So if I install and Free and Open Source OS and applications on my PC does the BSA count that as "lost sales?"
If not, why not? It would fit in perfectly with their perverse logic, and it would nicely light the blue touchpaper on all sorts of issues that would make it into mainstream politics.
Option 3: A smart phone that doesn't expect me to be a falsebook/twatter obsessive and just works as a smart phone with some computing & browsing (and farting if I choose) capabilities.
I don't know. The 80's were the early days of computing.
No, the 50s were the early days of computing. Programs were generally written in machine code and ran on the bare metal. Compilers for a few high-level languages were just being developed. LISP hadn't arrived. Only the very privileged had access to computers.
By the 1980s personal computers (microcomputers) were mass-market and high-level languages (C, Pascal, Modula-2, FORTH, BASIC) were replacing machine code and assembly language programming. Computers came with operating systems (often only a few kB and in ROM). They had graphics and sound and mass storage. RISC had come along and computer power mushroomed.
If you're writing code for a fixed-point DSP chip (used in most home receivers) the majority of RISC processors (ARM, MIPS, most Power) or any microcontroller, working in integer space will be much faster since floating point has to be emulated or issued to a separate floating point unit which requires extra overhead to move data to/from it.
All of the main RISC architectures: MIPS, SPARC, POWER, PowerPC, Alpha, PA-RISC do (or did) floating-point in hardware since the 1980s. Almost all of them had move to having the FPU and CPU on the same chip by the early 1990s. Usually the ADDs and ofter the MULs could be done in a single clock cycle and often in pairs or quadruplets. That's what made them all at least 10 times as fast the x86 CPUs of the day at the same clock frequency.
Low power, low cost RISC CPUs found in embedded devices such as PowerPC and ARM may not have floating-point hardware and need emulation libraries in software. I work on devices that use PowerPC processors. The lower-end products use integer-only CPUs, whereas the higher-end stuff have the CPUs with the integrated FPU. You can run the code from the low-end machine on the high-end, but not vice versa.
I bought an m100 (2MB, 16MHz, 160x160 greyscale display) on implulse for £50 when I bought my first mobile phone back in 2000.
The m100 was fun and surprisingly useful. It had a web browser that worked pretty well. I used to be able to sit in the pub with it and browse using the modem in the mobile phone over the IR link at a whopping 9600bps.
What was really cool, though, was the SDK which you could download along with all the documentation for free from Palm and use on whatever system you liked (Linux in my case) and develop whatever you liked for it. There was a pretty good emulator as well that you could test on. I spent a couple of hours one Saturday afternoon writing a little analogue clock program for it using the 16-bit floating-point. I had to write the trig. functions myself because the OS and SDK didn't have any.
Later I bought a Tungsten T3 which had a 400MHz ARM processor and a 480x320 16-bit colour display. The SDK was available for that but the cross-platform emulator went away. There was only a flaky "simulator" that only ran under Windows. I tried to get it running under Wine, but no banana. At that point, I gave up.
I've got a Tungsten E2 as well. They all still work and I used the E2 as an MP3 player and notepad. I broke the card slot on the T3.
I have a wireless card for the Tungsten ones and they have web browsers, but they're really flaky. Not reliable enough to use. No more sitting in the pub browsing my CVS on sourceforge...
Software is the most complex thing that human beings make. It is very difficult indeed to guarantee that software works perfectly. What is possible, though, is to make software as best we can to the highest standards, and promise to fix or mitigate any defects found in operation.
To "know exactly what you're getting" is very difficult (impossible?). It is only possible to make general assertions about what is known to work. I'm afraid that's the best we can do with software at present.
To be fair, in the real world (i.e. working for a living), if you come across an obscure function written in a "clever" way it is sometime a far better use of everyone's time to ask the person that wrote it what it does.
Otherwise you risk missing some subtlety about why it needed to be that way (and not another) and spending lots of time and effort writing a solution to the problem in hand that may work superficially but not in detail.
The Linux kernel isn't fun any more. It's corporate now. It's mature. It has nowhere left to go. We need something new.
You can't choose your parents, that's why civillised countries provide free education for children, free (or subsidised) medical care and a small amount of money for food and clothes if the families are very poor.
Parents have a great responsibility to their children, but as we all know, many irresponsible and incapable people have children.
Americans are far more likely to be ignorant religious loonies who refuse to believe scientific fact in favour of archaic superstition and myth and profess to follow the word of a deity, meanwhile trying as hard as they can to ensure that the poor and sick don't get the help they need.
