Why? Look at how successful patented derivatives of his ideas have been, or even patented ideas from the same era. Most fizzled to nothing. He may have made some money, but the most likely effect of patenting things in C would have been to make us all use some Pascal derivative. If UNIX had been patented, then we'd probably be using a CP/M clone.
Had he been a patent hound, he'd have died a rich man.
I doubt it. Most of what he created was part of the 'worse is better' philosophy. Given the choice between C and Algol, most people would have picked Algol in the '80s, but C compilers were cheaper (or free), so they went with C. The same with UNIX. There were much better operating systems around, but they were either expensive, required expensive hardware or, in many cases, both. C and UNIX were both good enough and free. That usually beats really good and expensive. If they'd tried to make a large profit from either, they'd have failed. In fact, the BSD lawsuit (AT&T vs UCB) came close to killing UNIX.
Why is this bad? I'd rather see ads for things I like and might consider buying than scattershot ads for shit I'd never use.
I take it you missed the Slashdot story a couple of years ago about Amazon giving different prices based on the browser you use? Targeted advertising isn't where it ends. Companies like Google and Facebook often record enough information to tell how much you shop around before buying things, for example. It doesn't take much data mining to work out how much you'd be willing to pay for a specific product. Next time you visit an online store, you may find it's exactly that amount. Meanwhile, it costs 20% less to the next visitor...
You and I know that 'publicly' means 'on any server you don't control'. The masses think 'publicly' means 'on a site that doesn't have access control'. People don't realise that 'private' communications on FaceBook may not be seen by the world at large, but that are seen by anyone who pays FaceBook (including the CIA, which seems a bit silly because you'd have to be a total moron to use something like FaceBook if you're planning the sort of things that the CIA cares about).
Are they? Recruiters seem to still find me, even though I'm not on any social networking sites (unless you count Slashdot). The most recent company to invite me for interview was... Facebook.
It doesn't matter. Judging by the trends in HP CEOs, Darl will only last 2 months. He will be replaced by Kim Jong Il, who will last a mere 3 weeks (and leave with a $100m golden parachute). The next CEO will be fired before she arrives in the building, and the one after her will be given her golden parachute before being offered the job. After that, things are going to get really bad...
My old Brother laser supported PostScript, but printing PCL was a lot faster because rasterising a complex PostScript document (e.g. the output from LaTeX) on a 50MHz MIPS processor in the printer took a long time. I've recently bought a cheap Dell all-in-one colour LED printer. It can print from and scan to PDFs, and doesn't even need a computer: you can print PDFs directly from a USB device. It also talks SMB and FTP, so I can scan to an SMB or FTP server.
The faster P4 chips were rarely faster than comparative Athlons in real workloads. They did very well in some benchmarks, but the cost of a branch predictor miss was insane (something on the order of 150 instructions), which killed their real-world performance. The Opteron and HyperTransport gave AMD an edge in 2+ socket systems and a big lead in 4+ socket systems for quite a while.
AMD was very late to focus on power consumption. This was probably due to their chips not being quite the space heaters that Intel chips were. Going from the late P4s to the Core 1 was a huge leap in power consumption for Intel, but not such a big leap for AMD. Since then, Intel has pushed power usage down, but AMD didn't really have anything competitive with Atom until the recent Fusion line.
Wait: you seriously expect a party that is part of a coalition to not make compromises?
No, but then I would expect people to read my post and the context in which it was written before posting a reply, meaning that I seem to have unreasonably high expectations...
Superheroes are people with special powers they use for good.
*nerd hat on*
What about Batman or the Green Arrow? Unless 'having lots of money' is a special power, they didn't have any. They just put on body armour and went and helped people - making them a much better role model than someone like Superman who is basically invulnerable.
I would say we need these guys a lot more than we need thugs assaulting each other or random people in the streets.
I only half agree. If groups of thugs want to kill each other off, I don't have a problem with that. It's only when they start attacking other people that it's a problem...
I feel represented. The Liberal Democrats are largely pushing policies I agree with, although they're limited by the government being bankrupt and the Tories having 3/5 of the seats that the coalition controls. My Labour MP (who I voted against) feels he has something to prove and has been very responsive since being elected. My Plaid Cymru MEP has been working hard to return IP laws to more sane levels.
They expected that a party with 40% of the seats controlled by the coalition would be the dominant force in policy making,would not have to make any compromises, and would be able to do everything that they said in their manifesto that they would do if they had a majority. The fact that they haven't done this is very disappointing to a lot of people. It also doesn't help that the labour press blame the Lib Dems and Tories equally for bad things, while the Tory press blames the Lib Dems for everything bad that the coalition does and praises the Tories for everything good.
