Incorrect -- the emulator is emulating ARM code. That's why it's an emulator. And why you can test NDK stuff as well on that emulator, which you can't using an x86 build of Android (well, at least not for ARM NDK).
If you don't care about NDK code, alternate methods of development exist. You can test on an x86 build of Android, either in VirtualBox, or (as we do at my company) on a netbook.
Oh, but there's that whole issue with Google's release process... you can't actually run Honeycomb on x86 yet (last I checked). So the emulator is the only option. This is one indication of why Google's release process is broken. They work with a single OEM and product, for Honeycomb it was Motorola and the Xoom. They don't release source code to one single other developer or OEM until that single product is actually released -- thus ensuring all early adopters get that wonderful "not quite beta yet" experience. They have formalized the process of eliminating as much testing as one can for the first release... the exact opposite of what you'd like to ensure a quality release. And, of course, they have put a short circuit on other development methods.
With that said, I think it's important to understand the difference between Android and iOS at the tablet stage. When the iPad shipped, there was one kind of app for iOS -- a plain old iPhone app, hard-coded for a 480x360 screen. So to do anything even slightly ok for the iPad, you needed an "iPad Optimized" app. After the iPhone 4 came out, you had the additional option of an "HD" app... not iPad optimized, but capable of using the iPad's higher resolution. And in fact, when Apple counts iPad apps, they count both HD and iPad-only apps.
For Android, a large percentage of 2.x apps already work on higher resolution screens. Maybe this goes back to development -- Android 2.x supports at least up to 1024x600 or so screen resolution, which is what you find on most netbooks, which are a fairly popular development target for Android development. So lots of apps already run just dandy on a tablet. The same app works fine on a smartphone, 2.x tablet, or 3.x tablet... much less incentive to move swiftly to Honeycomb-only applications. Not to mention the fact that for several months, the Xoom was the only 3.x tablet, and the only version of that available was around $800, on-contract at Verizon. And it was buggy. So right out of the gate, there's no pressing need for Honeycomb apps, and plenty of reasons to wait for more Honeycomb machines to ship.
A Commodore? Nope. A "Commodore" is when you have some crazy good technology, spend practically nothing developing or marketing it, and pay managers who know absolutely nothing about your industry way more than the guys at your much larger competitors make. And hey, while you're at it, replace one of your best assets -- the proven engineering managemet team --with q bunch of fools. Who then proceed to delay your next gen products by 6 months, and screw them up along the way. Oh, and sure, also keep huge volumes of the old product around, because you didn't actually believe your proven engineering team would deliver the delayed products in time, so you didn't order enough parts. That was Commodore, '91-'93.
I think Nokia's pulling more of an Osborne. They have a very successful product line, bit aging but still well liked. They could have just concentrated on the current stuff, maybe evolving SymbianOS into MeeGo via a compatibility layer, if SymbianOS was beyond modernization. Or even suggest that Win7Phone would simply join the Nokia lineup. But no, they felt compelled to tell everyone that every current Nokia smartphone is already obsolete, an evolutionary dead end. And that, not only are they going all MS, but basically, there's just no reason to buy a Nokia smartphone for a year or more.
That really was the message delivered if not intended... Adam Osborne didn't intend to starve his company either. Or send all that business to Kaypro, either. This message has not been lost on consumers or the industry, either.... Nokia is in crash and burn mode right now. They can only fall so far -- someone will want the patents.
Yup. I expected the Republicans to be idiots about healthcare. After all, they're well paid by the insurance companies, they're rich (and so, they already can afford it), and they can pull out the "don't trust the government" thing when it suits them. And ignore the terrible effect of employer-paid healthcare on every US business that's not an insurance company.
But it was the Dems who caved on single payer this time around, and they did that even before they got to the table with the Republicans. Both sides are for sale.
Incidently... thanks to big fat Chris Christie (Jersey governor), my wife's insurance, formerly a benefit (teachers have typically had great benefits and lousy pay... as an engineer working at startups for the last few years, I'm usually dealing with good pay and lousy benefits), may cost as much as $6,000 this next year, to cover our family of four. As with many of the Republican governors, Christie has been working hard to effectively increase taxes on teachers, firemen, police, and other public employees, so he can afford the tax breaks previously given to the richest in the state.
Of course, what he's effectively done is killed an untold number of small businesses. It's an on-going thing... we lost a couple locally already this year (Woodstown/Pilesgrove), even a liquor store. Those are supposed to do well in bad economies...
The rich are already the richest (compared to the average) they've been since before the great depression -- additional income does nothing. Same with businesses -- additional income does absolutely nothing to grow a business. Additional demand for a product or service is the thing that grows a business. And these new sideways taxes have just ensured that a big chunk of the New Jersey middle class will have thousands less to spend this year and next.
That's important to understand -- the poor and the middle class spend all of their income; the rich don't. So additional lower end income boosts an economy, additional higher end income alone does nothing to the local economy. Boosting everyone's income boosts the economy, but only for a very short time.. things do ultimately stabilize around supply and demand... the price of things in short supply go up, and eventually, no one sees that income boost any longer. But I digress...
The single largest cause of bankruptcy in the USA? Medical expenses incurred by people who were "fully covered". That's about half of all bankruptcies.
And chances are, you'll never know if you're fully covered or not. And even if you are, you may have some hours of taking at your insurance company to ensure they agree. There seems to be a standard policy at most insurance companies to just randomly reject claims, even those they know are covered.
I do know from where I speak. My wife was diagnosed with breast cancer in April of 2010. She had very good health insurance for someone in the US these days... at least a one-time advantage of being a teacher in New Jersey. Six surgeries, 16 weeks of chemotherapy, 6 weeks of radiotherapy down, and one surgery left to go, we haven't spent a great deal out of pocket (but some financial fallout from her missing summer school in 2010). But again, this is about as good as it gets, and you'll find millions of people who thought they were just as secure, but weren't.
