How does competing "AGAINST a charity" hurt that charity? Are you suggesting that there can be only one charity in any given field, and that any field with competing charities is necessarily worse off? Not to mention that any "charity" that has any non-volunteer staff isn't really a charity -- they may do charitable work, but they've in business just like anyone else.
I know people don't always pick the product best suited for their needs, but I don't see how offering two choices for computing platforms hurts anyone. If their laptop really is worse, then they won't sell many, and the OLPC program will be largely unaffected. And there may well be situations where Intel's laptop is better than the OLPC offering, even if you don't immediately foresee them -- it's hard for me to believe that the OLPC machine is the only one that could be useful for people who currently don't have a computer.
I really hope you're not buying CDs because they are "lossless" -- they aren't. Their encoding is uncompressed, but the very act of digital sampling is lossy. In particular it's lossy if you have to re-sample from recordings at 48kHz or 96kHz to CDs at 41.1kHz, as it's a non-integer conversion. There's no information loss as a result of sampling->encoding changes on a CD, but there is a loss with the sampling conversion, and that doesn't need to happen with with other music formats.
I'm denying that there are benefits to owning physical disks, or that it's possible to get higher quality audio from a CD than from a download, but you mean "higher quality" not "lossless", and you only mean "higher quality" if the format you download is less than 41.1 kHz, 16-bit.
More commonly I've seen that they obtain access to a merchant account an process ~$10 transactions themselves. THe hope is that they can use the merchant account for a couple of months before people notice -- a $10 transaction doesn't call much attention unless you really do accounting -- and then when they lose access to their merchant account they move on to another.
This can be done either by obtaining merchant accounts directly (not as difficult or traceable as you might think) or just convincing the clerk at any store with a valid account to process a bunch of bogus transactions and pay them out from the till.
Web servers are intended for the dissemination of information to third parties. Wi-fi gateways are basic infrastructure, and can be reasonably be considered intended for the use of authorized parties only, given most people are unlikely to want anonymous third parties using their network without permission.
First, I don't understand how you can say with a straight face that web servers indicate and intent to share and broadcasting gateways do not. For one thing, most web services specifically forbid all sorts of uses in their terms of use. For example: http://www.ostg.com/terms.htm. For another, many people assume that information they post on the web is private (they shouldn't, but they do) if it isn't linked in to a well-known web page.
Second, couldn't you make the same argument about not wanting anonymous third parties using email servers? I only want authorized people to email me, not Mexican pharmacy bots. But since I have a publicly accessible email address I'm likely to get some such email, whether I want it or not.
I don't support the "it's open so I have a right to use it" viewpoint, but there are reasonable technical and social measures that could be employed to indicate that you don't want to share your gateway. They could simply add a the WEP password "pasword". Or not broadcast the SSID. Or put up a terms-of-use page for the first port-80 request from a new MAC address. Or posted a sign outside that said "WiFi access for customers only". Or they could have walked out to his car and asked him to stop using their network.
If they had done any one of those things and he continued using the access point I'd have no trouble prosecuting him. But when people physically trespass, they must have bypassed reasonable security measures and/or be asked to leave before they've committed a crime, and I don't see the benefit of a stricter standard for access points.
He's making a distinction between giving priority based on type and giving priority based on source. The former lets you give interactive traffic priority over bulk traffic, the later lets you give priority to your partner's interactive traffic vs. your competitor's interactive traffic.
General priority-based processing just says that all packets are buffered for at least X ms, and if during that time more traffic arrives than can be de-queued, certain packets are preferred when de-queuing. This is incredibly handy for things like keeping your VoIP calls working even when you're downloading big files -- you can prefer SIP packets over HTTP packets when de-queuing. The great part about this plan is that it has almost no effect when you don't have any high-priority traffic; when you aren't routing any SIP packets the entire line can be used for HTTP.
It's also possible to use the same method to do fixed rate limiting -- over any X ms only allow the first Y packets of a through and drop the rest. This could be used to limit your overall transfer rate, or could be applied to specific types/sources/etc. of traffic to do things like limit HTTP to be no more than 50% of your overall connection rate.
