Google's announced a number of times that their long-term plan is to converge Android with Chrome OS in the long term.
If so, what's the holdup? It should be abundantly clear by now that the market for a browser-only laptop is vanishingly small. To sell Chromebook, Google needs to ship the Chromebook with Android and add a touchscreen so that the device is not perceived as inferior to a tablet + bluetooth keyboard. It's really time to stop beating the dead cloud-only horse.
Why ship Chromebook with Android? Not for any technical reason... KDE would be a better choice technically... but because the market knows Android and wants it.
Why google is maintaining 2 different OSes? Microsoft has gone in different direction to make same OS for its phone, Table and PC.
Excuse me, but which Googlers modded down that fair question? Not that I am any fan of Microsoft, quite the contrary, but I am very definitely not a fan of evil Googlers. At least, Microsofties are unabashedly evil.
If you ask designers about fonts hinting they say "yuck, ugly"
Only a clueless designer would say that, or an iFan defending Apple's inexplicable decision to punt on hinting. In fact, font hinting is entirely about improving the esthetics of fonts a lower point sizes. Even at 20 points (not generally considered a small font) you can see visible improvement in some cases... all vertical strokes of a lower case m guaranteed to be exactly the same width in pixels. How wouldn't want this? Oh right, an iFan defending Apple's inexplicable decision to punt on hinting.
And how's that variable screen resolution working out for Android software? Not very well.
Works great. I've got three Android devices now: 3.5", 7" and 10". Some games actually work perfectly on the whole range. Others require the tablet form factor, or are aimed at the phone therefore look sparse on the tablets, but they work. Rendering is perfect because scaling is handled by the GPU. Pretty well all the games are 3D rendered, no 80's looking pixel-shooting sprites as Apple seems to think people should want. On this, Android gets a slam dunk. I've heard occasional whining about Android multiresolution layout, but only from Apple fans.
Well, for one thing it's only 7 inch (1280Ã--800) rather than the 9.7 inch (2048Ã--1536) of an iPad 3.
iPad 3 is a power sucking monstrosity. The only reason Apple quadrupled the resolution was because of the scatterbrained design decision to let applications depend on fixed resolution. Exacerbated by the idiotic lack of hinting in Apple's font engine, so higher resolution is needed just to get equivalently sharp characters that Android gets with proper hinting. Oh, and the fixed resolution idiocy came back to bite Apple again with the iPhone 5 - forcing the funny looking too-long-and-skinny form factor just to keep the 640 dot display width. And letterboxing! Who was asleep at the wheel in the Apple's engineering department anyway? Well I'm not complaining of course. Strategy like this is the best and fastest way to transform Apple from a growth stock to a shrink stock. Which couldn't happen to a nicer company.
I put this in the same category as Apple refusing to adopt other standards, such as USB power. Reinforcing its reputation as an operation that doesn't play well with standards.
And some Apple spinmod just reinforced Apple's reputation for not playing well with people.
I put this in the same category as Apple refusing to adopt other standards, such as USB power. Reinforcing its reputation as an operation that doesn't play well with standards.
Steve Ballmer simply multiplied the price by the bazillion units the powerpoint slide says will ship and wow, it just makes Microsoft's market cap way bigger than Apple's. So obviously the price is justified.
It should cost much more for Amazon, Apple, Microsoft, Google, IBM, etc to file a patent vs an individual tinkerer in his home.
Bad idea. The deep pocketed corps will just buy the tinkerer and nothing will change. The cleanest, fastest, most effective fix is simply banning software patents. In other words, just put things back the way they were before this sorry chapter in American business history.
Someone is going to have to own the enterprise space. RIM and Microsoft are far better choices than Apple but both at least today have inferior product eco-systems.
In case you haven't noticed, enterprises are going all cloudy. The leaders in that space are Google and Amazon, who just happen to offer wildly popular tablet products as well. 2+2.
There are some niches, like airline pilots, who actually gain a real, tangible benefit from a compact, touchscreen computer, because it replaces their flight manuals. At the moment, the iPad is fitting that bill. In the future, we should expect (and hope) that there are alternatives.
Which practice did I advocate that reduces readability, takes longer to write, or betrays a lack of familiarity with modern compiler design.
You said: "you shouldn't assume a compiler is going to do something for you unless that something an optimization you are specifically turning on." That is incorrect. You should know in some detail what your compiler will do in its default configuration, then if that is not suitable to the task at hand, adjust its behavior. Document this behavior where it matters and is not obvious, or explicitly supply a compiler setting where you feel the default might be obvious to a typical practitioner, but try not to obfuscate your build setup with unnecessary settings.
