> If a single device brings down your entire data center, you've got design problems and your architect should be fired or retrained.
Please: if your data center has the time, and skill, and is willing to take the service interruptions to make the whole setup properly immune to single points of failure, that's great. But very, very few live business environments have that kind of resource, time, and willingness to enable critical switches with robust failover.
The good "tab protector" cables actually use a hood,, not just a fragile tab, second reversed tab above the connector tab. I've had some problems with even those where the recess for the connector was too deep and too tightly encased, making it impossible to get a hooded cable in place. Those are especially handy because they cost considerably more, and can require a small screwdriver to lever under the hoold and release the connector tab.
I'm sorry, I mentioned the wrong paradox. They have contributed to understanding the Fermi Paradox solution by setting an upper bound to the size and age of the perceptible universe, and a maximum age of the components for complex chemistry that might sustain life.
The paradox they really solved is Ober's Paradox, which involves a tradeoff between the density of glowing objects like stars, and how many are in a volume, and why the more distant stars do not radiate so much light to be seen, even faintly, at Earth that the night sky is uniformly as bright as the Sun. The answer is that, as more distant galaxies recede at greater and greater speeds, their light is red-shifted and less energetic. And they've established an initial start and thus a maximum radius for our universe, so there are no perceptible sources of light from outside that radius.
I'm not trying to create sympathy for drug manufacturers, but a real understanding of the economic risks for them. I'm noting real risks of drug manufacture. Thorough human testing is very expensive and lengthy, and it's possible to make horrific mistakes that would bankrupt a non-profit very easily.
Much of that delay is the reverse DNS done by the remote SSH daemon, especially when the reverse DNS is unavailable. Turn that off, especially for wandering sftp clients or git access, and you'll profoundly improve initial connection time.
Adding encryption means you can't export them to "non-approved" countries, and raises a great number of hoops to be able to export the product at all..
> But udev doesn't depend on systemd, so your slippery slope is not very slippery.
It only has to be a little slippery, it's the steepness that lets people fall in. See the kdbus and similar components for the ongoing integration of udev with systemd.
An appeal to ridicule is a logical fallacy: if it were an entertaining appeal to ridicule, it might be amusing, and I wouldn't expect pure logic on Slashdot. But please note that it doesn't address or even acknowledge a single one of the issues I mentioned. Many of those issues are architectural and core to systemd development.
And no, I don't "hate systemd with a white hot passion". It does a few things reasonably well, and there are some real benefits to getting faster boot times and kernel logging. The network integration is problematic, but shows some promise.
But systemd really needs to _stop_ trying to replace core system functions with yet another add-on module. And yes, it's trying to take on software packaging, as well.
Examples of systemd breaking the kernel include the "debug" logging option, and the inevitable failures of such a complex weave of components killing PID 1.
Unfortunately, "running syslogd in parallel" doesn't work well as new daemons or services are compiled for one or the other. And I'm afraid the code to integrate with systemd logging is a tar-baby: it becomes very difficult, very quickly, to maintain separate logging, but the logging is not portable to UNIX based operating systems. And that change is breaking portability for new projects even as I write. If you're willing, take a good look at the latest httpd source code to see what's happening to logging there.
And yes, systemd is trying to replace "su". See the comments by systemd's core author, Leonart Pottering, at:
It's particularly amusing in those comments that Lennart Potteroing thinks that Linux is UNIX. UNIX is trademarked, licensed, and applies only to systems that follow various POSIX standards, and there's a fascinating history of lawsuits about this involving the SCO Group, which tried to claim that Linux was a UNIX descendant. Old material on this is at:
I can understand why hearing these issues voiced again could be tiresome, and not all concerned developers are well informed. But rejecting all concerns as being "information from trolls" ignores the very real and often unnecessary problems systemd is creating.
> Cosmology serves as an object lesson in what happens when you let academics get entrenched
They get to see the handwriting of God, writ large in the universe around us? They explain the very origins of matter, and solve Fermi's Paradox? They help provide a sense of scale to our image of ourselves in the universe? They confirm the interactions of gravity and light, fundamental forces in physics? They explain the concentrations of different types of matter in the universe? They explain and reveal the nature of background radiation that affect electronics, and weather?
