Airplanes do not normally float in a medium of highly corrosive saltwater. And they can usually go _around_ storms, or land when the weather gets excessively dangerous. And even good quality, powerful engines need regular hand-on maintenance over the course of a 30 day trip.
> You opt for "married men make better performers"
Please: I didn't say that. It affects the role. For example, married men may be unwilling to perform 80 hour work weeks at crunch time. Being older, and willing to object to abusive contracts or work hours, certainly affects _my_ work performance. And my age is correlated with my work experience.
What I'm saying is that it's unrealistic, and unreasonable, to say these factors don't _affect_ work performance. In soome cases, they may _improve_ work performance! It depends on the situation and the role.
My point is that can it can _affect_ job performance, just as age, gender, religion, nationality, size, weight, skin color, and marital status can do so. The effects can be modest, and are often legislatively barred from affecting hiring. But to pretend that they are not factors, or that they do not alter people's chances of behing hired or affect evaluations of performance is disingenuous.
And let's be clear: many people consider "icky power differentials" to be attractive, both for the more powerful person to enjoy the power, and for the less powerful person to enjoy the attention of a more successful or powerful person. It can be wildly abused, but it's been common practice throughout human history. Ignoring it won't prevent it.
> Religion co-opted marriage and turned it into a religious thing millennia ago
Religions have been involved in major life events throughout the history of civilization. Can you point to any history or culture in which marriage, like birth and death, was +not+ treated as a religious matter? I suspect that any culture which handled marriage as a purely civil issue would integrate religion into it very, very quickly.
Actually, like being white, male, young, married, Christian, or born in the USA, it can certainly affect job performance. Being a member of the "power" ethnic or social jobs can certainly affect political power and perceived acceptability a social and political effect. But most relevant for job performance is the likelihood of being married, and having children, which also affect work availability and available weekly work hours. Then look at the number of business and political leaders who do _not_ have a spouse who takes care of their home and their children. And do _not_ be shocked that the possibility of dating, of or sexual interaction, effects the work place. And anytime there's a possible sexual interaction with a superior or a subordinate, it effects the work relationship.
That said, I'm delighted that we're more accepting of different lifestyles than we were at the start of my career. Let's just not ignore the direct effects of marriage, and of children, on work performance because they remain quite profound. Schedule a major product release for Christmas, Valentine's Day, or for graduation day and see who can do the work if you think it's irrelevant.
There is a great deal of science, and public policy, that would benefit from public exposure. But medical and sociological research benefits from the privacy of the subject, who then feel more free to be truthful. The same is true of political survey data, and "anonymizing" it can be a lengthy, expensive, and uncertain process, especially when coupled with various metadata that is being collected with the experiments or in parallel with it. It can also be very expensive to make public, even without privacy issues, because transforming it from obsolete media and making it available for public download often takes real engineering time. Long term science projects can span decades, and the first sets of data are often on obsolete media.
Overall, it seems an excellent policy, but exceptions will have to be made.
Those can be very useful for quick strikes. As long as you're not actually occupying foreign soil, your point is well mad But wars of occupatio0n take manpower, as demonstrated in Korea, Vietnam, Afghanistan, and the second Iraq war. And don't be mistaken, the USA didn't "win" any of those. In Iraq and Afghanistan now, the US is "declaring victory" and leaving a mess that is, in some ways, worse than when the wars started. Sadam Hussein, as much of a genocidal dictator as he was, didn't allow the Taliban to operate within Iraq. Now they're an integral part of Iraq politics.
Those are problems that air strikes and drone strikes don't solve, they exacerbate.
They want to reduce _personnel_. Personnel have to be trained, and are known to sometimes refuse illegal orders, or worse, to send home pictures of illegal orders.
I did not refer to the whole operating system. I referred specifically to the _kernel_. Do go read http://windowsitpro.com/window..., it's one of the better publicly available analyses I can find.
I'm afraid I don't have copies of both source trees for comparison, and would have legal difficulties if I did post them. DEC and Microsoft settled out of court on this: do look for keywords "Microsoft" and "Cutler" for more detailed legal and software analysis.
> Since Windows 2000 came out the stability has improved immensely
That's because much of that kernel was VMS, stolen from DEC. Thee's a readable description of the wholesale copying at http://windowsitpro.com/window....
s a business tactic, Microsoft's behaviors have been often very effective. I'd refrain from mentioning older thefts by Microsoft except these are at the core of what they sell now: The NT kernel, at the core of Windows 8 and Windows Server, was extensively VMS code stolen from DEC, and DEC bent bankrupt after that. The browser standards wars continue to include "embrace, extend, and break compatibility". The entire "OOXML" debacle of "publishing open standards" for Microsoft Office document standards, then ignoring them for actual MS Office software is an ongoing example. Microsoft Office violates its own standards, and the standards were themselves corrupted, to allow Microsoft to claim "open standards" compatibility which it doesn't actually have.