To the best of my knowledge, no one does anything to intentionally increase the delay of packets they route. It wouldn't be incredibly hard to do from a technical perspective, assuming you had equipment with sufficient buffer space, but I don't know what problem it would solve. Increasing the transit delay doesn't necessarily mean that you're using less bandwidth, it just means that there's more data in-flight at any given time. It would make short TCP transfers slower, but the overall transfer rate of long-running TCP connections and any sort of UDP traffic would be basically unaffected by an increase in transit time. From an economic standpoint I can only see this being a benefit if you have competitors with interactive traffic, and you want to specifically prevent their services from working well over lines you control; it wouldn't do much good for ISPs just trying to reduce their bandwidth use.
There's a whole range of microwave frequencies that are absorbed by water molecules. We picked ~2.4 GHz for home appliances because it offers a good balance of penetration vs. absorption and because it's relatively cheap to produce and to shield. But water absorbs radiation at other wavelengths as well; IIRC 900 nm and 1200 nm are absorption peaks, and there's a whole range of other wavelengths with varying degrees of absorption. We did choose 2.4 GHz for WiFi just because it's unlicensed, but we didn't choose the "most dangerous" frequency with respect to absorption, just one that happens to coincide with home appliances.
What is not known is: how much absorption of that radiation is bad for the kids?
That's not as unknown as you might think. Since we're talking about non-ionizing radiation here, "absorption" is the same as "heating", and "How much heating is bad for kids?" is a question we've studied for hundreds of years, at least informally. People ascribe magically properties to "radiation" even though we know from actual testing that the absorption of non-ionzing radition results either in heating or the re-transmission of long-band EM radition. Heating is something we've regularly experienced as humans, long before we discovered radio, and 2.4 GHz is too low a freqency for you to be emitting long-band EM radiation.
No, no no. Neighboring states would never build inter-connecting roads if they didn't get money from the federal government. Because it's not like anyone in the individual states would have an economic interest in continuing to fund the highway system.
Seriously, I wish states would just suck it up and do without the federal highway funding. But that's only sensible under a system where it's possible to get the federal government to stop taking the money from states in the first place -- the states wouldn't have to "get money" from the federal government if the feds didn't take it away.
No. You will not resample or compress my video and audio data. Maybe you don't care about fuzzy characters on your monitor, but I do, and I'm not going to let you compress the data just so that you can use a familar cable type.
Beside that Ethernet isn't any more friendly. Even if it worked on GigE, you'd still need an isolated network, because A) you need all the bandwidth for the A/V data and B) putting more than once source on the network would require that users setup some sort of addressing on the display to select the right source, and that would never fly -- most people can't even get back to the physical input line, let alone select a source from within that line.
Having had a ~$100 Harmony and being limited by its programming capabilities, I can tell you the H360 doesn't let you do everything you can with an MX-900, at least not without a *lot* of button pushing.
Also, you can get an MX-900 for ~$250, which is in-line with the middle of the Harmony range. The MSRP is ridiculous, but you don't have to buy at the price -- it's intended to allow professional installers give you a 25% discount while still gouging you.
Thanks for the recommendation. I've had a harmony and been moderately happy with it -- it lets anyone use my system and it does correctly switch between modes. But it dosen't let me have any input in what switching between modes means. It doesn't seem so terribly complicated to expose a limited scripting language to let me do some more complicated, customized things than are built in to the configuration system.
Windows-only is a bad time though;-(. I'm sure I can come up with a Windows system for the 2 times a year I need to re-program the thing, but it would be nice if I could at least generate the configuration file in a non-Windows app.
Limited-vocabulary speech recognition has been working -- without training -- for years now. Suitable engines are built-in to OS X and MS Office, and there are several choices for Linux as well. Not all tools provide good programmatic access, so it may or may not be easy to integrate with your favorite tools, but the actual speech recognition part is there.