And diseases like this are curable, but extremely expensive to cure. We're probably over $50,000 on surgeries alone, probably in the same ballpark on chemo drugs... just Adriamycin (Doxorubicin) costs over a million per kg just to produce (expensive to administer too -- only by infusion).. and there are some cancer drugs over 10x that price. And once you have the cancer drug, there are a half-to-a-dozen additional drugs just to mitigate the effect of the cancer drug. That's also part of the problem -- such things are done just as well in other parts of the world, but the expenses are much better controlled.
But we did receive thousands of dollars in bills that should have been covered, and had to argue about them. And honestly, this insurance company was amazingly well run compared to others we've both dealt with.
There are several reasons... and that's ignoring the fact that state run healthcare, everywhere else in the world, has proven more effective, more universal, and about half the cost of the US system today.
The first big problem is profit motive. A government organization filled with government workers (eg, like Congress, the Senate, the Military, etc) is not expected to turn a profit. And no multi-million dollar CEO salaries. So, as long as it's well run (yes, that's an argument, but it applies equally to private enterprise), most of the money going into the system is used for healthcare. Not profits. Not advertising. Not lobbying... and generally, lobbying against the best interests of those served by the insurance industry. This is one big reason so much money is spent in the US for lower quality healthcare, versus more civilized countries.
Another issues is price negotiations... countless smaller healthcare companies cannot as effectively negotiate on the price of medicine and other supplies as can the entire country. And even if you did open up healthcare insurance to competition, there's no real guarantee it'll be actual competition. Other types of insurance are openly competitive on the surface, but they're really practicing artificial competition -- they bring you in at a discounted rate that looks great next to what you're paying, but a year later, you're paying more or less what you did at the old company. And they spend hundreds of millions in advertising to grab each others customers on a rotating basis. Geico alone spent over US$900 million last year in advertising... and they just do car insurance. That's over 10% of their expenses right there, before a single car is repaired. The health insurance business is an order of magnitude (at least) larger. You really want to see tens of billions in healthcare money going into these kinds of campaign?
The US healthcare system is also rabidly anti-business... unless you happen to be in the insurance business. I've even heard Righties saying this about "Obamacare" recently... but the main flaw in the Patient Protection and Affordable Care Act is that if left this as a problem... it didn't create it. In all these other countries, businesses are not funding healthcare. So they have an immediate competitive advantage over US manufacturing. Universal healthcare gets the business out of the picture, and the hand out of their wallets on this. If the right were actually pro business (they're "pro" donation, primarily, which is why they've sided with the insurance lobby and against every other business in the USA... all those other guys are spending their lobbying money on other things), if the left actually had the balls to not cave in to the right before the negotiations even begin (could be a medical problem, but ball exams are probably not covered). And of course, all of the people actually making the laws, both those caving in and those rising up against any notion of public healthcare, enjoy the best public healthcare system in the country... all Senators and Congresscritters get this.
And finally... healthcare as insurance is fundamentally the wrong answer. The insurance companies can insure things that are properly insured. My house and car, for example... something bad might happen, but chances are, no. But each and every person on the planet will eventually have a health issue of some kind. If that issue isn't immediate death, they will need medical treatment. Insurance is the wrong answer for something so guaranteed. And for something that needs constant treatment. Actual health care includes a strong prevention component... stress health rather than medicine, and you lower costs AND increase the quality of life. US healthcare is largely about fixing what's already broken.
The all you can eat model is a great one -- you know precisely what you're going to pay, and don't feel like bandwidth is wasted. But the whole cellular industry builds their huge profits based on mistaken expected use. So it's no great shock to see the data model move that way.
An equally fair rate would be pay-per-MB or whatever -- precisely metered, just as you might pay for electricity or gasoline. As long as the per-MB fee is fair, I have no problem with it. But the telcos do -- they want me to pay for the occasional high-use month, every month. So they put up big penalties (I mean, hell, that first MB over the limit doesn't cost them one iota more than that least MB under the limit) to fear the user into paying too much on a regular basis.
I'm on Verizon today, and fairly happy with it, largely because it can actually reach my house. T-Mo and Sprint, not so much.. and that's going to the case for Sprint and T-Mo in any rural area -- they just don't have the low frequency 850MHz band to penetrate foliage well. Also why Verizon works better than AT&T -- they both have 850MHz and 1900MHz slots, but AT&T's 3G takes both, while Verizon's runs only at 850MHz though the forest.
On the other hand, given that the price going from $30/month for unlimited to $30/month for 2GB, that sure smells like they're just plain doubling the price in practical term (I have often gone over 2GB, rarely get over 4GB). For less than that doubling, I can put up a roof antenna and repeater/booster, which ought to solve the T-Mo or AT&T problem just dandy. If only they didn't suck in other ways... What we need is a new company doing a reasonably priced data-only service. Sadly, it takes a very, very long time for new companies to build out a generally useful network -- both MetroPCS and Cricket are limited to a small number of generally urban-only areas so far.
Looks like you're either mistaken or lived under a rock. Most mobile apps devs speak Objective-C now and it has been this way for quite some time.
True only for very small values of "quite some time" and "most mobile apps devs". Apple iPhone apps have only been "native" in Objective-C for three years and three months... the SDK was released in March 2008.
I guess YOU lived under a rock prior to then. Blackberry ran Java MDIP as far back as 2003, their basis for mobile apps. Countless "not so smart" phones ran Java ME. Nokia, Microsoft, and PalmOS smartphones and PDAs, dating back to the early 1990s, were usually programmed in C or C++, though plenty of other options existed. And even today, you can program iPhone apps otherwise... for example, in Adobe ActionScript or Flex. Check this out: http://en.wikipedia.org/wiki/Mobile_application_development
There is and has been a ton of mobile development unrelated to Apple's 10% or so share of the global mobile market (factoring in smartphones, iPads, and iPods).