It's amazing how looking at the largest scales of the universe leads back to information about the smallest scales of the universe, and both _do_ affect every day life. We just tend not to notice that from day to day.
> systemd's policy to drop them makes servers less secure
They have _replaced_, systlog with kernel resident binary logging utilities. This has advantages: you can generate monitoring from the kernel, itself, at boot time, before syslogd is running.
The big concern I've encountered is that systemd replaced stable, legible, parseable, well understood log output with t published but quite unique log format which no other tools in the world knew how to read. It wasn't necessary for enhancing init script management: if i had an engineer working with me who kept expanding projects this way, I'd let them know there's a problem and help them find another job.
systemd has already broken the kernel repeatedly, a decade of log collection and analysis tools that I personally use, decades of cross-platform tools, decades of network configuration tools, and now it's trying to replace "su" and "/etc" to create a "stateless Linux". I'm afraid that it's a system management tar baby, adhering to every system component that it touches and leaving them snarled in sticky debris that is difficult to wash off.
I was actually wondering if they were reported by the vice principal trying to monitor children in their homes with "educational spyware" installed on the laptops, much like that previously reported on Slashdot.
It does not work well with the plates covered with dried food, and it requires regular clearing of the debris traps on the dishwasher. It also tends to clog the holes in the spray mechanisms, which will need regular clearing and are awkward to access.
Yes, I've worked in as dormitory staff, in shared apartments, and workplaces where the students and my work colleagues followed your approach. Their mother didn't work there, but I did, which is why the dishwashers still worked.
If interviewing you to join my my workforce, or evaluating you as a potential friend or spouse for anyone I know, I would award you many, many points for such a sensible belief. It's too rare!
> The PD has always seemed to me to be a "I don't want to give a fuck" get out of jail free card
Oh, my. The Prime Directive has been violated, _repeatedly_ in every version of Star Trek. It's much like Constitutional law and like the amendments. It had to be balanced against the other directives, on the spot, by captains and other officers often outside of direct contact with StarFleet and with the Federation. And the interaction of it with justice, with morals, and with StarFleet made for fascinating plots.
Please note: these were fictional _plots_, in a fictional universe. They did not have to be completely consistent. One can hurt onself quite badly insisting on complete consistency in such a universe..
Don't forget the insurance. One mistake, even a purely accidental one or a malicious one from a third party, and the lawsuits can bankrupt such a group in moments. I'm old enough to remember the thalidomide birth defects, and the malicious poisoning of Tylenol. The manufacturers of both drugs were _horrified_ at these tragedies, and did their best to protect the public after the problems were discovered.
> So far the big reductions, such as the switch from coal to gas, LED lights, variable speed DC motors, more efficient HVAC, better insulation, have all been driven by the market, not "government regulation".
Look again, and look more carefully, please. The restrictions on building new coal fired plants with less pollution, coupled with many other factors, have raised electricity prices: that is one of the market forces" involved. Government support of the switch and numerous projects at every level of government have encouraged every one of those changes, including tax breaks on more energy efficient homes, the forced publication of fuel efficiency on home appliances to better _inform_ consumers of the efficiencies, and regulation on government funded construction.
I must admit, a 20 kg carnivorous snail is pretty cool. But they apparently breed very slowly, a fact for which many other species may be very grateful.
Eager masturbation to achieve prgas? Anal sex, where lubrication is normally applied as part of the act but may prove insufficient partway through the act?
> Humans have been promiscuous since before there was anything vaguely promiscuous.
Have you looked into bonobos, or dolphins? Bonobos are quite close primate relatives, and even more promiscuous in general than humans. And dolphins are _quite_ promiscuous. We're evolutionary latecomers, only a few million years old at most. Almost every physical trait and behavior we have was tried by many other species long before we evolved.
Edward Snowden is also a computer professional with direct experience of of federal failure to obey laws in the name of political benefit, and direct experience of how "secure" documents can be duplicated and leaked.
Slashdot Mirror
> If a single device brings down your entire data center, you've got design problems and your architect should be fired or retrained.
Please: if your data center has the time, and skill, and is willing to take the service interruptions to make the whole setup properly immune to single points of failure, that's great. But very, very few live business environments have that kind of resource, time, and willingness to enable critical switches with robust failover.