Yes, young people often act like they know everything. So do some older people. Unfortunately it's an occasional side-effect of being a person. Being arrogant does not imply that a person isn't or cannot do useful work, just that they're annoying to be around.
And you're right, we probably don't need many "Klingon Language" degrees. But even with such a silly degree, it's a bit of a stretch to assume that speaking Klingon is the only skill the student picked up while at school. I'm not pretending that undergraduate studies are a bastion of altruistic academics, but most people that go through a four-year program learn how to teach themselves new things, and how to commit to a multi-year project where the incremental goals are not strongly correlated to the ultimate goal.
Part of the reason they make announcements like this is to let other vendors in on their plans: MS would probably prefer to stop releasing a 32-bit OS and programs, but they need (at least the big) PC makers to stop selling 32-bit systems before that can happen. Similarly PC makers don't want to move to the (at least for now) more expensive 64-bit systems if MS is still offering up-to-date products for their more profitable 32-bit systems.
It's the same deal with ATI -- they'd love to only make one card type, and as soon as they see AGP sales drop off I'm sure they will. But that can't happen until PCIe has been the standard video slot on motherboards for at least a few years. So they announce their intent long before they're prepared to act, in the hopes of encouraging other people to move with them, thereby mitigating the chicken/egg problem.
First, they aren't leeching from your taxes any more than you are for driving on roads in other states -- there are certain things we decided should be publicly funded due to their public benefit. Education is one of those things, and it's probably one of the most important, at least if you believe in self-governance. You may not personally think that education is a worthwhile project to fund, but many people do, and you're not going to convince them otherwise by accusing students of leeching from "your" taxes.
But beyond that, Stanford is not part of the UC system, and is not particularly publicly funded. I'm sure they get some public money, but so do many other institutes, with or without students. For example, road construction companies derive a large amount of their income from public contracts, and very few construction companies enroll non-employee students.
Does your ignorance make you a loud mouth arrogant bastard with your head up your own ass WHO LEECHES FROM MY INTERWEBS, or should I just excuse you as someone that's angry about not going to college?
I think one could argue that the page itself is copyrighted content, and that the ads are part of the content protection mechanism. It's not entirely sensible, but I don't see how it's any more unreasonable than most DMCA claims.
Even if you set sudo to allow password-less access and allow it to launch a shell then yes, sudo can be dangerous. But that's not an entirely common configuration, and used correctly sudo can offer many advantages over a root account. The first few that pop in to my head are:
1. No shared credentials. Personal logins not only let you avoid the hassles and insecurity of a shared password, but also give you accountability when someone does something stupid. 2. Allow only certain commands to be run as root -- you can create a print-admin group with access to LPR-related commands but not full root access 3. Allow commands to be run as other non-root users -- you let users modify files created by print_queue_daemon_user without being root 4. Log each command issued as root (and yes, you can launch a shell, but only if sudo lets you, and even then your shell launching would be logged)
Earth is the center of the universe. If you just say "center" -- and not "center of mass" or some other more specific defintion -- you really just mean the point where you define the coordinate system to be 0,0,0. And since no one know the shape or mass of the universe there really aren't a lot of other "centers" you could define with any accuracy. If you're traveling outside of Earth orbit Earth is probably a bad reference point, but there's no technical reason you couldn't use it.
Likewise the sun does revolve around Earth every bit as much as the moon does, so drawing Earth at the center of the solar system is again just a different frame of reference, not an invalid model. The only part people got wrong was the planets revolving around Earth, and even that belief didn't last for long once observers were able to make accurate measurements to disprove the model.
That withstanding, your point about failing to accept the commonly held beliefs, even if the commonly held beliefs are inaccurate, has traditionally been seen as heresy or the like is perfectly valid.
User-privilege-based security is not intended to protect information a user controlls from processes the same user controlls, and can't even if you don't allow debugging. If you don't want your processes to know anything about each other you shouldn't run them as the same user.
Beside that, running a debugger is also far from the easiest way to get someone's passwords if you have the ability to launch programs.
I don't have a non-PO box mailing address. I do have a physical address if I call the fire department, but I can't get USPS mail there. Somehow I don't think you've though your cunning little plan all the way through.
