Maybe in 10 or 15 years Firefox will be production ready. So instead of crashing several times daily, it might only crash several times weekly.
Are you sure you're using the same Firefox as me? It crashes less than once a year, and that's on Debian unstable, with 33 extensions and hardly ever below 100 tabs. Firefox does have its flaws, such as dropping sound support, massive memory use and using lots of CPU even when idle, but crashiness isn't one of them.
If you experience crashes "several times daily", you'd better check your hardware. Or perhaps you're running some bogus DRM scheme.
The official Unix brand is granted to whoever pays the fee, actual compliance with the specs doesn't matter -- there's a huge leeway in testing so even stuff as insane as AIX can pass.
It's a car ride on "public" transportation. I have zero fucks to give about the driver's integrity. I care about getting from A to B quickly and the driver's driving ability and safety record.
Like, say, the driver lying up front to you how long the trip is and how long will it cost? Or the price suddenly rising by a factor of three the moment you arrive at the destination? Or them always "not having the change"?
Uber fixes all of this. Their execs might be strangling fluffy bunnies in all of their free time for all I care, if they can get us rid of the vermin that current cab mafia are, they're still good.
It's still the US, so people are far more mindboggingly car-crazy than in the rest of the world. And the study quoted an percentage decrease, so absolute values don't matter (as long as they're big enough for a meaningful difference).
Those "renewables" of yours are just a very, very inefficient way of scraping nuclear energy; it just happens that the reactor is 8 light minutes away. And the harvesting causes massive deaths of wildlife (constantly) or even humans (accidents). Compare how many more powers of magnitude of fatalities a single incident at Banqiao had than all deaths due to nuclear power together.
Now think what would happen if even a fraction of money put into the renewables drive went into fusion research...
What's worse is clicking on the titles randomly taking you to http://slashdot.org/ instead of the article, with reloading being apparently the only recourse.
qemu-user is nice when it works, yeah. Too bad, it notoriously interferes with debugging, has broken threading, and badly lags wrt compatibility with current kernels. For example, currently powerpc and powerpcspe guests on amd64 fail at network access (it used to work quite recently). qemu-system has none of such problems.
Also, qemu-user won't let me catch that Green Pickelhaube's kernels are now built with a seappgrarmor patch that disallows unprivileged user programs to add 2 to an even number. And obviously doesn't work with *BSD at all.
As a past long-time SourceForge user, I'm afraid I see no reason to come back and plenty of reasons to not do so.
There is, though, a feature SourceForge used to have that I dearly miss: the Compile Farm. At least two of TUI programs I maintain notoriously fail to build on some obscure platform, usually OS X; getting a report after a release and having to beg such a random user for help is not fun. No other platform requires a large monetary investment (you can run MiddayCockroachBSD and Green Pickelhaube in VirtualBox/KVM, MS used to give Insider Preview for gratis, etc) but figuring out how to install/upgrade all of them is also a significant chunk of time. Especially getting MIPS and the likes to run in QEMU is pretty involved. A typical developer owns an amd64 box and a couple of armhf/arm64 boards, but nothing more exotic.
Unlike 2007 when you shut down the Compile Farm, ways to contain untrusted users do exist nowadays and are mature: containers, kvm, etc. Thus, it'd be a great help if Compile Farm could be brought back.
The API bits can be mostly replaced with local versions by installing Decentraleyes.
For the rest, Smart Referer lets you block tracking that doesn't include explicit tokens. And Request Policy axes crap that you don't need with a default-deny.
These phones come from the same factories, the same workers, the same machines. Not paying the marketing fee is a pure win. In the past, the "non-counterfeit" products could at least claim to have superior quality control, but that has long since been sacrificed on the altar of profits (checking whether the product isn't utter shit vs an extra boat for an executive -- easy choice). Thus, fuck brands.
Unrelated: the Stylish snippet needed since yesterday is: @-moz-document domain("slashdot.org") { article[class*="-sponsored-"] {display:none !important} }
None. ALSA is strictly better for any single source of local audio: PulseAudio is physically unable of working when ALSA doesn't.
PulseAudio provides a clicky-clicky way to connect certain bluetooth headphones, and may provide software mixing for some sound cards that are otherwise limited to one sound at a time, but those are fringe uses. On the other hand, ALSA is way more powerful wrt channel routing: try for example reordering+remixing 5.1 surround: ALSA gives you an user-unfriendly 6x6 matrix which you can reconfigure for any possible way you may think of, while with Pulse all you can do is upmix/downmix stereo to 5.1 and that's it.
Slashdot Mirror
Maybe in 10 or 15 years Firefox will be production ready. So instead of crashing several times daily, it might only crash several times weekly.
Are you sure you're using the same Firefox as me? It crashes less than once a year, and that's on Debian unstable, with 33 extensions and hardly ever below 100 tabs. Firefox does have its flaws, such as dropping sound support, massive memory use and using lots of CPU even when idle, but crashiness isn't one of them.
If you experience crashes "several times daily", you'd better check your hardware. Or perhaps you're running some bogus DRM scheme.
The AI likely would have Chinese walls built into them to prevent collusion and insider trading.
Ha ha ha. Hahahahahaha. Well played, sir.
At least sound in such videos is no longer a concern on Linux.
The official Unix brand is granted to whoever pays the fee, actual compliance with the specs doesn't matter -- there's a huge leeway in testing so even stuff as insane as AIX can pass.
GNOME is a lateral one.
Compared to... what? Even from Metro the direction is slightly slanted down.
It's a car ride on "public" transportation. I have zero fucks to give about the driver's integrity. I care about getting from A to B quickly and the driver's driving ability and safety record.
