> What's the difference between an Android tablet docked to mouse, keyboard and 1080p screen and a "bigger computer"?
Roughly speaking, "1 ARM Gigahertz" is approximately equivalent to "400-500 Pentium III Xeon (or Pentium M, or Core-non-Duo) Megahertz".
This means the highest-end Android tablet money can buy, equipped with bluetooth keyboard and mouse, is roughly comparable to a hypothetical 750MHz Pentium III notebook from 2003, retrofitted with a SSD. On such a computer (with conventional hard drive and 1 gig), Netbeans 4 or 5 took about a minute to load itself. Even back in 2004, when it was still a somewhat respectable laptop, it was painful.
Fast forward to 2012. Netbeans 7 takes about 15 seconds to load itself on a 3GHz quadcore i7 with SSD and 6 gigs of ram. Visual Studio 2010 takes about the same amount of time to go from "totally not loaded" to "launched and ready to do something". Extrapolate accordingly. Even at the highest end, modern Java development on a tablet would be excruciatingly painful compared to even a low-end desktop PC. Yeah, it could be done... but it wouldn't be pleasant, and if you're paying a software engineer's salary, it would be a criminally inefficient waste of his time absent some compelling circumstance that required in the field software development without even a decent laptop.
Tablets lend themselves well to in-the-field tweaking of things like Python. They're not suitable for things like developing real heavyweight applications like "Angry Birds" or an Android/IOS Twitter client. The fact that you could theoretically do it doesn't mean it's a good idea or an efficient use of your time.
No, "thermonuclear" would be Google mustering all of its available resources to get Android ported to the iPhone 5 within a month of its release, complete with "Nexus A" boot animation featuring an Android eating an Apple and pooping out a phone whose display expands to fill the whole screen at the end of the animation.
Apple would still be selling hardware, of course... a lot of it, in fact... but Google would be skimming 100% of the "Market" revenue and "Maps ads" revenue from any iPhone 5 reflashed to Android. Apple is probably selling iPhones at a small loss, with the expectation of making back the difference from other fees. Imagine if they found themselves in the position of Sony -- selling devices at a small loss like hotcakes, knowing they'd never see another cent from those same devices once they left their hands.
The goal, of course, wouldn't be to just make iPhones a first-rate Android hardware platform -- it would be to goad Apple into locking down new iPhones against jailbreaking and reflashing really, REALLY HARD. As in, "Motorola hard". Instantly, Apple would alienate their most influential and enthusiastic group of hardcore users, and drive them away from the platform. Consumers would scramble to buy new old stock iPhones that weren't locked down, and angrily return the ones that ended up being locked down anyway. More consumer ire.
Keep in mind, Google would lose nothing from this. A phone running Android is a phone running Android, insofar as Google revenue is concerned. It would be a bit of a gamble, because it would obviously horrify Samsung and HTC (who might, or might not, buy into the logic of using the move to force Apple's hand and goad them into totally locking down IOS devices against reflashing unapproved firmware). Ultimately, though, this isn't about money for Apple -- it's about control. If it were just about money, Apple would shake down Samsung for royalties and move along to HTC & LG. I can't think of any single act of guerrilla terrorism Google could do to Apple that would more effectively undermine the control Apple is determined to exercise than porting Android to Apple's newest and best hardware.
> Who's in the enterprise world using Java 1.7 anyway?
Enterprise applications requiring Java 7 are rare. Enterprise applications requiring Java 6 or better are not.
Unfortunately, Java 6 doesn't exist for OS X (ie, Macintosh). Java 7 is the first real version of Java Mac users have had in literally *years*.
For Mac users, the next step down from Java 7 isn't Java 6... it's Apple's broken, obsolete, Steve-shackled Java 5. If a Mac user wants to run Netbeans 7, in particular, he has exactly two choices: install Java 7, or run it under Windows or Linux using VMware.
I'm sure Apple will try to patent this, but you heard it here first: what Android needs is the ability for users who don't want to actually *use* wi-fi for some reason to nevertheless allow it to be turned on long enough to listen for nearby SSIDs, and possibly send out a round of "are you there?" inquiries to access points that are on the phone's list of known, but non-broadcasting, SSIDs. There are plenty of times I might not want to actually USE wi-fi for some reason, but nevertheless wouldn't mind if apps could semi-passively use it to "check out the neighborhood" and "see which access points are nearby".
Ditto, for GPS. The power-intensive part of GPS (once you get the initial fix, and a few hours thereafter) isn't the telemetry-sampling, it's the math you have to do to get a fix based upon it. What Android needs is the ability to periodically turn on the GPS radio and sample GPS telemetry (becoming more and more aggressive when the accelerometers indicate motion and it's been a while since the last successful reading), but just buffer it in case something needs a GPS fix within the next minute or two. If something wants a location fix, THEN do the math to calculate the location fix. Otherwise, just sniff the air and log the raw telemetry data in a ring buffer for a few minutes, or until the next successful reading.
Another thing Android desperately needs: a setting that tells it, "Connect to these access points when available, but don't send a single byte of data once you do until (and UNLESS) you've ALSO connected to ${this_vpn} via that access point... and terminate the wi-fi session without sending further data over it if you can't. I have a $3/month VPN account I use with public wi-fi, but I hate, Hate, *HATE* the fact that Android sends out a torrent of pent-up http requests the *nanosecond* the wi-fi connection comes up, and there's no way to stop it. Not even Tasker can suppress them until it brings up the VPN... assuming it's even able to bring up the VPN. Somewhere along the line, newer versions of Android became hellbent on forcing users to password-protect their phones in order to store VPN credentials, and apparently at one point Google intentionally removed the ability to launch a VPN via intent in a way that conveyed the password through the intent (to make it impossible to auto-launch and auto-login to a VPN, and force users to use Android's built-in password storage instead, which forces you to enable additional annoying layers of security you might not want if your only goal is to keep others at McDonalds or Starbucks from casually slurping your non-https background Facebook logins by hiding them in a VPN stream).
Let's not go overboard here. Microsoft is far from being saintly, but at the end of the day, Microsoft is largely indifferent to the existence of desktop Linux.