Google wanted the basic Android apps to be machine independent. So the same stuff largely runs on ARM and x86... even a few MIPS fans are starting to do things with Android. Java was available, and of course, the language itself couldn't be patented or copyrighted, and there was this GPLed implementation, etc. Sure looked like a good idea at the time.
I had the same kind of plan when developing a video/internet set-top box back in the late 1990s. We wanted applications, but they needed to be insulated from the hardware, so in the future, the hardware could change completely without messing with compiled apps.
Sure, they could have written their own language as well as their own VM... would that have been enough to keep out of the crosshairs of companies like Oracle who now think they have ownership over any software VM, thanks to Java? Maybe, maybe not.
Gates did no charity, either, until Melinda came along. And maybe he started thinking about his legacy -- with all that money, and all he's remembered for is Windows, Office, and the BSOD? Far cooler to say, wipe out Malaria. That kind of cash is the very definition of a superpower... hmmm. All this started about the time he saw "Spiderman", too....
Or, if you're using an Android phone, load a version of the camera app/OS/etc. that doesn't have the "decode IR" function. Oh, wait... won't be an issue for 20 years anyway, since Apple will own the technology until then. Apple's just terrible at establishing public standards (think "Firewire", which is the best they've managed to date), so they'll be the only users. And they'll sell it as an advantage to the iPhone/iPod, since the typical use will be getting messages about the thing you're photographing.
This is one big reason that all modern cameras hook to PCs as plain old USB storage class devices, rather than "some custom thing" that you need a private drive for. Also helps eliminate the need for drivers on Windows, Linux, and MacOS (or complaints by those who don't get supported). And of course, even if your old camera uses the terrible old SmartMedia cards, you can still read those. But if you find that a common activity, you really do need a new camera. Digital cameras are great today, and they manufacturers are looking hard for new compelling features. Ten years ago, they pretty much sucked.
Cellphones are different -- they get replaced all the time. They are planned to be obsolescent almost exactly two years from date of purchase (sucks if you're on a Canadian three year contract, but that's pretty much the reality of the situation).
Actually, the cellphones all use CMOS sensors, not CCDs. But either way. The single sensor cameras, cheap or very expensive, all have an IR bandstop filter and a Bayer or similar color filter (well, ok, if you're using a Foveon sensor, you don't have the Bayer filter)... these can be the same (eg, the Bayer microfilters also block IR) or not ("night mode" on consumer camcorders is usually achieved by rotating out the IR bandpass filter).
You could add IR sensitivity without any additional expense by altering the Bayer matrix to go RGBI (Red Green Blue IR), with the microfilters being IR blocking and IR passing, as needed. Software, then, would have to ignore the IR pixels, at least outside of "night mode". This would lower the image quality a bit, but hey, it's a consumer camcorder. The pattern might be altered in ways to minimize this, too. For example, Sony's "ClearVid" sensor doesn't use Dr. Bayer's original RGBG pattern, but rotates the sensor 45 degrees, and uses a pattern that's using something like 6x as many green pixels as red or blue. Drop a few IRs in there instead of a few of the greens, and you're be really hard pressed to tell the difference.
Of course, this scheme delivers only a fairly low resolution IR output. I'm sure they want the IR messages time multiplexed, but you'd still need a fairly large emitter to catch it on camera. And they'd have to be spaced well or synchronized, or they'd mess each other up.
The TDM data can obviously be itself detected by a single sensor -- this is how IR remotes work, and I used a slightly more sophisticated version of that once in a gaming robot design. But Apple's scheme is going to require some reasonable localization of the IR sources. After all, if you read the patent, only one of the functions is "no photos"... they also plan to use the IR becons to send information. For example, you're in a museum, you point your camera at an exhibit, and the IR messaging delivers some information (probably akin to a bar code) to let your phone deliver augmented reality information/links about the thing you're pointing at.
Some Canon camcorders from the early 2000's used IR ranging in addition to contrast... the idea was that the IR could very quickly come up with a ballpark range, the contrast AF could go there immediately, then slowly seek to attain the final focus.
This is a classic issue with camcorders. On a still camera, you focus as quickly as possible, then snap the photo -- not too much of an issue. But for video, you don't want focus seeking to be visible in the shot. This is why some of the early video modes in P&S still cameras had such bad autofocus; they still used the still algorithms.
So I put a very weak IR emitter on the backside of my IR filter. Or better still, a diffraction grating -- I still get IR light though the filter, but the information is lost. Visible light is not compromised. Either way, trivial to work around... and that's just with a minute of thinking. Well, ok, they're using some kind of TDM encoding scheme, so the diffraction would eliminate any point source, but if you have just one IR source in the image, the signal might still get through.
The other problem is already kind of practical -- how to build an IR sensor that can actually read a point source and not destroy the visible light image. Today's 5Mpixel on 1/4" cell cameras are already diffraction limited at about f2.5, and it's fairly typical for camera makers to use f3 lenses on these. Depending on just where you put the IR, you're already getting some pretty serious blurring. Not to mention the fact that IR has to be filtered before the sensors.
The simplest solution would be to modify the Bayer matrix, probably by adding IR filtered pixels in with the R, G, and B filtered pixels. That's going to hurt photo quality a bit, but hey, it's a cellphone... anyone who thinks it's a replacement for a real camera won't notice. More sophisticated would be a diachroic prism to split off the IR, replacing today's typical IR bandstop filter, but that's not so practical in a tiny device like this.
The on-device IR emitter reflection scheme (basic principle behind the "IR Blaster" used in some STBs to control VCRs, back in the 90s) needs quite a bit of power to deliver a readable signal off random room contents, and that's just in a typical livingroom. This would be totally impractical on a portable device, and it would fail entirely in a large room, much less outdoors.
