Yes, it's crappy, broad-brush trolling journalism, designed to get a response (and page impressions) from the Linux crowd. Yes, the guy's an idiot. The main point being that the people who are providing the evidence against SCO's case (Bruse perens etc) are not, by his definition, Linux Zealots.
>Perhaps you missed the first sentence?
I gave him the benefit of the doubt that, taken in the context of the surrounding paragraphs, that statement didn't literally mean that "people who like linux are going to blow up buildings".
Much (not all) Slashdot (and pro-OSS) discussion never actually references sources, preferring to stick to anecdotal second hand knowledge - "Windows always crashes" etc.
'Linux Priests' can be blinded to flaws in OSS by their love of OSS in general. I'm guilty of this myself. Blind belief is never a good thing - admit flaws and fix them, don't just flame/ignore people who point them out.
His use of zealot is different from what a standard slashdotian would use it for. He means the real nutcases (check out arstechnica openforum battlefront for examples).
He doesn't call zealots terrorists. He says that the nutcases are dangerous to the OSS cause, just as islamic terrorists are dangerous to moderate Islam - the nutcases get the press coverage and we all get tarred with the same brush. Take the DoS attack on SCO, for example. That didn't do anyone any good and gave the other side ammunition to use against the Linux community
What is this? Holy war week on Slashdot? In the last week or so we've had stories on BSD vs Linux, Linux vs Solaris, PHP vs Java, Exchange vs Sendmail , x86 vs PPC and now IDE vs SCSI. All that's missing is Vi vs Emacs and I think we'll have pretty much every major computing disagreement covered.
Why do people even bother watching TV with that many adverts? 22 minutes per hour is insane. I find the 7.5 minutes allowed here (UK, terrestrial stations) annoying enough. If people are prepared to put up with that much crap to watch tv, maybe they'll just accept not being able to record it too.
No, the MHz is only relevant when you are comparing the same processor type. Otherwise it's completely meaningless. The only thing that matters is how quickly it can process instructions and at what price.
You should use Plucker books on your Palm. Only 2000 books so far, but no random line breaks (apart from those enforced by 160 pixel screen). And a new version (1.6) of Plucker is just out.
So, what Forbes are basically saying is that people who work on Open Source projects should just roll over and let any company benefit from their work for free, because then the company makes more money?
How many times do we have to say it:
If you don't want to abide by the GPL, don't use GPL code
Hardly rocket science, is it? Linksys (apparently) saves hundreds of thousands in software development by violating the chosen licence on the work of hundreds of volunteers, and that's meant to be acceptable?
Browsing the web should not be a dangerous activity for your computer. You should not have to think "better not follow that link, my box might get rooted and all my work since my last backup destroyed". People who think that this kind of thing is acceptable are why companies like microsoft do so well - everyone says "ah, it's not that important" right up until the point that it bites them in the arse.
What is the thing about calling Macs overpriced? I don't own a Mac, but go to dell.com and spec out a PC similar to a Mac G5. The price will be pretty close to what the Mac costs. Yes, you can get a PC for $299. Yes, you can build one yourself for not much money. But once you get into the corporate, workstation-quality and performance area, the prices are very similar. Personally, I dislike the Dell machine I have to use at work. The build quality is pretty poor - my CD drive has died, the keyboard/mouse are poor quality. The IDE cable popped out of the CD drive in the machine on the next desk because it wasn't quite long enough.
Roblimo doesn't need to give a sarcastic-toned list of points of why he hates Windows to show Linux in a good light. If he'd just stuck to simple facts he'd have a much better argument.
The things that annoy me most using Windows are: 1) Badly designed UI 2) Slower than my Linux install 3) few configuration options 4) Poor interactivity - often hard to do something in the foreground while something is compiling in the background, for example.
I find these problems on every windows box I've used.
Ah, now the question makes sense. I did wonder what the problem was about cheating on 'homeworks' was - surely they'd just fail the exam at the end. The only coursework (UK uni) I had that was worth any marks was either a group project, or individual reports where everyone in the year got different report titles. People would sometimes get copies of reports from the previous year, but that wasn't much help really.
I find the stats for my tiny site (average 50 visitors a day) even more depressing - it's an open-source project site, only linked to by freshmeat and palmopensource.com, and IE is still the most common browser with around ~45% IE, 32% mozilla.
Of course IE gets them revenues - it is basically only usable on Windows, and therefore another way of locking people into Windows. Yes, there was a Mac version, but it wasn't (from what I've read) a very good version, and has been left to die now. How many web sites still say "requires IE5+" or whatever? How many websites rely on IE's quirks? By abusing their monopoly position, MS made "the web" and IE synonymous for most users, and required for many things (online banking, for example, often requires IE). Of course customers want good browsers. They just can't see them past the big blue e on their desktop.
