The kernel hackers are not complaining about a lack of binary drivers; they don't want any binary drivers.
There are a lot of Linux users saying "please give me a driver, any driver!", but I think those people should ask themselves if it's a good idea to use an OS when they have such a fundamental disagreement with the developers of that OS.
There's only one problem: WiMax doesn't work in notebooks. WiMax requires the clients to be fixed, have a large(ish) antenna pointed directly at the base station, and (in a large cell) use so much power it'd drain your battery pretty quickly.
Fedora uses an auto-build system (inherited from RHL/RHEL) which tries to build every SRPM for every known architecture. Just because a bunch of RPMs exist for weird architectures doesn't mean Fedora actually works on those architectures.
AKA the Mozilla approach. Don't like our license? We'll give you a few more to choose from!
Re:Cut-and-replace takes longer in X than Windows
on
X.org and XFree86 Reform
·
· Score: 2, Informative
I would prefer the X developers to just implement the Windows method, instead of requiring users who migrate to learn a new method.
Ctrl-C and Ctrl-V work fine in all the modern (i.e. GNOME) apps I've tried.
In a related vein, does anyone know how to disable the regular CTRL-C in a KDE terminal window, perhaps by making it into a menu item instead? Then I could finally use CTRL-C in the terminal.
Of course, the correct solution is to have a separate Command key, and use Command-XCV for cut, copy, and paste. This works great on the Mac.:-)
Re:IP6s problem is much deeper than that.
on
The State of IPv6
·
· Score: 2, Insightful
What header? All but the first and last router don't care about anything but the source and destination address, and most times not about the source address at all. So what else is there to simplify?
You're ignoring header checksums and fragmentation.
Yes, a lot of IPv4 features were simply taken out of IPv6, but that doesn't make IPv6 unnecessary. Taking a feature out of IPv4 would result in something that's not compatible with IPv4, thus you'd have to give it a new name and upgrade all the equipment. So why not increase the address size while we're at it and call the result IPv6?
Ideally you would replace your NAT with one that can act as a 6to4 border router, but there aren't any of those yet.
You could run Teredo on your computers, but only Windows XP supports it AFAIK.
Re:IP6s problem is much deeper than that.
on
The State of IPv6
·
· Score: 1
People can yell "better routing" all day long, it doesn't make it so. Routing tables are not improved by being made larger. All it means is greater and greater functionality required of the networking hardware, which means just more and more ways for the network itself to fail.
IPv6 routing tables are much smaller because the address allocations are being done in a much more efficient manner. Also, IPv6 is simpler to route than IPv4, because header processing was simplified.
Even a $50 video card has DVI these days and quite a few cards have component adaptors. Sometimes it takes a bit of fiddling with Powerstrip to convince the card to output weird resolutions, but it's not impossible.
Red lasers haven't gone away; Microsoft's WMV HD format is essentially a WMV file on a regular DVD. Expect MS to push this very aggressively when it comes out.
If it's not on by default it doesn't matter. In order to really make the IPv6 transition happen, OSes will also need 6to4 and Teredo on by default, which Linux and OS X do not have.
Slashdot Mirror
The kernel hackers are not complaining about a lack of binary drivers; they don't want any binary drivers.
There are a lot of Linux users saying "please give me a driver, any driver!", but I think those people should ask themselves if it's a good idea to use an OS when they have such a fundamental disagreement with the developers of that OS.
There's only one problem: WiMax doesn't work in notebooks. WiMax requires the clients to be fixed, have a large(ish) antenna pointed directly at the base station, and (in a large cell) use so much power it'd drain your battery pretty quickly.
Fedora uses an auto-build system (inherited from RHL/RHEL) which tries to build every SRPM for every known architecture. Just because a bunch of RPMs exist for weird architectures doesn't mean Fedora actually works on those architectures.
AKA the Mozilla approach. Don't like our license? We'll give you a few more to choose from!
