The trick to taking advantage of future processors like the ones architecture futurists such as David Patterson envisions when he talks about "manycore" chips is to make parallel programming easy. Making the programmer puzzle out the parallelism for himself isn't the way to do that. We already know pre-emptive threading is too difficult for most; putting pervasively parallel programming (PPP) in human hands would be even worse. A proper approach to PPP involves inventing a new language, not adding warts to C++, which already has more than enough of its own, thank you very much.
Peripheral manufacturers being unwilling to bottleneck through a common hardware interface are merely trying to protect their ability to differentiate their products from those of their competitors. This is just the way marketing works. If you want an enemy, you don't have to go as far as greed; capitalism is far enough.
Scoble is still too close to Microsoft to be free of its private bizarro dictionary. At Microsoft, "to innovate" is to make any positive change. For example, improving web browser error messages is something which should be (and all-too-often can be) done for any application; it was a new idea only when the first computer program was written. My favorite example of this particular entry in the Microsoft Dictionary is a few years ago when BillG crowed about how Windows had finally incorporated a TCP/IP stack, killing off several third party products. He called that innovation even though Mac OS had TCP/IP for years by then. The Microsoft Dictionary needs to hire the sage Inigo Montoya, who said: "You keep using that word. I do not think it means what you think it means."
I don't know exactly what you mean by "programmer", but if you mean "HTML jockey", then I wouldn't be surprised if there were a glut. Too many people christened themselves programmers during the dot bomb era. On the other hand, if you want to hire, say, a good Windows audio device driver engineer -- or even just anyone who seems capable of becoming one within a few months -- good luck.
These people are not arguing about the origin of the species, emotionally. They don't care about the truth; that's ancillary. What they care about is that they demonstrate their membership in a particular tribe. It's not a coincidence that the supporters of ID are also rural, love both types of music (country and western), fundamentalist, blue collar, NASCAR-watching, conservative, SUV-driving, uneducated, and Republican all at once. This particular tribe, even though it now controls both branches of elected government and more or less controls the third, still feels its territory constricting around it due to the inexorable march of science and technology. The tribe will get more fierce before it finally succumbs. The utter humiliation of the Bush Administration now occurring will definitely not be the tribe's last gasp.
> You're telling me that they don't have internal > documentation anyway? How do they develop > their own drivers, etc?
In many cases, no, they don't have documentation. What they have is a guy who knows the hardware and a guy who knows the firmware and a guy who hopes to know the software and they all talk until they are blue in the face.
Sort of. It's far too easy to romanticize the excellence and forget that OpenStep was already on Intel at the time NeXT took over Apple, and Darwin has stayed on Intel all along. Keeping everything else compiling is just not that difficult. It was not even a secret within the reasonably well connected developer community. The most impressive achievement is Rosetta, and even it's not as cool as the 68K emulator because Rosetta only has to run a subset of application code as opposed to all applications *and* the operating system.
Shipping to millions of real people who might actually use your software is vastly inferior to shipping to thousands of geeks who'll install it just to reinforce their identities. Ron really blew it.
Multi-fork files are a separate issue. The new regime at Apple doesn't like them any more than meta-data invented in Cupertino, and they are deprecated for the same reason ("internet compatibility"), but they are nevertheless a separate issue. It has taken Apple years to upgrade command line utilities to support them. It seems likely the next file system Apple evangelizes will not support multi-fork files. Pretend NTFS doesn't exist, because that's what Apple is doing.
The supposed abandonment of metadata mentioned in the Slashdot article is a red herring. After the NeXT acquisition of Apple, Apple announced it no longer liked metadata invented in Cupertino. Only metadata invented in Berkeley was desirable. The announcement manifested several ways and generally took the form of a general denunciation of metadata. However, it was never any such thing, and the fact that Apple had no idea what it was actually saying just added insult to injury. The ostensible reasoning behind all this is that Mac OS should be "compatible with the internet." Since credible Mac internet applications had supported such things as BinHex and MacBinary for years, what this amounts to is saying that POSIX API compatibility was more important than Macintosh user experience.