Microsoft views desktop Linux kind of like the tiny black ants you see walking behind the toilet after a week of thunderstorms. You could go get the ant spray and wipe them out, but then the second floor will smell like ant spray for the rest of the day, and it would mean having to go downstairs, hunt for the ant spray, go back upstairs, and use it.
More importantly, from Microsoft's world view, half the computers running Linux have a paid OEM license for some version of Windows anyway, and the other half are owned by people who, if you backed them up against a wall and forced them at gunpoint to give up using Linux, would buy a Mac. If Linux annoys Microsoft, it's only because Microsoft is forced to stop and find some way to accommodate it.
Server Linux? Yes, that annoys Microsoft. Android? Hell yeah, that annoys Microsoft even more. But Desktop Linux? Meh. Barely even on the radar. Now, if somebody fights back against Metro by porting KDE or Gnome to Windows 9... well, THEN Microsoft might get really annoyed and notice...
From what I've read over at XDA, Chinese Rockchip-based media players and tablets are even less open, documented, or hackable than Broadcomm's stuff. Even if you can get documentation and/or source, the circuit boards have no development or test or test headers (or use weird, undocumented serialized Jtag-ish protocols), and you can't even get at non-exposed pins on the chip because the chip itself is an epoxied blob on the circuit board. They design it, get it to "sort of" work, crank out 20 million copies for pennies apiece, sell them for next to nothing, and just expect users to toss them and buy a new one next year. They literally have no "Plan B" to fix bugs or allow existing items to evolve after manufacture.
The problem is that the throw-away hardware is SO MUCH cheaper, it ends up driving better hardware off the market because even slightly-better designs end up costing 5-20 times as much money to buy. What happens is, the crap hardware gets 100% of the "economy of scale" benefit because stores like Walmart buy it exclusively, and the better hardware rapidly becomes an exotic botique item with zero economy of scale. In the race to the bottom, the crap hardware gets its capabilities squeezed to the point where it can't even be repurposed and integrated into better designs as a component of more-capable hardware.
If you're doing it with non-realtime content as a form of public performance art of finite duration, it's fairly straightforward: stream it slowly in advance, with each phone buffering its individual pixel value for each frame along with a timecode, then use the network to just transmit the clock & timecodes and have the phones step through their pre-buffered values on schedule.
It's kind of like an orchestra with a director -- unless they're all spontaneously improvising jazz, they have sheet music in front of them that was given to them (and well-practiced) long before the actual performance. The director isn't communicating the note and duration to each individual musician in realtime -- he's just conveying timing info to keep them all in sync.
At least wireless companies in Britain haven't started soldering embedded SIM cards to the circuit board to force users to pay criminally overpriced international roaming charges from (*cough*) "strategic global roaming partners" when their customers travel overseas, instead of buying a prepaid SIM from a local network.
Sadly, this isn't an artificial, contrived example. Sprint did it to their new "world" phone, the Motorola Photon Q. Apparently, Verizon is chomping at the bit to start doing the same. When I first read about it, all I could think of was the quote from 1984 about the boot stepping on a face.
Are you *sure* it was that low? I don't remember it *ever* being below 75 or 80 Canadian cents per US dollar in the 30 or so years I've been old enough to know or care.
> and how many real laptops have serial ports these days?
Don't remind me. Two weeks ago, I spent the better part of a day troubleshooting a nonexistent bug in an embedded hardware project that was caused by a malfunctioning USB virtual fucking serial port on the computer I was using to view the debugging info I was outputting to the MCU's serial port.
Now I know why so many guys have built their own serial terminals with a 4x20 LCD and a microcontroller. It's the only way left to reliably and infallably render serial input straight to a display without risking stupid errors caused by Windows or its USB subsystem.
When you're dealing with bare metal embedded hardware, USB anywhere along the signal chain really sucks.
As I understand it, most mainstream Linux distros (Ubuntu, Debian, and CentOS among them) have rendered "text mode" as bitmaps for years, probably a decade or more. I think the motive was partly to get proper support for UTF8, and partly so Linux could gracefully deal with use cases where a user runs one app that requres something like ISO-8859-1, then runs another app that uses Russian Cyrillic (ISO-8859-2?), and both are within scrollback distance and have output that could conceivably be visible to the user within a single window at the same time.
Side note: the reason I found this out is because I was wondering how Linux magically managed to render bitmap penguins onto a textmode display. I originally thought that somebody figured out how to bitbang a VGA card into changing modes mid-scanline a-la-c64/AppleII/Atari/Amiga through pure software-timed brute force. Apparently, though, even if you ignore the BIOS and try writing directly to the hardware registers for a specific known VGA chip, there's some hardwired implementation-level reason why it wouldn't work (at least, not on any video card from the past century). Anyway, that's the story of how I found out that most/all modern Linux distros actually fake character mode with a frame buffer.
> So why is downloaded software marked up by similar or greater amounts?
My guess? Historically, AUD$1 == US$0.50, and quite a few people haven't realized yet that they're now more or less at parity. As a result, I suspect quite a few American companies are just doubling the nominal US price out of habit.
~15 years ago (circa 1996 or so), touchpads emulated the kinetics of thumb-manipulated trackballs, and I didn't mind them too badly. Then, sometime around 1999, all the manufacturers started to de-tune their virtual kinetics to accommodate users who used their index fingers. The difference isn't subtle. Your thumb's range of motion is limited, and the first-gen touchpads used the same algorithms as trackballs to interpret a motion that can loosely be described as "kind of towards the left, somewhat downward" as "move the pointer directly left" (and vice-versa).
Once in a blue moon, I'll trip over a modern touchpad that *almost* gets it right, but manufacturers view them as commodities, so even when I find one in real life that doesn't suck, there's no guarantee that the identical laptop ordered from Newegg will have the same one.
I have a hunch that the touchpads in laptops can do more than the generic (or official) drivers offer, but low-level documentation for them is nonexistent (in the public space, at least). I still spend an occasional afternoon looking at Linux source hoping that someone has leaked a driver that gives a determined driver the ability to tweak it in nonstandard ways.