Can someone please explain why America is like this?
I'll see your Wall Street and raise you a Canary Wharf! I hear they are going to make a big pyramid of stock certificates and Bank of England £20 notes all the way up to the moon, for the first ever British manned lunar expedition.
How are other parties supposed to rise up and represent the people who share their values if the citizens won't vote for them "because they can't win?"
Hear, hear.
Put more plainly, although other parties may not have a chance of winning outright this time around (or even next time etc.) by their very existence and presence they let alternative views get aired.
A vote for these parties is not wasted.
A vote for either of the major two parties is a vote for the status quo and therefore stagnation.
The mainstream policies of today were considered "loony" 50 years ago, radical 30 years ago and progressive 20 years ago. 10 years ago they looked fresh and exciting and "a real possible alternative."
Political change is slow, but voting for the more progressive less popular parties lets these ideas get out into the mainstream sooner.
The huge amount of instruction-level parallelism (dependent on a very good compiler) really seems like the best way to do things. It's too bad it doesn't work out in practice.
It was never going to work out in practice. "A very good compiler" was a joke - a euphamism - for an impossible compiler. There's only so much scheduling that can be done at compile time. Due to the very nature of computation, there are many things that can only be decided dynamically, at run time, depending upon which paths of execution the code takes under encountered circumstances.
The only way to get maximum theoretical performance out of a design light itanium would be to have a program compiled in such a way that all possible paths of execution are executed at once in parallel. Clearly, this is impossible. It's impossible on any processor.
The Alpha, which the itanic was designed to kill, did things properly (and completely differently). It was well ahead of its time. Marketing killed the Alpha. It killed MIPS, and PA-RISC too.
Surely this is just a tool for the copyright infringement of the RIAA's music?
I thought you were joking about god deliberately putting the star there for the benefit of the pagan astrologers. Sorry.
Why did God make the half-lives of U-238 and U-235 just so?
You'll just have to use cracked pirate versions or that smelly, dirty, un-American long-haired-hippy Free Software.
So if I install and Free and Open Source OS and applications on my PC does the BSA count that as "lost sales?"
If not, why not? It would fit in perfectly with their perverse logic, and it would nicely light the blue touchpaper on all sorts of issues that would make it into mainstream politics.
Option 3: A smart phone that doesn't expect me to be a falsebook/twatter obsessive and just works as a smart phone with some computing & browsing (and farting if I choose) capabilities.
Keep taking the pills.
And is the number of semis integral? What about the DVDs? Do they sell them by fractions?
With a laundry basket to catch it in after it's been through the reader.
Yes, and much lower quality...
I used to get mine on cassette tapes.
What's not "digital" about CDs, DVDs or flash memory?
I don't know. The 80's were the early days of computing.
No, the 50s were the early days of computing. Programs were generally written in machine code and ran on the bare metal. Compilers for a few high-level languages were just being developed. LISP hadn't arrived. Only the very privileged had access to computers.
By the 1980s personal computers (microcomputers) were mass-market and high-level languages (C, Pascal, Modula-2, FORTH, BASIC) were replacing machine code and assembly language programming. Computers came with operating systems (often only a few kB and in ROM). They had graphics and sound and mass storage. RISC had come along and computer power mushroomed.
If you're writing code for a fixed-point DSP chip (used in most home receivers) the majority of RISC processors (ARM, MIPS, most Power) or any microcontroller, working in integer space will be much faster since floating point has to be emulated or issued to a separate floating point unit which requires extra overhead to move data to/from it.
All of the main RISC architectures: MIPS, SPARC, POWER, PowerPC, Alpha, PA-RISC do (or did) floating-point in hardware since the 1980s. Almost all of them had move to having the FPU and CPU on the same chip by the early 1990s. Usually the ADDs and ofter the MULs could be done in a single clock cycle and often in pairs or quadruplets. That's what made them all at least 10 times as fast the x86 CPUs of the day at the same clock frequency.
Low power, low cost RISC CPUs found in embedded devices such as PowerPC and ARM may not have floating-point hardware and need emulation libraries in software. I work on devices that use PowerPC processors. The lower-end products use integer-only CPUs, whereas the higher-end stuff have the CPUs with the integrated FPU. You can run the code from the low-end machine on the high-end, but not vice versa.
And is Sergei behind it?
