So you believe that language directly manipulates neurons to cause the correct thought to appear in the brain? If not, then you must conceed that language is interpreted. Sometimes reasonably, sometimes not.
Let enough time pass and you most certainly have to look to intent since the language itself will have changed. Even the mode of thought will have to a degree shifted given a couple hundred years.
What you call textualism is based on the absurd notion that a word has an intrinsic meaning not subject to human understanding.
Or do you believe that when a parent tells a crabby child to be nice they mean it's OK to continue snapping at people as long as he stands at a right angle to the ground?
That would be great, but won't happen. The unwritten law is that they and their buds on Wall Street are immune as long as they don't prey on each other.
They might as well have a stack of money outside the courtroom that says "you must be at least this rich to get a free ride". The only thing such interpretations do is make it impossible for the rest of us to even know if any given action of inaction might lead to felony charges.
I suppose it's mostly that people who are violently insane enough to do something newsworthy and yet organized enough to choose a target and actually plan against it are rare.
I do note that there are more businesses that put people behind heavy plexiglass than there used to be. It's either corporate paranoia or they actually have had increasing numbers of people jump over the desk and "register their dissatisfaction".
No, I am arguing that the rule of law requires an actual understanding of the law and a reasonable interpretation of language. There really are phrases that any reasonable person would understand one way even if the words might suggest something else if read by an AI or a non-native speaker.
That is quite different from rule of man.
I would go so far as to say that allowing every bizarre over-broad interpretation of law (such as applying SOX to the fishing boat case) is an example of allowing rule of man to take over. For example by applying a law that was never meant to be applied such that a minor infraction suddenly carries a major penalty.
Essentially I advocate calling bullshit when someone wants to bend the law to their will. Legislative intent is often quite discernible and it is perfectly valid to do so in order to squash these impermissible efforts by individuals to effectively re-write the law
I do not believe that inventing the stupidest possible interpretation of a law and enforcing it with vigor is a very good approach. Human language isn't sufficiently preceice for that to ever work. Besides, we have more than enough stupid in the world.
I base it on the actual title of the law: 'Public Company Accounting Reform and Investor Protection Act'.or if you prefer the House version: 'Corporate and Auditing Accountability and Responsibility Act'. It seems it was made fairly clear what was to be regulated and why.
Then there's the way the law keeps referring to corporate accounting and auditing procedures and the SEC and such while fish and wildlife is conspicuously left out of the rule making process.
No, there is a such thing as DELIBERATELY mis-understanding plain language through excessively literal interpretation. For example, by trying to apply a law that was very clearly intended to apply to accounting and finance and abusing it through technicality to apply it to fishing in hopes of securing a conviction for a much more serious offense than is warranted.
It's the same sort of liberalism that gives us zero tolerance idiocy in schools.
You just named a bunch of stuff that builds upon small interchangeable components. For example, you have a display manager that runs an external X server and then displays itself. It uses pam to determine if you entered a valid password. If so, it changes user and runs a window manager and session manager on top of the X server. All of those components can be mixed and matched and none of them are part of the same project. You could also choose to log in on the console and then startx, skipping the display manager. You could run X by itself and then a GUI application on top of that and have no window manager. You can run just the X app redirected by ssh (or directly) to another X display (which may or may not be the same version as the one on your usual desktop. That's not counting adding in some combination of xpra, vnc, and RDP. Often you can switch window managers out in a desktop environment.
Meanwhile, once you get that loose collection of parts up and running, you start your browser and connect to webmin (through apache). Then webmin (unless it has changed a LOT) shells out to those system utilities you didn't realize you were using in order to gather information to display and to carry out your commands.
Agreed that adding a print server to X was a mistake, but because it's all done the Unix way, you can at least pretend it doesn't exist and avoid the worst of that dane brammage.
