give us Aero Glass Dynamically-Translucent Rotating Cube, with support for LCD touchpad [which defaults to a normal desktop experience unless you feel like slumming with phone/tablet apps, and simultaneously has sufficiently-high 'wow!' factor that doesn't interfere with usability AND gives consumers a compelling reason to buy new laptops with 3GHz+ quadcore i7 CPUs and discrete GPUs that would have been jaw-dropping on a desktop 2-4 years ago], they will shift from market dominance to just another vendor (...)
There, fixed that for you.
The problem isn't that computers are "good enough" now, the problem is that Windows 8 is butt ugly, and offers consumers nothing compelling to take their breath away and say, "Wow, that totally kicks ass!". Nobody is going to go buy a new computer that looks like a dumbed-down beta version of their previous one. If manufacturers want to fight the race to the bottom, they have to give consumers a reason to care about higher-end specs and put them to visibly-compelling good use. MetroModern just isn't going to sell high-end discrete GPUs and i7 CPUs. And frankly, kick-ass eye candy that doesn't impede usability is just about the only high-end hardware frontier left to fight Apple-ization and laminated.7mm slabs with the approximate usability of Ubuntu Unity.
Seriously. Microsoft needs to give laptop and desktop manufacturers a reason to sell thousand-dollar high-end laptops, and desktops with 3+ displays that will make people with lower-end hardware feel like poor second-class users who can't run the latest software (well) until they upgrade, not cut Windows down to something that can run on an anonymous shit one-gigahertz tablet from Shenzhen with a tiled window or two and soft keyboard.
There are unclean hands on both sides, and as usual it's the end users who get caught in the crossfire.
If Linux's kernel developers made even the most half-assed token effort at not wantonly breaking the ABI with every new release, or allowed drivers to have an intermediate thunking layer that stabilized the ABI for at least a year or two at a time in exchange for a little more overhead, the driver problem would largely fix itself and become a non-issue. The problem is, Linux's stewards *want* kernel-dependent binaries to catastrophically break with every release. It's an annoyance with desktop linux, a *real* problem with laptop linux, and the biggest single source of misery in Android guerrilla-AOSP land.
Ironically, much of the NVidia-side misery is caused by licensing terms with the same viral recursiveness as the GPL. If NVidia or AMD want to support playback of Blu-Ray content under Windows, or DRM'ed content under Android, they're required to agree to viral licensing terms that force them to implement and enforce things like Cinavia watermark detection everywhere, across the board... and prohibit them from talking about it or releasing the source to the general public. So they end up in a no-win situation... they aren't allowed to support playback of non-DRM'ed content under Linux, because it would be an easy way to bypass things like Cinavia, but they aren't allowed to release open-source drivers that pretend to care about it, because they're prohibited from disclosing details about how it works.
A federal law banning recursive scorched-earth viral licensing terms might fix the problem... but we all know that's not going to happen during our lifetimes, because Hollywood would freak out, and even a small Hollywood studio has more money & influence than the entire computer industry combined when it comes to Congress.
The next-best compromise might be if someone like NVidia were to start building mobile (and desktop) GPUs with a well-documented FPGA of sufficient capacity to implement h.264 decompression. Then, they could release official opensource drivers that simply ignored it and left it idle (since the same FPGA could arguably be used to implement a 2D sprite engine, or bitcoin mining, or whatever), and we could include our own code to hijack the FPGA and turn it in to a DRM-unencumbered h.264 decoding engine that might very well stomp all over 300 of MPEG-LA's patents and wantonly infringe upon them in ways that would make Hollywood seethe in rage, but as long as the source were hosted on a repo in Kazakhstan & distros like Ubuntu kept it at arm's length and didn't "officially" recognize it, Hollywood couldn't do a thing to stop end users from using the driver.
It depends. Linux tends to be nicer than Windows for USB, but still kind of falls flat on its face when it comes to GPU support, including HD h.264 video. Brute-force CPU decoding can only get you *so* far before the memory-bandwidth realities of trying to shovel around realtime 60fps 1920x1080 32-bit video data via PIO are going to bite you and cause problems.
Linux on laptops has gotten *enormously* better over the past few years, but it seems to consistently still have one set of problems with laptops that are "too new" (no drivers), but distros made for older computers seem to keep omitting support for stuff like PCMCIA/Cardbus that older laptops still depend upon to work properly.
I've gone through more than a bit of frustration over the past few few days trying to get Linux working properly on an old laptop I pulled out of the closet to resurrect (Compaq Armada M700... 700MHz PIII, 544mb ram). USB sucks on it (1.x only... though I've contemplated trying a Cardbus USB card), and most of the distros I've tried are either clueless about dealing with PCMCIA/Cardbus ATA devices, or freeze and require power-cycling when inserted (Anti-X being a notable example) .
Puppy (Wary 5.5, their distro for older hardware) seems to be the best so far, but even its Seamonkey browser tends to completely choke on web sites that go off the deep end with Ajax (like returning a blank page that references Javascript, then using that Javascript to fetch content via Ajax & build the entire DOM from scratch one node at a time). Still, I'm toying with stuffing a 16gb no-name PATA SSD from eBay into it to see whether it works some magic. I have a hunch, though, that the cheap PATA SSDs from China might really just be a microSD card with some glue logic to make them look like a CF card with 44-pin PATA interface, and not really be any real improvement over the 7200RPM drive that's in it already.
