But that doesn't mean that I have to sit idly by waiting for the police while they decide my things would look nicer in their house. I'll confront the trespasser and shoot to kill if I feel threatened. It's up to the jury decide beyond reasonable doubt that I acted criminally - I don't have the time to conduct a trial if my life may be in danger.
And in most countries that aren't America, a jury will almost certainly find you partially - if not wholly - culpable for escalating a situation to violence without justifiable cause.
Someone carrying your TV out the door is not a good enough reason to kill them, nor risk the lives of everyone else in the house.
How do I determine if they are there to steal food?
Well, you could try waiting rather than just shooting first and asking questions later.
Though since you've already stated you don't consider intent relevant - "they've broken into my home, and that is all I need for proof" - it's clear you're just itching to kill someone and want an excuse.
But it's almost certainly true. Just look at OpenBSD's record. They went for a full decade without any vulnerabilities in the base system before one was eventually found.
Remote vulnerabilities. In the default install. Which isn't that hard to achieve when your default install doesn't really do much and hardly anyone uses your system.
Just imagine what you could get from programmers that are both paid and required to use secure coding practices.
Who didn't have to work towards any specific deadlines or goals ? And had essentially nothing to lose if they didn't get there ? I'd expect much the same.
When you have nothing in particular to achieve, all the time in the world to achieve it, and no real consequences if you don't, then you'd expect anything that was done would be done well. However, the real world doesn't work like that.
why? Because "some crazy fucker" is the only type of person who would defend himself and his family?
No, because fear for your life is a perfectly natural situation under which to consider getting the fuck out of there.
I'll tell you what, Doc. A person committing a home invasion robbery is not likely to have premeditated damaging your door or window to enter your private dwelling with a weapon in order to make you scrambled eggs for breakfast. If they break into an occupied home with a weapon, they are planning to be violent if necessary even if they claim otherwise.
Of course, they could also just have picked the lock on the back door and not expected to find anyone home at all until you came downstairs looking for a glass of water.
They've broken into my home, and that is all I need for proof. What the hell else should I do to confirm they desire to kill/rape/attack me or my family? Ask them politely?
I bet you heroes can't wait to bag your first street urchin who sneaks in the pet door looking for food. Do you think the police will let you put his head up above the mantle ?
True, and I agree with you. But in the end your definition of reasonable might not be shared by the police and the prosecutor's office. And if they convince the jury to see it their way, well, you're in trouble.
And if you have some examples of that happening, you might have an argument.
The counter-example is equally terrifying. Someone shoots a kid in their kitchen who had snuck in through the pet door to steal some food, and automatically gets off by claiming "self defense". Note that most of the cowboys in this discussions are insisting that they should be able to do this.
For example [...]
Your "example" is meaningless until it has wound it's way through the court system and an actual conclusion has been reached. For all you know the guy is lying through his teeth and fired a dozen shots over the heads of a "gang" of five 12 year olds who called him naughty names. It's almost certain the police *must* charge anyone who uses a gun without a direct threat from another gun be charged, and that those charges can later be dropped.
Vaguely making statements that essentially boil down to "they should just fix the problems" is not giving examples. Especially when they're already doing it and have been forever.
Brush up on your history, and look at the architectures of each.
I have. They were two of the examples used as part of my OS Architecture subjects at University. Heck, even the Wikipedia pages have all the information necessary in them to demonstrate you're wrong.
However, if you want to try and explain how the microkernel, modular, portable, 32-bit, pervasively multithreaded, SMP-capable, multiuser, ACLs-throughout Windows NT bears the slightest architectural resemblence to the monolithic, x86 only, 16-bit, single user, non-SMP capable, completely-without-security OS/2, I could do with a laugh.
Windows NT looks nothing like OS/2. It does look a lot like VMS, which is hardly surprising since the same group of people were basically responsible for both. In fact, it looks so much like VMS (more accurately, it's canned successor) that DEC sued Microsoft about it.
Don't expect slashdot to handhold you through your learning about these things though.
Hah. As if I'd trust Slashdot to provide any useful or accurate information about Windows.
Look, if you can find a single credible source that can identify meaningful similarities between Windows NT and OS/2, I'm more than happy to read it. But you won't, because it doesn't exist.
Nearly ANY change in state of a Windows machine will "edit the Registry". Any configuration with a shiny happy GUI tool certainly will.
*Applications* changing the Registry is an entirely different situation to manual hand-editing, which was the scenario you were describing requiring a revision control system.
What, pray tell, is baroque about "set it at the temperature you want"?
Absolutely nothing, but that's not the system you described.