And then there's "Trusted Computing". The entire ongoing project is not aimed at user privacy: it's aimed at vendor lockin for software, data, and even hardware. And the private keys, including keys to revoke other keys, are held almost entirely in escrow by Microsoft, with no usable guarantees of the keys protection from wholesale abuse.
As does gcc with the right flags, which detect "unreachable code". That can often generate a storm of warnings that require a great deal of additional time to sort through and resolve, and can be different depending on compile time definitions. So it's not often used.
> Long term acceleration on the highway? Nobody thought to just turn off the ignition switch? Turns off the fuel pump, care stops running, eventually you come to full stop
I know of at least one such case, http://www.theguardian.com/wor.... The car was modified for a disabled man, so it's not clear what the combination of controls he could manage was. But he wound up rocketing across France into Belgium, traveling over 100 MPH with quick acting police clearing traffic ahead of him.
And because a software error or "bit error" can cause the software to be stuck in the recursion, in an infinite loop. Even when software errors don't exist, "bit flips" due to electrical transients or even due to cosmic radiation are inevitable due to the quite small transistor size in modern processors. And cars are electrically quite noisy environments, so a certain amount of "bit flip" should be expected as normal for any such system.
> drone has yet to be involved in such an attack, except as an agent of state force and under color of something vaguely resembling 'law'.
I wasn't personally thinking of well focused assassination, but rather of attacks like the Boston Bombing. Even indiscriminate manslaughter among such large crowds has enormous political force, and the Olympics focus worldwide attention on a crowd aimed specially at international cooperation. I can vividly remember the Olympic hostage crisis in 1972: It focused worldwide attention on the politics of the terrorists who committed those murders.
The automatic recover features are worth mentioning. It reduces the risk of danger due to human error, or signal loss: those are inevitable in a crowded environment with thousands of cameras and radio transmitters saturating the airwaves legally and illegally.
My concern would be about the ability to hijack the drones, and about getting the crowds used to low flying remote controlled aircraft. Similar, unauthorized dromes could be reloaded with light and bring it near the crowds or athletes from well outside of any reasonable security perimeter.
It's become much more stable in the last 10 years, in my observed experience. My colleagues and I have seen issues activating downloadable drivers for Windows clients to use various unusual printers, and the client selection of single sided and double sided printing has been awkward. But we've normally simply set up one queue for single sided, the other for double sided, and that's worked well. And if a particular print driver isn't published in the available CUPS or other GUI configurations, we've had good success using old Apple Laserwriter Postscript drivers with modern printers.
Much of the technological problem for CUPS in complex environments has not been CUPS, itself, it's been the available drivers and limitations of the underlying Ghostscript technology that does much of the transformation to printable document format. It's been very difficult to keep up to date on all the subtly modified and "enhanced" features of new printers, and provide user access to those for CUPS and ghostscript. In my observation, ghostscript has gotten _much_ better at successfully providing the necessary tools. And some GUI's for print management have improved, as well. The most recent RHEL system-config-printer has gotten quite good: they seem to have read Eric Raymond's essay and taken it to heart.
The link is to Eric Raymond's "The Luxury of Ignorance: A CUPS Horror Story". It's just what I I point people to when they deal with Windows 8, the new Ubuntu interface, Gnome 3, and the new Fedora installer. Its follow up article, "The Luxury of Ignorance: Part Deux" is at:
The followup shows that the authors of CUPS, much like Mark Shuttleworth in his first responses to the nearly universal dislike for the new Ubuntu interface, showed the same response as CUPS authors made to Eric's complaints. I'm also afraid that the built-in CUPS configuration tool as not improved in any appreciable way since Eric's original essay. Fortunately, many OS developers have written their own and far superior wrappers for configuring CUPS. But such technically sophisticated but incomplete and ignorant of basic workflow tools still abound. Cleaning up after the chaos they leave when they overwrite hand-edited system configurations and disable critical features actually pays a considerable amount of my wages.
They're going to find the _exposed_ ports of deliberately enabled sercices, anyway. Script kiddie port scanning is a popular pasttime, with zombied and rootkitted hosts all over the world doing it as a matter of course for their entire accessible VLAN's. The key underlying advantage is that most of the ports _aren't_ exposed. Do take a look at the VLAN our own ISP, or your own office network uses, for exposed ports. Unless your firewalls are very scrupulously maintained, they will have exposed services directly exposed because they don't have NAT, and they don't have good firewalls set up.