Insofar as "multitouch" goes, the idea is actually nothing new. Users had the idea 10+ years ago. The whole problem was with the touchpad firmware itself, which reported only a single x,y location (or a single velocity vector). All you need to implement "pinch type" multitouch is a touchpad that reports raw events, and does it quickly enough to preserve the trigger order. If I see "finger down at (x,y)", then "finger down at (a or u, b or w)", it doesn't take a rocket scientist to put 2 and 2 together and figure out that the values closest to x,y probably are one point, and the remaining two values are another. From that point, it's just a matter of tracking them quickly enough to ensure that the jump from sample to sample doesn't exceed the difference between the two (so you can keep track of which value is which over time).
Also, multitouch touchpads that support 3+ fingers aren't necessarily able to support completely arbitrary touches, or touches that are truly simultaneous. For example, supporting a "3 fingers swiped vertically" gesture is one thing... supporting "3 fingers moving with completely independent motion along paths that cross" is another mater entirely. A "3 fingers to scroll" gesture would be implementable even with old touchpads with simple matrices, as long as they report ALL contact points instead of just the first or most dominant pair. You'd just have to notice that there are 3 different x values that differ by approximately N units, and notice that the 3 x values all have y values that are similar & changing at a similar rate.
I own two of them, but the mouse buttons on one are completely nonworking, and the mouse buttons on the other are so flaky I barely even try to use them anymore. The thing is, the buttons NEVER really worked well... they just kind of went from "annoyingly flaky" to "barely-working, or not working at all."
Can anybody actually fix them in a way that not only fixes my problem, but hopefully fixes it in a way that improves upon the original buttons? I love the keyboards (even though most of my coworkers hate the one I have at the office), but I really wish I could get the buttons fixed.
That said, I really wish somebody would make a keyboard like the M13 with Trackpoint, but give it a SECOND Trackpoint located directly below the spacebar, and configure it so that one can be used for pointer movement, and one can be used for scrolling & panning. Fujitsu got the location right with THEIR pointer, but their mechanism sucks. IBM got the mechanism right, but "GHB" is a poor location for fine-manipulation (your thumb is stronger, but has less range of motion... precisely the use case ideal for pointer sticks). I've seen a few Sony netbooks over the years that put the stick below the spacebar, but I can count them on one hand. I wish more companies would try putting sticks there (or hedging their bets and putting two sticks in place, so users can pick one for mouse and one for scrolling).
Not to mention the ability to charge for different levels of performance using the same underlying hardware, kind of like ATI & Nvidia do, and Intel was planning to do with their value-priced CPUs.
Here's an easy way to tell whether they're doing it for "safety", or just to increase their own profits -- if they give copies of the security key to end users, their motives are probably good. If they won't even give the code to mechanics, and force field replacement of expensive parts that could be repaired if the mechanic had the code, then they're doing it for their own benefit. It's just like UEFI. If I have a copy of the key, it's awesome. If the only copy of my key is held by Microsoft or Sony, it's a shameless pwnage of my consumer rights whose physical and political defeat is a moral imperative.
You didn't get the memo? iPhone 5w. "w", as in "Woz Edition".
Rumored features include
* Zigbee wireless mesh networking
* multiplexed pins on the headphone jack that can be repurposed for I2C, SPI, or GPIO (not at the same time, obviously). Oh, and legacy UART (3.3v logic) that also supports Atmel-friendly baudrates (125kbit, 250kbit, 1mbit), 9-bit word length (9/N/1, to be exact), and can use pin #3 for GPIO, RTS, CTS, or synchronous clocking in or out.
* fixed 4800mAH lithium cell
* gamepad wings that slide out from the underside (to the left and right when held in landscape orientation). One side has an analog stick & digital stick, the other side has an analog stick & 4 buttons.
* Volume button pair that also serves as the "left trigger button" in "gamepad" orientation, and 2-stage camera button that serves as the "right trigger button" in gamepad orientation. All of which can be intercepted, redefined, and repurposed by end users as they see fit.
The iPhone 5w's flagship applications will be a logic analyzer/DSO, which demonstrates the use of the bundled iProbe (4 clip-on leads, terminating at a 1/8" TRRRS headphone jack). Additional accessories will allow connectivity to most car ECUs, CANbus, and JTAG.
Oh, and the phone will also include a fully-unlocked bootloader. Of course, not even Woz will likely be able to get the Powers that Be at Apple to release the source to their crown jewel, but it won't matter. People will buy one, and reflash it to Android. Even Larry & Sergey will be spotted in public with it (running Android, of course).
The problem with that argument is that one might legitimately argue that if there's a safety issue that's mitigated through software, the need for modified firmware to do the same needs to be communicated. 99.9% of damage and hard-bricking is caused by either booby traps left by the manufacturer to trap the unwary, or important details like "always monitor the battery temperature, and back off if it exceeds N degrtees Celsuis" that aren't disclosed.
I'm happy to see that Moto is finally starting to become non-Evil under the ownership of Google. I'm disappointed as hell by the fact that the Photon Q was totally gimped by Sprint and/or Motorola. This would be a great phone if it were in a blister pack at Wal Mart being sold for use with some value-priced prepay network. It's not, however, a top-tier best of breed flagship Android phone:
* 540x960 qHD. WTF?!? 540x960? And not even OLED? The Q's display is a decisive step down from the Photon 4G. Note to Sprint & Moto: the next generation of any phone is supposed to AT LEAST as good as what it replaced, especially now that we're going to be stuck with the damn phone for almost two painfully long years thanks to last year's abolition of 12-month upgrades.
* tiny battery that can't even be swapped when it dies halfway through Friday night. If they'd put a huge battery inside like the one in the Razr Max, it might have been tolerable. But 1785mAH? You can't be fucking serious. I'd literally burn through that in 3 hours.
* Half the flash of its predecessor. WTF. Read the note above about how successor phones are supposed to be a step up, not down.
* Nonremovable SIM. Yay, it can roam on GSM in other countries... except at $2.50/minute, nobody is going to actually DO it knowingly and voluntarily once they find out how badly Sprint is going to rape them for doing it. Overall, this is just kind of like Sprint turning around and giving us a final kick in the balls, just for good measure.
