In a nutshell, you already know that half of their tests are conceptually wrong and the other hald conclusively show that IE sucks. What is that new thing you were expecting to read?
You apparently don't follow the history of GCC:
- 3.4 introduced huge ABI-breaking changes, and was not as standard-compliant as they felt it should have been
- 4.0 was their "screw it, let's throw a lot away and get it right" release, which sucked a little despite being more compliant and having some great new tech
- 4.1 was amazing, benefitting from 4.0's new technology and fixing (most of) the bugs
- 4.2 was not such a big deal compared to 4.1, as far as I know
- 4.3 again was fairly great, bringing great performance increases, and is now at least as mature as 3.3 was
My point is that the differences between these minor releases have been fairly dramatic at times. They have been at least as big as the changes between MSVC major releases. It's just that FSF thankfully doesn't have MS's marketing department that slaps a new year-number-version on every minor upgrade, calls it a big deal, and sells it for even more money.
This destroys Microsoft's claim that their intimate knowledge of the OS that runs IE will increase performance.
To be precise (by pulling numbers out of my ass), if IE had 50% of Firefox's performance to begin with, and embedding into the OS gave it a 50% advantage, it'd still only have 75% of Firefox's performance. But MS's claim could in theory still be true.
Of course, given their all-around incompetence it's probably not true.
As for Google Chrome, it makes perfect sense to bind the user to the webmaster's control. After all, for many important things like e-mail, calendaring, and many more, that webmaster is probably Google. (After all, how many yahoo.com or live.com users would install a Google browser?) And Google loves it when you can't block their cookies or stop them from doing whatever they want to spy on you.
I forgot: The reason why I use the GNU/Linux name (before I get bashed for it) is that I recently played a lot with Solaris. Believe me when I tell you - UNIX without the GNU tools is a bitch.
I don't agree with your last point. People will go out of their way to get Windows even on a computer that came with something else, depite those still being the exception rather than the norm.
For example, almost every Eee PC purchased by people around me gets its GNU/Linux removed and replaced by XP first thing in the morning. And that thing won't be running games or speciality apps.
Most people will not bother to go through the 60 minutes of learning (less if one has above average intelligence) required to get comfortable with a new UI, not only because they resist change, but also because they do not percieve any advantage.
The ones that have considered switching are the ones that have been bitten by viruses one too many times. In those cases, they've heard some Linux hype and they wonder what that might be like. Consequently, I think GNU/Linux needs a lot more hype. Hype is a great thing as long as one can deliver on it, which GNU/Linux certainly can.
First, the 4K includes the operating system, because with TinyOS, applications are statically compiled together with the OS into a single blob. And in these 4K, there's still space left for something that uses IPv6, so the implementation itself is most likely at least as small as the one in TFA.
Second, what do you call a "layer for transferring IPv6 packets"? I call it an IPv6 stack - something that takes your IPv6 packet and transfers it where it's supposed to go. I saw that thing transfer UDP packets during a demonstration.
I'd like to file a bug report on the US educational system.
Status: Resolved
Resolution: WORKSFORME
Seriously, the only way to be good at math is to love math. If you love math, then it does not matter what others may think or whether you get any media coverage.
Re:Building your own kernel these days ain't easy
on
Linux 2.6.27 Out
·
· Score: 4, Funny
You're looking for an attractive single woman that is also an OpenBSD core developer and would be willing to mate with a Gentoo core developer. And you're looking for her on Slashdot.
Good luck with that.:)
Re:Building your own kernel these days ain't easy
on
Linux 2.6.27 Out
·
· Score: 5, Informative
I see your point, but consider the fact that every option (if you use "make menuconfig" at least) has a context-based help message. For the most part, they are actually very useful. Just go through all the options and think whether you need something or not. If you're not sure, there's a recommended safe default right at the end of the help message.
And you really need to do this once. After that for each new version, you just do "make oldconfig" against the old.config file (the one that stored your choices) and that's typically 10-20 options tops for new major versions.
Changed hardware? New PC? Just reconfigure the "Drivers" section in a few minutes and you're golden. That's assuming of course you stripped down everything you don't need - if you left it in, you don't evenhave to do as much, it will just work.
BTW, if you're into tinkering, go all the way and try Gentoo. That project is alive and kicking, regardless of what the media have been saying recently.
It is pretty hard actually. Their financial models depend on vendor lock-in, lack of privacy (from them), precise control of their software (ads, "premium" features, etc.), or all of the above.
Free software could allow you to encrypt your data automatically from the service provider, to migrate yourself and your data to a compatible competitor, or to implement "premium" features yourself. But what's in it for the service provider?
They do something of the sort. A friend bought the nForce 680i SLI when it came out. It had a BIOS option to increase PCI-E bandwidth with certain high-end nVidia video cards by some 15%. The option would only work with a 8800-series card. Interestingly, the option disappeared in later BIOS revisions.
Yet another example: the only difference between a commodity GeForce and a really expensive Quadro is a switch in the driver. The driver enables special features for Quadros, including the ability to run SLI on non-Nvidia pre-X58 chipsets. So much for their "technical reasons" lie.