Nope. Compare the original Althlon to the comparable Intel chips at the time. It was both faster and cheaper, making it an obvious win. The K6-2 was about the price of a Pentium, but performance was comparable to a Pentium 2 or 3. My K6-2 350MHz cost about as much as a 266MHz Pentium 2 - much less if you included motherboard costs in both cases. The K6-2 and K6-3 were a bit slower than fast Intel chips, but it was very close. The Athlon was faster than the Pentium 3 and than the early Pentium 4s. The Opteron was the only serious game in town for server chips for a while. The Core 2 was the first chip that Intel released for a while that was really competitive and none of the AMD offering since then have really been that exciting.
There are still some niches where AMD is better. I just finished building a NAS box, and if you want a low power CPU that supports virtualisation extensions, then AMD is the only game in town.
Intel's Pentium was greeted with ridicule in the smoking hot 60Mhz incarnation (15 watts, can you believe that?) It went on to great success after a die shrink.
I talked to the head of the Pentium Pro to Pentium 4 projects (after he left Intel), and he said that their first power wakeup call came with that chip, when they were told by a company in New York that they couldn't upgrade their desktops because their building's power supply wouldn't be able to cope with the increased load. Sadly, it wasn't until after the Pentium 4 that they really learned this lesson to any degree.
Again, I ask: how is this gonna help decrease piracy?
Why do you care? Piracy is not a business problem. Potential customers not buying your product is a business problem. Pirates are irritating on an emotional level (you create something and people take it without paying, that sucks), but they are not a problem on a financial level. You have three categories of people:
Customers.
Potential customers.
Everyone else.
The first category is where you get your money. You want this set to be as big as possible. You want to try to move people from the second set into the first. Pirates overlap sets two and three, but it's an irrelevant designation. Would you rather have 10 paying customers and 10 pirates, or 100 paying customers and 1000 pirates? A rational business would pick the second option. The movie industry seems to prefer the first.
The only people who care about changing APIs are people who want to make proprietary drivers
No, the only people who care about changing APIs are people who don't want to modify the code that they've written and tested except to fix bugs or add features. This includes proprietary driver developers, but it also includes almost ever out-of-tree driver developer. Actually, it includes every driver developer, but in-tree driver developers don't have to because people who change the APIs in the kernel are required to modify (but not test, so they frequently introduce bugs) every in-tree driver when they do so.
Making Linux drivers isn't any different from any other OS, except that when you're done you submit it to the kernel for inclusion
Which means that you then need to go through the process again every time you want to add a feature, back-ports to older kernels are a pain (some of your customers use RHEL? Well, they may be able to use the new version of your driver in 3-5 years), and so on.
After that, the kernel devs do the maintenance for you, fixing things if there's any API changes
Well, they'll change your code. If they don't have access to your hardware then they won't test the changes. If they do, then they might test the changes. Check the LKML archives and you'll see a lot of cases where this kind of 'fix' has introduced bugs into drivers that used to work fine.
You realise that it's possible to implement a new ABI and provide a compatibility layer for the old one, right? Most non-toy kernels do this - even OS X. If you aren't using any drivers from the old ABI, then you don't load the compatibility layer and there are no problems. If you are, then the old drivers may not be quite as efficient as ones written for the new ABI, but they still work. Not doing this is not good engineering, it's just laziness.
Maybe people would be more willing to put effort into Linux drivers if they didn't keep moving the goalposts by changing the driver APIs. I notice that the OS X and Windows ports of VirtualBox don't seem to have these problems. I've only run FreeBSD as a guest, but apparently it works well as a VirtualBox host too...
How's its Java syntax highlighting? Objective-C autocompletion? Visual nib editor?
Visual Studio is a nice IDE (the debugger is still probably the best for any mainstream platform), but it is very much a Windows IDE. If you're targeting Windows, then it's probably the right tool for the job. If you aren't, then it almost certainly isn't.
Slashdot Mirror
Why? Look at how successful patented derivatives of his ideas have been, or even patented ideas from the same era. Most fizzled to nothing. He may have made some money, but the most likely effect of patenting things in C would have been to make us all use some Pascal derivative. If UNIX had been patented, then we'd probably be using a CP/M clone.
C is actually a subset of c++ as in all c programs will compile with a c++ compiler but C++ will not compile in a c compiler.
No it isn't. Some examples:
Valid C, not valid C++. How about a more complicated one?