The same way you stop people from using lights and sirens so they can get through traffic faster -- you make it a crime and enforce that rule against people that are obviously breaking it. If they are influencing traffic in any significant way you could see that effect and it wouldn't be terribly difficult to record the broadcasts in the area and correlate them with the vehicle weaving through traffic.
That's only true if you assume that two distant points share the same timescale -- a relativist might argue that "now, far away" is the same moment in time as "here, long ago", at least baring the discovery of macro-scale faster-than-light causation. That is to say, while an observer near the supernova might have seen the explosion long ago, his "long ago" and your "now" may be the same moment, not just two different perspectives of the same event that happened long ago with respect to all observers.
Beside that, even if there is a universal timescale unrelated to the speed of light, from our perspective it is happening "now", and since we don't often communicate with anyone more than a few thousand miles away it's silly to express things in any other timescale.
This is different from her standing in from of a classroom talking about being hammered because *that* would be a lack of seperation between her professional and personal life. She'd be taking time out of her profession to address her personal life, and that lack of a boundry would be a problem.
But I don't see why it's a problem for a teacher to get drunk from time to time, or why it's a problem that her students may be aware that she gets drunk from time to time. Would it be "even remotely appropriate" for her to publish pictures showing her smoking or is it just drinking that gets you all riled up?
The problem is that it always tries to save in XLS format. Try this: 1. Open a tab-delimited file in MS Excel 2. Edit the text in one cell without creating a formula or adding formatting or other non-text features 3. File->Save It will tell you that you're going to lose data unless you use another format. Or at least that's what it always tells me.
If it worked the way you describe it wouldn't be a big deal, but the way it actually works is a real hassle when you want to use Excel for one stage of an otherwise non-MS workflow.
Slashdot Mirror
How does competing "AGAINST a charity" hurt that charity? Are you suggesting that there can be only one charity in any given field, and that any field with competing charities is necessarily worse off? Not to mention that any "charity" that has any non-volunteer staff isn't really a charity -- they may do charitable work, but they've in business just like anyone else.
I know people don't always pick the product best suited for their needs, but I don't see how offering two choices for computing platforms hurts anyone. If their laptop really is worse, then they won't sell many, and the OLPC program will be largely unaffected. And there may well be situations where Intel's laptop is better than the OLPC offering, even if you don't immediately foresee them -- it's hard for me to believe that the OLPC machine is the only one that could be useful for people who currently don't have a computer.
I really hope you're not buying CDs because they are "lossless" -- they aren't. Their encoding is uncompressed, but the very act of digital sampling is lossy. In particular it's lossy if you have to re-sample from recordings at 48kHz or 96kHz to CDs at 41.1kHz, as it's a non-integer conversion. There's no information loss as a result of sampling->encoding changes on a CD, but there is a loss with the sampling conversion, and that doesn't need to happen with with other music formats.
I'm denying that there are benefits to owning physical disks, or that it's possible to get higher quality audio from a CD than from a download, but you mean "higher quality" not "lossless", and you only mean "higher quality" if the format you download is less than 41.1 kHz, 16-bit.
More commonly I've seen that they obtain access to a merchant account an process ~$10 transactions themselves. THe hope is that they can use the merchant account for a couple of months before people notice -- a $10 transaction doesn't call much attention unless you really do accounting -- and then when they lose access to their merchant account they move on to another.
This can be done either by obtaining merchant accounts directly (not as difficult or traceable as you might think) or just convincing the clerk at any store with a valid account to process a bunch of bogus transactions and pay them out from the till.
Web servers are intended for the dissemination of information to third parties. Wi-fi gateways are basic infrastructure, and can be reasonably be considered intended for the use of authorized parties only, given most people are unlikely to want anonymous third parties using their network without permission.
First, I don't understand how you can say with a straight face that web servers indicate and intent to share and broadcasting gateways do not. For one thing, most web services specifically forbid all sorts of uses in their terms of use. For example: http://www.ostg.com/terms.htm. For another, many people assume that information they post on the web is private (they shouldn't, but they do) if it isn't linked in to a well-known web page.