No. PalmOS in its pre-Cobalt form was fundamentally dead on arrival the moment realtime tcp/ip with a responsive UI became a big deal, just like MacOS9. A PalmOS phone was basically a Palm Pilot strapped onto a headless phone through a serial port. The phone subsystem was pretty much completely independent from the palm subsystem, sharing little more than an internal serial link and battery.
As fashionable as it is to hate Microsoft and gripe about how badly Windows 8 sucks, the fact IS that Windows Mobile WAS groundbreaking back in the early 00s. It might have been utterly dysfunctional out of the box as an operating system for making voice calls, but as the operating system for a pocket-sized laptop with wireless data capabilities, it picked up the ball where Palm dropped it and ran even harder. Had Microsoft left well enough alone, and reacted to Android by creating proper APIs for implementing alternative 'phone' and 'launcher/homescreen' thirdparty apps (instead of delegating the task to HTC and calling it a day, then later throwing the baby out with the bathwater so it could port Danger's OS from Java to C# and rebrand it as "Windows Phone"), WinMo8, 9, or 10 would have been a strong alternative to Android today instead of the crippled, unloved, locked-down joke we have now that's turned into a cancer destroying desktop Windows as well.
Lots of the things we take for granted in Android were "there" and worked fine in Windows Mobile 5/6, too... and more importantly (for patent purposes), did NOT work well AT ALL in PalmOS (if they worked at all), and barely worked in Android & IOS until 2010 and beyond. The biggest single problem high-end WinMo phones had was hardware -- US Carriers weren't in any hurry to push the envelope, and HTC was perfectly content to give them the minimum they asked for. And HTC made the ill-conceived decision to eliminate the 'windows' and 'ok' hardkeys in an effort to be more iPhone-like, without stopping to consider the fact that all of Microsoft's usability testing up to that point TOOK FOR GRANTED that the device would have two physical buttons that required at least a little bit of physical force to trigger (hence, the in-pocket touchscreen activations that caused endless misery if you got a text message or phone call that went straight to voicemail).
Anyway, the point is that once in a great while, Microsoft *does* manage to do something right, even if it completely drops the ball in other related areas. WinMo had plenty of warts, but circa 2005/2006, it WAS pretty much the best thing you could get if you wanted wireless internet connectivity in a device that could (sort of) limp along and make voice calls in a pinch. And it sure as HELL beat walking around with a Palm Vc or Handspring Visor and $129 18" cable to plug it into your clamshell phone for data a few years earlier, or limping with a later PalmOS phone that was good for making voice calls and managing an address book, but fell flat on its face the moment you tried doing anything that involved realtime network communication with a responsive UI (the UI froze whenever the phone was sending or receiving data due to the way PalmOS Garnet's network stack was stapled onto it as an afterthought).
Also, I believe a big chunk of Microsoft's patent portfolio came from its acquisition of Danger (the Sidekick's maker), which had plenty of its own innovations.
Because taskbar-pinning sucks. The taskbar is for minimized windows (and quicklaunch shortcuts for things like the command prompt), not ambiguously-running-not-running applications. When I close my applications, I want them to die and be metaphorically incinerated immediately by the garbage collector, not linger around as undead zombies consuming resources until the next time I reboot.
It utterly blows my mind that some people leave programs with embedded browsers churning away client-side Ajax running nonstop in the background, then wonder why their computer stutters and at least one of their cores is pegged. Or leave something like Word running, with 400 megs of cut & pasted bitmap data clogging the clipboard. Or Photoshop. Or Eclipse and the Android Emulator (shudder). Or 12 Firefox windows, half of which have more client-side Ajax churning away nonstop in the background, updating ads you aren't even looking at, leaking megabytes of allocated ram per hour, and pegging one or two more of your CPU cores for no good reason?
Taskbar-pinning was the *worst* part of Windows 7... but at least Microsoft had the decency to allow users to disable it without having to throw the baby out with the bathwater and give up Aero Glass, the start menu, and everything else, too.
The whole Metro-vs-Desktop debate reminds me of why everyone who's ever used a program developed for people whose life revolves around the use of that program is horrified by the thought of Microsoft ever acquiring ownership of it (programs like Dreamweaver, Photoshop, Chief Architect, EagleCAD, Blender, etc).
Microsoft truly doesn't *get* that hardcore users who interact with a piece of software all day have fundamentally different needs and expectations than a receptionist who wants to make a sign for the office party's dessert table using ComicSans and Nyancat clipart. I remember back when I used to do web design how completely fsck'ing tedious it was to edit tables with FrontPage, and how completely painless it was to do in Dreamweaver (once you learned how to do it). I shudder to think about what a Microsoft CAD program would be like.
With Microsoft-UI philosophy, you might be able to guess how something works in 12 seconds without reading the manual, but you'll feel like you're fighting with it for all eternity thereafter, because things that should be painlessly easy to do with a few minor gestures end up turning into 2 minutes of mouse-clicking and drilldown-traversal.