I mean, what do you think is more reasonable in the following situation: it's 90 degrees outside, and someone sets the thermostat in a large office to 60 degrees.
3) Someone wants to get the temperature down as quickly as possible.
What does it tell you, please? You are taking away one choice (not turning up), so how can this tell you/more/ about how people feel?
Because not turning up says _nothing_ about how they feel. It just says they didn't turn up. Having to turn up and do _something_ at least gives you an idea whether or not they want anything to change.
My argument is that something is better than nothing. You have yet to present an argument as to how nothing is better than something.
Are you proposing that the writing on any spoilt ballot is recorded?
It would certainly help with providing more useful information, but no.
For example, in my case I would write that I refuse to vote while voting is compulsory (actually, I lie, I would probably not turn up and pay the $20ish fine, if it becomes anything like the Australian system - and on the form explaining my reason for not voting, I would say "because I have to, and that is not democracy").
In no way is it "not democracy". There is nothing compulsory voting sacrifices, and it gives a *vastly* better insight into how the electorate feels.
I could say that about you. I agree with the GP and that comes from 25+ years in this industry.
Regardless of whether or not you agree, it's still wrong.
* The Registry has been protected by ACLs since the day it was implemented.
* There is no special "relationship" with "kernel, user space and hardware".
* It's no more "ridiculously easy to screw up" than editing text files in/etc, with the added bonus that directly modifying the Registry is exceptionally uncommon, whereas editing files in/etc is a regular occurrence.
* User space was in no way "redefined" in XP SP2. (Aside: It always fascinates me how some people see SP2 as this near-mythical turning point for Windows, when it's architecturally no different from the several major revision before it.)
Incidentally, on unix systems i make heavy use of revision control for my configuration files (simply check them in to cvs), how do you do that with registry keys?
The real question is why the hell are you editing the Registry so frequently this is even something worth worrying about ?
The registry is a foul design decision, and up to XP SP2, was accessible by anything for the worst of reasons. Because of its relationship to the kernel, user space, and hardware, it was ridiculously simple to screw it up, or make it the crux of bad behavior in strange, unusual, and bizarre ways. After XP SP2 when user-space was 'redefined', it continued to be the garbage pail for every bad programming mistake ever made in Windows.
Pretty much everything in this paragraph is wrong.
Exactly. It's a problem across many industries - just ask any auto mechanic. The people designing the product aren't thinking in terms of servicing the product.
Yes they are, they're just considering it as a secondary (well, tertiary or lower) priority to things like size, weight, battery life and "can pull it apart with nothing more than a pocket knife".
Which is generally a good engineering decision, when a machine spends 99.99% of its time as a laptop and 0.01% of its time disassembled on a workbench.
There should be no reason why a laptop couldn't be well designed like an Apple, and easy to service.
When you say "easy to service", do you mean "easy to service by a trained individual with a proper toolset and work environment", or "easy to service by some random person who put a computer together once with a philips screwdriver and a kitchen table" ? Because these are *very* different design constraints.
(i) I am happy with the status quo, and will leave it up to others to decide which candidate based on details which are unimportant to me;
These two statements are nearly contradictory. If you're happy with the status quo, you'll vote for the incumbent because you "have" to vote for someone. If you're not, you'll vote for someone else.
If you're happy for everyone else to decide, you'll lodge an invalid vote, or vote for something ridiculous like the Beer Drinker's Party.
Look, I didn't say compulsory voting was a perfect solution. I didn't even suggest it. I do, however, firmly believe it's better than a non-compulsory system, because it "forces the hand" of apathetic people to make _some_ indication of their opinion.
At least with voluntary voting you may be able to say something about the difference between someone who spoilt his ballot and somebody who does not turn up.
The distinction between these two options is nearly irrelevant, from the perspective of drawing useful conclusions. Knowing whether someone supports the incumbent or not, which a compulsory system goes much further towards demonstrating, is vastly more useful information than a shaky inference about whether they don't like the system, don't like the options, or simply can't be bothered.
In short, a voluntary system tells you *nothing useful at all* about the apathetic masses. It is impossible to make any useful conclusion whatsoever about people who don't turn up. It only tells you about the people who care enough to get out and vote, who are *generally* hot-button- or single-issue voters.
A compulsory system, at least, tells you _something_ about the majority, which - while far from a complete solution - is a massive improvement over nothing.
I've never seen an argument against compulsory voting that didn't ultimately boil down to "I don't want to have to get off my arse and make a decision, even if it's one as simple as 'screw the bastards'".
In this scenario you could build a reasonably powerful laptop for $300-$400 if you chose last gen's components instead of the latest and greatest.