There are both good and bad reasons for the crazy high price on an unlocked smartphone.
The good reason is simple: carriers (in theory) subsudize the price of the phone, in order to get consumers to agree to muli year contracts. If they don't pay, naturally you have to make up the difference.
The bad reason is a reflection of the immense power of the carriers, particularly in the USA. I mean, up until this point, Apple had only two actual customers ror the iPhone in the US: Verizon and AT&T. You know that neither is paying anything close to $649 for an iPhone.
In fact, carriers pay a fixed percentage of the MSRP for any phone they resell. If a manufacturer wants to set a reasonable price for a phone, they are welcome to do so.... but that would mean losing money on each telco sale, or not selling through the carriers at all.
What do I mean by a reasonable price? Something in keeping with other CE devices. A 32GB iPod Touch retails for $299. The 32GB iPhone costs maybe $50 more to make, so it should run $399, maybe $449 tops. Not $749.. that additional cost is basically a cell phone tax, levied by he carriers on anyone not playing the contract game.
Pretty happy with my Notion Ink Adam as an eReader. I don't need all week on a charge all day is fine. Indoors, backlit is handy in a dark room. Color is important for magazines, higher resolution useful for tech documents, as is first class PDF support and plenty of memory (32GB in extra flash, support for USB sticks, etc).
Outdoors, the Pixel Qi display is nice and crisp in full sunlight, tough not as high contrast as eink. With the backlight off, the device runs longer still. And of course, I can use Amazon, B&N, Adobe, and many other readers, no lock-in.
Sure, weight is an issue for long term hand holding. Like most tablets, the screen is glossy... the eBook folks seem to have the screen done better, most being a modest matte finish. The Adam has a bottom ridge along one side which makes hand holding a little easier, tough of course, not as crazy thin as a Kindle or iPad 2.
Hard drives are rated in bytes... base 10 bytes. That's MB and GB, not MiB or GiB. Sure, it favors the marketing, but it is also the proper unit for a hard drive, or a human. The power of two sizing is only appropriate for RAM and ROM integrated circuit devices. Don't blame your hard drive when its your OS using the inappropriate units.
There are some things you absolutely need at boot time. These happened sequentially in the Amiga's startup-sequence. Fairly early, the OS evolved a second initialization script, spawned as a separate process, for everything you might need eventually, but should not be a reason to not boot fast. And the user was very easily in charge of this -- all in nice, plain, everyday ASCII, not some deep hidden arcanity in the OS.
Second thing.. disc drives flushed their writes, and the OS recognized added drives. So there was never any reason to mount or unmount a portable drive -- plug it in, pull it out. That's far more useful than write-caching a write to some removable device for longer than it would take for a human to pull it out. Windows and Linux fail on this, completely, 26 years later.
Making things easier, particularly via GUIs, all kinds of formerly useful features of text shells, text languages, etc. have been tossed out. The GUI experts won the battle, and the didn't consult those who came before.
How about graphical shells... where's the pipe function? Redirection? Something so useful, simple, and fundamental... and missing from every GUI.
How about wordprocessors. Back in 1979, using Scribe and Emacs, I could put together a complex document in ways I haven't seen recently. For one, with few changes lines, I could change the format of citations, footnotes, index, etc.... maybe going from IEEE standard to ACS standard in 20 seconds. That's because the word processor understood such things at a high level, and could format these in your format of choice.
Compound documents, too.. I could wire a paper, then include it as a chapter in a book, without having to renumber or change much of anything.. and keep it as a separate file, for both evolving documents. I've seen WYSIWYG tools that supported this (Mentor's "Doc" comes to mind), but this fails completely in popular WPs like MS Word.
One reason new TVs are streaming video over Ethernet -- they can do an end run around the cable companies that way. Netflix is more likely than HTML5 video for overall success, since most content companies are not going to stream media without DRM. Of course, Flash and other technologies (Netflix is using MS's Silverlight) can provide DRM as well. And that's not to suggest it all will be pay, either.. most of the free-to-view stuff on the net is DRM protected.
The reason is simple: BD players and televisions are rapidly moving away from the fixed-function device model. Once you have the ability to accept new software, things like "streaming media box" or "TiVo" become a simple matter of software, perhaps with the addition of extra storage. Economies of scale will make TVs, game machines, and BD players a far better deal than the stand-alone boxes, while at least as capable. I mean, look at the PS3 (when it's not being hacked, anyway). This shipped as a first generation BD player (and game console). It's been upgraded to every version of Blu-ray so far released, as a free download, including the net-enable stuff, PiP addition, and 3D. It's also a photo organizer, a video editor, a Netflix players, and a whole slew of things that didn't come in the box.
TV and BD is following. The reasons are simple: competition among the CE companies is fierce, and software add-ons are a cheap way to offer new features. Even if the next guy has it, your TV or DVD/BD player of a year or two back probably doesn't. So they potentially get upgrades just based on the strength of the extra software features.
If you look at the sales of BD players (likely to beat out DVD players by the year's end) versus BD media (usually 20-30%, varies a bit by release, versus DVD), it's clear that things like integrated Netflix are already having an impact on buyer's habits.
Yup... Elon Musk is on record as claiming that $1.1K per Kg is achievable in the foreseeable future.... presumably, whatever's the next cost reduction after Falcon 5 (the Falcon X platform, perhaps). SpaceX is already scaring the Chinese on costs, so they could definitely have an effect here.
As cool as the tech is here, it's clear that private companies are and will have a big impact. Especially with multi-government run agencies like the ESA, they're taking forever on concepts, which makes it unclear that, if they actually do ever produce, whether the new rocket/spaceship will actually be current, or saddled with the last 20-30 years of design-by-committee. There's something nice about the "just do it" approach that SpaceX is taking (ok, sure, they've been at it awhile too, but they do seem to be doing it).