It seems unified to you because each part is designed to inter-operate the Unix way and the default stack of parts is well chosen to work cleanly. You may not even notice if you end up running components of 3 different desktops at the same time. But it sure can come in handy to be able to re-arrange those pieces nearly arbitrarily sometimes.
That's the elephant in the room. Is the poor turnout true voter apathy as you imagine or is the apathy for a process perceived to make no difference.
For example, when I spill salt, I don't throw any over my shoulder. Is it that I am apathetic towards the devil or is it simply a reflection of my strong conviction that throwing salt over my shoulder is a useless gesture?
Funny thing, runnit is in by default. It gets called by init. It just doesn't actually do anything by default but get called and see that there is nothing to do.
If you and others like systemd, I'm fine with that. My objection is to the way freedesktop tries to leverage dependency (and FUD about dependency) to cram it down my throat. Why is it such a problem for it to do its thing and leave the rest of us alone?
You are, naturally, welcome to your own philosophical view, but it seems odd then that you would be involved in an OS so fundamentally opposite to your philosophy.
The problem is that the merchant has no good way to prove ID and yet gets left holding the bag. It is possible to make the transactions safe and secure for both parties, but the credit card companies have no incentive to do so because they have managed to push all liabilities off on the merchant (ultimately reflected in higher prices to everyone to cover losses).
A clarification of uselessd may be in order. It does not appear to be removing functionality, just changing it form excessively tight coupling to loose coupling. It still supports logging, it just logs through the well understood and time tested syslog. It still does process monitoring, login management, and hardware management. In fact, it allows systemd (uselessd) to do process management without being the system init.
Personally, I would prefer to further split up process management by assigning a monitor process to each process to be monitored with a common API. Thatwould make it much easier to, for instance, write a new monitor that actually polls the service it monitors using it's standard interface in order to detect a process that is still technically running but is in a wedged state (for example, accepts connections but doesn't actually respond to queries). It would be nice to be able to just drop that in with no way that a bug in the new monitor will affect other monitors.
Such separation confines failures and that is a cornerstone of reliability.
I find it interesting that Debian has supported runsv to monitor services for some time and NOBODY seems to use it EVER, yet suddenly there is a burning need for process monitoring (apparently not runsv).
Only because none of them actually address the issue I pointed out or offer a way to do so. So sorry I gored your sacred cow. This is a technical matter, not politics. Answering at a question doesn't do it.
Now, look at that list and tell me where I am wrong or apologize (or slink away if you prefer).
You should understand from a reliability standpoint why I would be reluctant to switch away from a system I have been using successfully for years for something that has had practically no time in production, particularly something where I have serious reservations about the design. I would much rather deal with the devil I knopw for as long as it takes for a new system with a decent design to come along.
For maximum safety, I would like that something to be able and willing to work alongside 'old reliable' so I can phase the change in and quickly revert if it screws up. Perhaps even define a runlevel to chain to the new system and another for normal operation under the old system.
But let's be realistic here. I can pound out a reliable init script in 5 minutes or so. It's not that I am some sort of scripting super genius, it just isn't that hard. From a reliability standpoint, why am I going to adopt an architecturally unsound init system to avoid that?
From watching uselessd, I can see that it took a one man team about a month to hack out a multitude of mistakes in systemd. Rookie mistakes honestly.
I am looking at a real answer to the problem that would be at the same time simpler and more powerful than systemd. Honestly, the biggest problem I'm having is hitting the moving target that is cgroup fs thanks to the munging about due to systemd sticking it's nose in. Imagine my joy that the systemd team broke their own "peace treaty" in less than a year. Now they want to claim total ownership of the API. You may wonder why I would go on about that. Simple answer, it shows the real focus of the project and it is not winning through technical superiority and reliability. The good news is that some of their bluster about cgroup seems to be FUD.
To be clear, I'm all for a better init, but systemd isn't it. Some of the ideas there could be part of it, but not without a better grasp of proper design and dependency management. The kitchen sink approach is not going to cut it.