You can buy a reasonable powerful laptop today for $300-$400, with the added bonus that it's probably had at least a modicum of engineering effort put into making sure the whole package works together (cooling, power draw, etc)
There is only one type of person for whom building a computer vs buying something off the shelf is cheaper: those who have the requisite knowledge and put no dollar value on the time involved. For everyone else - and even for experts who value their time appropriately - buying something off the shelf is cheaper.
I worked at a facility where a thermostat set above seventy-something is in air conditioning mode and set below that is heating mode. And I worked with morons whom alternated it at extremes and then couldn't figure out why the HVAC didn't work. I get to work and its about 50 in the cubes... cow orker says "I'm freezing so I set it to 85"... "Well, don't you think 85 is kind of high for the airconditioner?" I turn it down to 70 and we warm right up. Same deal in the summer. Its 90 in the cubes because some clown set it to 60 placing us in heating mode, and god knows its well above 60 so nothing happens. I crank it up to 75 and we're soon chilling. And the amazing part is these people NEVER LEARNED. Ever. I would imagine they're still all screwed up.
One can easily imagine why given the ridiculously baroque and counter-intuitive system you've just described.
I'm amazed how many people think HVAC is strictly proportional and the thermostat tells the machinery how hard to work. That technology exists but is rare and expensive and you almost certainly don't have it.
A quite reasonable expectation with any thermostat is that when you set the temperature on it, that is the temperature the system will reach and maintain. A perfectly reasonable conclusion from that assumption is that the system will attempt to attain the initial temperature relatively quickly without "overshooting", and thus a larger delta will bring the temperature down quicker.
I remember when we first moved to Phoenix, and I saw air conditioners with thermostats that had to be put into either "heat" or "cool" mode (and then had separate sets of thresholds for each). All I could do it just shake my head and wonder what idiot ever came up with that interface, and why.
Slashdot Mirror
But that doesn't mean that I have to sit idly by waiting for the police while they decide my things would look nicer in their house. I'll confront the trespasser and shoot to kill if I feel threatened. It's up to the jury decide beyond reasonable doubt that I acted criminally - I don't have the time to conduct a trial if my life may be in danger.
And in most countries that aren't America, a jury will almost certainly find you partially - if not wholly - culpable for escalating a situation to violence without justifiable cause.
Someone carrying your TV out the door is not a good enough reason to kill them, nor risk the lives of everyone else in the house.
When you get a license to carry a lethal weapon, your attitude and manner changes.
And when you dare to venture outside into the perilous urban jungle without your gun, does your attitude and manner change back ?
How do I determine if they are there to steal food?
Well, you could try waiting rather than just shooting first and asking questions later.
Though since you've already stated you don't consider intent relevant - "they've broken into my home, and that is all I need for proof" - it's clear you're just itching to kill someone and want an excuse.
But it's almost certainly true. Just look at OpenBSD's record. They went for a full decade without any vulnerabilities in the base system before one was eventually found.
Remote vulnerabilities. In the default install. Which isn't that hard to achieve when your default install doesn't really do much and hardly anyone uses your system.
Just imagine what you could get from programmers that are both paid and required to use secure coding practices.
Who didn't have to work towards any specific deadlines or goals ? And had essentially nothing to lose if they didn't get there ? I'd expect much the same.
When you have nothing in particular to achieve, all the time in the world to achieve it, and no real consequences if you don't, then you'd expect anything that was done would be done well. However, the real world doesn't work like that.
why? Because "some crazy fucker" is the only type of person who would defend himself and his family?
No, because fear for your life is a perfectly natural situation under which to consider getting the fuck out of there.
I'll tell you what, Doc. A person committing a home invasion robbery is not likely to have premeditated damaging your door or window to enter your private dwelling with a weapon in order to make you scrambled eggs for breakfast. If they break into an occupied home with a weapon, they are planning to be violent if necessary even if they claim otherwise.
Of course, they could also just have picked the lock on the back door and not expected to find anyone home at all until you came downstairs looking for a glass of water.
My wife, kids and mother have nothing to fear from the Bernies of the world, why do you fear him?
How about if your kid was standing nearby, Bernie mistook him for a mugger and shot him, then one of the stray bullets killed your wife ?
They've broken into my home, and that is all I need for proof. What the hell else should I do to confirm they desire to kill/rape/attack me or my family? Ask them politely?
I bet you heroes can't wait to bag your first street urchin who sneaks in the pet door looking for food. Do you think the police will let you put his head up above the mantle ?
True, and I agree with you. But in the end your definition of reasonable might not be shared by the police and the prosecutor's office. And if they convince the jury to see it their way, well, you're in trouble.
And if you have some examples of that happening, you might have an argument.
