Well, yeah, if you aren't using GNU then you wouldn't call the OS GNU/Linux. However, if you are using GNU and you want to refer to a POSIX operating system, GNU/Linux is a more specific term than Linux.
Linux, by itself, describes a variety of systems including GNU/Linux, Android, and embedded systems, as you pointed out. The term GNU/Linux allows us to refer to a specific one of those.
It may be simply a matter of being specific about what you mean. My phone runs a Linux system, as much as my laptop does. However, when I write software I'm typically targeting a POSIX OS. Linux is not, by itself, such a system, but any GNU/Linux system is.
I agree with you, but "base" isn't a very well-defined term. GNU/Linux is a (mostly) POSIX system. Your argument is pretty solid when you use that term instead.
Because GNU/Linux implements a mostly compliant POSIX OS. Linux, by itself does not. And Mozilla, X.org, and Oracle are not covered by the POSIX specification.
I disagree with Stallman, glibc should drop the joke.
On the other hand, it's not a joke about abortion, it's a joke about censorship, which is why Stallman wants to keep it. And if advocates for its removal would acknowledge what Stallman likes about it, they'd probably have an easier time with that conversation.
Just two days ago I set up CentOS on a Windows laptop provided by my employer using WSL-Distribution-Switcher. It'll download and run any distro published as a docker image on Docker Hub.
Related: What terminal emulator are people using on Windows? I'm using wsl-terminal currently, but I'm curious if there are compelling alternatives.
The POSIX operating system you are familiar with is GNU/Linux. GNU and Linux provide the (mostly compliant) interfaces that POSIX requires. Everything else is an application.
If you're running GNU on WIndows, then the POSIX OS is GNU/Windows.
Yes, there are applications on top of GNU. There are also applications on top of Windows. Their presence doens't change the name of the OS.
I've had innumerable mobile computing devices over the last 25 years, and quite a few of them for a long time. I have NEVER seen one of them spontaneously reboot because the battery couldn't output enough power at some point before the charge, measured by the battery monitor, depleted.
If Apple's devices can't run at normal speed without crashing, 12-18 months after purchase, then they're possibly the shittiest device sold in decades.
Sure. I'm not saying it wasn't *any* better. Just that it wasn't the revolution that some people make it out to be. This article calls it a "remarkable feat of engineering," which I think is a stretch. Reading it, you'd think there had never been a laptop of that size.
As far as battery life goes: the X40 had two available, and you could swap it out whenever you wanted. You could choose to take on a little extra weight for more battery life. Or you could carry several batteries. Go nuts. Whatever you want. You had choices that you really didn't get from Apple.
How do you expect patching to be in place in advance for OpenBSD if the kernel developers weren't notified?
You're missing the point. The OpenBSD team would be notified if they cooperated with the temporary embargoes that are in place to provide vendors time to patch before attacks are developed and deployed. They haven't, in the past, so they're no longer in the group that gets advance notice.
No. Spectre affects AMD and ARM as well (and likely other architectures too).
AMD CPUs were demonstrated to be vulnerable to Spectre under Linux only in a non-standard kernel configuration. In the standard configuration, they demonstrated "the ability to read data within the same process, without crossing privilege boundaries."
It's possible that future research will reveal vulnerabilities on AMD CPUs, but as of now, I don't see that one has been verified under the standard kernel configuration. (So don't enable eBPF JIT)
I've found that, for the places where I spend the most time, OSM is amazingly detailed and accurate compared to Google Maps. UW campus for instance, where foot paths are much better mapped in OSM:
Red Hat's "tuned" package will set the CPU frequency governor to a high-performance governor if the word "server" or the word "computenode" appear in the system's CPE identifier. CentOS's CPE contains neither, so a low-performance profile is selected by default.
The short version is that CentOS "out of the box" will always benchmark poorly, and you must run "tuned-adm profile throughput-performance" to get a high performance profile.
You speak only for yourself. Many of use who operate large server farms are quite happy with systemd. And the evidence suggests that those who integrate systems prefer systemd, as there are vanishingly few distributions that don't use systemd either exclusively or by default.