I remember the UI debate well from the Palm/WinMo era... the Palm philosophy was that if something is going to be a seamless, integrated part of your daily life for the next year, it's worth spending an hour learning stuff that will pay dividends that multiply in value over time. The Microsoft philosophy was that everything blindly had to work like XP and Windows, regardless of how tedious and absurd it made the process of using it.
> Knowing Microsoft, they'll probably release SP2 for Win7, which puts the "Modern" UI on top of it too, > and then make SP2 a prerequisite for every security update that comes out after it.
And if they did, business school textbooks would have a new case study for corporate suicide, and a breathtaking example of how a company that managed to go from a dominant market share of the high-end mobile market to irrelevance within a matter of months was able to repeat it to throw away their desktop dominance as well (everyone had a major love-hate relationship with Windows Mobile, but if you wanted a pocket laptop with a useful browser that could be used for making voice calls in a pinch, WinMo was pretty much the best there WAS circa ~2007).
It would be the day I officially blew away Windows and promised God, Xenu, Thor, and the Flying Spaghetti Monster that I would never, EVER voluntarily run Windows as my real operating system again. And did my best to get everyone else whom I influence to do the same.
Put another way, for Microsoft to do something like that would constitute a full-frontal act of corporate warfare against its customers... and retribution from the consumers who matter would be swift, damning, and deadly. Look at the amount of hate Microsoft has taken from... well... everyone... over the past 6 months. Now imagine how much MORE hate they'd take if they loudly and proudly sank the lifeboat (Windows 7) that's keeping them alive right now. They'd have people burning computers on the sidewalk in front of their offices, hanging Ballmer in effigy, and Barnes & Noble would be filled with books about dumping Windows almost overnight.
A full frontal assault upon their customers would be the beginning of a rapid end for Microsoft. With their "influencer class" of users angrily gone, and thirdparty developers leaving in protest as well, Windows would degenerate into an inferior, second-rate OSX for consumers who buy a computer and use only the apps that were bundled with it.
Not necessarily. Realistically, any nuclear bomb detonated in an American city *today* would likely be ~5-10 kilotons, and most of the deaths and injuries would be due to panicked people running for their lives & getting hit by cars, injuries from wildlife or exposure to the elements if they literally "headed for the hills", or in the looting spree that would occur a few hours later.
That's not to say even a small bomb wouldn't be incredibly harmful... death is death regardless of whether it's due to nuclear incineration, radiation, a bear attacking people who fled and are hiding in the woods, freezing to death (winter attack), a rattlesnake bite, or a bad car wreck when the healthcare system is regionally-overwhelmed, and the federal government would probably spend tens of trillions of dollars on the cleanup (excavating a square-mile down to the bedrock and hauling the dirt 3,000 miles to a site in Nevada or Wyoming).
Regardless, anybody who lies awake at night worrying about cities getting multiply-nuked by 20+ megaton bombs can relax. The main reason they're still there is the perverse irony that we (in the US) are safer with the missiles pointing at us, but firmly under Russian government control and subject to inspection & auditing by the US, than we'd be if they were rapidly decommissioned. Decommission them *too* quickly, and there's a VERY real risk of uranium or bomb parts getting "lost" in the shuffle and ending up in the hands of people whom NEITHER the US *nor* Russia wants to have nuclear bombs. We're both gradually decommissioning missiles, and the main limiting factor is demand for their fissile uranium by nuclear power plants. The goal in both countries is for it to go straight from missile to power plant, with as little surplus floating around at any given time as possible.
>Without the ability to survive at that location there is no value gained in going there.
It's a good thing King Ferdinand & Queen Isabella didn't share your view. In a more recent context, I suspect there are quite a few people in Las Vegas and Dubai who'd beg to differ (not to mention Antarctica, undersea, the summit of Mt. Everest, and South Florida around August & September).
If you were exiled to what's now Las Vegas 500 years ago, and had to somehow live off the land without external supplies or support, how many days do you think you'd have survived under the best and most optimistic possible circumstances?
> So far PR is the only reason I have ever seen for sending humans.
And PR is what generates public (and ultimately political) support for NASA's budgets. If NASA decided to reinvent itself as an agency focused exclusively on pure science, and eliminated everything you might regard as "PR Fluff" from its mission, it wouldn't *have* a mission within 10 years. It would end up on the chopping block and get cut the next time there was any kind of budget crisis. Oh, ok... it might not get abolished and defunded *entirely*... it would probably get reassigned to some other agency, then get progressively starved to death until it was little more than a name with a director and a few university grants to manage per year.
Slashdot Mirror
From what year are you counting? The Amiga 1000 came out in 1985, and its graphics capabilities were pretty close to dead-even with the best you could get from a SGI Iris 1000 series graphics terminal from the same era.
From what I remember, the most impressive mainframe graphics demos from that era actually had to cheat, because even the best graphics terminals weren't fast enough to update an entire raster frame during VBLANK... and unlike Amiga, they didn't have Jay Miner's magic working behind the scenes to let them output graphics that were more impressive than the raw chipset specs implied. Instead, they'd synchronize the outputs of two or more SGI terminals, use them to render alternating frames, and use custom hardware to switch a single monitor or projector between them -- often, using additional vector projectors to overlay additional detail on top. Had Amiga ethernet cards existed & somebody written a suitable host app to turn it into a terminal in its own right, it would have made a pretty impressive mainframe graphics terminal itself (especially an A2000 with a FlickerFixer).