These benchmarks, by the guys who host most of Gentoo, were done on a dual 2.8Ghz P4 Xeon's, 2Gb of RAM machine (Dell PowerEdge 2650) and (AFAICS) reach broadly similar conclusions - 'best' depends on your usage, but JFS is pretty good, reiser uses a lot of CPU etc.
Well, annealing is an optimisation process that emulates the process of crystal formation in cooling molten materials. It avoids getting stuck in a local minima and missing the global minima by sometimes accepting a 'worse' solution than the current one during the search process with a certain probablity.
Genetic search techniques work like evolution. You generate a 'generation' of candidate solutions, from which the best are selected (like natural selection) and the next generation are bred and so on until you get close enough to the optimum.
Putting the two together, I would guess that genetic annealing is a genetic algorithm that has a certain probablity of selecting a solution from a generation that doesn't have the lowest objective function value in order to aviod getting stuck in local minima. In a way, I guess it's like preventing too much inbreeding in the solutions.
The API isn't the problem, it's the hardware. On a console the developers know exactly what the hardware configuration is, including cache sizes, relative speeds of different components, available memory, available machine instructions. They can therefore tweak the hell out of the engine. On a PC, even given one common API like OpenGL, there are so many variables that optimisation is far harder. Can you assume SSE1/2? Or should you switch to 3DNow on athlons? How big is your L1/2 cache? Bus speed to memory/video? How many texture units? How fast is the GPU - how much work can we do on the CPU in parallel with it? Texture memory size? Main memory size/speed/latency?
Yes, you can write different code paths for common configurations, but that is time that the console developers can spend optimising their single path.
Thinking about it, the average intelligence of managers does seem to at the lower end of the simian spectrum. Maybe a "GPL for dummies" book is needed.
It's very clearly defined what you can and can't do. Linksys have benefited hugely from GPLed code (how many thousands of developer-hours have they saved?).
They also (appear to) have broken the terms of the GPL. So they are liable to be sued by the copyright holders of the code, just as if it were proprietary code they had stolen. Not exactly complex, is it?
The GPL protects the free software community. If some big company wants to use GPLed code, then they have to become "part of the community" and give something back. It clearly works - would IBM, Sun or Apple be donating so much code for free if it wasn't for the GPL? I doubt it.
Slashdot Mirror
Yes, it's crappy, broad-brush trolling journalism, designed to get a response (and page impressions) from the Linux crowd. Yes, the guy's an idiot. The main point being that the people who are providing the evidence against SCO's case (Bruse perens etc) are not, by his definition, Linux Zealots.
>Perhaps you missed the first sentence?
I gave him the benefit of the doubt that, taken in the context of the surrounding paragraphs, that statement didn't literally mean that "people who like linux are going to blow up buildings".
Of course, it will if MS makes wearing a DRM Helmet part of the EULA.
He actually makes some very valid points:
You mean you're wasting 56k of hard disk space and memory?
/bin/echo* /bin/ed*
-rwxr-xr-x 1 root root 16K Sep 16 14:57
-rwxr-xr-x 1 root root 72K Aug 3 02:06
What is this? Holy war week on Slashdot? In the last week or so we've had stories on BSD vs Linux, Linux vs Solaris, PHP vs Java, Exchange vs Sendmail , x86 vs PPC and now IDE vs SCSI. All that's missing is Vi vs Emacs and I think we'll have pretty much every major computing disagreement covered.
Why do people even bother watching TV with that many adverts? 22 minutes per hour is insane. I find the 7.5 minutes allowed here (UK, terrestrial stations) annoying enough.
If people are prepared to put up with that much crap to watch tv, maybe they'll just accept not being able to record it too.
No, the MHz is only relevant when you are comparing the same processor type. Otherwise it's completely meaningless. The only thing that matters is how quickly it can process instructions and at what price.
You should use Plucker books on your Palm. Only 2000 books so far, but no random line breaks (apart from those enforced by 160 pixel screen). And a new version (1.6) of Plucker is just out.
So, what Forbes are basically saying is that people who work on Open Source projects should just roll over and let any company benefit from their work for free, because then the company makes more money?
How many times do we have to say it:
If you don't want to abide by the GPL, don't use GPL code
Hardly rocket science, is it? Linksys (apparently) saves hundreds of thousands in software development by violating the chosen licence on the work of hundreds of volunteers, and that's meant to be acceptable?
So, 3 years ago (october 2000), we had:
- Mozilla M18
- KDE 1.1.2
- GNOME 1.2
- Openoffice was barely released and was pretty much just staroffice 5.2.
I wonder what a Linux desktop will look like by 2006?
Browsing the web should not be a dangerous activity for your computer. You should not have to think "better not follow that link, my box might get rooted and all my work since my last backup destroyed".