Slashdot Mirror
Incorrect -- the emulator is emulating ARM code. That's why it's an emulator. And why you can test NDK stuff as well on that emulator, which you can't using an x86 build of Android (well, at least not for ARM NDK).
If you don't care about NDK code, alternate methods of development exist. You can test on an x86 build of Android, either in VirtualBox, or (as we do at my company) on a netbook.
Oh, but there's that whole issue with Google's release process... you can't actually run Honeycomb on x86 yet (last I checked). So the emulator is the only option. This is one indication of why Google's release process is broken. They work with a single OEM and product, for Honeycomb it was Motorola and the Xoom. They don't release source code to one single other developer or OEM until that single product is actually released -- thus ensuring all early adopters get that wonderful "not quite beta yet" experience. They have formalized the process of eliminating as much testing as one can for the first release... the exact opposite of what you'd like to ensure a quality release. And, of course, they have put a short circuit on other development methods.
With that said, I think it's important to understand the difference between Android and iOS at the tablet stage. When the iPad shipped, there was one kind of app for iOS -- a plain old iPhone app, hard-coded for a 480x360 screen. So to do anything even slightly ok for the iPad, you needed an "iPad Optimized" app. After the iPhone 4 came out, you had the additional option of an "HD" app... not iPad optimized, but capable of using the iPad's higher resolution. And in fact, when Apple counts iPad apps, they count both HD and iPad-only apps.
For Android, a large percentage of 2.x apps already work on higher resolution screens. Maybe this goes back to development -- Android 2.x supports at least up to 1024x600 or so screen resolution, which is what you find on most netbooks, which are a fairly popular development target for Android development. So lots of apps already run just dandy on a tablet. The same app works fine on a smartphone, 2.x tablet, or 3.x tablet... much less incentive to move swiftly to Honeycomb-only applications. Not to mention the fact that for several months, the Xoom was the only 3.x tablet, and the only version of that available was around $800, on-contract at Verizon. And it was buggy. So right out of the gate, there's no pressing need for Honeycomb apps, and plenty of reasons to wait for more Honeycomb machines to ship.
A Commodore? Nope. A "Commodore" is when you have some crazy good technology, spend practically nothing developing or marketing it, and pay managers who know absolutely nothing about your industry way more than the guys at your much larger competitors make. And hey, while you're at it, replace one of your best assets -- the proven engineering managemet team --with q bunch of fools. Who then proceed to delay your next gen products by 6 months, and screw them up along the way. Oh, and sure, also keep huge volumes of the old product around, because you didn't actually believe your proven engineering team would deliver the delayed products in time, so you didn't order enough parts. That was Commodore, '91-'93.
I think Nokia's pulling more of an Osborne. They have a very successful product line, bit aging but still well liked. They could have just concentrated on the current stuff, maybe evolving SymbianOS into MeeGo via a compatibility layer, if SymbianOS was beyond modernization. Or even suggest that Win7Phone would simply join the Nokia lineup. But no, they felt compelled to tell everyone that every current Nokia smartphone is already obsolete, an evolutionary dead end. And that, not only are they going all MS, but basically, there's just no reason to buy a Nokia smartphone for a year or more.
That really was the message delivered if not intended... Adam Osborne didn't intend to starve his company either. Or send all that business to Kaypro, either. This message has not been lost on consumers or the industry, either.... Nokia is in crash and burn mode right now. They can only fall so far -- someone will want the patents.
Yup. I expected the Republicans to be idiots about healthcare. After all, they're well paid by the insurance companies, they're rich (and so, they already can afford it), and they can pull out the "don't trust the government" thing when it suits them. And ignore the terrible effect of employer-paid healthcare on every US business that's not an insurance company.
But it was the Dems who caved on single payer this time around, and they did that even before they got to the table with the Republicans. Both sides are for sale.
Incidently... thanks to big fat Chris Christie (Jersey governor), my wife's insurance, formerly a benefit (teachers have typically had great benefits and lousy pay... as an engineer working at startups for the last few years, I'm usually dealing with good pay and lousy benefits), may cost as much as $6,000 this next year, to cover our family of four. As with many of the Republican governors, Christie has been working hard to effectively increase taxes on teachers, firemen, police, and other public employees, so he can afford the tax breaks previously given to the richest in the state.
Of course, what he's effectively done is killed an untold number of small businesses. It's an on-going thing... we lost a couple locally already this year (Woodstown/Pilesgrove), even a liquor store. Those are supposed to do well in bad economies...
The rich are already the richest (compared to the average) they've been since before the great depression -- additional income does nothing. Same with businesses -- additional income does absolutely nothing to grow a business. Additional demand for a product or service is the thing that grows a business. And these new sideways taxes have just ensured that a big chunk of the New Jersey middle class will have thousands less to spend this year and next.
That's important to understand -- the poor and the middle class spend all of their income; the rich don't. So additional lower end income boosts an economy, additional higher end income alone does nothing to the local economy. Boosting everyone's income boosts the economy, but only for a very short time.. things do ultimately stabilize around supply and demand... the price of things in short supply go up, and eventually, no one sees that income boost any longer. But I digress...
The single largest cause of bankruptcy in the USA? Medical expenses incurred by people who were "fully covered". That's about half of all bankruptcies.
And chances are, you'll never know if you're fully covered or not. And even if you are, you may have some hours of taking at your insurance company to ensure they agree. There seems to be a standard policy at most insurance companies to just randomly reject claims, even those they know are covered.
I do know from where I speak. My wife was diagnosed with breast cancer in April of 2010. She had very good health insurance for someone in the US these days... at least a one-time advantage of being a teacher in New Jersey. Six surgeries, 16 weeks of chemotherapy, 6 weeks of radiotherapy down, and one surgery left to go, we haven't spent a great deal out of pocket (but some financial fallout from her missing summer school in 2010). But again, this is about as good as it gets, and you'll find millions of people who thought they were just as secure, but weren't.