OK, the NEXT generation of SGI's best terminals that arrived about a year or two later pretty much toasted everything, Amiga or otherwise, and it was 15 years before PCs even started to come close, but it IS kind of satisfying to know that there was a brief period when the graphics capabilities of at least one popular home computer & the best mainframe graphics terminals money could buy briefly converged.
The difference is, if DIY'ers do it, there will be relatively little public data that can be mined, the various formats will usually be mutually-incompatible, and it'll be self-limiting in its ability to cause harm to random strangers.
It's another matter entirely when you have a quarter-million people streaming everything in sight (literally) to Google's servers for eternal retention... geotagged, face-recognized, and there for anyone to search.
Interesting.
So, why,exactly, doesn't anybody who sells such hardware actually include the magic "802.11y" or "802.11y-2008" *anyplace* where Google might find it and associate it with their products? Is all 3.6Ghz Wimax 802.11y-2008 by definition and reality, or is it kind of like the situation with Zigbee-vs-802.15.4, where you have hardware that could technically be compatible if the firmware supported it... but doesn't?
^^^ The problem is, you really don't have any choice in the matter. Most MegaHuge Corporations won't reimburse you for purchases not made with the official corporate card, so if that card is Amex, they can pretty much rape your ass any way they feel like doing, and you really don't have any choice in the matter besides quitting your job or getting fired for being unable to travel and refusing to cooperate.
In a way, it's even worse than the usual form of corporate serfdom, because you can be required to run up literally thousands of dollars in travel expenses as a condition of remaining employed, but your employer can then turn around, declare bankruptcy without warning, and leave you personally on the hook for more money than you actually earned in salary going all the way back to day one. The media loves to talk about executives making $20,000 day trips by private jet to have a staff meeting at a vineyard in Napa... it never talks about staff members who were ordered to run up $2,000 in last-minute travel expenses the day before a mandatory "all hands" meeting on the other side of the country where they were told that they're getting laid off, and the company's lawyer is filing for bankruptcy at that very minute. It happens. (textbook example: Worldcom)
A *lot* of Worldcom employees got burned by that seeming technicality. They made business trips during the company's final days with their Worldcom Amex card, then when the company declared bankruptcy, all reimbursements were frozen, and there were rumors that employees might even have to pay back expenses that were previously reimbursed over the previous 5 months.
In the end, the court had mercy on the non-executive employees who were at risk of getting laid off with an unreimbursed $5-10k Amex bill to boot, but American Express made their lives absolutely *miserable* for months while the auditors scrutinized the bills to make sure the charges weren't "excessive" (and thus excluded from reimbursement) unless they had the cash to pay their bills out of pocket while waiting to get their expenses approved.
The real story here is that American Express, unlike Visa and Mastercard, rarely -- if ever -- sides with a cardholder over a merchant, even in situations where it's blatantly obvious that a charge is wrong. It's how they're able to get merchants to accept higher swipe fees and transaction charges.
For the most part, if a cardholder files a chargeback with Mastercard or Visa that looks even slightly reasonable, they'll freeze the funds from the merchant account within a matter of minutes. They might not issue the refund to the cardholder immediately while they're investigating if there's a dispute, but the burden at that point is overwhelmingly on the merchant to prove the charge was legitimate and correct. And if there's still any doubt, Mastercard or Visa will issue the refund, furnish documentation to the merchant, and tell them to sue you in small claims court if they think payment is owed.
In stark contrast, American Express will demand copies of the receipt from YOU (the customer), demand nothing from the merchant until they're 100% satisfied, and will still probably side with the merchant absent overwhelming evidence to the contrary.
In my life, I've had two chargeback disputes with Amex... and lost both. One was careless stupidity on my part (though still blatantly wrong), the other was Amex showing just how anti-cardholder they really are.
The first time it happened (circa 2000), the clerk accidentally hit '0' twice when entering the amount into the credit card terminal, and turned my $13.90 purchase into a $139.00 purchase. I had a receipt showing that the price was ~$13.25 with ~$0.65 in sales tax, but because I didn't notice until a month later & signed the receipt, Amex refused to budge. When I challenged American Express to have the merchant produce anything resembling an itemized receipt showing PRECISELY $139.00 in purchases with a timestamp within an hour of my alleged $139.00 purchase, they refused. As far as they were concerned, all that mattered was the 2x4 inch nearly-illegible receipt printed by the credit card terminal with something that resembled a blue smudge of a signature on it that I admitted to having (in addition to the itemized receipt for 1/10 the amount). I cancelled my card over it in rage, and refused to do business with them for about 10 years.