It's been years since I've actually left the US, but the sealed-SIM anti-feature ALONE is enough to make me want to leave Sprint, because it demonstrates total and complete contempt towards us. I mean, really... once you factor in the administrative cost of roaming, and fighting with livid customers who just got a bill for $900 in roaming charges after spending 3 days in Montreal, how much extra is Sprint *really* going to make compared to what they would have not made by just giving it a normal SIM slot and charging a $35-50 one-time admin fee to unlock the SIM lock during the first year of a contract? Sprint could even offer an olive branch to deflect criticism and waive the fee if the customer puts down a $400 deposit that gets returned after the contract's 12th month (knowing that 99.9% of customers would just say 'fuck it' and pay the $35-50).
It's a shame, because I love Motorola's build quality and superior radios. But the harsh fact is, I'm going to be stuck with my next phone for 20 long, painful months now that Sprint has taken away our annual upgrades, so my next phone has to be damn near flawless & one I'm sure I'll be able to live with. The Photon Q is not that phone. I can only pray to the flying spaghetti monster that 18 months from now, Sprint + Moto will be unleashing an unlocked Nexus device with specs to die for.
They take advantage of human nature, which tends to underestimate personal risk and overestimate one's likelihood of successfully avoiding bad outcomes. Often, the individuals who are the most vehement in their condemnation of "bad drivers" are themselves some of the worst drivers out there.
Call it narcissism, delusion, or simply "an unbelievably creative ability to rationalize any point of view regardless of objective facts", but this is basically a way for insurers to identify a small (but expensive in terms of payouts) group of risk-takers that depends mostly upon that same group's willingness to roll the dice with their insurance company and risk higher rates.
> Driving without insurance in the UK will get your car seized and crushed.
Maybe Britain is radically different from the US in this regard, but I'd be shocked if any such law weren't written in a way that made it impossible to actually DO that to any car that was secured by a bank loan, as opposed to a car that was paid off and driven by an uninsured owner. At the *very* least, the law would be written with "safe harbor" provisions that exempted any car used to secure a loan as long as the lender made a good-faith effort to enforce compliance (regardless of whether or not the lender's efforts were actually successful and effective).
Not to mention the fact that it would instantly burn any chances China might have of cooperating with the US & Russia on any international space ventures going forward, something they want to do quite badly.
Yes, we know that "Chinese Hacker" != "Chinese Government", but it would set back the cause of Chinese-US-Russian space relations by decades if it were "merely" independent hackers, and really WOULD set it back forever and for all eternity if it were proven to have been in any way, shape, or form an official program of China's government. This brings us to one logical conclusion:
If you're Chinese, and you humiliate China & harm its reputation with NASA & Russia's space program, you'll be ruthlessly hunted down and executed at the first possible opportunity on live TV in a grand public relations exercise ordered by the highest ranks of the CCP. NASA will probably be horrified, and Russia will probably be unimpressed & see right through it, but the end result is that you'd have to be completely suicidal to be Chinese and even *try* hacking American or Russian space hardware. It's one thing to deface the National Parks Service website for Kennedy Space Center (or whomever runs it as a tourist attraction). It's another matter entirely to brick a piece of hardware that cost 1.000 times as much as you and your ancestors going all the way back to your oldest homo sapiens ancestor have collectively earned since the dawn of human history.
Would Al Qaeda (or someone influenced/inspired by it) try to do it if they could? Maybe. China? No.
I rarely see problems involving new releases of Java... with one specific exception: 64-bit Java. I've seen more than a few apps that die horrible deaths with 64-bit Java... and almost exactly the same number that specifically *require* 64-bit Java when running under 64-bit Windows.
I'm still trying to figure out what, exactly, causes some random Java app to specifically require one or the other when running under Win7/64. I suspect it has something to do with changes Microsoft made to 64-bit Windows that forced (or at least induced) Sun, then Oracle, to change something major with 64-bit Java... but I've never seen any white paper or article that identifies and explores the specific reasons why this might be so. I'm sure the mass layoffs and exodus of Sun's employees right around the time 64-bit Java became relevant, and 64-bit Windows became mainstream, didn't help.
But then, when you go to log into a site with wacky rules that requires a special password using a different computer/tablet/phone, how does it KNOW it has to use the alternate password scheme?
The LastPass keyboard is a nice idea, but AFAIK, Android doesn't allow on-the-fly keyboard switching (you have to launch settings, navigate to 'keyboard and input', set a new default keyboard, then start over). I use Graffiti for everything (I'm crippled without it), so I'd still be SOL. So would somebody who uses Swype, a split tablet keyboard, or even the funky keyboard whose name eludes me that has you compose English the way Koreans compose Hangul (ex: 'd' = 'c' + 'l').
Probably for a few years. I think you can STILL buy "New Old Stock" retail copies of XP Pro on Amazon. The copies that will dry up quickly are the $129 OEM copies of Win7 Pro. If you want Win7Pro 2 years from now, you'll probably have to buy a full-priced retail copy unless a home edition is good enough for you.
I'm pretty sure Microsoft allows distributors and volume customers to return unsold oem copies and exchange them for the new version for a nominal fee, if not free, because OEM copies always seem to dry up within weeks of any new version's release. Even XP dried up, before Microsoft backed down a few months later & allowed distributors to buy it again.
Can someone honestly remind me why so many people hated Win2k? To this day, I still remember it as one of the best versions of Windows, ever. I could upgrade or install MAJOR drivers & system-level components, then go days or weeks without having to reboot.
Maybe it's because NT4 already weeded out most of my software and hardware that Win2k would have objected to, so it ended up being a net improvement in every meaningful way, but it honestly wasn't until I got my first dual-CPU motherboard that I really felt any need to 'upgrade' to XP... then had a MAJOR love-hate relationship with it. Hugely-improved SMP support? Major plus. Endless reboots? Yuck. I reinstalled without SP1, and vehemently resisted SP1 for more than a year until Microsoft slipped another patch past me that basically bundled everything about SP1 I was trying to avoid, and instantly took away my ability to do reboot-free updates once and for all. RIP. Sigh.
Sadly, even Linux now seems to try and force reboots by default now for some updates. Oh, you can usually open a shell and force it to do a hot update, or ignore the warnings and it'll work anyway, but more and more, it feels like even Linux has abandoned the ideal of "never force a reboot". and instead embraced Microsoft's philosophy of, "If something's wrong, reboot and it'll probably fix it."
Slashdot Mirror
> What's the difference between an Android tablet docked to mouse, keyboard and 1080p screen and a "bigger computer"?
