Good point. In some of these countries a solution that cuts death rates from hundreds per year to dozens per year shouldn't be overlooked simply because it isn't the complete solution.
...and I still wouldn't go anywhere near that ground. Too many pink mist incidents in places that have been subject to GROUND searches.
It sounds like the target market is places where the problem is along the lines of "Hey, impoverished local government, you have a zillion acres of variously vegetated former combat zone and no idea where to even start sending in the deminers. Would you prefer to guess blindly or have a (relatively) cheap map with 'probably some bombing over here' marked where applicable?"
Good point. In some of these countries a solution that cuts death rates from hundreds per year to dozens per year shouldn't be overlooked simply because it isn't the complete solution.
Given how much of the world's UXO and especially ill-documented mines fall in places that are poor, somewhat weakly governed, and not necessarily equipped with even decent topographical maps for their entire area, there is probably a lot of room for solutions that can beat 'peasants finding them one limb at a time' as long as they don't cost too much.
It's like healthcare: Sure, "Go to a first world teaching hospital with a superb reputation" isn't a bad idea; but for the almost-everyone who would find that advice irrelevant, there's a lot to be said for trying to take on the low hanging fruit, given that the alternative is basically nothing.
...and I still wouldn't go anywhere near that ground. Too many pink mist incidents in places that have been subject to GROUND searches.
I read a really good article on this ages ago that went into some of the challenges. The military is very efficient at clearing minefields for military use. The thing is, that just means clearing a narrow path of the really nasty stuff so that things like armored vehicles can drive through with an acceptable level of losses (by combat standards). If going via an unmined route will cause you 10k casualties, but clearing an alternate route through a minefield will only get 100 soldiers killed, then the minefield is preferable. The troops would be travelling defined routes, would probably have some level of protection (even on foot), and are just going to be there for a short time. Getting 95% of the mines near the path might be a completely acceptable level of success.
Civilian demining is an entirely different ballpark. It isn't acceptable that only 100 kids die playing in a field, you can't just put up a sign that says "stay on this well-marked path" and ignore kids who deviate to go play in the grass, and people want to go back to normal life. 95% is no longer good enough.
In order to be effective at all mines cannot be trivially avoidable, which means they're generally not reliably detectable.
The real problem is that people are using web browsers to read their e-mail instead of a proper e-mail client that already supports the existing standards of pgp and s/mime This yahoo plugin is actually based on google's code for an end to end plugin. It implents pgp.
The problem is that the browser+javascript is the most ubiquitous platform around. It is also FAR more convenient to use.
I'd love to see a decent FOSS webmail application that supports encryption. The only options that exist right now are pretty weak compared to something like GMail.
Everybody except the headless victims in the mass graves.
True, and it is wrong to buy into the whole moral equivalence thing. Granted, the US was out of control in Iraq with torture and all that, but I imagine the US pillaging is fairly tame compared to ISIS.
The only thing that will get us out of this mode will be an attack on our soil that will force us to savage our opponent so that no other rival thinks that our territory can be struck without a lethal response
They can whip up a false flag any old time, or go all FBI and just encourage and equip some terrorists, and then they get their excuse.
Maybe, but why would the US want to conqueror Vietnam? The only reason we cared about it before was to try to slow Soviet expansion. That, and the US was still stuck in island-hopping mode a bit from WWII.
After Vietnam the US learned its lesson and just did what the USSR did in Vietnam - give tons of guns to the locals and turn it into a huge war of attrition. That is why everybody who used to be in the USSR has fond memories of Afghanistan.
I don't think the US really has any strategic interests there any longer.
In any case, the US really does seem to be fairly war-fatigued these days. Sure, Bush II would probably have been reluctant to go into Ukraine, but you can bet that there would be a huge deployment against ISIS back in that era. Now the US is more eager to let the Iraqis take care of themselves, and everybody is probably somewhat better off for it.
Your statement makes sense right up until someone flies one of these into the engine of a commercial jet and causes a crash that kills hundreds and does millions in property damage.
So, doing something like that in a reckless manner is likely already illegal under a bazillion other laws. Also, a drone isn't going to do much more to an airliner than a bird. As long as you're not buzzing around in the vicinity of airports such an event is extremely unlikely.
