Most "middle boxes" won't forward unknown protocols. They're too paranoid about security. Sure, they're willing to transmit absolutely anything over port 443. But an unknown IP type might be a HACKER! Oh noes!
Not only that, lots of them break unknown TCP options. Or TCP windows. I remember years when TCP ECN and window sizes wouldn't work on random internet sites because of their short-sighted, STUPID firewall boxes.
That's why QUIC is completely encrypted. Random hardware and software providers have proven they don't know what they're doing. So don't let them see into the protocol at all. They would only break it.
In actually poor countries with actual poor people, some of them don't have tax-funded mass transportation at all. And yet they manage to do it as private enterprise and make money doing it.
Some guy buys an old truck, fixes it up, drives it around the city with people hanging off the sides while his cousin collects fees or kicks them off the truck.
This doesn't require air conditioning, union drivers and union maintenance technicians. All it needs is free enterprise and an appalling lack of safety regulations.
Even with congestion this is still better than trying to find a taxi or waiting for a bus.
Maybe if mass transit systems weren't so terrible, more people would use them. Cities should try being competitive and offer a better service for once.
Because AMD is careful not to cross privilege levels but Spectre attacks are user mode to user mode. So even though they may be two different users they are still in Ring 3. Spectre can only be used against kernel code if the kernel is convinced to run a user's code for some reason. Like an eBPF byte-code, for example.
But it can work really well for a sandboxed program to steal information from outside the sandbox.
So AMD is still vulnerable to speculation attacks.
I can't convince anyone without a copy of Microsoft's Edge revision control. So I guess think whatever you like.
But if they didn't create a new revision history, then yes it is the same browser. Firefox is still Firefox even though there's nothing left of the original. Linux is still Linux. I think there was an old bit of TTY code and some floppy driver code in the latest Linux left from 1993, but all the rest of it changed, and it's still the same thing.
By replacing small pieces of Firefox with new pieces that use Rust, they are avoiding the Netscape problem. They aren't rewriting the whole thing. But one day the whole browser will be Rust, replaced a piece at a time.
You don't think Microsoft Edge still uses pieces of Spyglass or Mosaic do you? And yet, at no point did they rewrite the whole thing.
I've come to the conclusion that there's only about 500 Linux desktop users in the WORLD who have PulseAudio problems. They're all the same people posting on forums about it.
Every other Linux desktop user uses the distro default which is usually PulseAudio, and it works.
Once Firefox e10s (Electrolysis) with sandboxing is enabled by default, every Firefox content process will be independent and restricted.
Most ALSA devices cannot handle multiple open. And the ALSA solution, dmix, requires shared memory which is a thing sandboxes do not really want to have, and dmix does not understand sandboxes, so it would probably have to be forked and modified.
So Firefox can write their own sound server to get sound data from each independent content tab, or blow huge holes in the sandboxes for ALSA dmix, or they can just use the sound server that already exists and is used by 98% of Linux desktop users: PulseAudio.
It's more about the legal right to STOP OTHER WORKERS.
OF course Uber drivers could unionize, collectively stop working for Uber until their demands are met. The problem Unions have, and why I think most of them are out of date and stupid, is that there are plenty of other people willing to work for the company under the current conditions. The Union hates that. What THEY want is to be the only workers legally allowed to do the job.
My opinion is that if the Uber work conditions are bad enough that unionizing makes sense then there wouldn't BE any "scab" workers because they'd all see it makes sense to unionize.
But the way it actually is, this is a few drivers that think they can do better, but don't want the competition.
Look at what driver groups are trying to do in Austin after Uber left. It's possible to do, but the problems they're having show that it is a lot more difficult and expensive than most people think.
Even mSATA is faster than any SATA cable, internal or external. NVMe (which includes M.2 which is just a 4-lane PCIe connector) is much, much faster.
Maximum speed for SATA is 600 MB/s. Maximum speed for an M.2 drive is about 3000 MB/s. I have a workstation with a 1.2 TB Intel 750 that does 2150 MB/s sequential.
SATA isn't fast anymore. It's slower than iSCSI over 10 Gbps Ethernet!
I wasn't aware that anyone read "literature" these days. I don't. Science fiction, fantasy, mystery and some romance, sure. But boring, depressing stories about the "human condition" that try extra hard to use metaphor? Nope. No one reads that junk.
So who cares about these irrelevant literary awards?
That "crappy GPU" is more cores. Specialized cores, but even the Intel GPU is ridiculously fast for the right kind of code. Now that we're getting Vulkan and DX 12 software should be able to run GPU compute on the Intel or AMD integrated GPU while doing video on the discrete card.