Roughly speaking, "1 ARM Gigahertz" is approximately equivalent to "400-500 Pentium III Xeon (or Pentium M, or Core-non-Duo) Megahertz".
This means the highest-end Android tablet money can buy, equipped with bluetooth keyboard and mouse, is roughly comparable to a hypothetical 750MHz Pentium III notebook from 2003, retrofitted with a SSD. On such a computer (with conventional hard drive and 1 gig), Netbeans 4 or 5 took about a minute to load itself. Even back in 2004, when it was still a somewhat respectable laptop, it was painful.
Fast forward to 2012. Netbeans 7 takes about 15 seconds to load itself on a 3GHz quadcore i7 with SSD and 6 gigs of ram. Visual Studio 2010 takes about the same amount of time to go from "totally not loaded" to "launched and ready to do something". Extrapolate accordingly. Even at the highest end, modern Java development on a tablet would be excruciatingly painful compared to even a low-end desktop PC. Yeah, it could be done... but it wouldn't be pleasant, and if you're paying a software engineer's salary, it would be a criminally inefficient waste of his time absent some compelling circumstance that required in the field software development without even a decent laptop.
Tablets lend themselves well to in-the-field tweaking of things like Python. They're not suitable for things like developing real heavyweight applications like "Angry Birds" or an Android/IOS Twitter client. The fact that you could theoretically do it doesn't mean it's a good idea or an efficient use of your time.
No, "thermonuclear" would be Google mustering all of its available resources to get Android ported to the iPhone 5 within a month of its release, complete with "Nexus A" boot animation featuring an Android eating an Apple and pooping out a phone whose display expands to fill the whole screen at the end of the animation.
Apple would still be selling hardware, of course... a lot of it, in fact... but Google would be skimming 100% of the "Market" revenue and "Maps ads" revenue from any iPhone 5 reflashed to Android. Apple is probably selling iPhones at a small loss, with the expectation of making back the difference from other fees. Imagine if they found themselves in the position of Sony -- selling devices at a small loss like hotcakes, knowing they'd never see another cent from those same devices once they left their hands.
The goal, of course, wouldn't be to just make iPhones a first-rate Android hardware platform -- it would be to goad Apple into locking down new iPhones against jailbreaking and reflashing really, REALLY HARD. As in, "Motorola hard". Instantly, Apple would alienate their most influential and enthusiastic group of hardcore users, and drive them away from the platform. Consumers would scramble to buy new old stock iPhones that weren't locked down, and angrily return the ones that ended up being locked down anyway. More consumer ire.
Keep in mind, Google would lose nothing from this. A phone running Android is a phone running Android, insofar as Google revenue is concerned. It would be a bit of a gamble, because it would obviously horrify Samsung and HTC (who might, or might not, buy into the logic of using the move to force Apple's hand and goad them into totally locking down IOS devices against reflashing unapproved firmware). Ultimately, though, this isn't about money for Apple -- it's about control. If it were just about money, Apple would shake down Samsung for royalties and move along to HTC & LG. I can't think of any single act of guerrilla terrorism Google could do to Apple that would more effectively undermine the control Apple is determined to exercise than porting Android to Apple's newest and best hardware.
> Who's in the enterprise world using Java 1.7 anyway?
Enterprise applications requiring Java 7 are rare. Enterprise applications requiring Java 6 or better are not.
Unfortunately, Java 6 doesn't exist for OS X (ie, Macintosh). Java 7 is the first real version of Java Mac users have had in literally *years*.
For Mac users, the next step down from Java 7 isn't Java 6... it's Apple's broken, obsolete, Steve-shackled Java 5. If a Mac user wants to run Netbeans 7, in particular, he has exactly two choices: install Java 7, or run it under Windows or Linux using VMware.
I'm sure Apple will try to patent this, but you heard it here first: what Android needs is the ability for users who don't want to actually *use* wi-fi for some reason to nevertheless allow it to be turned on long enough to listen for nearby SSIDs, and possibly send out a round of "are you there?" inquiries to access points that are on the phone's list of known, but non-broadcasting, SSIDs. There are plenty of times I might not want to actually USE wi-fi for some reason, but nevertheless wouldn't mind if apps could semi-passively use it to "check out the neighborhood" and "see which access points are nearby".
Ditto, for GPS. The power-intensive part of GPS (once you get the initial fix, and a few hours thereafter) isn't the telemetry-sampling, it's the math you have to do to get a fix based upon it. What Android needs is the ability to periodically turn on the GPS radio and sample GPS telemetry (becoming more and more aggressive when the accelerometers indicate motion and it's been a while since the last successful reading), but just buffer it in case something needs a GPS fix within the next minute or two. If something wants a location fix, THEN do the math to calculate the location fix. Otherwise, just sniff the air and log the raw telemetry data in a ring buffer for a few minutes, or until the next successful reading.
Another thing Android desperately needs: a setting that tells it, "Connect to these access points when available, but don't send a single byte of data once you do until (and UNLESS) you've ALSO connected to ${this_vpn} via that access point... and terminate the wi-fi session without sending further data over it if you can't. I have a $3/month VPN account I use with public wi-fi, but I hate, Hate, *HATE* the fact that Android sends out a torrent of pent-up http requests the *nanosecond* the wi-fi connection comes up, and there's no way to stop it. Not even Tasker can suppress them until it brings up the VPN... assuming it's even able to bring up the VPN. Somewhere along the line, newer versions of Android became hellbent on forcing users to password-protect their phones in order to store VPN credentials, and apparently at one point Google intentionally removed the ability to launch a VPN via intent in a way that conveyed the password through the intent (to make it impossible to auto-launch and auto-login to a VPN, and force users to use Android's built-in password storage instead, which forces you to enable additional annoying layers of security you might not want if your only goal is to keep others at McDonalds or Starbucks from casually slurping your non-https background Facebook logins by hiding them in a VPN stream).
Let's not go overboard here. Microsoft is far from being saintly, but at the end of the day, Microsoft is largely indifferent to the existence of desktop Linux.
Microsoft views desktop Linux kind of like the tiny black ants you see walking behind the toilet after a week of thunderstorms. You could go get the ant spray and wipe them out, but then the second floor will smell like ant spray for the rest of the day, and it would mean having to go downstairs, hunt for the ant spray, go back upstairs, and use it.