Second, couldn't you make the same argument about not wanting anonymous third parties using email servers? I only want authorized people to email me, not Mexican pharmacy bots. But since I have a publicly accessible email address I'm likely to get some such email, whether I want it or not.
I don't support the "it's open so I have a right to use it" viewpoint, but there are reasonable technical and social measures that could be employed to indicate that you don't want to share your gateway. They could simply add a the WEP password "pasword". Or not broadcast the SSID. Or put up a terms-of-use page for the first port-80 request from a new MAC address. Or posted a sign outside that said "WiFi access for customers only". Or they could have walked out to his car and asked him to stop using their network.
If they had done any one of those things and he continued using the access point I'd have no trouble prosecuting him. But when people physically trespass, they must have bypassed reasonable security measures and/or be asked to leave before they've committed a crime, and I don't see the benefit of a stricter standard for access points.
He's making a distinction between giving priority based on type and giving priority based on source. The former lets you give interactive traffic priority over bulk traffic, the later lets you give priority to your partner's interactive traffic vs. your competitor's interactive traffic.
General priority-based processing just says that all packets are buffered for at least X ms, and if during that time more traffic arrives than can be de-queued, certain packets are preferred when de-queuing. This is incredibly handy for things like keeping your VoIP calls working even when you're downloading big files -- you can prefer SIP packets over HTTP packets when de-queuing. The great part about this plan is that it has almost no effect when you don't have any high-priority traffic; when you aren't routing any SIP packets the entire line can be used for HTTP.
It's also possible to use the same method to do fixed rate limiting -- over any X ms only allow the first Y packets of a through and drop the rest. This could be used to limit your overall transfer rate, or could be applied to specific types/sources/etc. of traffic to do things like limit HTTP to be no more than 50% of your overall connection rate.
To the best of my knowledge, no one does anything to intentionally increase the delay of packets they route. It wouldn't be incredibly hard to do from a technical perspective, assuming you had equipment with sufficient buffer space, but I don't know what problem it would solve. Increasing the transit delay doesn't necessarily mean that you're using less bandwidth, it just means that there's more data in-flight at any given time. It would make short TCP transfers slower, but the overall transfer rate of long-running TCP connections and any sort of UDP traffic would be basically unaffected by an increase in transit time. From an economic standpoint I can only see this being a benefit if you have competitors with interactive traffic, and you want to specifically prevent their services from working well over lines you control; it wouldn't do much good for ISPs just trying to reduce their bandwidth use.
There's a whole range of microwave frequencies that are absorbed by water molecules. We picked ~2.4 GHz for home appliances because it offers a good balance of penetration vs. absorption and because it's relatively cheap to produce and to shield. But water absorbs radiation at other wavelengths as well; IIRC 900 nm and 1200 nm are absorption peaks, and there's a whole range of other wavelengths with varying degrees of absorption. We did choose 2.4 GHz for WiFi just because it's unlicensed, but we didn't choose the "most dangerous" frequency with respect to absorption, just one that happens to coincide with home appliances.
What is not known is: how much absorption of that radiation is bad for the kids?
That's not as unknown as you might think. Since we're talking about non-ionizing radiation here, "absorption" is the same as "heating", and "How much heating is bad for kids?" is a question we've studied for hundreds of years, at least informally. People ascribe magically properties to "radiation" even though we know from actual testing that the absorption of non-ionzing radition results either in heating or the re-transmission of long-band EM radition. Heating is something we've regularly experienced as humans, long before we discovered radio, and 2.4 GHz is too low a freqency for you to be emitting long-band EM radiation.
No, no no. Neighboring states would never build inter-connecting roads if they didn't get money from the federal government. Because it's not like anyone in the individual states would have an economic interest in continuing to fund the highway system.
Seriously, I wish states would just suck it up and do without the federal highway funding. But that's only sensible under a system where it's possible to get the federal government to stop taking the money from states in the first place -- the states wouldn't have to "get money" from the federal government if the feds didn't take it away.