Valid C, not valid C++. Or another simple one:
Once again, valid C, not valid C++. The semantics of inline are very different in C and C++. And here's a really fun one:
If sizeof(int) is 4 and alignof(int) is 4, this prints 4 in C and 12 in C++.
Why am I such a geek?
I didn't know that the definition of 'geek' had been changed to 'someone who believes falsehoods'.
Had he been a patent hound, he'd have died a rich man.
I doubt it. Most of what he created was part of the 'worse is better' philosophy. Given the choice between C and Algol, most people would have picked Algol in the '80s, but C compilers were cheaper (or free), so they went with C. The same with UNIX. There were much better operating systems around, but they were either expensive, required expensive hardware or, in many cases, both. C and UNIX were both good enough and free. That usually beats really good and expensive. If they'd tried to make a large profit from either, they'd have failed. In fact, the BSD lawsuit (AT&T vs UCB) came close to killing UNIX.
Most of it is written in C++. A lot of it is written in the C-like subset of C++, but it is not C.
Why is this bad? I'd rather see ads for things I like and might consider buying than scattershot ads for shit I'd never use.
I take it you missed the Slashdot story a couple of years ago about Amazon giving different prices based on the browser you use? Targeted advertising isn't where it ends. Companies like Google and Facebook often record enough information to tell how much you shop around before buying things, for example. It doesn't take much data mining to work out how much you'd be willing to pay for a specific product. Next time you visit an online store, you may find it's exactly that amount. Meanwhile, it costs 20% less to the next visitor...
It would have to be. Dotsheep are hidden.
You and I know that 'publicly' means 'on any server you don't control'. The masses think 'publicly' means 'on a site that doesn't have access control'. People don't realise that 'private' communications on FaceBook may not be seen by the world at large, but that are seen by anyone who pays FaceBook (including the CIA, which seems a bit silly because you'd have to be a total moron to use something like FaceBook if you're planning the sort of things that the CIA cares about).
Are they? Recruiters seem to still find me, even though I'm not on any social networking sites (unless you count Slashdot). The most recent company to invite me for interview was... Facebook.
They may be, but without Bruce Wayne's fortune to buy them kevlar suits and bullet proof cars, they don't last long...
It doesn't matter. Judging by the trends in HP CEOs, Darl will only last 2 months. He will be replaced by Kim Jong Il, who will last a mere 3 weeks (and leave with a $100m golden parachute). The next CEO will be fired before she arrives in the building, and the one after her will be given her golden parachute before being offered the job. After that, things are going to get really bad...
My old Brother laser supported PostScript, but printing PCL was a lot faster because rasterising a complex PostScript document (e.g. the output from LaTeX) on a 50MHz MIPS processor in the printer took a long time. I've recently bought a cheap Dell all-in-one colour LED printer. It can print from and scan to PDFs, and doesn't even need a computer: you can print PDFs directly from a USB device. It also talks SMB and FTP, so I can scan to an SMB or FTP server.
The faster P4 chips were rarely faster than comparative Athlons in real workloads. They did very well in some benchmarks, but the cost of a branch predictor miss was insane (something on the order of 150 instructions), which killed their real-world performance. The Opteron and HyperTransport gave AMD an edge in 2+ socket systems and a big lead in 4+ socket systems for quite a while.
AMD was very late to focus on power consumption. This was probably due to their chips not being quite the space heaters that Intel chips were. Going from the late P4s to the Core 1 was a huge leap in power consumption for Intel, but not such a big leap for AMD. Since then, Intel has pushed power usage down, but AMD didn't really have anything competitive with Atom until the recent Fusion line.
Wait: you seriously expect a party that is part of a coalition to not make compromises?
No, but then I would expect people to read my post and the context in which it was written before posting a reply, meaning that I seem to have unreasonably high expectations...
Superheroes are people with special powers they use for good.
*nerd hat on*
What about Batman or the Green Arrow? Unless 'having lots of money' is a special power, they didn't have any. They just put on body armour and went and helped people - making them a much better role model than someone like Superman who is basically invulnerable.
*nerd hat off*
I would say we need these guys a lot more than we need thugs assaulting each other or random people in the streets.
I only half agree. If groups of thugs want to kill each other off, I don't have a problem with that. It's only when they start attacking other people that it's a problem...
I feel represented. The Liberal Democrats are largely pushing policies I agree with, although they're limited by the government being bankrupt and the Tories having 3/5 of the seats that the coalition controls. My Labour MP (who I voted against) feels he has something to prove and has been very responsive since being elected. My Plaid Cymru MEP has been working hard to return IP laws to more sane levels.