More importantly, from Microsoft's world view, half the computers running Linux have a paid OEM license for some version of Windows anyway, and the other half are owned by people who, if you backed them up against a wall and forced them at gunpoint to give up using Linux, would buy a Mac. If Linux annoys Microsoft, it's only because Microsoft is forced to stop and find some way to accommodate it.
Server Linux? Yes, that annoys Microsoft. Android? Hell yeah, that annoys Microsoft even more. But Desktop Linux? Meh. Barely even on the radar. Now, if somebody fights back against Metro by porting KDE or Gnome to Windows 9... well, THEN Microsoft might get really annoyed and notice...
From what I've read over at XDA, Chinese Rockchip-based media players and tablets are even less open, documented, or hackable than Broadcomm's stuff. Even if you can get documentation and/or source, the circuit boards have no development or test or test headers (or use weird, undocumented serialized Jtag-ish protocols), and you can't even get at non-exposed pins on the chip because the chip itself is an epoxied blob on the circuit board. They design it, get it to "sort of" work, crank out 20 million copies for pennies apiece, sell them for next to nothing, and just expect users to toss them and buy a new one next year. They literally have no "Plan B" to fix bugs or allow existing items to evolve after manufacture.
The problem is that the throw-away hardware is SO MUCH cheaper, it ends up driving better hardware off the market because even slightly-better designs end up costing 5-20 times as much money to buy. What happens is, the crap hardware gets 100% of the "economy of scale" benefit because stores like Walmart buy it exclusively, and the better hardware rapidly becomes an exotic botique item with zero economy of scale. In the race to the bottom, the crap hardware gets its capabilities squeezed to the point where it can't even be repurposed and integrated into better designs as a component of more-capable hardware.
If you're doing it with non-realtime content as a form of public performance art of finite duration, it's fairly straightforward: stream it slowly in advance, with each phone buffering its individual pixel value for each frame along with a timecode, then use the network to just transmit the clock & timecodes and have the phones step through their pre-buffered values on schedule.
It's kind of like an orchestra with a director -- unless they're all spontaneously improvising jazz, they have sheet music in front of them that was given to them (and well-practiced) long before the actual performance. The director isn't communicating the note and duration to each individual musician in realtime -- he's just conveying timing info to keep them all in sync.
At least wireless companies in Britain haven't started soldering embedded SIM cards to the circuit board to force users to pay criminally overpriced international roaming charges from (*cough*) "strategic global roaming partners" when their customers travel overseas, instead of buying a prepaid SIM from a local network.
Sadly, this isn't an artificial, contrived example. Sprint did it to their new "world" phone, the Motorola Photon Q. Apparently, Verizon is chomping at the bit to start doing the same. When I first read about it, all I could think of was the quote from 1984 about the boot stepping on a face.
Are you *sure* it was that low? I don't remember it *ever* being below 75 or 80 Canadian cents per US dollar in the 30 or so years I've been old enough to know or care.
> and how many real laptops have serial ports these days?
Don't remind me. Two weeks ago, I spent the better part of a day troubleshooting a nonexistent bug in an embedded hardware project that was caused by a malfunctioning USB virtual fucking serial port on the computer I was using to view the debugging info I was outputting to the MCU's serial port.
Now I know why so many guys have built their own serial terminals with a 4x20 LCD and a microcontroller. It's the only way left to reliably and infallably render serial input straight to a display without risking stupid errors caused by Windows or its USB subsystem.
When you're dealing with bare metal embedded hardware, USB anywhere along the signal chain really sucks.
As I understand it, most mainstream Linux distros (Ubuntu, Debian, and CentOS among them) have rendered "text mode" as bitmaps for years, probably a decade or more. I think the motive was partly to get proper support for UTF8, and partly so Linux could gracefully deal with use cases where a user runs one app that requres something like ISO-8859-1, then runs another app that uses Russian Cyrillic (ISO-8859-2?), and both are within scrollback distance and have output that could conceivably be visible to the user within a single window at the same time.
Side note: the reason I found this out is because I was wondering how Linux magically managed to render bitmap penguins onto a textmode display. I originally thought that somebody figured out how to bitbang a VGA card into changing modes mid-scanline a-la-c64/AppleII/Atari/Amiga through pure software-timed brute force. Apparently, though, even if you ignore the BIOS and try writing directly to the hardware registers for a specific known VGA chip, there's some hardwired implementation-level reason why it wouldn't work (at least, not on any video card from the past century). Anyway, that's the story of how I found out that most/all modern Linux distros actually fake character mode with a frame buffer.
> So why is downloaded software marked up by similar or greater amounts?
My guess? Historically, AUD$1 == US$0.50, and quite a few people haven't realized yet that they're now more or less at parity. As a result, I suspect quite a few American companies are just doubling the nominal US price out of habit.
~15 years ago (circa 1996 or so), touchpads emulated the kinetics of thumb-manipulated trackballs, and I didn't mind them too badly. Then, sometime around 1999, all the manufacturers started to de-tune their virtual kinetics to accommodate users who used their index fingers. The difference isn't subtle. Your thumb's range of motion is limited, and the first-gen touchpads used the same algorithms as trackballs to interpret a motion that can loosely be described as "kind of towards the left, somewhat downward" as "move the pointer directly left" (and vice-versa).
Once in a blue moon, I'll trip over a modern touchpad that *almost* gets it right, but manufacturers view them as commodities, so even when I find one in real life that doesn't suck, there's no guarantee that the identical laptop ordered from Newegg will have the same one.
I have a hunch that the touchpads in laptops can do more than the generic (or official) drivers offer, but low-level documentation for them is nonexistent (in the public space, at least). I still spend an occasional afternoon looking at Linux source hoping that someone has leaked a driver that gives a determined driver the ability to tweak it in nonstandard ways.