I predict a future with a lot more OpenCL code in it. I also predict a future with more idiot gamers who complain that using all of the CPU cores plus the integrated GPU ruins their 4.6 GHz overclocks.
Windows 8.1 and Windows 10 both use a fast boot mode that is essentially hibernation. But if it detects a disk change since last boot it has to abort and start over from scratch. That's where the long delay comes from.
I don't recall how OS X reports memory use but double check the difference between virtual and real. For example, right now my Firefox is using 3.5 GB virtual but only 1.1 GB real. And I've got piles of stuff open at the moment.
Odd. I installed a new Thunderbird on a Windows 10 install just a couple months ago, and it picked up the self-signed cert on my mail server just fine. Unless by "recent builds" you mean something newer than October.
"Make sure a white male won" You appear to be spewing back whatever blog you last read that claimed the Puppies were racist and sexist.
Right there in the Wired article is a woman who was on the Sad Puppies recommended list. She excused herself when the politics showed up.
That Correia and Togerson are racists is also laughable. If you believe it, show some evidence.
The Sad Puppies opposition worked hard to push the narrative that Puppies are racist and sexist.
Their racist/sexist accusation comes down to "If Puppies are not FOR promoting Hugos for authors because of non-white race and non-male gender, then Puppies must be racist/sexist." Now there is a fallacious argument. The old "if you're not for us you're against us" nonsense.
Go executables are static by design. It is pretty great to be able to copy it into place and it just works. Add a couple other build options to remove the ability to bind to external C code and include a Go DNS resolver, and the binary can be put into a completely empty container. Now that's great. It's the difference between 8 MB and 300 MB containers.
The arguments about fixing bugs in shared libraries do apply, but that's a problem with containers too, so you need a policy for rebuilding containers with bugs as well as Go binaries with bugs.
Slashdot Mirror
Most "middle boxes" won't forward unknown protocols. They're too paranoid about security. Sure, they're willing to transmit absolutely anything over port 443. But an unknown IP type might be a HACKER! Oh noes!
Not only that, lots of them break unknown TCP options. Or TCP windows. I remember years when TCP ECN and window sizes wouldn't work on random internet sites because of their short-sighted, STUPID firewall boxes.
That's why QUIC is completely encrypted. Random hardware and software providers have proven they don't know what they're doing. So don't let them see into the protocol at all. They would only break it.
In actually poor countries with actual poor people, some of them don't have tax-funded mass transportation at all. And yet they manage to do it as private enterprise and make money doing it.
Some guy buys an old truck, fixes it up, drives it around the city with people hanging off the sides while his cousin collects fees or kicks them off the truck.
This doesn't require air conditioning, union drivers and union maintenance technicians. All it needs is free enterprise and an appalling lack of safety regulations.
Even with congestion this is still better than trying to find a taxi or waiting for a bus.
Maybe if mass transit systems weren't so terrible, more people would use them. Cities should try being competitive and offer a better service for once.
Because AMD is careful not to cross privilege levels but Spectre attacks are user mode to user mode. So even though they may be two different users they are still in Ring 3. Spectre can only be used against kernel code if the kernel is convinced to run a user's code for some reason. Like an eBPF byte-code, for example.
But it can work really well for a sandboxed program to steal information from outside the sandbox.
So AMD is still vulnerable to speculation attacks.
I can't convince anyone without a copy of Microsoft's Edge revision control. So I guess think whatever you like.
But if they didn't create a new revision history, then yes it is the same browser. Firefox is still Firefox even though there's nothing left of the original. Linux is still Linux. I think there was an old bit of TTY code and some floppy driver code in the latest Linux left from 1993, but all the rest of it changed, and it's still the same thing.
And as far as I know, Edge was not a rewrite, it was IE with everything old and legacy ripped out.
By replacing small pieces of Firefox with new pieces that use Rust, they are avoiding the Netscape problem. They aren't rewriting the whole thing. But one day the whole browser will be Rust, replaced a piece at a time.
You don't think Microsoft Edge still uses pieces of Spyglass or Mosaic do you? And yet, at no point did they rewrite the whole thing.
They had this in 2005. It was called DPI. It scaled really well. Then some IDIOTS decided it should be force set to 92.
/vmlinux and /dev/mem are just wasting space. I deleted them and nothing bad happened. . .
I've come to the conclusion that there's only about 500 Linux desktop users in the WORLD who have PulseAudio problems. They're all the same people posting on forums about it.
Every other Linux desktop user uses the distro default which is usually PulseAudio, and it works.
This is how I understand it.