Fast forward a decade. My employer required me to get an Amex corporate card as a condition of getting reimbursed on business trips. One morning, the cashier at a Waffle House ran the card through, then somehow screwed up the machine between the moment the charge went through and the moment the printer produced the receipt. She insisted that the charge didn't go through. I argued with her for 3 minutes (I heard the printer start, before the power cord short or whatever rebooted it), then gave in because I didn't have much of a choice. Sure enough, I got charged twice, about 3 minutes apart, for the same amount. This time, I was sure Amex would take my side, because the restaurant obviously didn't have a signed receipt from me for the first one, and there was no sane reason why somebody would have two ~$6.00 charges approximately 3 minutes apart. Goddamn if they didn't do it to me again, and refused to reverse the first charge. Their official excuse? I didn't have an unsigned copy of the first receipt. What. The. FUCK. That really, REALLY pissed me off. Yeah, my boss signed off on the override and told me it wasn't worth fighting with them over $6, but it was the moral principle of it.
So, if you're an AMEX cardholder, be warned: if anything goes wrong, American Express WILL NOT take your side. In fact, they won't necessarily even lift a finger to make the merchant defend the charge. They'll press YOU for receipts, and will disqualify your chargeback on the slightest technicality, but will let the merchant get away with almost anything short of blatant, systematic intentional fraud.
Or, if you can find any actual company with a real product to sell, buy some 802.11y gear, pay the (theoretical) hundred-buck fee for a 10-year license to use the 3.6GHz spectrum, and do it with complete unambiguous legality.
Now, finding actual 802.11y hardware might be a bit of a challenge... the last time I looked, it was *still* commercially-nonexistent circa 2-3 months ago.
Were they literally brand new isolated segments, or were they new lines that extended (or at least directly shared a station with) existing lines? I'm talking about the American situation where a city with no meaningful transit to begin with builds its first rail line, at least half of which passes through areas that are industrial wastelands, dangerous slums, and/or just plain stupid places to build it (generally, surrounded by an ocean of single-family homes whose owners will fight any attempts at skyscraper-density redevelopment next door).
In America, skyscrapers next to the station are almost mandatory for viability, because Americans have very little tolerance for extended outdoor walks. If their destination isn't next door or across the street from the station, Americans generally won't take the train (however, transit systems can cheat a bit, and build underground air-conditioned corridors that psychologically extend the station's useful footprint, because "next door" is measured from the point where the rider has to actually go outside. The underground corridors at Crystal City (Arlington, VA) can get kind of creepy (it almost feels like you stumbled onto a movie set or into some area where you aren't really allowed, and you're going to get run down by security guards at any moment), but they DO serve the useful purpose of making a fairly large area feel like it's all adjacent to the station.
Washington. Atlanta. Miami. San Francisco.
All were mostly useless the day they opened, and were seriously limited in their usefulness for at least a decade... but now, roughly 30-40 years after the first segment opened, all four cities have at least gotten to the point where somebody who lives within walking distance of a station can find just about everything he or she needs within walking distance of another station, and depending upon the local job market and one's career field, has roughly 50-50 odds of being able to work within "viable" distance of a station.
More importantly, all four rail lines now have at least a few segments that are of sufficient importance that none of the cities would *dare* to take up an evil genie on his offer to refund the lines' original construction cost (not adjusted for inflation) in return for making them disappear forever. That's not to say that all four don't have examples of glaring mistakes that are painfully obvious today... but they each offer examples of the circumstances under which they can prosper and flourish in their respective markets (most of which can be summed up as, "embrace sprawl, proudly take ownership of it, and direct it into new suburban skyscraper clusters around outlying stations").
The problem isn't that LA needs wider freeways OR an expanded rail/subway network... LA needs wider freeways *AND* an expanded rail/subway network. People who argue for one to the exclusion of the other are missing the point.
A new rail line won't do a damn thing for gridlock *today* -- it takes a minimum of 25-40 years before a new rail line really starts to pull its own weight (40 when it's the first segment in a city, 25 when you're extending one that's already established). That doesn't mean the rail line shouldn't get built... it means the rail line should get built today, along with the freeway's reconstruction, so that 25-40 years from now, the rail line will have matured enough to start absorbing travel loads from the freeway, and it won't be necessary to demolish and rebuild the whole freeway yet again.
Verizon's bigger problem is that their implementation of EVDO is different from everyone else, so a phone from any other EVDO network on earth that's unlocked and made to work on Verizon will never be able to do better than 1xRTT. That's why Sprint/Metro/other CDMA phones roaming on Verizon can only do 1xRTT.
So, how many gigabytes per month can you use with an average data speed of 23kbps down (the actual real-world speeds I was getting from Sprint the day I finally left them in disgust), and 500kbps being cause for a standing ovation? OK, there might be 2 or 3 LTE towers somewhere within 10 miles, but right now, their LTE coverage is thinly-spread to a degree that makes T-Mobile's 2007 HSPA+ coverage look downright *saturated*. Sprint screwed up badly by doing a clean break from wimax, instead of using dualmode wimax-LTE chipsets for 2-3 years, deploying LTE into places without wimax, and THEN migrating the wimax to LTE. Instead, they took a relatively-functional wimax network, and pretty much threw it away the moment everyone bought a new phone last year. Just about everyone I know left Sprint after they got their new phone, and went from "ok" wimax that compensated for dysfunctional 3G/EVDO, to having no meaningful mobile data service at all because their 3G network was (and from what I've heard, pretty much still IS) in a state of dysfunctional meltdown.
I'm afraid to look at the rest of the specs to see which non-negotiable spec ended up getting gimped this time, since it seems to be a grand tradition among manufacturers that the closer they get to perfection, the more horribly they have to ruin at least one specific thing.