And diseases like this are curable, but extremely expensive to cure. We're probably over $50,000 on surgeries alone, probably in the same ballpark on chemo drugs... just Adriamycin (Doxorubicin) costs over a million per kg just to produce (expensive to administer too -- only by infusion).. and there are some cancer drugs over 10x that price. And once you have the cancer drug, there are a half-to-a-dozen additional drugs just to mitigate the effect of the cancer drug. That's also part of the problem -- such things are done just as well in other parts of the world, but the expenses are much better controlled.
But we did receive thousands of dollars in bills that should have been covered, and had to argue about them. And honestly, this insurance company was amazingly well run compared to others we've both dealt with.
There are several reasons... and that's ignoring the fact that state run healthcare, everywhere else in the world, has proven more effective, more universal, and about half the cost of the US system today.
The first big problem is profit motive. A government organization filled with government workers (eg, like Congress, the Senate, the Military, etc) is not expected to turn a profit. And no multi-million dollar CEO salaries. So, as long as it's well run (yes, that's an argument, but it applies equally to private enterprise), most of the money going into the system is used for healthcare. Not profits. Not advertising. Not lobbying... and generally, lobbying against the best interests of those served by the insurance industry. This is one big reason so much money is spent in the US for lower quality healthcare, versus more civilized countries.
Another issues is price negotiations... countless smaller healthcare companies cannot as effectively negotiate on the price of medicine and other supplies as can the entire country. And even if you did open up healthcare insurance to competition, there's no real guarantee it'll be actual competition. Other types of insurance are openly competitive on the surface, but they're really practicing artificial competition -- they bring you in at a discounted rate that looks great next to what you're paying, but a year later, you're paying more or less what you did at the old company. And they spend hundreds of millions in advertising to grab each others customers on a rotating basis. Geico alone spent over US$900 million last year in advertising... and they just do car insurance. That's over 10% of their expenses right there, before a single car is repaired. The health insurance business is an order of magnitude (at least) larger. You really want to see tens of billions in healthcare money going into these kinds of campaign?
The US healthcare system is also rabidly anti-business... unless you happen to be in the insurance business. I've even heard Righties saying this about "Obamacare" recently... but the main flaw in the Patient Protection and Affordable Care Act is that if left this as a problem... it didn't create it. In all these other countries, businesses are not funding healthcare. So they have an immediate competitive advantage over US manufacturing. Universal healthcare gets the business out of the picture, and the hand out of their wallets on this. If the right were actually pro business (they're "pro" donation, primarily, which is why they've sided with the insurance lobby and against every other business in the USA... all those other guys are spending their lobbying money on other things), if the left actually had the balls to not cave in to the right before the negotiations even begin (could be a medical problem, but ball exams are probably not covered). And of course, all of the people actually making the laws, both those caving in and those rising up against any notion of public healthcare, enjoy the best public healthcare system in the country... all Senators and Congresscritters get this.
And finally... healthcare as insurance is fundamentally the wrong answer. The insurance companies can insure things that are properly insured. My house and car, for example... something bad might happen, but chances are, no. But each and every person on the planet will eventually have a health issue of some kind. If that issue isn't immediate death, they will need medical treatment. Insurance is the wrong answer for something so guaranteed. And for something that needs constant treatment. Actual health care includes a strong prevention component... stress health rather than medicine, and you lower costs AND increase the quality of life. US healthcare is largely about fixing what's already broken.
The all you can eat model is a great one -- you know precisely what you're going to pay, and don't feel like bandwidth is wasted. But the whole cellular industry builds their huge profits based on mistaken expected use. So it's no great shock to see the data model move that way.
An equally fair rate would be pay-per-MB or whatever -- precisely metered, just as you might pay for electricity or gasoline. As long as the per-MB fee is fair, I have no problem with it. But the telcos do -- they want me to pay for the occasional high-use month, every month. So they put up big penalties (I mean, hell, that first MB over the limit doesn't cost them one iota more than that least MB under the limit) to fear the user into paying too much on a regular basis.
I'm on Verizon today, and fairly happy with it, largely because it can actually reach my house. T-Mo and Sprint, not so much.. and that's going to the case for Sprint and T-Mo in any rural area -- they just don't have the low frequency 850MHz band to penetrate foliage well. Also why Verizon works better than AT&T -- they both have 850MHz and 1900MHz slots, but AT&T's 3G takes both, while Verizon's runs only at 850MHz though the forest.
On the other hand, given that the price going from $30/month for unlimited to $30/month for 2GB, that sure smells like they're just plain doubling the price in practical term (I have often gone over 2GB, rarely get over 4GB). For less than that doubling, I can put up a roof antenna and repeater/booster, which ought to solve the T-Mo or AT&T problem just dandy. If only they didn't suck in other ways... What we need is a new company doing a reasonably priced data-only service. Sadly, it takes a very, very long time for new companies to build out a generally useful network -- both MetroPCS and Cricket are limited to a small number of generally urban-only areas so far.
Looks like you're either mistaken or lived under a rock. Most mobile apps devs speak Objective-C now and it has been this way for quite some time.
True only for very small values of "quite some time" and "most mobile apps devs". Apple iPhone apps have only been "native" in Objective-C for three years and three months... the SDK was released in March 2008.
I guess YOU lived under a rock prior to then. Blackberry ran Java MDIP as far back as 2003, their basis for mobile apps. Countless "not so smart" phones ran Java ME. Nokia, Microsoft, and PalmOS smartphones and PDAs, dating back to the early 1990s, were usually programmed in C or C++, though plenty of other options existed. And even today, you can program iPhone apps otherwise... for example, in Adobe ActionScript or Flex. Check this out: http://en.wikipedia.org/wiki/Mobile_application_development
There is and has been a ton of mobile development unrelated to Apple's 10% or so share of the global mobile market (factoring in smartphones, iPads, and iPods).