Back to the scripts for a moment. One claim I keep seeing is of a translator that can take a systemd config and generate an equivalent script for init. Why not use that and skip the rest?
They ground the government to a halty over Obamacare, but haven't even made a peep about curbing the NSA. They easily have to power to set the NSA's budget to $0. but haven't.
You are clearly uneducable. I for one welcome our all singing all dancing date time display that edits photos, does my taxes, booots the system, makes ice cream, and starts world war three but can;'t actually display the date or the time because it's far too busy for that.
You are desperately twisting logic and semantics until they break to try to claim systemd is the ned God while showing that you don't actuially know enough to make the call. I have no idea why.
I'm guessing you aren't a developer. If an individual app wants te capabilities, it can link against the library directly. It is rather ugly to link it in at a low level such as the widget toolkit.
Why, for example would it be a good idea for the date time display widget at the top of the screen to have a built-in spreadsheet, web browser, photo editor, mp3 player, and minesweeper?
But the scripts exist now, why so anxious to throw that work away?
However, when I write my own scripts, they generally do take about 10 lines and get the job done. However, compare a couple scripts from the same distro. You'll see that it is mostly boilerplate with names filled in, so not actually that much work. If it makes you feel better, write the ten lines and run it through a translator, perhaps even at runtime if you must.
Or, install OpenRC. As near as I can tell it is at worst unobjectionable (I have only glanced at it).
You DO realizew that gtk is a set of gimp libraries, yes? You know Gimp is a photoshop like app, yes? Would you care to have photoshop load every time yuou use notepad?
Do you really need image editing capabilities in mtr?
Slashdot Mirror
Of interest is the "52 people between the ages of 19 and 45 that have received a diagnosis of type 1 diabetes within the previous three months"
I imagine minors are excluded based on the many legal issues of research on minors rather than an expectation that it can't work on younger patients.
So you believe that language directly manipulates neurons to cause the correct thought to appear in the brain? If not, then you must conceed that language is interpreted. Sometimes reasonably, sometimes not.
Let enough time pass and you most certainly have to look to intent since the language itself will have changed. Even the mode of thought will have to a degree shifted given a couple hundred years.
What you call textualism is based on the absurd notion that a word has an intrinsic meaning not subject to human understanding.
Or do you believe that when a parent tells a crabby child to be nice they mean it's OK to continue snapping at people as long as he stands at a right angle to the ground?
That would be great, but won't happen. The unwritten law is that they and their buds on Wall Street are immune as long as they don't prey on each other.
They might as well have a stack of money outside the courtroom that says "you must be at least this rich to get a free ride". The only thing such interpretations do is make it impossible for the rest of us to even know if any given action of inaction might lead to felony charges.
So it's like a rainbow in the dark...
I suppose it's mostly that people who are violently insane enough to do something newsworthy and yet organized enough to choose a target and actually plan against it are rare.
I do note that there are more businesses that put people behind heavy plexiglass than there used to be. It's either corporate paranoia or they actually have had increasing numbers of people jump over the desk and "register their dissatisfaction".
No, I am arguing that the rule of law requires an actual understanding of the law and a reasonable interpretation of language. There really are phrases that any reasonable person would understand one way even if the words might suggest something else if read by an AI or a non-native speaker.
That is quite different from rule of man.
I would go so far as to say that allowing every bizarre over-broad interpretation of law (such as applying SOX to the fishing boat case) is an example of allowing rule of man to take over. For example by applying a law that was never meant to be applied such that a minor infraction suddenly carries a major penalty.
Essentially I advocate calling bullshit when someone wants to bend the law to their will. Legislative intent is often quite discernible and it is perfectly valid to do so in order to squash these impermissible efforts by individuals to effectively re-write the law
I do not believe that inventing the stupidest possible interpretation of a law and enforcing it with vigor is a very good approach. Human language isn't sufficiently preceice for that to ever work. Besides, we have more than enough stupid in the world.