Please, sweet baby Jesus, fierce Xenu, and most noodly FSM, let this phone have a removable battery whose stock capacity exceeds 2500mAH, microSD, and a 5" display having a minimum of 1920x1080 resolution. With abundant ram and DVI (possibly via MLH), of course.
> Unless they really right the ship with Windows 9
(...) So far, their effort has been mostly a flop. Unless they really right the ship with Windows 9 by giving us back Aero Glass, or raise the bar several notches and give us Aero Glass Dynamically-Translucent Rotating Cube, with support for LCD touchpad [which defaults to a normal desktop experience unless you feel like slumming with phone/tablet apps, and simultaneously has sufficiently-high 'wow!' factor that doesn't interfere with usability AND gives consumers a compelling reason to buy new laptops with 3GHz+ quadcore i7 CPUs and discrete GPUs that would have been jaw-dropping on a desktop 2-4 years ago], they will shift from market dominance to just another vendor (...)
There, fixed that for you.
The problem isn't that computers are "good enough" now, the problem is that Windows 8 is butt ugly, and offers consumers nothing compelling to take their breath away and say, "Wow, that totally kicks ass!". Nobody is going to go buy a new computer that looks like a dumbed-down beta version of their previous one. If manufacturers want to fight the race to the bottom, they have to give consumers a reason to care about higher-end specs and put them to visibly-compelling good use. MetroModern just isn't going to sell high-end discrete GPUs and i7 CPUs. And frankly, kick-ass eye candy that doesn't impede usability is just about the only high-end hardware frontier left to fight Apple-ization and laminated .7mm slabs with the approximate usability of Ubuntu Unity.
Seriously. Microsoft needs to give laptop and desktop manufacturers a reason to sell thousand-dollar high-end laptops, and desktops with 3+ displays that will make people with lower-end hardware feel like poor second-class users who can't run the latest software (well) until they upgrade, not cut Windows down to something that can run on an anonymous shit one-gigahertz tablet from Shenzhen with a tiled window or two and soft keyboard.
There are unclean hands on both sides, and as usual it's the end users who get caught in the crossfire.
If Linux's kernel developers made even the most half-assed token effort at not wantonly breaking the ABI with every new release, or allowed drivers to have an intermediate thunking layer that stabilized the ABI for at least a year or two at a time in exchange for a little more overhead, the driver problem would largely fix itself and become a non-issue. The problem is, Linux's stewards *want* kernel-dependent binaries to catastrophically break with every release. It's an annoyance with desktop linux, a *real* problem with laptop linux, and the biggest single source of misery in Android guerrilla-AOSP land.
Ironically, much of the NVidia-side misery is caused by licensing terms with the same viral recursiveness as the GPL. If NVidia or AMD want to support playback of Blu-Ray content under Windows, or DRM'ed content under Android, they're required to agree to viral licensing terms that force them to implement and enforce things like Cinavia watermark detection everywhere, across the board... and prohibit them from talking about it or releasing the source to the general public. So they end up in a no-win situation... they aren't allowed to support playback of non-DRM'ed content under Linux, because it would be an easy way to bypass things like Cinavia, but they aren't allowed to release open-source drivers that pretend to care about it, because they're prohibited from disclosing details about how it works.
A federal law banning recursive scorched-earth viral licensing terms might fix the problem... but we all know that's not going to happen during our lifetimes, because Hollywood would freak out, and even a small Hollywood studio has more money & influence than the entire computer industry combined when it comes to Congress.
The next-best compromise might be if someone like NVidia were to start building mobile (and desktop) GPUs with a well-documented FPGA of sufficient capacity to implement h.264 decompression. Then, they could release official opensource drivers that simply ignored it and left it idle (since the same FPGA could arguably be used to implement a 2D sprite engine, or bitcoin mining, or whatever), and we could include our own code to hijack the FPGA and turn it in to a DRM-unencumbered h.264 decoding engine that might very well stomp all over 300 of MPEG-LA's patents and wantonly infringe upon them in ways that would make Hollywood seethe in rage, but as long as the source were hosted on a repo in Kazakhstan & distros like Ubuntu kept it at arm's length and didn't "officially" recognize it, Hollywood couldn't do a thing to stop end users from using the driver.
It depends. Linux tends to be nicer than Windows for USB, but still kind of falls flat on its face when it comes to GPU support, including HD h.264 video. Brute-force CPU decoding can only get you *so* far before the memory-bandwidth realities of trying to shovel around realtime 60fps 1920x1080 32-bit video data via PIO are going to bite you and cause problems.
Linux on laptops has gotten *enormously* better over the past few years, but it seems to consistently still have one set of problems with laptops that are "too new" (no drivers), but distros made for older computers seem to keep omitting support for stuff like PCMCIA/Cardbus that older laptops still depend upon to work properly.
I've gone through more than a bit of frustration over the past few few days trying to get Linux working properly on an old laptop I pulled out of the closet to resurrect (Compaq Armada M700... 700MHz PIII, 544mb ram). USB sucks on it (1.x only... though I've contemplated trying a Cardbus USB card), and most of the distros I've tried are either clueless about dealing with PCMCIA/Cardbus ATA devices, or freeze and require power-cycling when inserted (Anti-X being a notable example) .
