It's not much of a problem because unlike a typical Windows install where you've got too much third party legacy cruft, a Linux install might be 100% free of 32-bit binaries.
Come to think of it, Slashdot probably has one of the highest ad-blocking rates out there due to the highly technical audience, but is somehow still around.
The really scummy part is that they harbor DDoS-for-hire sites, which in turn helps drive their business. It would be like if a radar gun company also sold radar detectors.
Things I can list off the top of my head (these may have changed in more recent USB specs):
1. Lack of DMA support. In the age of the IOMMU, it's no longer a security risk.
2. It's a purely polled interface. The host must initiate all communications with the device, rather than the device being able to do it. In many cases, the polling rate has to be quite high such as to not introduce unacceptable latency, but then this increases power usage. There's a reason laptop keyboards tend to use PS/2 rather than USB.
3. Introduces new connectors every time they feel like it.
The problem is that it's a jack of all trades, master of none. From a technical standpoint, PS/2 is better for simple keyboards and mice, SATA/eSATA is better for storage, and USB just plain doesn't work for many things such as eGPUs because of the lack of DMA.
ISIS is great at hiding their communications in plain sight, which our (un)intelligence agencies seem to be awful at detecting. See: Paris terrorists' unencrypted communications.
Even if it were decided that the government must respect the second amendment in all govt-owned buildings, you could just do a complete end run as such:
1. Transfer ownership of building to private entity
2. Lease building to original govt entity
3. Landlord enacts a no-gun rule
Is Steam's 2 week / 2 hours gameplay limit too strict for Aussie law?
I think that's too strict in general. The problem is it's not 2 hours of gameplay, it's 2 hours of having the game open. I might easily spend 2 hours trying to get a buggy game to work. This can be especially problematic with games that use their own download/update system rather than distributing their payload via steam, since Steam would track that time where the launcher is open, downloading the game/updates.
For simple wifi stuff, wicd is a nice lightweight solution.
The most ironic part of the whole thing is that the problems described here as justification for systemd's awful interface naming are all things that should be taken care of by a higher level tool such as network-manager.
You're right, a phone shouldn't be different. So why is it that the FBI is expecting special treatment from Apple? Proper search and seizure doesn't mean they can force manufacturers to bend to their will.
You only need the PIN/passphrase to unlock the phone after reboot or after a certain period of time. All other times, you just use the fingerprint scanner.
Probably good for Apple, actually. The problem with using this case as a precedent for whether or not a phone manufacturer should be required to unlock a phone or not is that Apple could have assisted them with the unlocking by doing what the FBI requested (writing a custom OS to facilitate unlocking). But now, Apple has already plugged that hole in newer models, which means if the same case were to happen with a newer iPhone, it would be even more in Apple's favor. On top of that, it wouldn't surprise me if the reason the FBI has backed out of the case is because they didn't think they were going to win and so didn't want to set a precedent which would be unfavorable to them.
Assuming there isn't some similar hole on the newer phones, I'm pretty sure Apple (and privacy, for that matter) is the big winner here.
It's a good idea, but poor execution. Ideally, it should all happen automatically, where the ISP can indicate to the service provider (better yet, the client, so that privacy issues are avoided) to switch to a lower-bandwidth mode. The problem with T-Mobile's plan is that if everyone did it, you would have to call up every ISP on the planet and ask to be zero-rated, which is an undue burden on the provider. I do think the best solution would be a less extreme version of the "soft cap" BS that AT&T does: the more bandwidth you use, the lower your QoS priority goes. It would never deny you any unused bandwidth or throttle you if there was room on the airwaves, but would still serve to discourage bandwidth hogging.
The summary is worded extremely poorly. The normal rate is 94.6 (adjusted) or 76 (unadjusted), but this study suggests it's a 72% rate for programmers (adjusted).
The Mac Pro specifically went from being a dual-CPU workstation that was easily competitive with any other workstation to a single-CPU machine crammed into a tiny case with minimal expansion and insufficient cooling.
From what I've noticed in Python, the most important packages tend to be available in a distro's package manager anyway. But packages (in both Python and node) can manually un-break such dependency issues by directly specifying the repo location of the package rather than relying on pip/npm.
You don't? I don't remember the last time I wanted to run a graphical program but forgot the name entirely. Best part is it accepts both the full name of the program (e.g. Computer Management) and the exact filename (compmgmt.msc), with or without the extension. I certainly don't use every little tool installed on the machine, but if I never use it then I don't need to know the name.