At the Nvision event in San Jose, California, Nvidia outlined another plan: it will certify certain Intel X58-based mainboards for SLI compliance and will provide âoeapproval keys that will be integrated into the system BIOS for boards that pass certificationâ. The company said that it will charge mainboard makers for SLI compliance, but right now the terms are unknown.
This smells like yet another "we'll put arbitrary software restrictions in our stuff because we're greedy. Wonder why they are the only ones with no free drivers whatsoever?
I call bullshit. I stopped using their chipsets long ago, but now I'll actually switch to AMD for video. No more NVidia - hello, software freedom.
Now if I could only find a P45-based board that can run FreeBIOS....
OK, first of all stop saying detained and call it by its name: arrested. Second, what the hell was this guy doing in China? He should've seen it coming.
Ouch. You shouldn't be working as root all the time. In fact, you should never be working as root. Use sudo. Otherwise, you've just defeated one of Linux's security mechanisms right there.
Linux forsaking its free usage model to embrace more paid distros
The two are not mutually exclusive.
where you get free Linux along with (much-needed) licenses to use patent-restricted codecs
As if. Just because the US has a broken patent system. Or was that the whole world? Ih wait, the US is the whole world.
Also predicted is an advance for the desktop based on â" surprise â" good acceptance for KDE 4.
Whether you like it or not, GNOME will be the big one, because nobody controls it. With Nokia owning Trolltech, no other company (whose primary business is not Linux itself) will touch KDE. I know that's not justified, but don't expect large corporations to care.
Finally, Linux is seen as making its biggest imprint not on the PC, but on mobile devices, eventually powering 40 million smartphones and netbooks.
That's clearly the future. The question is - besides having Linux as the kernel, will the phone of the future be any different? Will free userspace triumph on phones, or are we going to see locked-down Linux? That's the interesting and harder question.
And what do you see for Linux in 4 years?
Let's see... a kernel that supports the latest hardware and runs the latest software?
The article is nonsense, but the discussion should be good.
Slashdot Mirror
In a nutshell, you already know that half of their tests are conceptually wrong and the other hald conclusively show that IE sucks. What is that new thing you were expecting to read?
You apparently don't follow the history of GCC:
- 3.4 introduced huge ABI-breaking changes, and was not as standard-compliant as they felt it should have been
- 4.0 was their "screw it, let's throw a lot away and get it right" release, which sucked a little despite being more compliant and having some great new tech
- 4.1 was amazing, benefitting from 4.0's new technology and fixing (most of) the bugs
- 4.2 was not such a big deal compared to 4.1, as far as I know
- 4.3 again was fairly great, bringing great performance increases, and is now at least as mature as 3.3 was
My point is that the differences between these minor releases have been fairly dramatic at times. They have been at least as big as the changes between MSVC major releases. It's just that FSF thankfully doesn't have MS's marketing department that slaps a new year-number-version on every minor upgrade, calls it a big deal, and sells it for even more money.
This destroys Microsoft's claim that their intimate knowledge of the OS that runs IE will increase performance.
To be precise (by pulling numbers out of my ass), if IE had 50% of Firefox's performance to begin with, and embedding into the OS gave it a 50% advantage, it'd still only have 75% of Firefox's performance. But MS's claim could in theory still be true.
Of course, given their all-around incompetence it's probably not true.
As for Google Chrome, it makes perfect sense to bind the user to the webmaster's control. After all, for many important things like e-mail, calendaring, and many more, that webmaster is probably Google. (After all, how many yahoo.com or live.com users would install a Google browser?) And Google loves it when you can't block their cookies or stop them from doing whatever they want to spy on you.
I forgot: The reason why I use the GNU/Linux name (before I get bashed for it) is that I recently played a lot with Solaris. Believe me when I tell you - UNIX without the GNU tools is a bitch.
I don't agree with your last point. People will go out of their way to get Windows even on a computer that came with something else, depite those still being the exception rather than the norm.
For example, almost every Eee PC purchased by people around me gets its GNU/Linux removed and replaced by XP first thing in the morning. And that thing won't be running games or speciality apps.
Most people will not bother to go through the 60 minutes of learning (less if one has above average intelligence) required to get comfortable with a new UI, not only because they resist change, but also because they do not percieve any advantage.
The ones that have considered switching are the ones that have been bitten by viruses one too many times. In those cases, they've heard some Linux hype and they wonder what that might be like. Consequently, I think GNU/Linux needs a lot more hype. Hype is a great thing as long as one can deliver on it, which GNU/Linux certainly can.
Apple and "low-cost or free" hardware? What have you been smoking?
And this is different from the Great Firewall of Chine how?
Not that being similar to the GFC makes it any more acceptable of course.
It's not about the money. It's about the precedent and the fear.
First, the 4K includes the operating system, because with TinyOS, applications are statically compiled together with the OS into a single blob. And in these 4K, there's still space left for something that uses IPv6, so the implementation itself is most likely at least as small as the one in TFA.