Makes me wonder about our next generation. It really does.
Yeah, kids these days. With their new-fangled gadgets and loose morales. There's just nothing you can do to keep them off your lawn.
Seriously, stop wondering about the next generation and own up to the fact that you were retarded when you were 14 too.
No. You will not resample or compress my video and audio data. Maybe you don't care about fuzzy characters on your monitor, but I do, and I'm not going to let you compress the data just so that you can use a familar cable type.
Beside that Ethernet isn't any more friendly. Even if it worked on GigE, you'd still need an isolated network, because A) you need all the bandwidth for the A/V data and B) putting more than once source on the network would require that users setup some sort of addressing on the display to select the right source, and that would never fly -- most people can't even get back to the physical input line, let alone select a source from within that line.
Having had a ~$100 Harmony and being limited by its programming capabilities, I can tell you the H360 doesn't let you do everything you can with an MX-900, at least not without a *lot* of button pushing.
Also, you can get an MX-900 for ~$250, which is in-line with the middle of the Harmony range. The MSRP is ridiculous, but you don't have to buy at the price -- it's intended to allow professional installers give you a 25% discount while still gouging you.
Thanks for the recommendation. I've had a harmony and been moderately happy with it -- it lets anyone use my system and it does correctly switch between modes. But it dosen't let me have any input in what switching between modes means. It doesn't seem so terribly complicated to expose a limited scripting language to let me do some more complicated, customized things than are built in to the configuration system.
;-(. I'm sure I can come up with a Windows system for the 2 times a year I need to re-program the thing, but it would be nice if I could at least generate the configuration file in a non-Windows app.
Windows-only is a bad time though
Limited-vocabulary speech recognition has been working -- without training -- for years now. Suitable engines are built-in to OS X and MS Office, and there are several choices for Linux as well. Not all tools provide good programmatic access, so it may or may not be easy to integrate with your favorite tools, but the actual speech recognition part is there.
Yes, young people often act like they know everything. So do some older people. Unfortunately it's an occasional side-effect of being a person. Being arrogant does not imply that a person isn't or cannot do useful work, just that they're annoying to be around.
And you're right, we probably don't need many "Klingon Language" degrees. But even with such a silly degree, it's a bit of a stretch to assume that speaking Klingon is the only skill the student picked up while at school. I'm not pretending that undergraduate studies are a bastion of altruistic academics, but most people that go through a four-year program learn how to teach themselves new things, and how to commit to a multi-year project where the incremental goals are not strongly correlated to the ultimate goal.
Part of the reason they make announcements like this is to let other vendors in on their plans: MS would probably prefer to stop releasing a 32-bit OS and programs, but they need (at least the big) PC makers to stop selling 32-bit systems before that can happen. Similarly PC makers don't want to move to the (at least for now) more expensive 64-bit systems if MS is still offering up-to-date products for their more profitable 32-bit systems.
It's the same deal with ATI -- they'd love to only make one card type, and as soon as they see AGP sales drop off I'm sure they will. But that can't happen until PCIe has been the standard video slot on motherboards for at least a few years. So they announce their intent long before they're prepared to act, in the hopes of encouraging other people to move with them, thereby mitigating the chicken/egg problem.
First, they aren't leeching from your taxes any more than you are for driving on roads in other states -- there are certain things we decided should be publicly funded due to their public benefit. Education is one of those things, and it's probably one of the most important, at least if you believe in self-governance. You may not personally think that education is a worthwhile project to fund, but many people do, and you're not going to convince them otherwise by accusing students of leeching from "your" taxes.
But beyond that, Stanford is not part of the UC system, and is not particularly publicly funded. I'm sure they get some public money, but so do many other institutes, with or without students. For example, road construction companies derive a large amount of their income from public contracts, and very few construction companies enroll non-employee students.
Does your ignorance make you a loud mouth arrogant bastard with your head up your own ass WHO LEECHES FROM MY INTERWEBS, or should I just excuse you as someone that's angry about not going to college?