The FAA would do better to promote ADS-B/UAT for drones and such, but they can't even get that right for piloted small aircraft. For starters, the government should just bless a ADS-B/UAT transmitter design and let anybody build them without further regulation. It shouldn't require more than what you'd find in any cell phone on the planet (including $30 feature phones from China) - it is just a GPS and a modem/radio.
And if you're talking about somebody intentionally piloting a drone into an aircraft, then no regulation is going to do anything to stop it.
Can someone point out to me which part of the 1st Amendment it is in?
Heck, I'll settle for the regulation that prohibits commercial operation of drones.
Links to US Laws and CFRs only, please. I know the FAA has a bunch of stuff on their website which the only court to date has ruled is not legally binding. That of course won't stop the FAA from suing you.
Agree. Basically at ranges where speed of light is a factor the battle is the attacker expending a lethal amount of energy across as much space as possible, with the defender trying to be able to either withstand shots or not be inside that area of space. As range closes, the advantage moves towards the attacker.
And of course hollywood never gets this. The only place I've seen it done right is a Charles Stross novel (where speed of light delay in long-range tactics is accounted for both in terms of detection and fire).
As Stross also exhibits, a guided weapon is of course much slower, but doesn't suffer the same problems since as it approaches the target the position uncertainty drops.
There isn't anything you can do with linux+bash that you can't do on an original IBM PC with the ROM Basic interpreter.
So, I'm replying to someone who has no clue about why 'the original PC' hardware can't support the memory models and process/thread handling necessary for today's modern O/Ss (and I'll include Windows in there as well). Nevertheless:
The original IBM PC can support every mode of operation present in the computer you're using right now. It might not support it in hardware, but it DEFINITELY can emulate it. If you can boot the linux kernel in a javascript emulator, you can certainly do it from BASIC.
Granted, you might have to fit in some kind of external storage to make up for the RAM capacity, but I'm sure that if you add an external hard drive over RS232 you could emulate all the RAM in a modern PC (at a bazillionth the speed). The only real constraint in a "universal" turing machine is the size of your tape.
I didn't compose my reply not understanding what protected mode is. One of my frustrations with talking about systemd with people is that many seem to assume that if you prefer systemd that it must be because you don't understand how computers actually work or something.
when I can just put 3 lines in a text file
Because that only covers the simple cases. The conditions upon which some action must be triggered can be much more complex than 3 lines will cover. Conditional branches, loops and other structures are needed. And now you are back to some sort of scripting language.
Based on my poking around in many init script setups, many of the daemon start/stop scripts utilize rather complex structures, testing various system parameters and passing variable arguments to the executable being started. All this stuff has been worked out by the people who build various distros plus the admins that have to customize systems to suit their applications. systemd might look neato to people running gaming boxes, tablets and those who think an IBM 5150 is the penultimate computing platform. But not to the people that support more complex systems.
So, on the system I'm typing this on, most of the scripting just accomplishes stuff that is already built into systemd. My apache2 stop function doesn't need to poll to check that it cleanly shut down - systemd will automatically kill it after the timeout passes for graceful shutdown, etc.
I'm sure that in some application situations you might need to do a bunch of scripting, but that is completely possible with systemd, while still retaining systemd's management of core functionality. You just script the stuff that actually NEEDS to be scripted.
I think the systemd devs have done a pretty good job of building on the richness of the whole dependency/event-based paradigm.
... for people who can't be buggered to read a man page and figure out how to do all of this with existing tools.
There isn't anything you can do with linux+bash that you can't do on an original IBM PC with the ROM Basic interpreter.
That doesn't mean that they're both equally-useful tools for the job.
Why would I want to figure out how to hack together a bunch of stuff using fnotify and a bunch of scripts to launch an executable anytime a file changes when I can just put 3 lines in a text file and symlink it in a directory and systemd does it for me?
Gentoo, unless you choose a multilib profile, will be 64-bit clean. In addition to the speed benefit, this is also a security benefit, as most rootkits and exploits are 32-bit and just won't run.
The kernel will still run 64-bit code unless you disable that feature. That means that a static-linked 32-bit binary or a package (like steam) that bundles its own 32-bit libs may still work, even on no-multilib Gentoo.
I don't know for sure that this is what is happening in your case, but I wouldn't rule it out without checking.
