Systemd has plenty of advantages. People just hate change and having to type systemctl instead of rc.d. It does suck that BSD is getting increasingly left out in the cold with this and udev, though.
Academic positions being highly sought after are why the whole PhD/post doc/tenure thing sucks so hard and is so abusive.
It's not a bad life when you get tenure. A comfortable living and you get to do what you want to do. Unfortunately, universities sell that unattainable dream to get cheap labors, with TAs/RAs getting paid less than a typical Wal-Mart cashier.
If you want to use it in a commercial application, you have to give your improvements back to the community.
It's about cooperation and giving back, rather than just leeching other's work. Companies put lots of work into GPL software all the time, not because they're nice guys, but because they want to modify and expand GPL software for their own purposes. This benefits everyone. Sure, the companies might want to keep their derivative works closed, but it benefits everyone more to have it open. It's not like they have anything to complain about. It's free code.
BSD licenses, on the other hand, just encourage leeching.
The average computer user is not going to be monkeying around in the BIOS. This is about making life more difficult for non-MS OSes, and reverting the mistake that was the open x86 platform.
Last I heard he uses Fedora and XFCE4 after leaving the sinking ship that is GNOME 3. (Choice quote: "Who do I need to fuck to get standard font size and panel options, instead of having to wade through this kind of "unsupported and random extensions that look ugly as hell and break randomly" crap?")
Arch Linux (obviously) is using Python 3 by default for a long time. On the scientific side, numpy/scipy and all that stuff are ported, but a lot of other scientific modules still aren't, which sucks.
No one is saying Alcubierre is wrong, but his scheme would violate causality, like every other FTL method. If the universe is non-causal, this wouldn't be a problem. But it probably is.
Still irrelevant. These machines don't violate SR, instead use apparent loopholes on GR. Another explanation. More.
Even in cases where SR doesn't apply, it still allows for causality violations (ie time travel) in GR. This is an absolute feature of any method of ftl. It does not matter whether it's a wormhole or a warpdrive or whatever. It sucks, but it's probably telling that every apparent superluminal phenomena (ie quantum tunnelling, entanglement) can't transfer informaton.
FTL implies at least backwards communication is possible under any method you can think of. If you get get to Alpha Centauri by stuffing yourself in your ass, it will still allow backwards time travel.
Microsoft Bob is about 15 years old or so. I don't think it's quite right to say that it's the penultimate in skeumorphic design since there's been a lot of skeumorphic stuff developed since then.
Oh yeah. Arch significantly more bleeding edge than Debian Unstable (IIRC, about 40% out of date, which is really good as these things go). Not making any pretense of patching does help things move quickly.
Contrary to many people, I've had great luck with Arch. Never had any serious breakage. I have on the other hand had huge clusterfucks with Mint and Fedora. If something does break, it's easy to revert back. I think the key is to pacman -Syu on at least a weekly basis to keep the number of updates at any one time small. Another advantage is that I run pretty spare systems, so there is less too break (that's the main reason I use Arch: bloat avoidance).
Arch is generally easier to configure, since most stuff is in a few plaintext files. I find the way Debian does things very difficult to understand in comparison.
Ah, the old linguistic superiority thing. English of $INSERT PAST TIME was the greatest flowering of human speech ever and represented the most precise and beautiful way to communicate.
Actually, "color" was oringally spelled "color." The Brits changed it to "colour" in the 19th century, along with several other spellings to make it sound more French.
England English has actually changed much more rapidly than American. You want to hear old 18th century English spoken today? Go to some of the more remote places in Appalachia. It's the closest. Not what your stereotype would have you believe, huh?
Ok, fair enough, although tags can still do *most* of that. I never really figured activities out. Tiling WMs are difficult to get used to, but you really can work more efficiently if you aren't having to move windows around. Getting windows in/out of view is just so easy in comparison, at least after you've been using it for a week or so.
I actually like KDE4, at least its more recent versions. It just takes too much memory. Also, I still have some persistent bugs, including one time where akonadi or whatever decided to fill up my entire home partition with.Xsession-errors. Oh, yeah, and it's incredibly difficult to actually disable it. That was fun.
After being a long time GNOME user, I bounced around DEs and WMs for quite a while before finally settling on DWM. I tried ratpoison a long time ago, and didn't like it, but I guess I just didn't give it enough of a chance.
If you use a tiling WM like DWM or Awesome, you have a similar concept (as far as I can tell) with tags. Each window is assigned a tag, and you can have one or more tags in your current view.
I can go to an Indian buffet or get Phad Thai or a gyro for $4 to $8. Why would I spend the same on a big mac, a thing of sugar water, and greasy fries.
People's taste buds are all kinds of fucked up these days.
Looking at "six sigma" is stupid. If you are talking about the management fad, it assumes the data follows a normal distribution. Generally, frequentist statistics is misleading. It's not wrong, but it is very commonly used improperly. For example, if you hear that a null hypothesis that the mean of a distribution is less than zero, H0:mu H0 is true, and that the data follow the assumed distribution, 99% of the sample means you get would be less than 0.
This article actually uses Bayesian statistics (samples the posterior PDF using MCMC), rather than frequentist.
Slashdot Mirror
This explains it pretty well:
Systemd has plenty of advantages. People just hate change and having to type systemctl instead of rc.d. It does suck that BSD is getting increasingly left out in the cold with this and udev, though.
Academic positions being highly sought after are why the whole PhD/post doc/tenure thing sucks so hard and is so abusive.