Puppy (Wary 5.5, their distro for older hardware) seems to be the best so far, but even its Seamonkey browser tends to completely choke on web sites that go off the deep end with Ajax (like returning a blank page that references Javascript, then using that Javascript to fetch content via Ajax & build the entire DOM from scratch one node at a time). Still, I'm toying with stuffing a 16gb no-name PATA SSD from eBay into it to see whether it works some magic. I have a hunch, though, that the cheap PATA SSDs from China might really just be a microSD card with some glue logic to make them look like a CF card with 44-pin PATA interface, and not really be any real improvement over the 7200RPM drive that's in it already.
No. PalmOS in its pre-Cobalt form was fundamentally dead on arrival the moment realtime tcp/ip with a responsive UI became a big deal, just like MacOS9. A PalmOS phone was basically a Palm Pilot strapped onto a headless phone through a serial port. The phone subsystem was pretty much completely independent from the palm subsystem, sharing little more than an internal serial link and battery.
As fashionable as it is to hate Microsoft and gripe about how badly Windows 8 sucks, the fact IS that Windows Mobile WAS groundbreaking back in the early 00s. It might have been utterly dysfunctional out of the box as an operating system for making voice calls, but as the operating system for a pocket-sized laptop with wireless data capabilities, it picked up the ball where Palm dropped it and ran even harder. Had Microsoft left well enough alone, and reacted to Android by creating proper APIs for implementing alternative 'phone' and 'launcher/homescreen' thirdparty apps (instead of delegating the task to HTC and calling it a day, then later throwing the baby out with the bathwater so it could port Danger's OS from Java to C# and rebrand it as "Windows Phone"), WinMo8, 9, or 10 would have been a strong alternative to Android today instead of the crippled, unloved, locked-down joke we have now that's turned into a cancer destroying desktop Windows as well.
Lots of the things we take for granted in Android were "there" and worked fine in Windows Mobile 5/6, too... and more importantly (for patent purposes), did NOT work well AT ALL in PalmOS (if they worked at all), and barely worked in Android & IOS until 2010 and beyond. The biggest single problem high-end WinMo phones had was hardware -- US Carriers weren't in any hurry to push the envelope, and HTC was perfectly content to give them the minimum they asked for. And HTC made the ill-conceived decision to eliminate the 'windows' and 'ok' hardkeys in an effort to be more iPhone-like, without stopping to consider the fact that all of Microsoft's usability testing up to that point TOOK FOR GRANTED that the device would have two physical buttons that required at least a little bit of physical force to trigger (hence, the in-pocket touchscreen activations that caused endless misery if you got a text message or phone call that went straight to voicemail).
Anyway, the point is that once in a great while, Microsoft *does* manage to do something right, even if it completely drops the ball in other related areas. WinMo had plenty of warts, but circa 2005/2006, it WAS pretty much the best thing you could get if you wanted wireless internet connectivity in a device that could (sort of) limp along and make voice calls in a pinch. And it sure as HELL beat walking around with a Palm Vc or Handspring Visor and $129 18" cable to plug it into your clamshell phone for data a few years earlier, or limping with a later PalmOS phone that was good for making voice calls and managing an address book, but fell flat on its face the moment you tried doing anything that involved realtime network communication with a responsive UI (the UI froze whenever the phone was sending or receiving data due to the way PalmOS Garnet's network stack was stapled onto it as an afterthought).
Also, I believe a big chunk of Microsoft's patent portfolio came from its acquisition of Danger (the Sidekick's maker), which had plenty of its own innovations.
> Windows Key + Tab, it's better.
No-go on a Lexmark Model M13 with Trackpoint -- it doesn't *have* a Windows key. The Endura Pro has a Windows key, but uses an inferior pointer stick.
Because taskbar-pinning sucks. The taskbar is for minimized windows (and quicklaunch shortcuts for things like the command prompt), not ambiguously-running-not-running applications. When I close my applications, I want them to die and be metaphorically incinerated immediately by the garbage collector, not linger around as undead zombies consuming resources until the next time I reboot.
It utterly blows my mind that some people leave programs with embedded browsers churning away client-side Ajax running nonstop in the background, then wonder why their computer stutters and at least one of their cores is pegged. Or leave something like Word running, with 400 megs of cut & pasted bitmap data clogging the clipboard. Or Photoshop. Or Eclipse and the Android Emulator (shudder). Or 12 Firefox windows, half of which have more client-side Ajax churning away nonstop in the background, updating ads you aren't even looking at, leaking megabytes of allocated ram per hour, and pegging one or two more of your CPU cores for no good reason?
Taskbar-pinning was the *worst* part of Windows 7... but at least Microsoft had the decency to allow users to disable it without having to throw the baby out with the bathwater and give up Aero Glass, the start menu, and everything else, too.
The whole Metro-vs-Desktop debate reminds me of why everyone who's ever used a program developed for people whose life revolves around the use of that program is horrified by the thought of Microsoft ever acquiring ownership of it (programs like Dreamweaver, Photoshop, Chief Architect, EagleCAD, Blender, etc).