Please explain how installing 32-bit libraries magically makes your system slower when they are not in use.
It doesn't. MAYBE the game MIGHT run a tiny bit slower than it might otherwise, but the rest of his system would not. Of course, the 32bit game runs a lot better than the hypothetical 64bit game that doesn't actually exist.
You forget that the tracking system has a 1/10th of a second delay as well, or if it is radar like 2/10th of a second.
Agree - the round trip time matters from a tracking perspective.
Obviously the target can not start maneuvering as soon as it realizes the it is hit, as that is to late. However your idea of 1000nds of g's is wrong.
A laser at that hight has perhaps a diameter of a meter. So sidestepping a meter is enough.
Even a human can step 1 yard/1 meter to the side in a 1/10th of a second.
First, it isn't enough to move a meter in 1/10th of a second. It has to accelerate such that ends up a meter away from where it would otherwise be.
That is an acceleration of 200 m/s^2, assuming we stick with a meter and 0.1 seconds. That is quite a bit of acceleration. It isn't actually enough to evade the fire though even with your numbers.
What happens if the laser misses? The computer fires again. It would fire in a random pattern around you, so it would hit if your RNG maneuvered you into the same place the turret's RNG picked to shoot at. Maybe it would take 5 shots to hit you, maybe 15. You're constantly accelerating at 20G the whole time in random directions to try to avoid it.
To really evade fire you'd need considerably more acceleration so that you're not just a meter away from your last predicted position when the laser arrives.
I actually outlined this on slashdot before, though it was probably years ago. Basically armor and acceleration are both forms of defense in space combat at long range, since both increase the total amount of energy an attacker has to expend into space in order to destroy you. If I can be anywhere in a circle a mile across by the time the laser arrives, you have to put enough energy into every point in that circle to destroy me no matter where I'm at. As range closes, the error in my position drops until I'm dead, or vice-versa.
Keep in mind that uniform motion is no defense at all, since the laser would be leading you. You need continuous random acceleration.
Okay, I admit it, I don't get the punchline. I even added one to the cart expecting that as the price per stick (even then, unbelievably low, but maybe for tested pulls), but no, $79 for the whole bulk pack, new???
You can't even get no-name sticks of non-ECC labelled in Chinese for that. That can't count as a real price, can it?
Sure. You just have to pay with bitcoin in advance. You know, just in case you're not a reliable purchaser.
I was looking at motherboards and even finding ones that support ECC is difficult, unless this is one of those situations where any motherboard works as long as the CPU supports it.
As far as I understand it, most AMD processors do support ECC, and Intel only supports it in their upper-end products (artificial restriction to segment the market - i7/Xeon/etc). What I don't understand is where the motherboard fits in.
Heck, Newegg doesn't even track that as an option on their product selector.
One of the things I learned early on as a tweenager within the FOSS was that if you have a good idea and can communicate well & civilly, nobody actually seems to care terribly much if you possibly are a dog that somehow learned how to type.
If somebody can communicate well and civilly, then they aren't a dog that somehow learned how to type. I think the whole point of the article is that some people don't communicate well and civilly.
The thing is, being an asshole shouldn't be relevant. Is the person contributing good code? That's all that should matter.
That is hardly true. If nobody wants to work for you because one of your top contributors drives everybody nuts, then you have to decide whether their sole contributions really do outweigh everything you're losing.
Sometimes the asshole really is a special snowflake and you just move them to their own side of the office and insulate them from everybody else and pander to them. Maybe you pay everybody who has to work with them an extra 20% to get them to put up with it. However, more often than not they're not nearly as special as they think they are, and you're better off without them.
This actually has something to do with why so many people hate Systemd. It turns out that Systemd is professional-quality work done by competent salaried engineers. Our problem with it is that we're used to beautiful code made by geniuses. Going all of the way back to DMR.
I can't say I agree. The reason I run systemd (on a distro where it isn't even the default) is that I find it a rather elegant and powerful solution.
I think that the problem is that people conflate good solutions with simple solutions. I'll certainly agree that the old way was simpler. That has a certain beauty to it. The problem is that it doesn't actually solve all the problems well.