The counter-example is equally terrifying. Someone shoots a kid in their kitchen who had snuck in through the pet door to steal some food, and automatically gets off by claiming "self defense". Note that most of the cowboys in this discussions are insisting that they should be able to do this.
For example [...]
Your "example" is meaningless until it has wound it's way through the court system and an actual conclusion has been reached. For all you know the guy is lying through his teeth and fired a dozen shots over the heads of a "gang" of five 12 year olds who called him naughty names. It's almost certain the police *must* charge anyone who uses a gun without a direct threat from another gun be charged, and that those charges can later be dropped.
In the U.S., you can justifiably kill someone if they've broken into your house and you could convince a jury that you were in fear of your life.
This is true for every country I'm aware of. Despite what you may have seen on TV.
If they are under aim and make a sudden movement, I can assume they are reaching for a weapon, [...]
Or maybe they're just scared shitless because some crazy fucker has a gun pointed at them and are turning to run away.
Vaguely making statements that essentially boil down to "they should just fix the problems" is not giving examples. Especially when they're already doing it and have been forever.
You're simply wrong.
No, I'm not.
Brush up on your history, and look at the architectures of each.
I have. They were two of the examples used as part of my OS Architecture subjects at University. Heck, even the Wikipedia pages have all the information necessary in them to demonstrate you're wrong.
However, if you want to try and explain how the microkernel, modular, portable, 32-bit, pervasively multithreaded, SMP-capable, multiuser, ACLs-throughout Windows NT bears the slightest architectural resemblence to the monolithic, x86 only, 16-bit, single user, non-SMP capable, completely-without-security OS/2, I could do with a laugh.
Windows NT looks nothing like OS/2. It does look a lot like VMS, which is hardly surprising since the same group of people were basically responsible for both. In fact, it looks so much like VMS (more accurately, it's canned successor) that DEC sued Microsoft about it.
Don't expect slashdot to handhold you through your learning about these things though.
Hah. As if I'd trust Slashdot to provide any useful or accurate information about Windows.
Look, if you can find a single credible source that can identify meaningful similarities between Windows NT and OS/2, I'm more than happy to read it. But you won't, because it doesn't exist.
Nearly ANY change in state of a Windows machine will "edit the Registry". Any configuration with a shiny happy GUI tool certainly will.
*Applications* changing the Registry is an entirely different situation to manual hand-editing, which was the scenario you were describing requiring a revision control system.
I have no idea where that came from, but I'm pretty sure you missed my point.
What, pray tell, is baroque about "set it at the temperature you want"?
Absolutely nothing, but that's not the system you described.
I mean, what do you think is more reasonable in the following situation: it's 90 degrees outside, and someone sets the thermostat in a large office to 60 degrees.
3) Someone wants to get the temperature down as quickly as possible.
What does it tell you, please? You are taking away one choice (not turning up), so how can this tell you /more/ about how people feel?
Because not turning up says _nothing_ about how they feel. It just says they didn't turn up. Having to turn up and do _something_ at least gives you an idea whether or not they want anything to change.
My argument is that something is better than nothing. You have yet to present an argument as to how nothing is better than something.
Are you proposing that the writing on any spoilt ballot is recorded?
It would certainly help with providing more useful information, but no.
For example, in my case I would write that I refuse to vote while voting is compulsory (actually, I lie, I would probably not turn up and pay the $20ish fine, if it becomes anything like the Australian system - and on the form explaining my reason for not voting, I would say "because I have to, and that is not democracy").
In no way is it "not democracy". There is nothing compulsory voting sacrifices, and it gives a *vastly* better insight into how the electorate feels.
I could say that about you. I agree with the GP and that comes from 25+ years in this industry.
Regardless of whether or not you agree, it's still wrong.
* The Registry has been protected by ACLs since the day it was implemented.
* There is no special "relationship" with "kernel, user space and hardware".
* It's no more "ridiculously easy to screw up" than editing text files in /etc, with the added bonus that directly modifying the Registry is exceptionally uncommon, whereas editing files in /etc is a regular occurrence.
* User space was in no way "redefined" in XP SP2. (Aside: It always fascinates me how some people see SP2 as this near-mythical turning point for Windows, when it's architecturally no different from the several major revision before it.)
Incidentally, on unix systems i make heavy use of revision control for my configuration files (simply check them in to cvs), how do you do that with registry keys?
The real question is why the hell are you editing the Registry so frequently this is even something worth worrying about ?