I would prefer the X developers to just implement the Windows method, instead of requiring users who migrate to learn a new method.
:-)
Ctrl-C and Ctrl-V work fine in all the modern (i.e. GNOME) apps I've tried.
In a related vein, does anyone know how to disable the regular CTRL-C in a KDE terminal window, perhaps by making it into a menu item instead? Then I could finally use CTRL-C in the terminal.
Of course, the correct solution is to have a separate Command key, and use Command-XCV for cut, copy, and paste. This works great on the Mac.
What header? All but the first and last router don't care about anything but the source and destination address, and most times not about the source address at all. So what else is there to simplify?
You're ignoring header checksums and fragmentation.
Yes, a lot of IPv4 features were simply taken out of IPv6, but that doesn't make IPv6 unnecessary. Taking a feature out of IPv4 would result in something that's not compatible with IPv4, thus you'd have to give it a new name and upgrade all the equipment. So why not increase the address size while we're at it and call the result IPv6?
In case anyone wants some hard facts:
A. Hartstein and Thomas R. Puzak (IBM): The Optimum Pipeline Depth for a Microprocessor, ISCA 2002.
M.S. Hrishikesh, Norman P. Jouppi, Keith I. Farkas, Doug Burger, Stephen W. Keckler, Premkishore Shivakumar (UT Austin, Compaq): The Optimal Logic Depth Per Pipeline Stage is 6 to 8 FO4 Inverter Delays, ISCA 2002.
Eric Sprangle , Doug Carmean (Intel): Increasing Processor Performance by Implementing Deeper Pipelines, ISCA 2002.
A. Hartstein and Thomas R. Puzak (IBM): Optimum Power/Performance Pipeline Depth, MICRO 2003.
What all these papers have in common is that they find that increasing the pipeline depth past 20 stages increases performance.
Hello, Pentium M?
Sure, Oracle can "just type make" to create a Linux/PowerPC version of Oracle, but offering a fully tested and supported version takes effort.
Yellow Dog for the G5 has been available for a while; it's working OK in my office.
Sure, you can stick your head in the sand if you want to.
This isn't Sun-specific; the IETF recommends that people use stateless address autoconfiguration and that creates MAC-address-based IPv6 addresses.
Ideally you would replace your NAT with one that can act as a 6to4 border router, but there aren't any of those yet.
You could run Teredo on your computers, but only Windows XP supports it AFAIK.
People can yell "better routing" all day long, it doesn't make it so. Routing tables are not improved by being made larger. All it means is greater and greater functionality required of the networking hardware, which means just more and more ways for the network itself to fail.
IPv6 routing tables are much smaller because the address allocations are being done in a much more efficient manner. Also, IPv6 is simpler to route than IPv4, because header processing was simplified.
At one point Napster planned to add fingerprinting to the peer software itself. But this gets back to "never trust the client".
Never, because then you might copy it.
Even a $50 video card has DVI these days and quite a few cards have component adaptors. Sometimes it takes a bit of fiddling with Powerstrip to convince the card to output weird resolutions, but it's not impossible.
The Bravo D3 is supposed to be $350.
I don't think MPEG-4 adopted the interactivity parts of the QuickTime file format. MPEG-4 part 10 (aka H.264) is being considered for HD DVD.
Superdrive Extreme, obviously.
Red lasers haven't gone away; Microsoft's WMV HD format is essentially a WMV file on a regular DVD. Expect MS to push this very aggressively when it comes out.
With that much space and modern codecs I would expect 1080p format video on most discs. (The format will probably allow a variety of resolutions.)
Even if HD DVD is "official", it will have to compete with WMV HD (coming in July) and maybe Blu-ray.
What the article does not cover, is when we will be able to purchase non-Opteron Dual processors.
Because the answer is never.
If it's not on by default it doesn't matter. In order to really make the IPv6 transition happen, OSes will also need 6to4 and Teredo on by default, which Linux and OS X do not have.