This is by no means a software issue exclusively. Take your favorite political soundbite, and most likely it involves taking some really complicated issue like war in the middle east or national healthcare or socialism or the finance industry or global warming, and then proposes some solution that could fit on a single powerpoint slide. People like solutions that are easier to understand, but that doesn't automatically make them the best solutions.
Wanting a positive atmosphere within the project is one thing. Worrying about angry blog posts is insecure and petty at best.
I think the concern was more with angry blog posts and other passive-aggressive behavior on community-provided sites.
If you want to rant on your random wordpress page, that is your right. If you want it to be on some blog aggregator provided by the people you're complaining about, that is another matter.
Agree completely. One of the things that really impresses me with systemd is just how event-driven it actually is, and how it separates events from services.
I can have a service that performs some task. I can trigger it to run when another service runs. I can trigger it to run when a target/"runlevel" is activated (the traditional rc.d approach). I can trigger it to run at certain times. I can trigger it to run when a particular udev rule is satisfied. I can trigger it to run when when a file is modified. I can probably trigger it to run when a filesystem is mounted. I can trigger it to run when somebody connects to a socket. I can even trigger it to run conditionally if one of those things happens and the device is in a particular power state, etc.
I think the systemd devs have done a pretty good job of building on the richness of the whole dependency/event-based paradigm.
I've worked in this industry for 32 years. I have never seen someone get a raise without moving to a new job. Not once. It just doesn't happen.
What industry is that? The software industry? I just told you that I don't work in the software industry, and I don't work for a start up. I work in IT. You do realize that virtually every company on Earth hires people that work in IT in some form. I'd probably prefer to work in the software industry but stuff like this is one of the reasons I don't.
I would. The current construction doesn't stop racism/whateverism you choose and moreover denies us the knowledge of who is who.
Your logic assumes that people can never change, and that racists should basically just live off of tax dollars or become criminals to survive.
I'm not really a big fan of solutions that involve blacklisting people for life anytime they do anything that a large number of people find offensive. I don't even support that if somebody commits homicide or any other heinous crime.
Slashdot Mirror
Ugh, butchered the quoting. My reply was:
Good point. In some of these countries a solution that cuts death rates from hundreds per year to dozens per year shouldn't be overlooked simply because it isn't the complete solution.
...and I still wouldn't go anywhere near that ground. Too many pink mist incidents in places that have been subject to GROUND searches.
It sounds like the target market is places where the problem is along the lines of "Hey, impoverished local government, you have a zillion acres of variously vegetated former combat zone and no idea where to even start sending in the deminers. Would you prefer to guess blindly or have a (relatively) cheap map with 'probably some bombing over here' marked where applicable?"
Good point. In some of these countries a solution that cuts death rates from hundreds per year to dozens per year shouldn't be overlooked simply because it isn't the complete solution.
Given how much of the world's UXO and especially ill-documented mines fall in places that are poor, somewhat weakly governed, and not necessarily equipped with even decent topographical maps for their entire area, there is probably a lot of room for solutions that can beat 'peasants finding them one limb at a time' as long as they don't cost too much.
It's like healthcare: Sure, "Go to a first world teaching hospital with a superb reputation" isn't a bad idea; but for the almost-everyone who would find that advice irrelevant, there's a lot to be said for trying to take on the low hanging fruit, given that the alternative is basically nothing.
...and I still wouldn't go anywhere near that ground. Too many pink mist incidents in places that have been subject to GROUND searches.
I read a really good article on this ages ago that went into some of the challenges. The military is very efficient at clearing minefields for military use. The thing is, that just means clearing a narrow path of the really nasty stuff so that things like armored vehicles can drive through with an acceptable level of losses (by combat standards). If going via an unmined route will cause you 10k casualties, but clearing an alternate route through a minefield will only get 100 soldiers killed, then the minefield is preferable. The troops would be travelling defined routes, would probably have some level of protection (even on foot), and are just going to be there for a short time. Getting 95% of the mines near the path might be a completely acceptable level of success.
Civilian demining is an entirely different ballpark. It isn't acceptable that only 100 kids die playing in a field, you can't just put up a sign that says "stay on this well-marked path" and ignore kids who deviate to go play in the grass, and people want to go back to normal life. 95% is no longer good enough.
In order to be effective at all mines cannot be trivially avoidable, which means they're generally not reliably detectable.