It's the not deliberately exposed ports and services that NAT is such an easy and useful first line of defense for. Old, never updated printers and routers, laptops and desktops and servers wehre the default password is still in place, Nagios servers with default "nagiosadmin" user and default "nagiosadmin" password that allow intruders detailed overviewss of your infrastructure: NFS shares and CIFS shares set up without authentication, which can be very difficult to get people to turn off, NTP services that have never been properly filtered, etc, etc.
May I presume that your account with them is quite old? Or that you are paying them extra for such an IP address? It really has fallen out of favor, partly due to IPv4 limitations, an dpartly to idscourage customers from hosting conntent from their home machines.
That is an excellent point, and I was negligent to not review which is the RFC specified VLAN for such connecitons.
I';m afraid, however, that it's not the case where I'm sitting right now nor in a lot of the less skilled ISP's I've had to deal with personally and with clients over the years.
NAT is not a firewall. But by reducing the exposed attack vector, and helping keep traffic from coming _in_ to your network that is not specifically permitted, it's a tremendous beneficial in reducing the constant clamor of script kiddies hammering on your network, hoping for an open router port or a misconfigured firewall. It also discourages careless security, such as running SSH on every single machine on a network, exposed to the Internet, or allowing unencrypted HTTP and FTP to arbitrary hosts inside your network when those can be packet sniffed for passwords on the incoming connection.
Basically, for something behind a NAT to support an externally exposed service such as DNS, SMTP, NTP, FTP, HTTP, NFS, CIFS, or SSH, you have to _select_ to allow it. For various firewall configurations, I'm afraid that too many people leave them far too open, and accidentally allow services they had no intention or plan to expose. Managing a firewall in a modest environment can, very quickly, become a nightmare of argument about what is and is not denied, and configuration changes can be extremely awkward for undocumented workflow requirements. The NAT default behavior is tremendously helpful against this.
It wasn't "insider trading", It was a "free to attend" presentation of a student's work that was being presented to potential investors. No NDA was signed, and the paper was publicly available.
Slashdot Mirror
Airplanes do not normally float in a medium of highly corrosive saltwater. And they can usually go _around_ storms, or land when the weather gets excessively dangerous. And even good quality, powerful engines need regular hand-on maintenance over the course of a 30 day trip.
> You opt for "married men make better performers"
Please: I didn't say that. It affects the role. For example, married men may be unwilling to perform 80 hour work weeks at crunch time. Being older, and willing to object to abusive contracts or work hours, certainly affects _my_ work performance. And my age is correlated with my work experience.
What I'm saying is that it's unrealistic, and unreasonable, to say these factors don't _affect_ work performance. In soome cases, they may _improve_ work performance! It depends on the situation and the role.
My point is that can it can _affect_ job performance, just as age, gender, religion, nationality, size, weight, skin color, and marital status can do so. The effects can be modest, and are often legislatively barred from affecting hiring. But to pretend that they are not factors, or that they do not alter people's chances of behing hired or affect evaluations of performance is disingenuous.
And let's be clear: many people consider "icky power differentials" to be attractive, both for the more powerful person to enjoy the power, and for the less powerful person to enjoy the attention of a more successful or powerful person. It can be wildly abused, but it's been common practice throughout human history. Ignoring it won't prevent it.
> Religion co-opted marriage and turned it into a religious thing millennia ago
Religions have been involved in major life events throughout the history of civilization. Can you point to any history or culture in which marriage, like birth and death, was +not+ treated as a religious matter? I suspect that any culture which handled marriage as a purely civil issue would integrate religion into it very, very quickly.
Actually, like being white, male, young, married, Christian, or born in the USA, it can certainly affect job performance. Being a member of the "power" ethnic or social jobs can certainly affect political power and perceived acceptability a social and political effect. But most relevant for job performance is the likelihood of being married, and having children, which also affect work availability and available weekly work hours. Then look at the number of business and political leaders who do _not_ have a spouse who takes care of their home and their children. And do _not_ be shocked that the possibility of dating, of or sexual interaction, effects the work place. And anytime there's a possible sexual interaction with a superior or a subordinate, it effects the work relationship.
That said, I'm delighted that we're more accepting of different lifestyles than we were at the start of my career. Let's just not ignore the direct effects of marriage, and of children, on work performance because they remain quite profound. Schedule a major product release for Christmas, Valentine's Day, or for graduation day and see who can do the work if you think it's irrelevant.