Simple: because on any non-garbage UI, the buttons on the taskbar will display the window title, so that the user can easily switch between any open windows. If you put it on the side, you end up with a ton of screen space used to get any good amount of text displayed (at least 200px), while the rest of the taskbar goes unused unless you have a ton of app-launching buttons like the Windows taskbar. On a bottom taskbar, you end up with a lot less wasted space with only a few windows open, and the window buttons can shorten themselves if there isn't room for all of them.
Now, having a taskbar that is needlessly thick is another issue, and is one of the many issues I have with the default Win7+ configuration.
Unfortunately this kind of crap hasn't lost enough popularity. Where I work, it's a pretty even distribution between Win, Mac, and Linux. Unfortunately, all the Linux users other than me run Ubuntu, and all but the two that I installed Xfce for are running Unity. I think one thing that might behoove Xfce would be to have additional starting configs (instead of just "one empty panel" and "default"), for example a "Gnome 2 Layout" and a "Windows-like Layout". A properly configured Win7 taskbar is actually a very good UI, shame the defaults are so awful.
But by far the most efficient way to launch things from the start menu is to just press the windows key and start typing the name of the program. It could be a completely unorganized flat mess for all I care.
Because Xubuntu users want to have only the Xfce desktop, without having to install the Unity desktop first, which, if you're never going to use it, means you're just wasting hard disk space.
Well, it can just be uninstalled. Or you can do it like other distros do and ask which one you want to have installed during the installation.
For one, there's the fact that they actually used to make a server, but specifically decided to get out of the server business. Apple is good at selling things to end consumers, not so much servers.
Not to mention the fact that you can run a cable through a tight space first, and then put the connector on after. If the connector is pre-attached, even if it fits, it has a higher risk of getting damage if you have to pull it through things.
QSFP+ does just that: 4 pairs of 10Gb each. The problem with huge bundles of cables is that they tend to be less flexible, have more points of failure, and have much more expensive equipment.
Slashdot Mirror
It's not much of a problem because unlike a typical Windows install where you've got too much third party legacy cruft, a Linux install might be 100% free of 32-bit binaries.
Come to think of it, Slashdot probably has one of the highest ad-blocking rates out there due to the highly technical audience, but is somehow still around.
The really scummy part is that they harbor DDoS-for-hire sites, which in turn helps drive their business. It would be like if a radar gun company also sold radar detectors.
Things I can list off the top of my head (these may have changed in more recent USB specs):
1. Lack of DMA support. In the age of the IOMMU, it's no longer a security risk.
2. It's a purely polled interface. The host must initiate all communications with the device, rather than the device being able to do it. In many cases, the polling rate has to be quite high such as to not introduce unacceptable latency, but then this increases power usage. There's a reason laptop keyboards tend to use PS/2 rather than USB.
3. Introduces new connectors every time they feel like it.
The problem is that it's a jack of all trades, master of none. From a technical standpoint, PS/2 is better for simple keyboards and mice, SATA/eSATA is better for storage, and USB just plain doesn't work for many things such as eGPUs because of the lack of DMA.
ISIS is great at hiding their communications in plain sight, which our (un)intelligence agencies seem to be awful at detecting. See: Paris terrorists' unencrypted communications.
Even if it were decided that the government must respect the second amendment in all govt-owned buildings, you could just do a complete end run as such:
1. Transfer ownership of building to private entity
2. Lease building to original govt entity
3. Landlord enacts a no-gun rule
Is Steam's 2 week / 2 hours gameplay limit too strict for Aussie law?
I think that's too strict in general. The problem is it's not 2 hours of gameplay, it's 2 hours of having the game open. I might easily spend 2 hours trying to get a buggy game to work. This can be especially problematic with games that use their own download/update system rather than distributing their payload via steam, since Steam would track that time where the launcher is open, downloading the game/updates.
For simple wifi stuff, wicd is a nice lightweight solution.
The most ironic part of the whole thing is that the problems described here as justification for systemd's awful interface naming are all things that should be taken care of by a higher level tool such as network-manager.
Ironically, there's no actual Linux (as in the kernel) to be found here. Just userspace stuff.
You're right, a phone shouldn't be different. So why is it that the FBI is expecting special treatment from Apple? Proper search and seizure doesn't mean they can force manufacturers to bend to their will.