Google wanted the basic Android apps to be machine independent. So the same stuff largely runs on ARM and x86... even a few MIPS fans are starting to do things with Android. Java was available, and of course, the language itself couldn't be patented or copyrighted, and there was this GPLed implementation, etc. Sure looked like a good idea at the time.
I had the same kind of plan when developing a video/internet set-top box back in the late 1990s. We wanted applications, but they needed to be insulated from the hardware, so in the future, the hardware could change completely without messing with compiled apps.
Sure, they could have written their own language as well as their own VM... would that have been enough to keep out of the crosshairs of companies like Oracle who now think they have ownership over any software VM, thanks to Java? Maybe, maybe not.
Gates did no charity, either, until Melinda came along. And maybe he started thinking about his legacy -- with all that money, and all he's remembered for is Windows, Office, and the BSOD? Far cooler to say, wipe out Malaria. That kind of cash is the very definition of a superpower... hmmm. All this started about the time he saw "Spiderman", too....
Or, if you're using an Android phone, load a version of the camera app/OS/etc. that doesn't have the "decode IR" function. Oh, wait... won't be an issue for 20 years anyway, since Apple will own the technology until then. Apple's just terrible at establishing public standards (think "Firewire", which is the best they've managed to date), so they'll be the only users. And they'll sell it as an advantage to the iPhone/iPod, since the typical use will be getting messages about the thing you're photographing.
This is one big reason that all modern cameras hook to PCs as plain old USB storage class devices, rather than "some custom thing" that you need a private drive for. Also helps eliminate the need for drivers on Windows, Linux, and MacOS (or complaints by those who don't get supported). And of course, even if your old camera uses the terrible old SmartMedia cards, you can still read those. But if you find that a common activity, you really do need a new camera. Digital cameras are great today, and they manufacturers are looking hard for new compelling features. Ten years ago, they pretty much sucked.
Cellphones are different -- they get replaced all the time. They are planned to be obsolescent almost exactly two years from date of purchase (sucks if you're on a Canadian three year contract, but that's pretty much the reality of the situation).
Actually, the cellphones all use CMOS sensors, not CCDs. But either way. The single sensor cameras, cheap or very expensive, all have an IR bandstop filter and a Bayer or similar color filter (well, ok, if you're using a Foveon sensor, you don't have the Bayer filter)... these can be the same (eg, the Bayer microfilters also block IR) or not ("night mode" on consumer camcorders is usually achieved by rotating out the IR bandpass filter).
You could add IR sensitivity without any additional expense by altering the Bayer matrix to go RGBI (Red Green Blue IR), with the microfilters being IR blocking and IR passing, as needed. Software, then, would have to ignore the IR pixels, at least outside of "night mode". This would lower the image quality a bit, but hey, it's a consumer camcorder. The pattern might be altered in ways to minimize this, too. For example, Sony's "ClearVid" sensor doesn't use Dr. Bayer's original RGBG pattern, but rotates the sensor 45 degrees, and uses a pattern that's using something like 6x as many green pixels as red or blue. Drop a few IRs in there instead of a few of the greens, and you're be really hard pressed to tell the difference.
Of course, this scheme delivers only a fairly low resolution IR output. I'm sure they want the IR messages time multiplexed, but you'd still need a fairly large emitter to catch it on camera. And they'd have to be spaced well or synchronized, or they'd mess each other up.
The TDM data can obviously be itself detected by a single sensor -- this is how IR remotes work, and I used a slightly more sophisticated version of that once in a gaming robot design. But Apple's scheme is going to require some reasonable localization of the IR sources. After all, if you read the patent, only one of the functions is "no photos"... they also plan to use the IR becons to send information. For example, you're in a museum, you point your camera at an exhibit, and the IR messaging delivers some information (probably akin to a bar code) to let your phone deliver augmented reality information/links about the thing you're pointing at.
Some Canon camcorders from the early 2000's used IR ranging in addition to contrast... the idea was that the IR could very quickly come up with a ballpark range, the contrast AF could go there immediately, then slowly seek to attain the final focus.
This is a classic issue with camcorders. On a still camera, you focus as quickly as possible, then snap the photo -- not too much of an issue. But for video, you don't want focus seeking to be visible in the shot. This is why some of the early video modes in P&S still cameras had such bad autofocus; they still used the still algorithms.
So I put a very weak IR emitter on the backside of my IR filter. Or better still, a diffraction grating -- I still get IR light though the filter, but the information is lost. Visible light is not compromised. Either way, trivial to work around... and that's just with a minute of thinking. Well, ok, they're using some kind of TDM encoding scheme, so the diffraction would eliminate any point source, but if you have just one IR source in the image, the signal might still get through.
The other problem is already kind of practical -- how to build an IR sensor that can actually read a point source and not destroy the visible light image. Today's 5Mpixel on 1/4" cell cameras are already diffraction limited at about f2.5, and it's fairly typical for camera makers to use f3 lenses on these. Depending on just where you put the IR, you're already getting some pretty serious blurring. Not to mention the fact that IR has to be filtered before the sensors.
The simplest solution would be to modify the Bayer matrix, probably by adding IR filtered pixels in with the R, G, and B filtered pixels. That's going to hurt photo quality a bit, but hey, it's a cellphone... anyone who thinks it's a replacement for a real camera won't notice. More sophisticated would be a diachroic prism to split off the IR, replacing today's typical IR bandstop filter, but that's not so practical in a tiny device like this.
The on-device IR emitter reflection scheme (basic principle behind the "IR Blaster" used in some STBs to control VCRs, back in the 90s) needs quite a bit of power to deliver a readable signal off random room contents, and that's just in a typical livingroom. This would be totally impractical on a portable device, and it would fail entirely in a large room, much less outdoors.
There are both good and bad reasons for the crazy high price on an unlocked smartphone.