I think one could argue that the page itself is copyrighted content, and that the ads are part of the content protection mechanism. It's not entirely sensible, but I don't see how it's any more unreasonable than most DMCA claims.
Even if you set sudo to allow password-less access and allow it to launch a shell then yes, sudo can be dangerous. But that's not an entirely common configuration, and used correctly sudo can offer many advantages over a root account. The first few that pop in to my head are:
1. No shared credentials. Personal logins not only let you avoid the hassles and insecurity of a shared password, but also give you accountability when someone does something stupid.
2. Allow only certain commands to be run as root -- you can create a print-admin group with access to LPR-related commands but not full root access
3. Allow commands to be run as other non-root users -- you let users modify files created by print_queue_daemon_user without being root
4. Log each command issued as root (and yes, you can launch a shell, but only if sudo lets you, and even then your shell launching would be logged)
Earth is the center of the universe. If you just say "center" -- and not "center of mass" or some other more specific defintion -- you really just mean the point where you define the coordinate system to be 0,0,0. And since no one know the shape or mass of the universe there really aren't a lot of other "centers" you could define with any accuracy. If you're traveling outside of Earth orbit Earth is probably a bad reference point, but there's no technical reason you couldn't use it.
Likewise the sun does revolve around Earth every bit as much as the moon does, so drawing Earth at the center of the solar system is again just a different frame of reference, not an invalid model. The only part people got wrong was the planets revolving around Earth, and even that belief didn't last for long once observers were able to make accurate measurements to disprove the model.
That withstanding, your point about failing to accept the commonly held beliefs, even if the commonly held beliefs are inaccurate, has traditionally been seen as heresy or the like is perfectly valid.
User-privilege-based security is not intended to protect information a user controlls from processes the same user controlls, and can't even if you don't allow debugging. If you don't want your processes to know anything about each other you shouldn't run them as the same user.
Beside that, running a debugger is also far from the easiest way to get someone's passwords if you have the ability to launch programs.
I don't have a non-PO box mailing address. I do have a physical address if I call the fire department, but I can't get USPS mail there. Somehow I don't think you've though your cunning little plan all the way through.
The same way you stop people from using lights and sirens so they can get through traffic faster -- you make it a crime and enforce that rule against people that are obviously breaking it. If they are influencing traffic in any significant way you could see that effect and it wouldn't be terribly difficult to record the broadcasts in the area and correlate them with the vehicle weaving through traffic.
That's only true if you assume that two distant points share the same timescale -- a relativist might argue that "now, far away" is the same moment in time as "here, long ago", at least baring the discovery of macro-scale faster-than-light causation. That is to say, while an observer near the supernova might have seen the explosion long ago, his "long ago" and your "now" may be the same moment, not just two different perspectives of the same event that happened long ago with respect to all observers.
Beside that, even if there is a universal timescale unrelated to the speed of light, from our perspective it is happening "now", and since we don't often communicate with anyone more than a few thousand miles away it's silly to express things in any other timescale.
This is different from her standing in from of a classroom talking about being hammered because *that* would be a lack of seperation between her professional and personal life. She'd be taking time out of her profession to address her personal life, and that lack of a boundry would be a problem.
But I don't see why it's a problem for a teacher to get drunk from time to time, or why it's a problem that her students may be aware that she gets drunk from time to time. Would it be "even remotely appropriate" for her to publish pictures showing her smoking or is it just drinking that gets you all riled up?
The problem is that it always tries to save in XLS format. Try this:
1. Open a tab-delimited file in MS Excel
2. Edit the text in one cell without creating a formula or adding formatting or other non-text features
3. File->Save
It will tell you that you're going to lose data unless you use another format. Or at least that's what it always tells me.
If it worked the way you describe it wouldn't be a big deal, but the way it actually works is a real hassle when you want to use Excel for one stage of an otherwise non-MS workflow.
Actually it's legal to say it period, because it's completely subjective unless you provide an objective definition for "running experience".
http://en.wikipedia.org/wiki/Puffery