For example, In the old days compilers did not unroll loops, so skilled programmers did it by hand. Today, the compilers you are working with probably do a better job of it than you can by hand. For example, the compiler can take the marchine architecture into account when deciding whether to unroll or not, or how deep to unroll. Always unrolling loops sometimes slows code down by increasing pressure on L1 cache. Best practice today is to simply write your performance critical loop as a simple, readable loop.
Another example: you should not always lift constant expressions out of loops. Modern compilers do that well, so unless it improves readability, let the compiler do it. Or you might need to do it explicitly if the expression is not opaque to automatic optimization, for example if it accesses a memory object that you know is constant but the compiler doesn't. (Better: declare the object constant.)
If you are writing in C/C++ then you are probablly using Microsoft VCC or GCC, both of which have sophisticated optimizers. If you are an embedded programmer then it is possible you are working with some ancient, substandard vendor compiler that might force you to compensate for its weaknesses. But I doubt that case applies to you.
Best optimization for Perl: use a different language.
"Premative optimization is the root or all evil". Optimize where it is useful and necessary, otherwise you only waste your time and sacrifice readability. And chances are, you betray less than ideal familiarty with the state of modern compiler design. You are using a modern compiler, aren't you?
Slashdot Mirror
who is gonna want an OS that is constantly blowing through bandwidth and is worthless without it when their ISPs are being stingy
And who wants a laptop that is basically a brick when the net connection drops?
Google's announced a number of times that their long-term plan is to converge Android with Chrome OS in the long term.
If so, what's the holdup? It should be abundantly clear by now that the market for a browser-only laptop is vanishingly small. To sell Chromebook, Google needs to ship the Chromebook with Android and add a touchscreen so that the device is not perceived as inferior to a tablet + bluetooth keyboard. It's really time to stop beating the dead cloud-only horse.
Why ship Chromebook with Android? Not for any technical reason... KDE would be a better choice technically... but because the market knows Android and wants it.
Why google is maintaining 2 different OSes? Microsoft has gone in different direction to make same OS for its phone, Table and PC.
Excuse me, but which Googlers modded down that fair question? Not that I am any fan of Microsoft, quite the contrary, but I am very definitely not a fan of evil Googlers. At least, Microsofties are unabashedly evil.
Not trying to derail a good rant but any text small enough to benefit from font hinting would be too small to read on an ios device.
Utter bullshit. Even 20 pt fonts benefit from hinting. I am sure you can read a 12 pt quite easily on your magical retina display, or are you blind?
Can't we believe what Apple says?
On this, Android gets a slam dunk. I've heard occasional whining about Android multiresolution layout, but only from Apple fans.
Guess what. All of your whining we've only ever heard from Android fans.
Oh, I'm not whining about Apple's design blunders. On the contrary, I applaud them. Apple needs to keep making as many blunders as possible.
Way to put on the richly deserved smackdown.
If you ask designers about fonts hinting they say "yuck, ugly"
Only a clueless designer would say that, or an iFan defending Apple's inexplicable decision to punt on hinting. In fact, font hinting is entirely about improving the esthetics of fonts a lower point sizes. Even at 20 points (not generally considered a small font) you can see visible improvement in some cases... all vertical strokes of a lower case m guaranteed to be exactly the same width in pixels. How wouldn't want this? Oh right, an iFan defending Apple's inexplicable decision to punt on hinting.
And how's that variable screen resolution working out for Android software? Not very well.
Works great. I've got three Android devices now: 3.5", 7" and 10". Some games actually work perfectly on the whole range. Others require the tablet form factor, or are aimed at the phone therefore look sparse on the tablets, but they work. Rendering is perfect because scaling is handled by the GPU. Pretty well all the games are 3D rendered, no 80's looking pixel-shooting sprites as Apple seems to think people should want. On this, Android gets a slam dunk. I've heard occasional whining about Android multiresolution layout, but only from Apple fans.