The real problem is that people are using web browsers to read their e-mail instead of a proper e-mail client that already supports the existing standards of pgp and s/mime This yahoo plugin is actually based on google's code for an end to end plugin. It implents pgp.
The problem is that the browser+javascript is the most ubiquitous platform around. It is also FAR more convenient to use.
I'd love to see a decent FOSS webmail application that supports encryption. The only options that exist right now are pretty weak compared to something like GMail.
Everybody except the headless victims in the mass graves.
True, and it is wrong to buy into the whole moral equivalence thing. Granted, the US was out of control in Iraq with torture and all that, but I imagine the US pillaging is fairly tame compared to ISIS.
The only thing that will get us out of this mode will be an attack on our soil that will force us to savage our opponent so that no other rival thinks that our territory can be struck without a lethal response
They can whip up a false flag any old time, or go all FBI and just encourage and equip some terrorists, and then they get their excuse.
Maybe, but why would the US want to conqueror Vietnam? The only reason we cared about it before was to try to slow Soviet expansion. That, and the US was still stuck in island-hopping mode a bit from WWII.
After Vietnam the US learned its lesson and just did what the USSR did in Vietnam - give tons of guns to the locals and turn it into a huge war of attrition. That is why everybody who used to be in the USSR has fond memories of Afghanistan.
I don't think the US really has any strategic interests there any longer.
In any case, the US really does seem to be fairly war-fatigued these days. Sure, Bush II would probably have been reluctant to go into Ukraine, but you can bet that there would be a huge deployment against ISIS back in that era. Now the US is more eager to let the Iraqis take care of themselves, and everybody is probably somewhat better off for it.
Your statement makes sense right up until someone flies one of these into the engine of a commercial jet and causes a crash that kills hundreds and does millions in property damage.
So, doing something like that in a reckless manner is likely already illegal under a bazillion other laws. Also, a drone isn't going to do much more to an airliner than a bird. As long as you're not buzzing around in the vicinity of airports such an event is extremely unlikely.
The FAA would do better to promote ADS-B/UAT for drones and such, but they can't even get that right for piloted small aircraft. For starters, the government should just bless a ADS-B/UAT transmitter design and let anybody build them without further regulation. It shouldn't require more than what you'd find in any cell phone on the planet (including $30 feature phones from China) - it is just a GPS and a modem/radio.
And if you're talking about somebody intentionally piloting a drone into an aircraft, then no regulation is going to do anything to stop it.
Can someone point out to me which part of the 1st Amendment it is in?
Heck, I'll settle for the regulation that prohibits commercial operation of drones.
Links to US Laws and CFRs only, please. I know the FAA has a bunch of stuff on their website which the only court to date has ruled is not legally binding. That of course won't stop the FAA from suing you.
It was Singularity Sky.
Agree. Basically at ranges where speed of light is a factor the battle is the attacker expending a lethal amount of energy across as much space as possible, with the defender trying to be able to either withstand shots or not be inside that area of space. As range closes, the advantage moves towards the attacker.
And of course hollywood never gets this. The only place I've seen it done right is a Charles Stross novel (where speed of light delay in long-range tactics is accounted for both in terms of detection and fire).
As Stross also exhibits, a guided weapon is of course much slower, but doesn't suffer the same problems since as it approaches the target the position uncertainty drops.
There isn't anything you can do with linux+bash that you can't do on an original IBM PC with the ROM Basic interpreter.
So, I'm replying to someone who has no clue about why 'the original PC' hardware can't support the memory models and process/thread handling necessary for today's modern O/Ss (and I'll include Windows in there as well). Nevertheless:
The original IBM PC can support every mode of operation present in the computer you're using right now. It might not support it in hardware, but it DEFINITELY can emulate it. If you can boot the linux kernel in a javascript emulator, you can certainly do it from BASIC.
Granted, you might have to fit in some kind of external storage to make up for the RAM capacity, but I'm sure that if you add an external hard drive over RS232 you could emulate all the RAM in a modern PC (at a bazillionth the speed). The only real constraint in a "universal" turing machine is the size of your tape.