They expected that a party with 40% of the seats controlled by the coalition would be the dominant force in policy making,would not have to make any compromises, and would be able to do everything that they said in their manifesto that they would do if they had a majority. The fact that they haven't done this is very disappointing to a lot of people. It also doesn't help that the labour press blame the Lib Dems and Tories equally for bad things, while the Tory press blames the Lib Dems for everything bad that the coalition does and praises the Tories for everything good.
Nope. Compare the original Althlon to the comparable Intel chips at the time. It was both faster and cheaper, making it an obvious win. The K6-2 was about the price of a Pentium, but performance was comparable to a Pentium 2 or 3. My K6-2 350MHz cost about as much as a 266MHz Pentium 2 - much less if you included motherboard costs in both cases. The K6-2 and K6-3 were a bit slower than fast Intel chips, but it was very close. The Athlon was faster than the Pentium 3 and than the early Pentium 4s. The Opteron was the only serious game in town for server chips for a while. The Core 2 was the first chip that Intel released for a while that was really competitive and none of the AMD offering since then have really been that exciting.
There are still some niches where AMD is better. I just finished building a NAS box, and if you want a low power CPU that supports virtualisation extensions, then AMD is the only game in town.
Intel's Pentium was greeted with ridicule in the smoking hot 60Mhz incarnation (15 watts, can you believe that?) It went on to great success after a die shrink.
I talked to the head of the Pentium Pro to Pentium 4 projects (after he left Intel), and he said that their first power wakeup call came with that chip, when they were told by a company in New York that they couldn't upgrade their desktops because their building's power supply wouldn't be able to cope with the increased load. Sadly, it wasn't until after the Pentium 4 that they really learned this lesson to any degree.
Again, I ask: how is this gonna help decrease piracy?
Why do you care? Piracy is not a business problem. Potential customers not buying your product is a business problem. Pirates are irritating on an emotional level (you create something and people take it without paying, that sucks), but they are not a problem on a financial level. You have three categories of people:
The first category is where you get your money. You want this set to be as big as possible. You want to try to move people from the second set into the first. Pirates overlap sets two and three, but it's an irrelevant designation. Would you rather have 10 paying customers and 10 pirates, or 100 paying customers and 1000 pirates? A rational business would pick the second option. The movie industry seems to prefer the first.
The only people who care about changing APIs are people who want to make proprietary drivers
No, the only people who care about changing APIs are people who don't want to modify the code that they've written and tested except to fix bugs or add features. This includes proprietary driver developers, but it also includes almost ever out-of-tree driver developer. Actually, it includes every driver developer, but in-tree driver developers don't have to because people who change the APIs in the kernel are required to modify (but not test, so they frequently introduce bugs) every in-tree driver when they do so.
Making Linux drivers isn't any different from any other OS, except that when you're done you submit it to the kernel for inclusion
Which means that you then need to go through the process again every time you want to add a feature, back-ports to older kernels are a pain (some of your customers use RHEL? Well, they may be able to use the new version of your driver in 3-5 years), and so on.
After that, the kernel devs do the maintenance for you, fixing things if there's any API changes
Well, they'll change your code. If they don't have access to your hardware then they won't test the changes. If they do, then they might test the changes. Check the LKML archives and you'll see a lot of cases where this kind of 'fix' has introduced bugs into drivers that used to work fine.
You realise that it's possible to implement a new ABI and provide a compatibility layer for the old one, right? Most non-toy kernels do this - even OS X. If you aren't using any drivers from the old ABI, then you don't load the compatibility layer and there are no problems. If you are, then the old drivers may not be quite as efficient as ones written for the new ABI, but they still work. Not doing this is not good engineering, it's just laziness.
Maybe people would be more willing to put effort into Linux drivers if they didn't keep moving the goalposts by changing the driver APIs. I notice that the OS X and Windows ports of VirtualBox don't seem to have these problems. I've only run FreeBSD as a guest, but apparently it works well as a VirtualBox host too...
Sounds about right. It's easy to turn one bug into ten bugs if you write each piece of code on the assumption that every other bit is bug-free...
How's its Java syntax highlighting? Objective-C autocompletion? Visual nib editor?
Visual Studio is a nice IDE (the debugger is still probably the best for any mainstream platform), but it is very much a Windows IDE. If you're targeting Windows, then it's probably the right tool for the job. If you aren't, then it almost certainly isn't.