I bought an m100 (2MB, 16MHz, 160x160 greyscale display) on implulse for £50 when I bought my first mobile phone back in 2000.
The m100 was fun and surprisingly useful. It had a web browser that worked pretty well. I used to be able to sit in the pub with it and browse using the modem in the mobile phone over the IR link at a whopping 9600bps.
What was really cool, though, was the SDK which you could download along with all the documentation for free from Palm and use on whatever system you liked (Linux in my case) and develop whatever you liked for it. There was a pretty good emulator as well that you could test on. I spent a couple of hours one Saturday afternoon writing a little analogue clock program for it using the 16-bit floating-point. I had to write the trig. functions myself because the OS and SDK didn't have any.
Later I bought a Tungsten T3 which had a 400MHz ARM processor and a 480x320 16-bit colour display. The SDK was available for that but the cross-platform emulator went away. There was only a flaky "simulator" that only ran under Windows. I tried to get it running under Wine, but no banana. At that point, I gave up.
I've got a Tungsten E2 as well. They all still work and I used the E2 as an MP3 player and notepad. I broke the card slot on the T3.
I have a wireless card for the Tungsten ones and they have web browsers, but they're really flaky. Not reliable enough to use. No more sitting in the pub browsing my CVS on sourceforge...
I, for my part, am a libertarian
And thus you disqualify yourself from any rational discussion.
Software is the most complex thing that human beings make. It is very difficult indeed to guarantee that software works perfectly. What is possible, though, is to make software as best we can to the highest standards, and promise to fix or mitigate any defects found in operation.
To "know exactly what you're getting" is very difficult (impossible?). It is only possible to make general assertions about what is known to work. I'm afraid that's the best we can do with software at present.
Ethics are relative.
Ethics are thouth of Thuffolk, not to be confused with Thuthicks, and north of Kent.
To be fair, in the real world (i.e. working for a living), if you come across an obscure function written in a "clever" way it is sometime a far better use of everyone's time to ask the person that wrote it what it does.
Otherwise you risk missing some subtlety about why it needed to be that way (and not another) and spending lots of time and effort writing a solution to the problem in hand that may work superficially but not in detail.
The Linux kernel isn't fun any more. It's corporate now. It's mature. It has nowhere left to go. We need something new.
You can't choose your parents, that's why civillised countries provide free education for children, free (or subsidised) medical care and a small amount of money for food and clothes if the families are very poor.
Parents have a great responsibility to their children, but as we all know, many irresponsible and incapable people have children.
Americans are far more likely to be ignorant religious loonies who refuse to believe scientific fact in favour of archaic superstition and myth and profess to follow the word of a deity, meanwhile trying as hard as they can to ensure that the poor and sick don't get the help they need.
Can someone please explain why America is like this?
I'll see your Wall Street and raise you a Canary Wharf! I hear they are going to make a big pyramid of stock certificates and Bank of England £20 notes all the way up to the moon, for the first ever British manned lunar expedition.
How are other parties supposed to rise up and represent the people who share their values if the citizens won't vote for them "because they can't win?"
Hear, hear.
Put more plainly, although other parties may not have a chance of winning outright this time around (or even next time etc.) by their very existence and presence they let alternative views get aired.
A vote for these parties is not wasted.
A vote for either of the major two parties is a vote for the status quo and therefore stagnation.
The mainstream policies of today were considered "loony" 50 years ago, radical 30 years ago and progressive 20 years ago. 10 years ago they looked fresh and exciting and "a real possible alternative."
Political change is slow, but voting for the more progressive less popular parties lets these ideas get out into the mainstream sooner.
What the hell is going on with our country?!
You gave up to chase stock markets instead.
The huge amount of instruction-level parallelism (dependent on a very good compiler) really seems like the best way to do things. It's too bad it doesn't work out in practice.
It was never going to work out in practice. "A very good compiler" was a joke - a euphamism - for an impossible compiler. There's only so much scheduling that can be done at compile time. Due to the very nature of computation, there are many things that can only be decided dynamically, at run time, depending upon which paths of execution the code takes under encountered circumstances.
The only way to get maximum theoretical performance out of a design light itanium would be to have a program compiled in such a way that all possible paths of execution are executed at once in parallel. Clearly, this is impossible. It's impossible on any processor.
The Alpha, which the itanic was designed to kill, did things properly (and completely differently). It was well ahead of its time. Marketing killed the Alpha. It killed MIPS, and PA-RISC too.