Like, say, the driver lying up front to you how long the trip is and how long will it cost? Or the price suddenly rising by a factor of three the moment you arrive at the destination? Or them always "not having the change"?
Uber fixes all of this. Their execs might be strangling fluffy bunnies in all of their free time for all I care, if they can get us rid of the vermin that current cab mafia are, they're still good.
There's not been any shortage of cabs in NYC and people weren't reluctant to pay for it.
I take it you haven't walked within ten meters of a cab in the last, say, 40 years, I guess? Cabbies have Microsoft marketroid levels of integrity.
It's either a coincidence, or a bogus finding.
You mean, an unbiased story slipping through the anti-Uber organized shit flinging fest we have in the media recently?
It's still the US, so people are far more mindboggingly car-crazy than in the rest of the world. And the study quoted an percentage decrease, so absolute values don't matter (as long as they're big enough for a meaningful difference).
But what about the GEOS windowing system on the C64. It was the bomb!
I'm comparing GNOME3 only with the default built-in UI, it obviously loses vs GEOS.
Even Metro is slightly better, and that's like comparing whether Temujin or Attila the Hun would be the better baby-sitter for your kid.
Why Gnome 3? The summary says it clearly: "to the superior GNOME". Ie, 2, that is, MATE.
Gnome 3 is maybe superior to, uhm, Commodore 64's user interface with it's LOAD "*",8,1 -- but perhaps even that is unfair to C64.
Those "renewables" of yours are just a very, very inefficient way of scraping nuclear energy; it just happens that the reactor is 8 light minutes away. And the harvesting causes massive deaths of wildlife (constantly) or even humans (accidents). Compare how many more powers of magnitude of fatalities a single incident at Banqiao had than all deaths due to nuclear power together.
Now think what would happen if even a fraction of money put into the renewables drive went into fusion research...
No one is suicidal enough to write critical code in C. What would happen if someone wrote, say, a kernel, in C!?!
What's worse is clicking on the titles randomly taking you to http://slashdot.org/ instead of the article, with reloading being apparently the only recourse.
qemu-user is nice when it works, yeah. Too bad, it notoriously interferes with debugging, has broken threading, and badly lags wrt compatibility with current kernels. For example, currently powerpc and powerpcspe guests on amd64 fail at network access (it used to work quite recently). qemu-system has none of such problems.
Also, qemu-user won't let me catch that Green Pickelhaube's kernels are now built with a seappgrarmor patch that disallows unprivileged user programs to add 2 to an even number. And obviously doesn't work with *BSD at all.
As a past long-time SourceForge user, I'm afraid I see no reason to come back and plenty of reasons to not do so.
There is, though, a feature SourceForge used to have that I dearly miss: the Compile Farm. At least two of TUI programs I maintain notoriously fail to build on some obscure platform, usually OS X; getting a report after a release and having to beg such a random user for help is not fun. No other platform requires a large monetary investment (you can run MiddayCockroachBSD and Green Pickelhaube in VirtualBox/KVM, MS used to give Insider Preview for gratis, etc) but figuring out how to install/upgrade all of them is also a significant chunk of time. Especially getting MIPS and the likes to run in QEMU is pretty involved. A typical developer owns an amd64 box and a couple of armhf/arm64 boards, but nothing more exotic.
Unlike 2007 when you shut down the Compile Farm, ways to contain untrusted users do exist nowadays and are mature: containers, kvm, etc. Thus, it'd be a great help if Compile Farm could be brought back.
Why would someone run a Microsoft web server vs. Nginx on OpenBSD?
Just asking, cuz I honestly can't fathom a situation where this would be desirable??? Maybe I'm missing something?
You're missing the baseball/handegg/etc tickets someone high in your company got.
runs the same "windows apps (non win32 variety)"
You mean, is there any person outside Microsoft's own marketing team who opens the Windows Store on purpose?
The Store isn't even a solution, it's a problem looking for one. And cpl. Hicks has a good suggestion.
The API bits can be mostly replaced with local versions by installing Decentraleyes.
For the rest, Smart Referer lets you block tracking that doesn't include explicit tokens. And Request Policy axes crap that you don't need with a default-deny.
These phones come from the same factories, the same workers, the same machines. Not paying the marketing fee is a pure win. In the past, the "non-counterfeit" products could at least claim to have superior quality control, but that has long since been sacrificed on the altar of profits (checking whether the product isn't utter shit vs an extra boat for an executive -- easy choice). Thus, fuck brands.
Unrelated: the Stylish snippet needed since yesterday is:
@-moz-document domain("slashdot.org") {
article[class*="-sponsored-"] {display:none !important}
}
They have more fans!
That's why it's time to get a fan-less computer that, incidentally, can't run Windows (Windows IoT is neither Windows nor useful).
and runs on modern windows.
I don't think anyone would run something as sensitive as an ATM on Windows.
WTF... someone actually pays these lowlife pieces of harassing scum?
I don't think you can make a power cord long enough to fly out of Europe.
Current technical reason: Firefox 52 no longer supports sound on any sane Linux system.
None. ALSA is strictly better for any single source of local audio: PulseAudio is physically unable of working when ALSA doesn't.
PulseAudio provides a clicky-clicky way to connect certain bluetooth headphones, and may provide software mixing for some sound cards that are otherwise limited to one sound at a time, but those are fringe uses. On the other hand, ALSA is way more powerful wrt channel routing: try for example reordering+remixing 5.1 surround: ALSA gives you an user-unfriendly 6x6 matrix which you can reconfigure for any possible way you may think of, while with Pulse all you can do is upmix/downmix stereo to 5.1 and that's it.