It's not a bad life when you get tenure. A comfortable living and you get to do what you want to do. Unfortunately, universities sell that unattainable dream to get cheap labors, with TAs/RAs getting paid less than a typical Wal-Mart cashier.
Enjoy being fat and out of shape then, if you don't like exercise.
If you want to use it in a commercial application, you have to give your improvements back to the community.
It's about cooperation and giving back, rather than just leeching other's work. Companies put lots of work into GPL software all the time, not because they're nice guys, but because they want to modify and expand GPL software for their own purposes. This benefits everyone. Sure, the companies might want to keep their derivative works closed, but it benefits everyone more to have it open. It's not like they have anything to complain about. It's free code.
BSD licenses, on the other hand, just encourage leeching.
The average computer user is not going to be monkeying around in the BIOS. This is about making life more difficult for non-MS OSes, and reverting the mistake that was the open x86 platform.
Last I heard he uses Fedora and XFCE4 after leaving the sinking ship that is GNOME 3. (Choice quote: "Who do I need to fuck to get standard font size and panel options, instead of having to wade through this kind of "unsupported and random extensions that look ugly as hell and break randomly" crap?")
Wow we're going to have 19th century steampunk after a solar flare? I'm so hard right now.
Arch Linux (obviously) is using Python 3 by default for a long time. On the scientific side, numpy/scipy and all that stuff are ported, but a lot of other scientific modules still aren't, which sucks.
No one is saying Alcubierre is wrong, but his scheme would violate causality, like every other FTL method. If the universe is non-causal, this wouldn't be a problem. But it probably is.
Still irrelevant. These machines don't violate SR, instead use apparent loopholes on GR.
Another explanation.
More.
Even in cases where SR doesn't apply, it still allows for causality violations (ie time travel) in GR. This is an absolute feature of any method of ftl. It does not matter whether it's a wormhole or a warpdrive or whatever. It sucks, but it's probably telling that every apparent superluminal phenomena (ie quantum tunnelling, entanglement) can't transfer informaton.
Sigh.
Here
You are wrong. Go take a physics class.
For chrissake:
Here
FTL implies at least backwards communication is possible under any method you can think of. If you get get to Alpha Centauri by stuffing yourself in your ass, it will still allow backwards time travel.
IQ is defined as being normally distributed. Thus the mean and the median are the same.
Microsoft Bob is about 15 years old or so. I don't think it's quite right to say that it's the penultimate in skeumorphic design since there's been a lot of skeumorphic stuff developed since then.
It's ok, RMS actually hates Debian.
Oh yeah. Arch significantly more bleeding edge than Debian Unstable (IIRC, about 40% out of date, which is really good as these things go). Not making any pretense of patching does help things move quickly.
Contrary to many people, I've had great luck with Arch. Never had any serious breakage. I have on the other hand had huge clusterfucks with Mint and Fedora. If something does break, it's easy to revert back. I think the key is to pacman -Syu on at least a weekly basis to keep the number of updates at any one time small. Another advantage is that I run pretty spare systems, so there is less too break (that's the main reason I use Arch: bloat avoidance).
Arch is generally easier to configure, since most stuff is in a few plaintext files. I find the way Debian does things very difficult to understand in comparison.
Ah, the old linguistic superiority thing. English of $INSERT PAST TIME was the greatest flowering of human speech ever and represented the most precise and beautiful way to communicate.
Actually, "color" was oringally spelled "color." The Brits changed it to "colour" in the 19th century, along with several other spellings to make it sound more French.
England English has actually changed much more rapidly than American. You want to hear old 18th century English spoken today? Go to some of the more remote places in Appalachia. It's the closest. Not what your stereotype would have you believe, huh?
Maths=
Mathematical
Anti
Telharsic
Harfatum
Septomin
Ok, fair enough, although tags can still do *most* of that. I never really figured activities out. Tiling WMs are difficult to get used to, but you really can work more efficiently if you aren't having to move windows around. Getting windows in/out of view is just so easy in comparison, at least after you've been using it for a week or so.
I actually like KDE4, at least its more recent versions. It just takes too much memory. Also, I still have some persistent bugs, including one time where akonadi or whatever decided to fill up my entire home partition with .Xsession-errors. Oh, yeah, and it's incredibly difficult to actually disable it. That was fun.
After being a long time GNOME user, I bounced around DEs and WMs for quite a while before finally settling on DWM. I tried ratpoison a long time ago, and didn't like it, but I guess I just didn't give it enough of a chance.
If you use a tiling WM like DWM or Awesome, you have a similar concept (as far as I can tell) with tags. Each window is assigned a tag, and you can have one or more tags in your current view.
Because it isn't fucking disgusting?
I can go to an Indian buffet or get Phad Thai or a gyro for $4 to $8. Why would I spend the same on a big mac, a thing of sugar water, and greasy fries.
People's taste buds are all kinds of fucked up these days.
Put down the bong.
Looking at "six sigma" is stupid. If you are talking about the management fad, it assumes the data follows a normal distribution. Generally, frequentist statistics is misleading. It's not wrong, but it is very commonly used improperly. For example, if you hear that a null hypothesis that the mean of a distribution is less than zero, H0:mu H0 is true, and that the data follow the assumed distribution, 99% of the sample means you get would be less than 0.
This article actually uses Bayesian statistics (samples the posterior PDF using MCMC), rather than frequentist.
Haha, no it's not a badge of honor in the gay community. There's a huge stigma that actually keeps people from getting tested.
Because Congress listens to people like us more than Microsoft.