People who think that this kind of thing is acceptable are why companies like microsoft do so well - everyone says "ah, it's not that important" right up until the point that it bites them in the arse.
What is the thing about calling Macs overpriced? I don't own a Mac, but go to dell.com and spec out a PC similar to a Mac G5. The price will be pretty close to what the Mac costs.
Yes, you can get a PC for $299. Yes, you can build one yourself for not much money. But once you get into the corporate, workstation-quality and performance area, the prices are very similar.
Personally, I dislike the Dell machine I have to use at work. The build quality is pretty poor - my CD drive has died, the keyboard/mouse are poor quality. The IDE cable popped out of the CD drive in the machine on the next desk because it wasn't quite long enough.
Roblimo doesn't need to give a sarcastic-toned list of points of why he hates Windows to show Linux in a good light. If he'd just stuck to simple facts he'd have a much better argument.
The things that annoy me most using Windows are:
1) Badly designed UI
2) Slower than my Linux install
3) few configuration options
4) Poor interactivity - often hard to do something in the foreground while something is compiling in the background, for example.
I find these problems on every windows box I've used.
Ah, now the question makes sense. I did wonder what the problem was about cheating on 'homeworks' was - surely they'd just fail the exam at the end. The only coursework (UK uni) I had that was worth any marks was either a group project, or individual reports where everyone in the year got different report titles. People would sometimes get copies of reports from the previous year, but that wasn't much help really.
This method relies on installing a device driver, so surely it won't work unless you're logged in as Administrator?
The google zeitgeist is a reasonable estimate.
I find the stats for my tiny site (average 50 visitors a day) even more depressing - it's an open-source project site, only linked to by freshmeat and palmopensource.com, and IE is still the most common browser with around ~45% IE, 32% mozilla.
Of course IE gets them revenues - it is basically only usable on Windows, and therefore another way of locking people into Windows. Yes, there was a Mac version, but it wasn't (from what I've read) a very good version, and has been left to die now.
How many web sites still say "requires IE5+" or whatever? How many websites rely on IE's quirks? By abusing their monopoly position, MS made "the web" and IE synonymous for most users, and required for many things (online banking, for example, often requires IE).
Of course customers want good browsers. They just can't see them past the big blue e on their desktop.
These benchmarks, by the guys who host most of Gentoo, were done on a dual 2.8Ghz P4 Xeon's, 2Gb of RAM machine (Dell PowerEdge 2650) and (AFAICS) reach broadly similar conclusions - 'best' depends on your usage, but JFS is pretty good, reiser uses a lot of CPU etc.
But of course, this is Slashdot, the geek pro-Linux/Open source site. So surely nobody uses IE anyway ;).
Well, annealing is an optimisation process that emulates the process of crystal formation in cooling molten materials. It avoids getting stuck in a local minima and missing the global minima by sometimes accepting a 'worse' solution than the current one during the search process with a certain probablity.
Genetic search techniques work like evolution. You generate a 'generation' of candidate solutions, from which the best are selected (like natural selection) and the next generation are bred and so on until you get close enough to the optimum.
Putting the two together, I would guess that genetic annealing is a genetic algorithm that has a certain probablity of selecting a solution from a generation that doesn't have the lowest objective function value in order to aviod getting stuck in local minima. In a way, I guess it's like preventing too much inbreeding in the solutions.
That's only a guess though.
The API isn't the problem, it's the hardware. On a console the developers know exactly what the hardware configuration is, including cache sizes, relative speeds of different components, available memory, available machine instructions. They can therefore tweak the hell out of the engine.
On a PC, even given one common API like OpenGL, there are so many variables that optimisation is far harder. Can you assume SSE1/2? Or should you switch to 3DNow on athlons? How big is your L1/2 cache? Bus speed to memory/video? How many texture units? How fast is the GPU - how much work can we do on the CPU in parallel with it? Texture memory size? Main memory size/speed/latency?
Yes, you can write different code paths for common configurations, but that is time that the console developers can spend optimising their single path.
Another good reason to run a kernel with the grsecurity patches on servers?
Thinking about it, the average intelligence of managers does seem to at the lower end of the simian spectrum. Maybe a "GPL for dummies" book is needed.
How is investing in GPL code risky?
It's very clearly defined what you can and can't do. Linksys have benefited hugely from GPLed code (how many thousands of developer-hours have they saved?).
They also (appear to) have broken the terms of the GPL. So they are liable to be sued by the copyright holders of the code, just as if it were proprietary code they had stolen. Not exactly complex, is it?
The GPL protects the free software community. If some big company wants to use GPLed code, then they have to become "part of the community" and give something back. It clearly works - would IBM, Sun or Apple be donating so much code for free if it wasn't for the GPL? I doubt it.