Once Firefox e10s (Electrolysis) with sandboxing is enabled by default, every Firefox content process will be independent and restricted.
Most ALSA devices cannot handle multiple open. And the ALSA solution, dmix, requires shared memory which is a thing sandboxes do not really want to have, and dmix does not understand sandboxes, so it would probably have to be forked and modified.
So Firefox can write their own sound server to get sound data from each independent content tab, or blow huge holes in the sandboxes for ALSA dmix, or they can just use the sound server that already exists and is used by 98% of Linux desktop users: PulseAudio.
It's more about the legal right to STOP OTHER WORKERS.
OF course Uber drivers could unionize, collectively stop working for Uber until their demands are met. The problem Unions have, and why I think most of them are out of date and stupid, is that there are plenty of other people willing to work for the company under the current conditions. The Union hates that. What THEY want is to be the only workers legally allowed to do the job.
My opinion is that if the Uber work conditions are bad enough that unionizing makes sense then there wouldn't BE any "scab" workers because they'd all see it makes sense to unionize.
But the way it actually is, this is a few drivers that think they can do better, but don't want the competition.
Look at what driver groups are trying to do in Austin after Uber left. It's possible to do, but the problems they're having show that it is a lot more difficult and expensive than most people think.
All of that software you listed runs great on a Surface Pro 4, aka "a tablet", aka "fondleslab."
Probably not what you're thinking of though since the price on that plus keyboard is $1,200+
I don't think it should be public. That just provides a handle for people to harass the domain owner.
Next I suppose people will want IP packets to have unique machine identities attached, or for print shops to get ID before doing print runs.
Even mSATA is faster than any SATA cable, internal or external. NVMe (which includes M.2 which is just a 4-lane PCIe connector) is much, much faster.
Maximum speed for SATA is 600 MB/s. Maximum speed for an M.2 drive is about 3000 MB/s. I have a workstation with a 1.2 TB Intel 750 that does 2150 MB/s sequential.
SATA isn't fast anymore. It's slower than iSCSI over 10 Gbps Ethernet!
I wasn't aware that anyone read "literature" these days. I don't. Science fiction, fantasy, mystery and some romance, sure. But boring, depressing stories about the "human condition" that try extra hard to use metaphor? Nope. No one reads that junk.
So who cares about these irrelevant literary awards?
They look amazing even upscaled. They make great PC monitors. They're affordable. I recently saw several that were on sale cheaper than 1080 models.
That "crappy GPU" is more cores. Specialized cores, but even the Intel GPU is ridiculously fast for the right kind of code. Now that we're getting Vulkan and DX 12 software should be able to run GPU compute on the Intel or AMD integrated GPU while doing video on the discrete card.
I predict a future with a lot more OpenCL code in it. I also predict a future with more idiot gamers who complain that using all of the CPU cores plus the integrated GPU ruins their 4.6 GHz overclocks.
Windows 8.1 and Windows 10 both use a fast boot mode that is essentially hibernation. But if it detects a disk change since last boot it has to abort and start over from scratch. That's where the long delay comes from.
There's some way to turn that off.
I don't recall how OS X reports memory use but double check the difference between virtual and real. For example, right now my Firefox is using 3.5 GB virtual but only 1.1 GB real. And I've got piles of stuff open at the moment.
Probably because they were inventing Rust at the same time they were using it to write Servo.
Odd. I installed a new Thunderbird on a Windows 10 install just a couple months ago, and it picked up the self-signed cert on my mail server just fine. Unless by "recent builds" you mean something newer than October.
"Make sure a white male won" You appear to be spewing back whatever blog you last read that claimed the Puppies were racist and sexist.
Right there in the Wired article is a woman who was on the Sad Puppies recommended list. She excused herself when the politics showed up.
That Correia and Togerson are racists is also laughable. If you believe it, show some evidence.
The Sad Puppies opposition worked hard to push the narrative that Puppies are racist and sexist.
Their racist/sexist accusation comes down to "If Puppies are not FOR promoting Hugos for authors because of non-white race and non-male gender, then Puppies must be racist/sexist." Now there is a fallacious argument. The old "if you're not for us you're against us" nonsense.
Go executables are static by design. It is pretty great to be able to copy it into place and it just works. Add a couple other build options to remove the ability to bind to external C code and include a Go DNS resolver, and the binary can be put into a completely empty container. Now that's great. It's the difference between 8 MB and 300 MB containers.
The arguments about fixing bugs in shared libraries do apply, but that's a problem with containers too, so you need a policy for rebuilding containers with bugs as well as Go binaries with bugs.