I base it on the actual title of the law: 'Public Company Accounting Reform and Investor Protection Act' .or if you prefer the House version: 'Corporate and Auditing Accountability and Responsibility Act'. It seems it was made fairly clear what was to be regulated and why.
Then there's the way the law keeps referring to corporate accounting and auditing procedures and the SEC and such while fish and wildlife is conspicuously left out of the rule making process.
No, there is a such thing as DELIBERATELY mis-understanding plain language through excessively literal interpretation. For example, by trying to apply a law that was very clearly intended to apply to accounting and finance and abusing it through technicality to apply it to fishing in hopes of securing a conviction for a much more serious offense than is warranted.
It's the same sort of liberalism that gives us zero tolerance idiocy in schools.
You just named a bunch of stuff that builds upon small interchangeable components. For example, you have a display manager that runs an external X server and then displays itself. It uses pam to determine if you entered a valid password. If so, it changes user and runs a window manager and session manager on top of the X server. All of those components can be mixed and matched and none of them are part of the same project. You could also choose to log in on the console and then startx, skipping the display manager. You could run X by itself and then a GUI application on top of that and have no window manager. You can run just the X app redirected by ssh (or directly) to another X display (which may or may not be the same version as the one on your usual desktop. That's not counting adding in some combination of xpra, vnc, and RDP. Often you can switch window managers out in a desktop environment.
Meanwhile, once you get that loose collection of parts up and running, you start your browser and connect to webmin (through apache). Then webmin (unless it has changed a LOT) shells out to those system utilities you didn't realize you were using in order to gather information to display and to carry out your commands.
Agreed that adding a print server to X was a mistake, but because it's all done the Unix way, you can at least pretend it doesn't exist and avoid the worst of that dane brammage.
It seems unified to you because each part is designed to inter-operate the Unix way and the default stack of parts is well chosen to work cleanly. You may not even notice if you end up running components of 3 different desktops at the same time. But it sure can come in handy to be able to re-arrange those pieces nearly arbitrarily sometimes.
Perhaps that was a vote of no confidence.
That's the elephant in the room. Is the poor turnout true voter apathy as you imagine or is the apathy for a process perceived to make no difference.
For example, when I spill salt, I don't throw any over my shoulder. Is it that I am apathetic towards the devil or is it simply a reflection of my strong conviction that throwing salt over my shoulder is a useless gesture?
Funny thing, runnit is in by default. It gets called by init. It just doesn't actually do anything by default but get called and see that there is nothing to do.
If you and others like systemd, I'm fine with that. My objection is to the way freedesktop tries to leverage dependency (and FUD about dependency) to cram it down my throat. Why is it such a problem for it to do its thing and leave the rest of us alone?
You are, naturally, welcome to your own philosophical view, but it seems odd then that you would be involved in an OS so fundamentally opposite to your philosophy.
I want a candelabra. When I turn the switch on, gas jets should light the candles. When I turn the switch off, a snuffer should put them out.
But I'm not willing to spend the kind of money it would take for a novelty item, so I guess nothing.
The problem is that the merchant has no good way to prove ID and yet gets left holding the bag. It is possible to make the transactions safe and secure for both parties, but the credit card companies have no incentive to do so because they have managed to push all liabilities off on the merchant (ultimately reflected in higher prices to everyone to cover losses).
A clarification of uselessd may be in order. It does not appear to be removing functionality, just changing it form excessively tight coupling to loose coupling. It still supports logging, it just logs through the well understood and time tested syslog. It still does process monitoring, login management, and hardware management. In fact, it allows systemd (uselessd) to do process management without being the system init.
Personally, I would prefer to further split up process management by assigning a monitor process to each process to be monitored with a common API. Thatwould make it much easier to, for instance, write a new monitor that actually polls the service it monitors using it's standard interface in order to detect a process that is still technically running but is in a wedged state (for example, accepts connections but doesn't actually respond to queries). It would be nice to be able to just drop that in with no way that a bug in the new monitor will affect other monitors.