Microsoft truly doesn't *get* that hardcore users who interact with a piece of software all day have fundamentally different needs and expectations than a receptionist who wants to make a sign for the office party's dessert table using ComicSans and Nyancat clipart. I remember back when I used to do web design how completely fsck'ing tedious it was to edit tables with FrontPage, and how completely painless it was to do in Dreamweaver (once you learned how to do it). I shudder to think about what a Microsoft CAD program would be like.
With Microsoft-UI philosophy, you might be able to guess how something works in 12 seconds without reading the manual, but you'll feel like you're fighting with it for all eternity thereafter, because things that should be painlessly easy to do with a few minor gestures end up turning into 2 minutes of mouse-clicking and drilldown-traversal.
I remember the UI debate well from the Palm/WinMo era... the Palm philosophy was that if something is going to be a seamless, integrated part of your daily life for the next year, it's worth spending an hour learning stuff that will pay dividends that multiply in value over time. The Microsoft philosophy was that everything blindly had to work like XP and Windows, regardless of how tedious and absurd it made the process of using it.
> Knowing Microsoft, they'll probably release SP2 for Win7, which puts the "Modern" UI on top of it too,
> and then make SP2 a prerequisite for every security update that comes out after it.
And if they did, business school textbooks would have a new case study for corporate suicide, and a breathtaking example of how a company that managed to go from a dominant market share of the high-end mobile market to irrelevance within a matter of months was able to repeat it to throw away their desktop dominance as well (everyone had a major love-hate relationship with Windows Mobile, but if you wanted a pocket laptop with a useful browser that could be used for making voice calls in a pinch, WinMo was pretty much the best there WAS circa ~2007).
It would be the day I officially blew away Windows and promised God, Xenu, Thor, and the Flying Spaghetti Monster that I would never, EVER voluntarily run Windows as my real operating system again. And did my best to get everyone else whom I influence to do the same.
Put another way, for Microsoft to do something like that would constitute a full-frontal act of corporate warfare against its customers... and retribution from the consumers who matter would be swift, damning, and deadly. Look at the amount of hate Microsoft has taken from... well... everyone... over the past 6 months. Now imagine how much MORE hate they'd take if they loudly and proudly sank the lifeboat (Windows 7) that's keeping them alive right now. They'd have people burning computers on the sidewalk in front of their offices, hanging Ballmer in effigy, and Barnes & Noble would be filled with books about dumping Windows almost overnight.
A full frontal assault upon their customers would be the beginning of a rapid end for Microsoft. With their "influencer class" of users angrily gone, and thirdparty developers leaving in protest as well, Windows would degenerate into an inferior, second-rate OSX for consumers who buy a computer and use only the apps that were bundled with it.
Not necessarily. Realistically, any nuclear bomb detonated in an American city *today* would likely be ~5-10 kilotons, and most of the deaths and injuries would be due to panicked people running for their lives & getting hit by cars, injuries from wildlife or exposure to the elements if they literally "headed for the hills", or in the looting spree that would occur a few hours later.
That's not to say even a small bomb wouldn't be incredibly harmful... death is death regardless of whether it's due to nuclear incineration, radiation, a bear attacking people who fled and are hiding in the woods, freezing to death (winter attack), a rattlesnake bite, or a bad car wreck when the healthcare system is regionally-overwhelmed, and the federal government would probably spend tens of trillions of dollars on the cleanup (excavating a square-mile down to the bedrock and hauling the dirt 3,000 miles to a site in Nevada or Wyoming).
Regardless, anybody who lies awake at night worrying about cities getting multiply-nuked by 20+ megaton bombs can relax. The main reason they're still there is the perverse irony that we (in the US) are safer with the missiles pointing at us, but firmly under Russian government control and subject to inspection & auditing by the US, than we'd be if they were rapidly decommissioned. Decommission them *too* quickly, and there's a VERY real risk of uranium or bomb parts getting "lost" in the shuffle and ending up in the hands of people whom NEITHER the US *nor* Russia wants to have nuclear bombs. We're both gradually decommissioning missiles, and the main limiting factor is demand for their fissile uranium by nuclear power plants. The goal in both countries is for it to go straight from missile to power plant, with as little surplus floating around at any given time as possible.
>Without the ability to survive at that location there is no value gained in going there.
It's a good thing King Ferdinand & Queen Isabella didn't share your view. In a more recent context, I suspect there are quite a few people in Las Vegas and Dubai who'd beg to differ (not to mention Antarctica, undersea, the summit of Mt. Everest, and South Florida around August & September).
If you were exiled to what's now Las Vegas 500 years ago, and had to somehow live off the land without external supplies or support, how many days do you think you'd have survived under the best and most optimistic possible circumstances?
> So far PR is the only reason I have ever seen for sending humans.
And PR is what generates public (and ultimately political) support for NASA's budgets. If NASA decided to reinvent itself as an agency focused exclusively on pure science, and eliminated everything you might regard as "PR Fluff" from its mission, it wouldn't *have* a mission within 10 years. It would end up on the chopping block and get cut the next time there was any kind of budget crisis. Oh, ok... it might not get abolished and defunded *entirely*... it would probably get reassigned to some other agency, then get progressively starved to death until it was little more than a name with a director and a few university grants to manage per year.