I picked up a copy of O'Reilly's "Ansible: Up & Running" this weekend. At a meetup last night, I got a very strong recommendation for "Ansible for DevOps," so that will be the next item on the list. Both seem to be well liked by others who've read them.
No, it hasn't. Fold change represents doubling. 2.5-fold is roughly 5.6 times as much as the original value. Python has seen 2.5 times as much traffic, not 2.5-fold as much traffic.
GNU/x.org/Gnome/Firefox/Linux just doesn't roll off the tongue too well
No one would suggest that you call it that, because the combination of GNU, X.org, GNOME, and Firefox doesn't implement any specification you can name.
You do, however, know the name of the specification that's implemented by GNU/Linux: POSIX. (Yes, I know that GNU is not strictly conformant).
Linux, by itself, does not implement POSIX. It implements its own proprietary interface.
Android implements its own interface, and when you refer to Android, it's clear what interface you mean. The same reasoning that justifies calling Android "Android" would conclude that you should call the desktop and server OS "GNU/Linux."
Slashdot Mirror
Well, yeah, if you aren't using GNU then you wouldn't call the OS GNU/Linux. However, if you are using GNU and you want to refer to a POSIX operating system, GNU/Linux is a more specific term than Linux.
Linux, by itself, describes a variety of systems including GNU/Linux, Android, and embedded systems, as you pointed out. The term GNU/Linux allows us to refer to a specific one of those.
So, why would GNU get credit when others do not?
It may be simply a matter of being specific about what you mean. My phone runs a Linux system, as much as my laptop does. However, when I write software I'm typically targeting a POSIX OS. Linux is not, by itself, such a system, but any GNU/Linux system is.
I agree with you, but "base" isn't a very well-defined term. GNU/Linux is a (mostly) POSIX system. Your argument is pretty solid when you use that term instead.
Because GNU/Linux implements a mostly compliant POSIX OS. Linux, by itself does not. And Mozilla, X.org, and Oracle are not covered by the POSIX specification.
I disagree with Stallman, glibc should drop the joke.
On the other hand, it's not a joke about abortion, it's a joke about censorship, which is why Stallman wants to keep it. And if advocates for its removal would acknowledge what Stallman likes about it, they'd probably have an easier time with that conversation.
You're a poser, friend.
You mean you found one of the people the article is about? They'll probably claim they were merely demonstrating the aforementioned bluffing. :)
https://github.com/RoliSoft/WS...
Just two days ago I set up CentOS on a Windows laptop provided by my employer using WSL-Distribution-Switcher. It'll download and run any distro published as a docker image on Docker Hub.
Related: What terminal emulator are people using on Windows? I'm using wsl-terminal currently, but I'm curious if there are compelling alternatives.
https://github.com/goreliu/wsl...
You know there is more to FOSS than Gnu?
The POSIX operating system you are familiar with is GNU/Linux. GNU and Linux provide the (mostly compliant) interfaces that POSIX requires. Everything else is an application.
If you're running GNU on WIndows, then the POSIX OS is GNU/Windows.
Yes, there are applications on top of GNU. There are also applications on top of Windows. Their presence doens't change the name of the OS.
Their priorities make the service an interesting alternative to Quad9: https://www.globalcyberallianc...
Are they also going to offer DNS over TLS?
Red Hat continues to buy good software companies and turn their products into Free Software. Most recent good example: Ansible Tower.
Red Hat does Free Software right, and does right by Free Software.
I've had innumerable mobile computing devices over the last 25 years, and quite a few of them for a long time. I have NEVER seen one of them spontaneously reboot because the battery couldn't output enough power at some point before the charge, measured by the battery monitor, depleted.
If Apple's devices can't run at normal speed without crashing, 12-18 months after purchase, then they're possibly the shittiest device sold in decades.
I can't believe people accept such claptrap.
Sure. I'm not saying it wasn't *any* better. Just that it wasn't the revolution that some people make it out to be. This article calls it a "remarkable feat of engineering," which I think is a stretch. Reading it, you'd think there had never been a laptop of that size.