The registry is a foul design decision, and up to XP SP2, was accessible by anything for the worst of reasons. Because of its relationship to the kernel, user space, and hardware, it was ridiculously simple to screw it up, or make it the crux of bad behavior in strange, unusual, and bizarre ways. After XP SP2 when user-space was 'redefined', it continued to be the garbage pail for every bad programming mistake ever made in Windows.
Pretty much everything in this paragraph is wrong.
People who care about security hate it too.
They shouldn't. The Registry is a vastly more securable system than anything built around textfiles.
None of your other "arguments" have anything to do with the Registry, they apply equally to textfiles.
Exactly. It's a problem across many industries - just ask any auto mechanic. The people designing the product aren't thinking in terms of servicing the product.
Yes they are, they're just considering it as a secondary (well, tertiary or lower) priority to things like size, weight, battery life and "can pull it apart with nothing more than a pocket knife".
Which is generally a good engineering decision, when a machine spends 99.99% of its time as a laptop and 0.01% of its time disassembled on a workbench.
There should be no reason why a laptop couldn't be well designed like an Apple, and easy to service.
When you say "easy to service", do you mean "easy to service by a trained individual with a proper toolset and work environment", or "easy to service by some random person who put a computer together once with a philips screwdriver and a kitchen table" ? Because these are *very* different design constraints.
(i) I am happy with the status quo, and will leave it up to others to decide which candidate based on details which are unimportant to me;
These two statements are nearly contradictory. If you're happy with the status quo, you'll vote for the incumbent because you "have" to vote for someone. If you're not, you'll vote for someone else.
If you're happy for everyone else to decide, you'll lodge an invalid vote, or vote for something ridiculous like the Beer Drinker's Party.
Look, I didn't say compulsory voting was a perfect solution. I didn't even suggest it. I do, however, firmly believe it's better than a non-compulsory system, because it "forces the hand" of apathetic people to make _some_ indication of their opinion.
At least with voluntary voting you may be able to say something about the difference between someone who spoilt his ballot and somebody who does not turn up.
The distinction between these two options is nearly irrelevant, from the perspective of drawing useful conclusions. Knowing whether someone supports the incumbent or not, which a compulsory system goes much further towards demonstrating, is vastly more useful information than a shaky inference about whether they don't like the system, don't like the options, or simply can't be bothered.
In short, a voluntary system tells you *nothing useful at all* about the apathetic masses. It is impossible to make any useful conclusion whatsoever about people who don't turn up. It only tells you about the people who care enough to get out and vote, who are *generally* hot-button- or single-issue voters.
A compulsory system, at least, tells you _something_ about the majority, which - while far from a complete solution - is a massive improvement over nothing.
I've never seen an argument against compulsory voting that didn't ultimately boil down to "I don't want to have to get off my arse and make a decision, even if it's one as simple as 'screw the bastards'".
In this scenario you could build a reasonably powerful laptop for $300-$400 if you chose last gen's components instead of the latest and greatest.
You can buy a reasonable powerful laptop today for $300-$400, with the added bonus that it's probably had at least a modicum of engineering effort put into making sure the whole package works together (cooling, power draw, etc)
There is only one type of person for whom building a computer vs buying something off the shelf is cheaper: those who have the requisite knowledge and put no dollar value on the time involved. For everyone else - and even for experts who value their time appropriately - buying something off the shelf is cheaper.
I worked at a facility where a thermostat set above seventy-something is in air conditioning mode and set below that is heating mode. And I worked with morons whom alternated it at extremes and then couldn't figure out why the HVAC didn't work. I get to work and its about 50 in the cubes ... cow orker says "I'm freezing so I set it to 85" ... "Well, don't you think 85 is kind of high for the airconditioner?" I turn it down to 70 and we warm right up. Same deal in the summer. Its 90 in the cubes because some clown set it to 60 placing us in heating mode, and god knows its well above 60 so nothing happens. I crank it up to 75 and we're soon chilling. And the amazing part is these people NEVER LEARNED. Ever. I would imagine they're still all screwed up.
One can easily imagine why given the ridiculously baroque and counter-intuitive system you've just described.
I'm amazed how many people think HVAC is strictly proportional and the thermostat tells the machinery how hard to work. That technology exists but is rare and expensive and you almost certainly don't have it.
A quite reasonable expectation with any thermostat is that when you set the temperature on it, that is the temperature the system will reach and maintain. A perfectly reasonable conclusion from that assumption is that the system will attempt to attain the initial temperature relatively quickly without "overshooting", and thus a larger delta will bring the temperature down quicker.
I remember when we first moved to Phoenix, and I saw air conditioners with thermostats that had to be put into either "heat" or "cool" mode (and then had separate sets of thresholds for each). All I could do it just shake my head and wonder what idiot ever came up with that interface, and why.