I didn't compose my reply not understanding what protected mode is. One of my frustrations with talking about systemd with people is that many seem to assume that if you prefer systemd that it must be because you don't understand how computers actually work or something.
when I can just put 3 lines in a text file
Because that only covers the simple cases. The conditions upon which some action must be triggered can be much more complex than 3 lines will cover. Conditional branches, loops and other structures are needed. And now you are back to some sort of scripting language.
Based on my poking around in many init script setups, many of the daemon start/stop scripts utilize rather complex structures, testing various system parameters and passing variable arguments to the executable being started. All this stuff has been worked out by the people who build various distros plus the admins that have to customize systems to suit their applications. systemd might look neato to people running gaming boxes, tablets and those who think an IBM 5150 is the penultimate computing platform. But not to the people that support more complex systems.
So, on the system I'm typing this on, most of the scripting just accomplishes stuff that is already built into systemd. My apache2 stop function doesn't need to poll to check that it cleanly shut down - systemd will automatically kill it after the timeout passes for graceful shutdown, etc.
I'm sure that in some application situations you might need to do a bunch of scripting, but that is completely possible with systemd, while still retaining systemd's management of core functionality. You just script the stuff that actually NEEDS to be scripted.
I think the systemd devs have done a pretty good job of building on the richness of the whole dependency/event-based paradigm.
There isn't anything you can do with linux+bash that you can't do on an original IBM PC with the ROM Basic interpreter.
That doesn't mean that they're both equally-useful tools for the job.
Why would I want to figure out how to hack together a bunch of stuff using fnotify and a bunch of scripts to launch an executable anytime a file changes when I can just put 3 lines in a text file and symlink it in a directory and systemd does it for me?
Gentoo, unless you choose a multilib profile, will be 64-bit clean.
In addition to the speed benefit, this is also a security benefit, as most rootkits and exploits are 32-bit and just won't run.
The kernel will still run 64-bit code unless you disable that feature. That means that a static-linked 32-bit binary or a package (like steam) that bundles its own 32-bit libs may still work, even on no-multilib Gentoo.
I don't know for sure that this is what is happening in your case, but I wouldn't rule it out without checking.
I have Steam installed on a pure 64bit Linux distro and I've had no issues so far.
Hmm, wonder if they are static linking or you're getting 32-bit libs that you don't realize you're getting.
You're sure that 32-bit support is actually disabled in your kernel? I'm not aware of any distros that do this out of the box.
Please explain how installing 32-bit libraries magically makes your system slower when they are not in use.
It doesn't. MAYBE the game MIGHT run a tiny bit slower than it might otherwise, but the rest of his system would not. Of course, the 32bit game runs a lot better than the hypothetical 64bit game that doesn't actually exist.
You forget that the tracking system has a 1/10th of a second delay as well, or if it is radar like 2/10th of a second.
Agree - the round trip time matters from a tracking perspective.
Obviously the target can not start maneuvering as soon as it realizes the it is hit, as that is to late. However your idea of 1000nds of g's is wrong.
A laser at that hight has perhaps a diameter of a meter. So sidestepping a meter is enough.
Even a human can step 1 yard/1 meter to the side in a 1/10th of a second.
First, it isn't enough to move a meter in 1/10th of a second. It has to accelerate such that ends up a meter away from where it would otherwise be.
That is an acceleration of 200 m/s^2, assuming we stick with a meter and 0.1 seconds. That is quite a bit of acceleration. It isn't actually enough to evade the fire though even with your numbers.
What happens if the laser misses? The computer fires again. It would fire in a random pattern around you, so it would hit if your RNG maneuvered you into the same place the turret's RNG picked to shoot at. Maybe it would take 5 shots to hit you, maybe 15. You're constantly accelerating at 20G the whole time in random directions to try to avoid it.
To really evade fire you'd need considerably more acceleration so that you're not just a meter away from your last predicted position when the laser arrives.
I actually outlined this on slashdot before, though it was probably years ago. Basically armor and acceleration are both forms of defense in space combat at long range, since both increase the total amount of energy an attacker has to expend into space in order to destroy you. If I can be anywhere in a circle a mile across by the time the laser arrives, you have to put enough energy into every point in that circle to destroy me no matter where I'm at. As range closes, the error in my position drops until I'm dead, or vice-versa.
Keep in mind that uniform motion is no defense at all, since the laser would be leading you. You need continuous random acceleration.