The good reason is simple: carriers (in theory) subsudize the price of the phone, in order to get consumers to agree to muli year contracts. If they don't pay, naturally you have to make up the difference.
The bad reason is a reflection of the immense power of the carriers, particularly in the USA. I mean, up until this point, Apple had only two actual customers ror the iPhone in the US: Verizon and AT&T. You know that neither is paying anything close to $649 for an iPhone.
In fact, carriers pay a fixed percentage of the MSRP for any phone they resell. If a manufacturer wants to set a reasonable price for a phone, they are welcome to do so.... but that would mean losing money on each telco sale, or not selling through the carriers at all.
What do I mean by a reasonable price? Something in keeping with other CE devices. A 32GB iPod Touch retails for $299. The 32GB iPhone costs maybe $50 more to make, so it should run $399, maybe $449 tops. Not $749.. that additional cost is basically a cell phone tax, levied by he carriers on anyone not playing the contract game.
Pretty happy with my Notion Ink Adam as an eReader. I don't need all week on a charge all day is fine. Indoors, backlit is handy in a dark room. Color is important for magazines, higher resolution useful for tech documents, as is first class PDF support and plenty of memory (32GB in extra flash, support for USB sticks, etc).
Outdoors, the Pixel Qi display is nice and crisp in full sunlight, tough not as high contrast as eink. With the backlight off, the device runs longer still. And of course, I can use Amazon, B&N, Adobe, and many other readers, no lock-in.
Sure, weight is an issue for long term hand holding. Like most tablets, the screen is glossy... the eBook folks seem to have the screen done better, most being a modest matte finish. The Adam has a bottom ridge along one side which makes hand holding a little easier, tough of course, not as crazy thin as a Kindle or iPad 2.
Hard drives are rated in bytes... base 10 bytes. That's MB and GB, not MiB or GiB. Sure, it favors the marketing, but it is also the proper unit for a hard drive, or a human. The power of two sizing is only appropriate for RAM and ROM integrated circuit devices. Don't blame your hard drive when its your OS using the inappropriate units.
Agreed. And the AmigaOS did both of these right.
There are some things you absolutely need at boot time. These happened sequentially in the Amiga's startup-sequence. Fairly early, the OS evolved a second initialization script, spawned as a separate process, for everything you might need eventually, but should not be a reason to not boot fast. And the user was very easily in charge of this -- all in nice, plain, everyday ASCII, not some deep hidden arcanity in the OS.
Second thing.. disc drives flushed their writes, and the OS recognized added drives. So there was never any reason to mount or unmount a portable drive -- plug it in, pull it out. That's far more useful than write-caching a write to some removable device for longer than it would take for a human to pull it out. Windows and Linux fail on this, completely, 26 years later.
Making things easier, particularly via GUIs, all kinds of formerly useful features of text shells, text languages, etc. have been tossed out. The GUI experts won the battle, and the didn't consult those who came before.
How about graphical shells... where's the pipe function? Redirection? Something so useful, simple, and fundamental... and missing from every GUI.
How about wordprocessors. Back in 1979, using Scribe and Emacs, I could put together a complex document in ways I haven't seen recently. For one, with few changes lines, I could change the format of citations, footnotes, index, etc.... maybe going from IEEE standard to ACS standard in 20 seconds. That's because the word processor understood such things at a high level, and could format these in your format of choice.
Compound documents, too.. I could wire a paper, then include it as a chapter in a book, without having to renumber or change much of anything.. and keep it as a separate file, for both evolving documents. I've seen WYSIWYG tools that supported this (Mentor's "Doc" comes to mind), but this fails completely in popular WPs like MS Word.
Are we not men?
One reason new TVs are streaming video over Ethernet -- they can do an end run around the cable companies that way. Netflix is more likely than HTML5 video for overall success, since most content companies are not going to stream media without DRM. Of course, Flash and other technologies (Netflix is using MS's Silverlight) can provide DRM as well. And that's not to suggest it all will be pay, either .. most of the free-to-view stuff on the net is DRM protected.
The reason is simple: BD players and televisions are rapidly moving away from the fixed-function device model. Once you have the ability to accept new software, things like "streaming media box" or "TiVo" become a simple matter of software, perhaps with the addition of extra storage. Economies of scale will make TVs, game machines, and BD players a far better deal than the stand-alone boxes, while at least as capable. I mean, look at the PS3 (when it's not being hacked, anyway). This shipped as a first generation BD player (and game console). It's been upgraded to every version of Blu-ray so far released, as a free download, including the net-enable stuff, PiP addition, and 3D. It's also a photo organizer, a video editor, a Netflix players, and a whole slew of things that didn't come in the box.
TV and BD is following. The reasons are simple: competition among the CE companies is fierce, and software add-ons are a cheap way to offer new features. Even if the next guy has it, your TV or DVD/BD player of a year or two back probably doesn't. So they potentially get upgrades just based on the strength of the extra software features.
If you look at the sales of BD players (likely to beat out DVD players by the year's end) versus BD media (usually 20-30%, varies a bit by release, versus DVD), it's clear that things like integrated Netflix are already having an impact on buyer's habits.
Yup... Elon Musk is on record as claiming that $1.1K per Kg is achievable in the foreseeable future.... presumably, whatever's the next cost reduction after Falcon 5 (the Falcon X platform, perhaps). SpaceX is already scaring the Chinese on costs, so they could definitely have an effect here.
As cool as the tech is here, it's clear that private companies are and will have a big impact. Especially with multi-government run agencies like the ESA, they're taking forever on concepts, which makes it unclear that, if they actually do ever produce, whether the new rocket/spaceship will actually be current, or saddled with the last 20-30 years of design-by-committee. There's something nice about the "just do it" approach that SpaceX is taking (ok, sure, they've been at it awhile too, but they do seem to be doing it).
Sukat, his eyes uncovered !