Such separation confines failures and that is a cornerstone of reliability.
I find it interesting that Debian has supported runsv to monitor services for some time and NOBODY seems to use it EVER, yet suddenly there is a burning need for process monitoring (apparently not runsv).
Only because none of them actually address the issue I pointed out or offer a way to do so. So sorry I gored your sacred cow. This is a technical matter, not politics. Answering at a question doesn't do it.
Now, look at that list and tell me where I am wrong or apologize (or slink away if you prefer).
You should understand from a reliability standpoint why I would be reluctant to switch away from a system I have been using successfully for years for something that has had practically no time in production, particularly something where I have serious reservations about the design. I would much rather deal with the devil I knopw for as long as it takes for a new system with a decent design to come along.
For maximum safety, I would like that something to be able and willing to work alongside 'old reliable' so I can phase the change in and quickly revert if it screws up. Perhaps even define a runlevel to chain to the new system and another for normal operation under the old system.
But let's be realistic here. I can pound out a reliable init script in 5 minutes or so. It's not that I am some sort of scripting super genius, it just isn't that hard. From a reliability standpoint, why am I going to adopt an architecturally unsound init system to avoid that?
From watching uselessd, I can see that it took a one man team about a month to hack out a multitude of mistakes in systemd. Rookie mistakes honestly.
I am looking at a real answer to the problem that would be at the same time simpler and more powerful than systemd. Honestly, the biggest problem I'm having is hitting the moving target that is cgroup fs thanks to the munging about due to systemd sticking it's nose in. Imagine my joy that the systemd team broke their own "peace treaty" in less than a year. Now they want to claim total ownership of the API. You may wonder why I would go on about that. Simple answer, it shows the real focus of the project and it is not winning through technical superiority and reliability. The good news is that some of their bluster about cgroup seems to be FUD.
To be clear, I'm all for a better init, but systemd isn't it. Some of the ideas there could be part of it, but not without a better grasp of proper design and dependency management. The kitchen sink approach is not going to cut it.
Back to the scripts for a moment. One claim I keep seeing is of a translator that can take a systemd config and generate an equivalent script for init. Why not use that and skip the rest?
You would need to line up the black box with dispatcher logs. Police often (read always) speed when there's no emergency.
They ground the government to a halty over Obamacare, but haven't even made a peep about curbing the NSA. They easily have to power to set the NSA's budget to $0. but haven't.
So vote green or write in the pirate party.
You are clearly uneducable. I for one welcome our all singing all dancing date time display that edits photos, does my taxes, booots the system, makes ice cream, and starts world war three but can;'t actually display the date or the time because it's far too busy for that.
You are desperately twisting logic and semantics until they break to try to claim systemd is the ned God while showing that you don't actuially know enough to make the call. I have no idea why.
I'm guessing you aren't a developer. If an individual app wants te capabilities, it can link against the library directly. It is rather ugly to link it in at a low level such as the widget toolkit.
Why, for example would it be a good idea for the date time display widget at the top of the screen to have a built-in spreadsheet, web browser, photo editor, mp3 player, and minesweeper?
But the scripts exist now, why so anxious to throw that work away?
However, when I write my own scripts, they generally do take about 10 lines and get the job done. However, compare a couple scripts from the same distro. You'll see that it is mostly boilerplate with names filled in, so not actually that much work. If it makes you feel better, write the ten lines and run it through a translator, perhaps even at runtime if you must.
Or, install OpenRC. As near as I can tell it is at worst unobjectionable (I have only glanced at it).
You DO realizew that gtk is a set of gimp libraries, yes? You know Gimp is a photoshop like app, yes? Would you care to have photoshop load every time yuou use notepad?
Do you really need image editing capabilities in mtr?
You say that every time I ask a question you don't like the answer to.
It's quite a tell.