As far as battery life goes: the X40 had two available, and you could swap it out whenever you wanted. You could choose to take on a little extra weight for more battery life. Or you could carry several batteries. Go nuts. Whatever you want. You had choices that you really didn't get from Apple.
MacBook Air (2008):
1280 x 800 display, 802.11 a/b/g (draft N), 12.8" x 8.94" x 0.76", 3.0 lbs
ThinkPad X40 (2004):
1024 x 768 display, 802.11 a/b/g, 10.5" x 8.3" x 1.0", 2.7 lbs
Considering they had four years, it's a pretty modest improvement over IBM's lightweight notebook.
How do you expect patching to be in place in advance for OpenBSD if the kernel developers weren't notified?
You're missing the point. The OpenBSD team would be notified if they cooperated with the temporary embargoes that are in place to provide vendors time to patch before attacks are developed and deployed. They haven't, in the past, so they're no longer in the group that gets advance notice.
No. Spectre affects AMD and ARM as well (and likely other architectures too).
AMD CPUs were demonstrated to be vulnerable to Spectre under Linux only in a non-standard kernel configuration. In the standard configuration, they demonstrated "the ability to read data within the same process, without crossing privilege boundaries."
It's possible that future research will reveal vulnerabilities on AMD CPUs, but as of now, I don't see that one has been verified under the standard kernel configuration. (So don't enable eBPF JIT)
This bug has been known and reported about since early November
November, nothing. Theo de Raadt called Intel "a nasty x86 architecture which barely has correct page protection" in 2007. He was right.
https://marc.info/?l=openbsd-m...
https://www.theregister.co.uk/...
I've found that, for the places where I spend the most time, OSM is amazingly detailed and accurate compared to Google Maps. UW campus for instance, where foot paths are much better mapped in OSM:
https://www.openstreetmap.org/...
https://www.google.com/maps/@4...
Judging the two by the specific criteria laid out by the submitter, OSM appears to have the advantage.
Red Hat's "tuned" package will set the CPU frequency governor to a high-performance governor if the word "server" or the word "computenode" appear in the system's CPE identifier. CentOS's CPE contains neither, so a low-performance profile is selected by default.
Details are available here: http://jperrin.org/centos/boos...
The short version is that CentOS "out of the box" will always benchmark poorly, and you must run "tuned-adm profile throughput-performance" to get a high performance profile.
Exactly. And what better authority than git's author? If you want to send him a patch, you just make a pull request, right?
Wait, the Linux team doesn't use github? Isn't that peculiar? :)
You speak only for yourself. Many of use who operate large server farms are quite happy with systemd. And the evidence suggests that those who integrate systems prefer systemd, as there are vanishingly few distributions that don't use systemd either exclusively or by default.
Gosh, you're going to tell me that people want to try things in person before they commit to buying(/keeping) them? I'm shocked! Shocked, I tell you!
I picked up a copy of O'Reilly's "Ansible: Up & Running" this weekend. At a meetup last night, I got a very strong recommendation for "Ansible for DevOps," so that will be the next item on the list. Both seem to be well liked by others who've read them.
My thoughts exactly. Blockchain credit history? Great! Now every fraudulent entry is there permanently, and can't be removed! Brilliant!
and has grown by 2.5-fold in that time
No, it hasn't. Fold change represents doubling. 2.5-fold is roughly 5.6 times as much as the original value. Python has seen 2.5 times as much traffic, not 2.5-fold as much traffic.
GNU/x.org/Gnome/Firefox/Linux just doesn't roll off the tongue too well
No one would suggest that you call it that, because the combination of GNU, X.org, GNOME, and Firefox doesn't implement any specification you can name.
You do, however, know the name of the specification that's implemented by GNU/Linux: POSIX. (Yes, I know that GNU is not strictly conformant).
Linux, by itself, does not implement POSIX. It implements its own proprietary interface.
Android implements its own interface, and when you refer to Android, it's clear what interface you mean. The same reasoning that justifies calling Android "Android" would conclude that you should call the desktop and server OS "GNU/Linux."