There is a great deal of science, and public policy, that would benefit from public exposure. But medical and sociological research benefits from the privacy of the subject, who then feel more free to be truthful. The same is true of political survey data, and "anonymizing" it can be a lengthy, expensive, and uncertain process, especially when coupled with various metadata that is being collected with the experiments or in parallel with it. It can also be very expensive to make public, even without privacy issues, because transforming it from obsolete media and making it available for public download often takes real engineering time. Long term science projects can span decades, and the first sets of data are often on obsolete media.
Overall, it seems an excellent policy, but exceptions will have to be made.
Those can be very useful for quick strikes. As long as you're not actually occupying foreign soil, your point is well mad But wars of occupatio0n take manpower, as demonstrated in Korea, Vietnam, Afghanistan, and the second Iraq war. And don't be mistaken, the USA didn't "win" any of those. In Iraq and Afghanistan now, the US is "declaring victory" and leaving a mess that is, in some ways, worse than when the wars started. Sadam Hussein, as much of a genocidal dictator as he was, didn't allow the Taliban to operate within Iraq. Now they're an integral part of Iraq politics.
Those are problems that air strikes and drone strikes don't solve, they exacerbate.
They want to reduce _personnel_. Personnel have to be trained, and are known to sometimes refuse illegal orders, or worse, to send home pictures of illegal orders.
I did not refer to the whole operating system. I referred specifically to the _kernel_. Do go read http://windowsitpro.com/window..., it's one of the better publicly available analyses I can find.
I'm afraid I don't have copies of both source trees for comparison, and would have legal difficulties if I did post them. DEC and Microsoft settled out of court on this: do look for keywords "Microsoft" and "Cutler" for more detailed legal and software analysis.
> Since Windows 2000 came out the stability has improved immensely
That's because much of that kernel was VMS, stolen from DEC. Thee's a readable description of the wholesale copying at http://windowsitpro.com/window....
s a business tactic, Microsoft's behaviors have been often very effective. I'd refrain from mentioning older thefts by Microsoft except these are at the core of what they sell now: The NT kernel, at the core of Windows 8 and Windows Server, was extensively VMS code stolen from DEC, and DEC bent bankrupt after that. The browser standards wars continue to include "embrace, extend, and break compatibility". The entire "OOXML" debacle of "publishing open standards" for Microsoft Office document standards, then ignoring them for actual MS Office software is an ongoing example. Microsoft Office violates its own standards, and the standards were themselves corrupted, to allow Microsoft to claim "open standards" compatibility which it doesn't actually have.
And then there's "Trusted Computing". The entire ongoing project is not aimed at user privacy: it's aimed at vendor lockin for software, data, and even hardware. And the private keys, including keys to revoke other keys, are held almost entirely in escrow by Microsoft, with no usable guarantees of the keys protection from wholesale abuse.
As does gcc with the right flags, which detect "unreachable code". That can often generate a storm of warnings that require a great deal of additional time to sort through and resolve, and can be different depending on compile time definitions. So it's not often used.
> Long term acceleration on the highway? Nobody thought to just turn off the ignition switch? Turns off the fuel pump, care stops running, eventually you come to full stop
I know of at least one such case, http://www.theguardian.com/wor.... The car was modified for a disabled man, so it's not clear what the combination of controls he could manage was. But he wound up rocketing across France into Belgium, traveling over 100 MPH with quick acting police clearing traffic ahead of him.
And because a software error or "bit error" can cause the software to be stuck in the recursion, in an infinite loop. Even when software errors don't exist, "bit flips" due to electrical transients or even due to cosmic radiation are inevitable due to the quite small transistor size in modern processors. And cars are electrically quite noisy environments, so a certain amount of "bit flip" should be expected as normal for any such system.
Thank you for the corruption: my eyes are apparently still bleeding from having seen the beta!
I meant to write "light weaponry".
> drone has yet to be involved in such an attack, except as an agent of state force and under color of something vaguely resembling 'law'.
I wasn't personally thinking of well focused assassination, but rather of attacks like the Boston Bombing. Even indiscriminate manslaughter among such large crowds has enormous political force, and the Olympics focus worldwide attention on a crowd aimed specially at international cooperation. I can vividly remember the Olympic hostage crisis in 1972: It focused worldwide attention on the politics of the terrorists who committed those murders.
The automatic recover features are worth mentioning. It reduces the risk of danger due to human error, or signal loss: those are inevitable in a crowded environment with thousands of cameras and radio transmitters saturating the airwaves legally and illegally.