Well, for one thing it's only 7 inch (1280Ã--800) rather than the 9.7 inch (2048Ã--1536) of an iPad 3.
iPad 3 is a power sucking monstrosity. The only reason Apple quadrupled the resolution was because of the scatterbrained design decision to let applications depend on fixed resolution. Exacerbated by the idiotic lack of hinting in Apple's font engine, so higher resolution is needed just to get equivalently sharp characters that Android gets with proper hinting. Oh, and the fixed resolution idiocy came back to bite Apple again with the iPhone 5 - forcing the funny looking too-long-and-skinny form factor just to keep the 640 dot display width. And letterboxing! Who was asleep at the wheel in the Apple's engineering department anyway? Well I'm not complaining of course. Strategy like this is the best and fastest way to transform Apple from a growth stock to a shrink stock. Which couldn't happen to a nicer company.
And only people with Apple products will be able to buy food.
I put this in the same category as Apple refusing to adopt other standards, such as USB power. Reinforcing its reputation as an operation that doesn't play well with standards.
And some Apple spinmod just reinforced Apple's reputation for not playing well with people.
USB power has a fundamental limitation...
Oh really. Then why doesn't it limit my quad core Nexus 7? Are you saying that Apple's power design is bad?
I put this in the same category as Apple refusing to adopt other standards, such as USB power. Reinforcing its reputation as an operation that doesn't play well with standards.
Steve Ballmer simply multiplied the price by the bazillion units the powerpoint slide says will ship and wow, it just makes Microsoft's market cap way bigger than Apple's. So obviously the price is justified.
I hope you realize that when you make fun of MS-Bob you make fun of Bill Gates' wife.
Jail terms for those guilty of reckless endangement by selling or using medical devices running Windows.
It should cost much more for Amazon, Apple, Microsoft, Google, IBM, etc to file a patent vs an individual tinkerer in his home.
Bad idea. The deep pocketed corps will just buy the tinkerer and nothing will change. The cleanest, fastest, most effective fix is simply banning software patents. In other words, just put things back the way they were before this sorry chapter in American business history.
If you can find any name that's not related to murdering your wife, go for it. Bonus points if it's catchy.
I though they were changing it to "Open Journaling FileSystem".
Ninafs has been suggested and I for one would find that entirely appropriate.
Someone is going to have to own the enterprise space. RIM and Microsoft are far better choices than Apple but both at least today have inferior product eco-systems.
In case you haven't noticed, enterprises are going all cloudy. The leaders in that space are Google and Amazon, who just happen to offer wildly popular tablet products as well. 2+2.
There are some niches, like airline pilots, who actually gain a real, tangible benefit from a compact, touchscreen computer, because it replaces their flight manuals. At the moment, the iPad is fitting that bill. In the future, we should expect (and hope) that there are alternatives.
Eh, there are already alternatives.
I still get accused of being a paid shill because I think Microsoft is going the right thing with Windows 8.
I happen to agree with you on this point, because the right thing for Microsoft to do is to destroy itself as quickly as possible.
Microsoft should by all means copy Apple's walled garden model. Then they can both proceed straight to hell, holding hands.
Which practice did I advocate that reduces readability, takes longer to write, or betrays a lack of familiarity with modern compiler design.
You said: "you shouldn't assume a compiler is going to do something for you unless that something an optimization you are specifically turning on." That is incorrect. You should know in some detail what your compiler will do in its default configuration, then if that is not suitable to the task at hand, adjust its behavior. Document this behavior where it matters and is not obvious, or explicitly supply a compiler setting where you feel the default might be obvious to a typical practitioner, but try not to obfuscate your build setup with unnecessary settings.
For example, In the old days compilers did not unroll loops, so skilled programmers did it by hand. Today, the compilers you are working with probably do a better job of it than you can by hand. For example, the compiler can take the marchine architecture into account when deciding whether to unroll or not, or how deep to unroll. Always unrolling loops sometimes slows code down by increasing pressure on L1 cache. Best practice today is to simply write your performance critical loop as a simple, readable loop.
Another example: you should not always lift constant expressions out of loops. Modern compilers do that well, so unless it improves readability, let the compiler do it. Or you might need to do it explicitly if the expression is not opaque to automatic optimization, for example if it accesses a memory object that you know is constant but the compiler doesn't. (Better: declare the object constant.)
If you are writing in C/C++ then you are probablly using Microsoft VCC or GCC, both of which have sophisticated optimizers. If you are an embedded programmer then it is possible you are working with some ancient, substandard vendor compiler that might force you to compensate for its weaknesses. But I doubt that case applies to you.
Best optimization for Perl: use a different language.
"Premative optimization is the root or all evil". Optimize where it is useful and necessary, otherwise you only waste your time and sacrifice readability. And chances are, you betray less than ideal familiarty with the state of modern compiler design. You are using a modern compiler, aren't you?