Slashdot Mirror
The trick to taking advantage of future processors like the ones architecture futurists such as David Patterson envisions when he talks about "manycore" chips is to make parallel programming easy. Making the programmer puzzle out the parallelism for himself isn't the way to do that. We already know pre-emptive threading is too difficult for most; putting pervasively parallel programming (PPP) in human hands would be even worse. A proper approach to PPP involves inventing a new language, not adding warts to C++, which already has more than enough of its own, thank you very much.
Peripheral manufacturers being unwilling to bottleneck through a common hardware interface are merely trying to protect their ability to differentiate their products from those of their competitors. This is just the way marketing works. If you want an enemy, you don't have to go as far as greed; capitalism is far enough.
Scoble is still too close to Microsoft to be free of its private bizarro dictionary. At Microsoft, "to innovate" is to make any positive change. For example, improving web browser error messages is something which should be (and all-too-often can be) done for any application; it was a new idea only when the first computer program was written. My favorite example of this particular entry in the Microsoft Dictionary is a few years ago when BillG crowed about how Windows had finally incorporated a TCP/IP stack, killing off several third party products. He called that innovation even though Mac OS had TCP/IP for years by then. The Microsoft Dictionary needs to hire the sage Inigo Montoya, who said: "You keep using that word. I do not think it means what you think it means."
I don't know exactly what you mean by "programmer", but if you mean "HTML jockey", then I wouldn't be surprised if there were a glut. Too many people christened themselves programmers during the dot bomb era. On the other hand, if you want to hire, say, a good Windows audio device driver engineer -- or even just anyone who seems capable of becoming one within a few months -- good luck.
These people are not arguing about the origin of the species, emotionally. They don't care about the truth; that's ancillary. What they care about is that they demonstrate their membership in a particular tribe. It's not a coincidence that the supporters of ID are also rural, love both types of music (country and western), fundamentalist, blue collar, NASCAR-watching, conservative, SUV-driving, uneducated, and Republican all at once. This particular tribe, even though it now controls both branches of elected government and more or less controls the third, still feels its territory constricting around it due to the inexorable march of science and technology. The tribe will get more fierce before it finally succumbs. The utter humiliation of the Bush Administration now occurring will definitely not be the tribe's last gasp.
You must not be a Mac person if you're realizing this only now.
Either that or my sarcasm detector is offline.
It's Orlando, after all. They must have been wishing they were Manhattan.
> You're telling me that they don't have internal
> documentation anyway? How do they develop
> their own drivers, etc?
In many cases, no, they don't have documentation. What they have is a guy who knows the hardware and a guy who knows the firmware and a guy who hopes to know the software and they all talk until they are blue in the face.
Sort of. It's far too easy to romanticize the excellence and forget that OpenStep was already on Intel at the time NeXT took over Apple, and Darwin has stayed on Intel all along. Keeping everything else compiling is just not that difficult. It was not even a secret within the reasonably well connected developer community. The most impressive achievement is Rosetta, and even it's not as cool as the 68K emulator because Rosetta only has to run a subset of application code as opposed to all applications *and* the operating system.
Damn straight. Money is the only valuable goal. It makes everyone who has it happy.
What do you mean by "already"? Graphing Calculator has been part of Mac OS for more than ten years now.
Shipping to millions of real people who might actually use your software is vastly inferior to shipping to thousands of geeks who'll install it just to reinforce their identities. Ron really blew it.
Multi-fork files are a separate issue. The new regime at Apple doesn't like them any more than meta-data invented in Cupertino, and they are deprecated for the same reason ("internet compatibility"), but they are nevertheless a separate issue. It has taken Apple years to upgrade command line utilities to support them. It seems likely the next file system Apple evangelizes will not support multi-fork files. Pretend NTFS doesn't exist, because that's what Apple is doing.
The supposed abandonment of metadata mentioned in the Slashdot article is a red herring. After the NeXT acquisition of Apple, Apple announced it no longer liked metadata invented in Cupertino. Only metadata invented in Berkeley was desirable. The announcement manifested several ways and generally took the form of a general denunciation of metadata. However, it was never any such thing, and the fact that Apple had no idea what it was actually saying just added insult to injury. The ostensible reasoning behind all this is that Mac OS should be "compatible with the internet." Since credible Mac internet applications had supported such things as BinHex and MacBinary for years, what this amounts to is saying that POSIX API compatibility was more important than Macintosh user experience.