My concern would be about the ability to hijack the drones, and about getting the crowds used to low flying remote controlled aircraft. Similar, unauthorized dromes could be reloaded with light and bring it near the crowds or athletes from well outside of any reasonable security perimeter.
It's become much more stable in the last 10 years, in my observed experience. My colleagues and I have seen issues activating downloadable drivers for Windows clients to use various unusual printers, and the client selection of single sided and double sided printing has been awkward. But we've normally simply set up one queue for single sided, the other for double sided, and that's worked well. And if a particular print driver isn't published in the available CUPS or other GUI configurations, we've had good success using old Apple Laserwriter Postscript drivers with modern printers.
Much of the technological problem for CUPS in complex environments has not been CUPS, itself, it's been the available drivers and limitations of the underlying Ghostscript technology that does much of the transformation to printable document format. It's been very difficult to keep up to date on all the subtly modified and "enhanced" features of new printers, and provide user access to those for CUPS and ghostscript. In my observation, ghostscript has gotten _much_ better at successfully providing the necessary tools. And some GUI's for print management have improved, as well. The most recent RHEL system-config-printer has gotten quite good: they seem to have read Eric Raymond's essay and taken it to heart.
The link is to Eric Raymond's "The Luxury of Ignorance: A CUPS Horror Story". It's just what I I point people to when they deal with Windows 8, the new Ubuntu interface, Gnome 3, and the new Fedora installer. Its follow up article, "The Luxury of Ignorance: Part Deux" is at:
http://www.catb.org/esr/writin...
The followup shows that the authors of CUPS, much like Mark Shuttleworth in his first responses to the nearly universal dislike for the new Ubuntu interface, showed the same response as CUPS authors made to Eric's complaints. I'm also afraid that the built-in CUPS configuration tool as not improved in any appreciable way since Eric's original essay. Fortunately, many OS developers have written their own and far superior wrappers for configuring CUPS. But such technically sophisticated but incomplete and ignorant of basic workflow tools still abound. Cleaning up after the chaos they leave when they overwrite hand-edited system configurations and disable critical features actually pays a considerable amount of my wages.
They're going to find the _exposed_ ports of deliberately enabled sercices, anyway. Script kiddie port scanning is a popular pasttime, with zombied and rootkitted hosts all over the world doing it as a matter of course for their entire accessible VLAN's. The key underlying advantage is that most of the ports _aren't_ exposed. Do take a look at the VLAN our own ISP, or your own office network uses, for exposed ports. Unless your firewalls are very scrupulously maintained, they will have exposed services directly exposed because they don't have NAT, and they don't have good firewalls set up.
It's the not deliberately exposed ports and services that NAT is such an easy and useful first line of defense for. Old, never updated printers and routers, laptops and desktops and servers wehre the default password is still in place, Nagios servers with default "nagiosadmin" user and default "nagiosadmin" password that allow intruders detailed overviewss of your infrastructure: NFS shares and CIFS shares set up without authentication, which can be very difficult to get people to turn off, NTP services that have never been properly filtered, etc, etc.
May I presume that your account with them is quite old? Or that you are paying them extra for such an IP address? It really has fallen out of favor, partly due to IPv4 limitations, an dpartly to idscourage customers from hosting conntent from their home machines.
That is an excellent point, and I was negligent to not review which is the RFC specified VLAN for such connecitons.
I';m afraid, however, that it's not the case where I'm sitting right now nor in a lot of the less skilled ISP's I've had to deal with personally and with clients over the years.
NAT is not a firewall. But by reducing the exposed attack vector, and helping keep traffic from coming _in_ to your network that is not specifically permitted, it's a tremendous beneficial in reducing the constant clamor of script kiddies hammering on your network, hoping for an open router port or a misconfigured firewall. It also discourages careless security, such as running SSH on every single machine on a network, exposed to the Internet, or allowing unencrypted HTTP and FTP to arbitrary hosts inside your network when those can be packet sniffed for passwords on the incoming connection.
Basically, for something behind a NAT to support an externally exposed service such as DNS, SMTP, NTP, FTP, HTTP, NFS, CIFS, or SSH, you have to _select_ to allow it. For various firewall configurations, I'm afraid that too many people leave them far too open, and accidentally allow services they had no intention or plan to expose. Managing a firewall in a modest environment can, very quickly, become a nightmare of argument about what is and is not denied, and configuration changes can be extremely awkward for undocumented workflow requirements. The NAT default behavior is tremendously helpful against this.
It wasn't "insider trading", It was a "free to attend" presentation of a student's work that was being presented to potential investors. No NDA was signed, and the paper was publicly available.