The good "tab protector" cables actually use a hood,, not just a fragile tab, second reversed tab above the connector tab. I've had some problems with even those where the recess for the connector was too deep and too tightly encased, making it impossible to get a hooded cable in place. Those are especially handy because they cost considerably more, and can require a small screwdriver to lever under the hoold and release the connector tab.
I'm sorry, I mentioned the wrong paradox. They have contributed to understanding the Fermi Paradox solution by setting an upper bound to the size and age of the perceptible universe, and a maximum age of the components for complex chemistry that might sustain life.
The paradox they really solved is Ober's Paradox, which involves a tradeoff between the density of glowing objects like stars, and how many are in a volume, and why the more distant stars do not radiate so much light to be seen, even faintly, at Earth that the night sky is uniformly as bright as the Sun. The answer is that, as more distant galaxies recede at greater and greater speeds, their light is red-shifted and less energetic. And they've established an initial start and thus a maximum radius for our universe, so there are no perceptible sources of light from outside that radius.
I'm not trying to create sympathy for drug manufacturers, but a real understanding of the economic risks for them. I'm noting real risks of drug manufacture. Thorough human testing is very expensive and lengthy, and it's possible to make horrific mistakes that would bankrupt a non-profit very easily.
Apparently so. See http://0pointer.net/blog/revis... .
Much of that delay is the reverse DNS done by the remote SSH daemon, especially when the reverse DNS is unavailable. Turn that off, especially for wandering sftp clients or git access, and you'll profoundly improve initial connection time.
Adding encryption means you can't export them to "non-approved" countries, and raises a great number of hoops to be able to export the product at all..
https://en.wikipedia.org/wiki/...
Also, encryption algorithms take more space in the very limited space on firmware and small controller chipsets.
> But udev doesn't depend on systemd, so your slippery slope is not very slippery.
It only has to be a little slippery, it's the steepness that lets people fall in. See the kdbus and similar components for the ongoing integration of udev with systemd.
An appeal to ridicule is a logical fallacy: if it were an entertaining appeal to ridicule, it might be amusing, and I wouldn't expect pure logic on Slashdot. But please note that it doesn't address or even acknowledge a single one of the issues I mentioned. Many of those issues are architectural and core to systemd development.
And no, I don't "hate systemd with a white hot passion". It does a few things reasonably well, and there are some real benefits to getting faster boot times and kernel logging. The network integration is problematic, but shows some promise.
But systemd really needs to _stop_ trying to replace core system functions with yet another add-on module. And yes, it's trying to take on software packaging, as well.
Examples of systemd breaking the kernel include the "debug" logging option, and the inevitable failures of such a complex weave of components killing PID 1.
https://bugs.freedesktop.org/s...
http://ewontfix.com/14/
Unfortunately, "running syslogd in parallel" doesn't work well as new daemons or services are compiled for one or the other. And I'm afraid the code to integrate with systemd logging is a tar-baby: it becomes very difficult, very quickly, to maintain separate logging, but the logging is not portable to UNIX based operating systems. And that change is breaking portability for new projects even as I write. If you're willing, take a good look at the latest httpd source code to see what's happening to logging there.
And yes, systemd is trying to replace "su". See the comments by systemd's core author, Leonart Pottering, at:
https://github.com/systemd/sys...
It's particularly amusing in those comments that Lennart Potteroing thinks that Linux is UNIX. UNIX is trademarked, licensed, and applies only to systems that follow various POSIX standards, and there's a fascinating history of lawsuits about this involving the SCO Group, which tried to claim that Linux was a UNIX descendant. Old material on this is at:
https://www.groklaw.net/
I can understand why hearing these issues voiced again could be tiresome, and not all concerned developers are well informed. But rejecting all concerns as being "information from trolls" ignores the very real and often unnecessary problems systemd is creating.
> Cosmology serves as an object lesson in what happens when you let academics get entrenched
They get to see the handwriting of God, writ large in the universe around us? They explain the very origins of matter, and solve Fermi's Paradox? They help provide a sense of scale to our image of ourselves in the universe? They confirm the interactions of gravity and light, fundamental forces in physics? They explain the concentrations of different types of matter in the universe? They explain and reveal the nature of background radiation that affect electronics, and weather?