Okay, I admit it, I don't get the punchline. I even added one to the cart expecting that as the price per stick (even then, unbelievably low, but maybe for tested pulls), but no, $79 for the whole bulk pack, new???
You can't even get no-name sticks of non-ECC labelled in Chinese for that. That can't count as a real price, can it?
Sure. You just have to pay with bitcoin in advance. You know, just in case you're not a reliable purchaser.
I was looking at motherboards and even finding ones that support ECC is difficult, unless this is one of those situations where any motherboard works as long as the CPU supports it.
As far as I understand it, most AMD processors do support ECC, and Intel only supports it in their upper-end products (artificial restriction to segment the market - i7/Xeon/etc). What I don't understand is where the motherboard fits in.
Heck, Newegg doesn't even track that as an option on their product selector.
One of the things I learned early on as a tweenager within the FOSS was that if you have a good idea and can communicate well & civilly, nobody actually seems to care terribly much if you possibly are a dog that somehow learned how to type.
If somebody can communicate well and civilly, then they aren't a dog that somehow learned how to type. I think the whole point of the article is that some people don't communicate well and civilly.
The thing is, being an asshole shouldn't be relevant. Is the person contributing good code? That's all that should matter.
That is hardly true. If nobody wants to work for you because one of your top contributors drives everybody nuts, then you have to decide whether their sole contributions really do outweigh everything you're losing.
Sometimes the asshole really is a special snowflake and you just move them to their own side of the office and insulate them from everybody else and pander to them. Maybe you pay everybody who has to work with them an extra 20% to get them to put up with it. However, more often than not they're not nearly as special as they think they are, and you're better off without them.
This actually has something to do with why so many people hate Systemd. It turns out that Systemd is professional-quality work done by competent salaried engineers. Our problem with it is that we're used to beautiful code made by geniuses. Going all of the way back to DMR.
I can't say I agree. The reason I run systemd (on a distro where it isn't even the default) is that I find it a rather elegant and powerful solution.
I think that the problem is that people conflate good solutions with simple solutions. I'll certainly agree that the old way was simpler. That has a certain beauty to it. The problem is that it doesn't actually solve all the problems well.
This is by no means a software issue exclusively. Take your favorite political soundbite, and most likely it involves taking some really complicated issue like war in the middle east or national healthcare or socialism or the finance industry or global warming, and then proposes some solution that could fit on a single powerpoint slide. People like solutions that are easier to understand, but that doesn't automatically make them the best solutions.
Wanting a positive atmosphere within the project is one thing. Worrying about angry blog posts is insecure and petty at best.
I think the concern was more with angry blog posts and other passive-aggressive behavior on community-provided sites.
If you want to rant on your random wordpress page, that is your right. If you want it to be on some blog aggregator provided by the people you're complaining about, that is another matter.
Agree completely. One of the things that really impresses me with systemd is just how event-driven it actually is, and how it separates events from services.
I can have a service that performs some task. I can trigger it to run when another service runs. I can trigger it to run when a target/"runlevel" is activated (the traditional rc.d approach). I can trigger it to run at certain times. I can trigger it to run when a particular udev rule is satisfied. I can trigger it to run when when a file is modified. I can probably trigger it to run when a filesystem is mounted. I can trigger it to run when somebody connects to a socket. I can even trigger it to run conditionally if one of those things happens and the device is in a particular power state, etc.
I think the systemd devs have done a pretty good job of building on the richness of the whole dependency/event-based paradigm.
I've worked in this industry for 32 years. I have never seen someone get a raise without moving to a new job. Not once. It just doesn't happen.
What industry is that? The software industry? I just told you that I don't work in the software industry, and I don't work for a start up. I work in IT. You do realize that virtually every company on Earth hires people that work in IT in some form. I'd probably prefer to work in the software industry but stuff like this is one of the reasons I don't.
I would. The current construction doesn't stop racism/whateverism you choose and moreover denies us the knowledge of who is who.
Your logic assumes that people can never change, and that racists should basically just live off of tax dollars or become criminals to survive.
I'm not really a big fan of solutions that involve blacklisting people for life anytime they do anything that a large number of people find offensive. I don't even support that if somebody commits homicide or any other heinous crime.