Insofar as "multitouch" goes, the idea is actually nothing new. Users had the idea 10+ years ago. The whole problem was with the touchpad firmware itself, which reported only a single x,y location (or a single velocity vector). All you need to implement "pinch type" multitouch is a touchpad that reports raw events, and does it quickly enough to preserve the trigger order. If I see "finger down at (x,y)", then "finger down at (a or u, b or w)", it doesn't take a rocket scientist to put 2 and 2 together and figure out that the values closest to x,y probably are one point, and the remaining two values are another. From that point, it's just a matter of tracking them quickly enough to ensure that the jump from sample to sample doesn't exceed the difference between the two (so you can keep track of which value is which over time).
Also, multitouch touchpads that support 3+ fingers aren't necessarily able to support completely arbitrary touches, or touches that are truly simultaneous. For example, supporting a "3 fingers swiped vertically" gesture is one thing... supporting "3 fingers moving with completely independent motion along paths that cross" is another mater entirely. A "3 fingers to scroll" gesture would be implementable even with old touchpads with simple matrices, as long as they report ALL contact points instead of just the first or most dominant pair. You'd just have to notice that there are 3 different x values that differ by approximately N units, and notice that the 3 x values all have y values that are similar & changing at a similar rate.
I own two of them, but the mouse buttons on one are completely nonworking, and the mouse buttons on the other are so flaky I barely even try to use them anymore. The thing is, the buttons NEVER really worked well... they just kind of went from "annoyingly flaky" to "barely-working, or not working at all."
Can anybody actually fix them in a way that not only fixes my problem, but hopefully fixes it in a way that improves upon the original buttons? I love the keyboards (even though most of my coworkers hate the one I have at the office), but I really wish I could get the buttons fixed.
That said, I really wish somebody would make a keyboard like the M13 with Trackpoint, but give it a SECOND Trackpoint located directly below the spacebar, and configure it so that one can be used for pointer movement, and one can be used for scrolling & panning. Fujitsu got the location right with THEIR pointer, but their mechanism sucks. IBM got the mechanism right, but "GHB" is a poor location for fine-manipulation (your thumb is stronger, but has less range of motion... precisely the use case ideal for pointer sticks). I've seen a few Sony netbooks over the years that put the stick below the spacebar, but I can count them on one hand. I wish more companies would try putting sticks there (or hedging their bets and putting two sticks in place, so users can pick one for mouse and one for scrolling).
Or worse... they start treating oil filters like laser toner, and declaring them to be "expired" the moment your odometer ticks off 3,000 miles.
Not to mention the ability to charge for different levels of performance using the same underlying hardware, kind of like ATI & Nvidia do, and Intel was planning to do with their value-priced CPUs.
Here's an easy way to tell whether they're doing it for "safety", or just to increase their own profits -- if they give copies of the security key to end users, their motives are probably good. If they won't even give the code to mechanics, and force field replacement of expensive parts that could be repaired if the mechanic had the code, then they're doing it for their own benefit. It's just like UEFI. If I have a copy of the key, it's awesome. If the only copy of my key is held by Microsoft or Sony, it's a shameless pwnage of my consumer rights whose physical and political defeat is a moral imperative.
> When is Apple following suit?
You didn't get the memo? iPhone 5w. "w", as in "Woz Edition".
Rumored features include
* Zigbee wireless mesh networking
* multiplexed pins on the headphone jack that can be repurposed for I2C, SPI, or GPIO (not at the same time, obviously). Oh, and legacy UART (3.3v logic) that also supports Atmel-friendly baudrates (125kbit, 250kbit, 1mbit), 9-bit word length (9/N/1, to be exact), and can use pin #3 for GPIO, RTS, CTS, or synchronous clocking in or out.
* fixed 4800mAH lithium cell
* gamepad wings that slide out from the underside (to the left and right when held in landscape orientation). One side has an analog stick & digital stick, the other side has an analog stick & 4 buttons.
* Volume button pair that also serves as the "left trigger button" in "gamepad" orientation, and 2-stage camera button that serves as the "right trigger button" in gamepad orientation. All of which can be intercepted, redefined, and repurposed by end users as they see fit.
The iPhone 5w's flagship applications will be a logic analyzer/DSO, which demonstrates the use of the bundled iProbe (4 clip-on leads, terminating at a 1/8" TRRRS headphone jack). Additional accessories will allow connectivity to most car ECUs, CANbus, and JTAG.
Oh, and the phone will also include a fully-unlocked bootloader. Of course, not even Woz will likely be able to get the Powers that Be at Apple to release the source to their crown jewel, but it won't matter. People will buy one, and reflash it to Android. Even Larry & Sergey will be spotted in public with it (running Android, of course).
The problem with that argument is that one might legitimately argue that if there's a safety issue that's mitigated through software, the need for modified firmware to do the same needs to be communicated. 99.9% of damage and hard-bricking is caused by either booby traps left by the manufacturer to trap the unwary, or important details like "always monitor the battery temperature, and back off if it exceeds N degrtees Celsuis" that aren't disclosed.
I'm happy to see that Moto is finally starting to become non-Evil under the ownership of Google. I'm disappointed as hell by the fact that the Photon Q was totally gimped by Sprint and/or Motorola. This would be a great phone if it were in a blister pack at Wal Mart being sold for use with some value-priced prepay network. It's not, however, a top-tier best of breed flagship Android phone:
* 540x960 qHD. WTF?!? 540x960? And not even OLED? The Q's display is a decisive step down from the Photon 4G. Note to Sprint & Moto: the next generation of any phone is supposed to AT LEAST as good as what it replaced, especially now that we're going to be stuck with the damn phone for almost two painfully long years thanks to last year's abolition of 12-month upgrades.
* tiny battery that can't even be swapped when it dies halfway through Friday night. If they'd put a huge battery inside like the one in the Razr Max, it might have been tolerable. But 1785mAH? You can't be fucking serious. I'd literally burn through that in 3 hours.
* Half the flash of its predecessor. WTF. Read the note above about how successor phones are supposed to be a step up, not down.
* Nonremovable SIM. Yay, it can roam on GSM in other countries... except at $2.50/minute, nobody is going to actually DO it knowingly and voluntarily once they find out how badly Sprint is going to rape them for doing it. Overall, this is just kind of like Sprint turning around and giving us a final kick in the balls, just for good measure.