It's amazing how looking at the largest scales of the universe leads back to information about the smallest scales of the universe, and both _do_ affect every day life. We just tend not to notice that from day to day.
> systemd's policy to drop them makes servers less secure
They have _replaced_, systlog with kernel resident binary logging utilities. This has advantages: you can generate monitoring from the kernel, itself, at boot time, before syslogd is running.
The big concern I've encountered is that systemd replaced stable, legible, parseable, well understood log output with t published but quite unique log format which no other tools in the world knew how to read. It wasn't necessary for enhancing init script management: if i had an engineer working with me who kept expanding projects this way, I'd let them know there's a problem and help them find another job.
systemd has already broken the kernel repeatedly, a decade of log collection and analysis tools that I personally use, decades of cross-platform tools, decades of network configuration tools, and now it's trying to replace "su" and "/etc" to create a "stateless Linux". I'm afraid that it's a system management tar baby, adhering to every system component that it touches and leaving them snarled in sticky debris that is difficult to wash off.
I'm afraid that the "insult" would like in _failing_ to laugh with him.
I was actually wondering if they were reported by the vice principal trying to monitor children in their homes with "educational spyware" installed on the laptops, much like that previously reported on Slashdot.
http://www.pcmag.com/article2/...
It does not work well with the plates covered with dried food, and it requires regular clearing of the debris traps on the dishwasher. It also tends to clog the holes in the spray mechanisms, which will need regular clearing and are awkward to access.
Yes, I've worked in as dormitory staff, in shared apartments, and workplaces where the students and my work colleagues followed your approach. Their mother didn't work there, but I did, which is why the dishwashers still worked.
If interviewing you to join my my workforce, or evaluating you as a potential friend or spouse for anyone I know, I would award you many, many points for such a sensible belief. It's too rare!
> The PD has always seemed to me to be a "I don't want to give a fuck" get out of jail free card
Oh, my. The Prime Directive has been violated, _repeatedly_ in every version of Star Trek. It's much like Constitutional law and like the amendments. It had to be balanced against the other directives, on the spot, by captains and other officers often outside of direct contact with StarFleet and with the Federation. And the interaction of it with justice, with morals, and with StarFleet made for fascinating plots.
Please note: these were fictional _plots_, in a fictional universe. They did not have to be completely consistent. One can hurt onself quite badly insisting on complete consistency in such a universe..
Don't forget the insurance. One mistake, even a purely accidental one or a malicious one from a third party, and the lawsuits can bankrupt such a group in moments. I'm old enough to remember the thalidomide birth defects, and the malicious poisoning of Tylenol. The manufacturers of both drugs were _horrified_ at these tragedies, and did their best to protect the public after the problems were discovered.
> So far the big reductions, such as the switch from coal to gas, LED lights, variable speed DC motors, more efficient HVAC, better insulation, have all been driven by the market, not "government regulation".
Look again, and look more carefully, please. The restrictions on building new coal fired plants with less pollution, coupled with many other factors, have raised electricity prices: that is one of the market forces" involved. Government support of the switch and numerous projects at every level of government have encouraged every one of those changes, including tax breaks on more energy efficient homes, the forced publication of fuel efficiency on home appliances to better _inform_ consumers of the efficiencies, and regulation on government funded construction.
I must admit, a 20 kg carnivorous snail is pretty cool. But they apparently breed very slowly, a fact for which many other species may be very grateful.
s/prgas/orgasm/g
Eager masturbation to achieve prgas? Anal sex, where lubrication is normally applied as part of the act but may prove insufficient partway through the act?
> Humans have been promiscuous since before there was anything vaguely promiscuous.
Have you looked into bonobos, or dolphins? Bonobos are quite close primate relatives, and even more promiscuous in general than humans. And dolphins are _quite_ promiscuous. We're evolutionary latecomers, only a few million years old at most. Almost every physical trait and behavior we have was tried by many other species long before we evolved.
Edward Snowden is also a computer professional with direct experience of of federal failure to obey laws in the name of political benefit, and direct experience of how "secure" documents can be duplicated and leaked.