Second, what do you call a "layer for transferring IPv6 packets"? I call it an IPv6 stack - something that takes your IPv6 packet and transfers it where it's supposed to go. I saw that thing transfer UDP packets during a demonstration.
I know a guy who coded a smaller IPv6 stack. It runs on TinyOS and requires 4 KB RAM in total together with the OS. That was an year ago. Check this out: http://www.inf.ethz.ch/personal/mharvan/research.html
I'd like to file a bug report on the US educational system.
Status: Resolved
Resolution: WORKSFORME
Seriously, the only way to be good at math is to love math. If you love math, then it does not matter what others may think or whether you get any media coverage.
You're looking for an attractive single woman that is also an OpenBSD core developer and would be willing to mate with a Gentoo core developer. And you're looking for her on Slashdot.
:)
Good luck with that.
I see your point, but consider the fact that every option (if you use "make menuconfig" at least) has a context-based help message. For the most part, they are actually very useful. Just go through all the options and think whether you need something or not. If you're not sure, there's a recommended safe default right at the end of the help message.
.config file (the one that stored your choices) and that's typically 10-20 options tops for new major versions.
And you really need to do this once. After that for each new version, you just do "make oldconfig" against the old
Changed hardware? New PC? Just reconfigure the "Drivers" section in a few minutes and you're golden. That's assuming of course you stripped down everything you don't need - if you left it in, you don't evenhave to do as much, it will just work.
BTW, if you're into tinkering, go all the way and try Gentoo. That project is alive and kicking, regardless of what the media have been saying recently.
It is pretty hard actually. Their financial models depend on vendor lock-in, lack of privacy (from them), precise control of their software (ads, "premium" features, etc.), or all of the above.
Free software could allow you to encrypt your data automatically from the service provider, to migrate yourself and your data to a compatible competitor, or to implement "premium" features yourself. But what's in it for the service provider?
Who says you have to pay with money? You're paying with data about you. Google must know a lot about you by now.
To rephrase the judge's decision:
<singing>Hit the road, Jack, and don't you come back no more no more no more no more, hit the road, Jack, and don't you come back no more!</singing>
What, you mean that was a META-Slashvertisement?! I think my head just exploded...
They do something of the sort. A friend bought the nForce 680i SLI when it came out. It had a BIOS option to increase PCI-E bandwidth with certain high-end nVidia video cards by some 15%. The option would only work with a 8800-series card. Interestingly, the option disappeared in later BIOS revisions.
Yet another example: the only difference between a commodity GeForce and a really expensive Quadro is a switch in the driver. The driver enables special features for Quadros, including the ability to run SLI on non-Nvidia pre-X58 chipsets. So much for their "technical reasons" lie.
I've had it with nVidia.
From AFA:
At the Nvision event in San Jose, California, Nvidia outlined another plan: it will certify certain Intel X58-based mainboards for SLI compliance and will provide âoeapproval keys that will be integrated into the system BIOS for boards that pass certificationâ. The company said that it will charge mainboard makers for SLI compliance, but right now the terms are unknown.
This smells like yet another "we'll put arbitrary software restrictions in our stuff because we're greedy. Wonder why they are the only ones with no free drivers whatsoever?
I call bullshit. I stopped using their chipsets long ago, but now I'll actually switch to AMD for video. No more NVidia - hello, software freedom.
Now if I could only find a P45-based board that can run FreeBIOS....
Nice to meat you
Given that you're talking about lawyers, I don't think that's a spelling mistake...
OK, first of all stop saying detained and call it by its name: arrested. Second, what the hell was this guy doing in China? He should've seen it coming.
-1? What the hell is wrong with you, people?
Ouch. You shouldn't be working as root all the time. In fact, you should never be working as root. Use sudo. Otherwise, you've just defeated one of Linux's security mechanisms right there.
Linux forsaking its free usage model to embrace more paid distros
The two are not mutually exclusive.
where you get free Linux along with (much-needed) licenses to use patent-restricted codecs
As if. Just because the US has a broken patent system. Or was that the whole world? Ih wait, the US is the whole world.
Also predicted is an advance for the desktop based on â" surprise â" good acceptance for KDE 4.
Whether you like it or not, GNOME will be the big one, because nobody controls it. With Nokia owning Trolltech, no other company (whose primary business is not Linux itself) will touch KDE. I know that's not justified, but don't expect large corporations to care.
Finally, Linux is seen as making its biggest imprint not on the PC, but on mobile devices, eventually powering 40 million smartphones and netbooks.
That's clearly the future. The question is - besides having Linux as the kernel, will the phone of the future be any different? Will free userspace triumph on phones, or are we going to see locked-down Linux? That's the interesting and harder question.
And what do you see for Linux in 4 years?
Let's see... a kernel that supports the latest hardware and runs the latest software?
The article is nonsense, but the discussion should be good.
BTW, what's the meaning of the 9-5? I'm missing some terminology, it seems.