You only need the PIN/passphrase to unlock the phone after reboot or after a certain period of time. All other times, you just use the fingerprint scanner.
Probably good for Apple, actually. The problem with using this case as a precedent for whether or not a phone manufacturer should be required to unlock a phone or not is that Apple could have assisted them with the unlocking by doing what the FBI requested (writing a custom OS to facilitate unlocking). But now, Apple has already plugged that hole in newer models, which means if the same case were to happen with a newer iPhone, it would be even more in Apple's favor. On top of that, it wouldn't surprise me if the reason the FBI has backed out of the case is because they didn't think they were going to win and so didn't want to set a precedent which would be unfavorable to them.
Assuming there isn't some similar hole on the newer phones, I'm pretty sure Apple (and privacy, for that matter) is the big winner here.
It's a good idea, but poor execution. Ideally, it should all happen automatically, where the ISP can indicate to the service provider (better yet, the client, so that privacy issues are avoided) to switch to a lower-bandwidth mode. The problem with T-Mobile's plan is that if everyone did it, you would have to call up every ISP on the planet and ask to be zero-rated, which is an undue burden on the provider. I do think the best solution would be a less extreme version of the "soft cap" BS that AT&T does: the more bandwidth you use, the lower your QoS priority goes. It would never deny you any unused bandwidth or throttle you if there was room on the airwaves, but would still serve to discourage bandwidth hogging.
The summary is worded extremely poorly. The normal rate is 94.6 (adjusted) or 76 (unadjusted), but this study suggests it's a 72% rate for programmers (adjusted).
The Mac Pro specifically went from being a dual-CPU workstation that was easily competitive with any other workstation to a single-CPU machine crammed into a tiny case with minimal expansion and insufficient cooling.
From what I've noticed in Python, the most important packages tend to be available in a distro's package manager anyway. But packages (in both Python and node) can manually un-break such dependency issues by directly specifying the repo location of the package rather than relying on pip/npm.
You don't? I don't remember the last time I wanted to run a graphical program but forgot the name entirely. Best part is it accepts both the full name of the program (e.g. Computer Management) and the exact filename (compmgmt.msc), with or without the extension. I certainly don't use every little tool installed on the machine, but if I never use it then I don't need to know the name.
16GB is the 1366x768 of the phone world. I have 16GB...on my ancient iPhone 4s.
Simple: because on any non-garbage UI, the buttons on the taskbar will display the window title, so that the user can easily switch between any open windows. If you put it on the side, you end up with a ton of screen space used to get any good amount of text displayed (at least 200px), while the rest of the taskbar goes unused unless you have a ton of app-launching buttons like the Windows taskbar. On a bottom taskbar, you end up with a lot less wasted space with only a few windows open, and the window buttons can shorten themselves if there isn't room for all of them.
Now, having a taskbar that is needlessly thick is another issue, and is one of the many issues I have with the default Win7+ configuration.
Unfortunately this kind of crap hasn't lost enough popularity. Where I work, it's a pretty even distribution between Win, Mac, and Linux. Unfortunately, all the Linux users other than me run Ubuntu, and all but the two that I installed Xfce for are running Unity. I think one thing that might behoove Xfce would be to have additional starting configs (instead of just "one empty panel" and "default"), for example a "Gnome 2 Layout" and a "Windows-like Layout". A properly configured Win7 taskbar is actually a very good UI, shame the defaults are so awful.
But by far the most efficient way to launch things from the start menu is to just press the windows key and start typing the name of the program. It could be a completely unorganized flat mess for all I care.
Because Xubuntu users want to have only the Xfce desktop, without having to install the Unity desktop first, which, if you're never going to use it, means you're just wasting hard disk space.
Well, it can just be uninstalled. Or you can do it like other distros do and ask which one you want to have installed during the installation.
For one, there's the fact that they actually used to make a server, but specifically decided to get out of the server business. Apple is good at selling things to end consumers, not so much servers.
Not to mention the fact that you can run a cable through a tight space first, and then put the connector on after. If the connector is pre-attached, even if it fits, it has a higher risk of getting damage if you have to pull it through things.
QSFP+ does just that: 4 pairs of 10Gb each. The problem with huge bundles of cables is that they tend to be less flexible, have more points of failure, and have much more expensive equipment.