It's been years since I've actually left the US, but the sealed-SIM anti-feature ALONE is enough to make me want to leave Sprint, because it demonstrates total and complete contempt towards us. I mean, really... once you factor in the administrative cost of roaming, and fighting with livid customers who just got a bill for $900 in roaming charges after spending 3 days in Montreal, how much extra is Sprint *really* going to make compared to what they would have not made by just giving it a normal SIM slot and charging a $35-50 one-time admin fee to unlock the SIM lock during the first year of a contract? Sprint could even offer an olive branch to deflect criticism and waive the fee if the customer puts down a $400 deposit that gets returned after the contract's 12th month (knowing that 99.9% of customers would just say 'fuck it' and pay the $35-50).
It's a shame, because I love Motorola's build quality and superior radios. But the harsh fact is, I'm going to be stuck with my next phone for 20 long, painful months now that Sprint has taken away our annual upgrades, so my next phone has to be damn near flawless & one I'm sure I'll be able to live with. The Photon Q is not that phone. I can only pray to the flying spaghetti monster that 18 months from now, Sprint + Moto will be unleashing an unlocked Nexus device with specs to die for.
They take advantage of human nature, which tends to underestimate personal risk and overestimate one's likelihood of successfully avoiding bad outcomes. Often, the individuals who are the most vehement in their condemnation of "bad drivers" are themselves some of the worst drivers out there.
Call it narcissism, delusion, or simply "an unbelievably creative ability to rationalize any point of view regardless of objective facts", but this is basically a way for insurers to identify a small (but expensive in terms of payouts) group of risk-takers that depends mostly upon that same group's willingness to roll the dice with their insurance company and risk higher rates.
> Driving without insurance in the UK will get your car seized and crushed.
Maybe Britain is radically different from the US in this regard, but I'd be shocked if any such law weren't written in a way that made it impossible to actually DO that to any car that was secured by a bank loan, as opposed to a car that was paid off and driven by an uninsured owner. At the *very* least, the law would be written with "safe harbor" provisions that exempted any car used to secure a loan as long as the lender made a good-faith effort to enforce compliance (regardless of whether or not the lender's efforts were actually successful and effective).
Not to mention the fact that it would instantly burn any chances China might have of cooperating with the US & Russia on any international space ventures going forward, something they want to do quite badly.
Yes, we know that "Chinese Hacker" != "Chinese Government", but it would set back the cause of Chinese-US-Russian space relations by decades if it were "merely" independent hackers, and really WOULD set it back forever and for all eternity if it were proven to have been in any way, shape, or form an official program of China's government. This brings us to one logical conclusion:
If you're Chinese, and you humiliate China & harm its reputation with NASA & Russia's space program, you'll be ruthlessly hunted down and executed at the first possible opportunity on live TV in a grand public relations exercise ordered by the highest ranks of the CCP. NASA will probably be horrified, and Russia will probably be unimpressed & see right through it, but the end result is that you'd have to be completely suicidal to be Chinese and even *try* hacking American or Russian space hardware. It's one thing to deface the National Parks Service website for Kennedy Space Center (or whomever runs it as a tourist attraction). It's another matter entirely to brick a piece of hardware that cost 1.000 times as much as you and your ancestors going all the way back to your oldest homo sapiens ancestor have collectively earned since the dawn of human history.
Would Al Qaeda (or someone influenced/inspired by it) try to do it if they could? Maybe. China? No.
I rarely see problems involving new releases of Java... with one specific exception: 64-bit Java. I've seen more than a few apps that die horrible deaths with 64-bit Java... and almost exactly the same number that specifically *require* 64-bit Java when running under 64-bit Windows.
I'm still trying to figure out what, exactly, causes some random Java app to specifically require one or the other when running under Win7/64. I suspect it has something to do with changes Microsoft made to 64-bit Windows that forced (or at least induced) Sun, then Oracle, to change something major with 64-bit Java... but I've never seen any white paper or article that identifies and explores the specific reasons why this might be so. I'm sure the mass layoffs and exodus of Sun's employees right around the time 64-bit Java became relevant, and 64-bit Windows became mainstream, didn't help.
But then, when you go to log into a site with wacky rules that requires a special password using a different computer/tablet/phone, how does it KNOW it has to use the alternate password scheme?
The LastPass keyboard is a nice idea, but AFAIK, Android doesn't allow on-the-fly keyboard switching (you have to launch settings, navigate to 'keyboard and input', set a new default keyboard, then start over). I use Graffiti for everything (I'm crippled without it), so I'd still be SOL. So would somebody who uses Swype, a split tablet keyboard, or even the funky keyboard whose name eludes me that has you compose English the way Koreans compose Hangul (ex: 'd' = 'c' + 'l').
Probably for a few years. I think you can STILL buy "New Old Stock" retail copies of XP Pro on Amazon. The copies that will dry up quickly are the $129 OEM copies of Win7 Pro. If you want Win7Pro 2 years from now, you'll probably have to buy a full-priced retail copy unless a home edition is good enough for you.
I'm pretty sure Microsoft allows distributors and volume customers to return unsold oem copies and exchange them for the new version for a nominal fee, if not free, because OEM copies always seem to dry up within weeks of any new version's release. Even XP dried up, before Microsoft backed down a few months later & allowed distributors to buy it again.
Can someone honestly remind me why so many people hated Win2k? To this day, I still remember it as one of the best versions of Windows, ever. I could upgrade or install MAJOR drivers & system-level components, then go days or weeks without having to reboot.
Maybe it's because NT4 already weeded out most of my software and hardware that Win2k would have objected to, so it ended up being a net improvement in every meaningful way, but it honestly wasn't until I got my first dual-CPU motherboard that I really felt any need to 'upgrade' to XP... then had a MAJOR love-hate relationship with it. Hugely-improved SMP support? Major plus. Endless reboots? Yuck. I reinstalled without SP1, and vehemently resisted SP1 for more than a year until Microsoft slipped another patch past me that basically bundled everything about SP1 I was trying to avoid, and instantly took away my ability to do reboot-free updates once and for all. RIP. Sigh.
Sadly, even Linux now seems to try and force reboots by default now for some updates. Oh, you can usually open a shell and force it to do a hot update, or ignore the warnings and it'll work anyway, but more and more, it feels like even Linux has abandoned the ideal of "never force a reboot". and instead embraced Microsoft's philosophy of, "If something's wrong, reboot and it'll probably fix it."