Third parties drivers aren't microsoft windows hardware support they are third party support of windows despite microsoft. You won't see microsoft taking responsibility if those drivers don't work will you? Of course not, if they don't take the responsibility they certainly aren't entitled to the credit if the drivers do work.
"When I said it can't be both - I meant that both of the above can't be true. You can buy any PC - even one preloaded with Linux and there is zero doubt in my mind that Windows will be able to run on that hardware."
Both can be true. I've never seen a non-preloaded windows system where windows supported all the hardware. In every case full hardware support required downloading third party drivers. Ubuntu may or may not support the hardware but if it is going to work at all, it most likely worked out of the box with no additional configuration or third party downloads required. In the few cases where they are needed the system uses detects it and prompts you to download them.
The difference might not be especially troublesome for you today but it will be when that hardware is a few years old. For instance I guarantee when many windows users "upgrade" to vista aka windows 7 their perfectly functional printers/scanners/multi-functions/digital cameras/web cams that are a few years old will have to be replaced to accommodate the upgrade. Ubuntu will continue to support nearly every piece of hardware it supported with the last release on into the future until some compelling TECHNICAL reason makes it infeasible.
If they don't want their data to be an ISP's hostage they do. And if they don't want their inbox searched by the FBI without warrants.
Of course the anti-spam effort have pretty much outlawed private mail servers so downloading/deleting the mail as fast as you can is about all you have left.
"Wardrivers and dictionary attacks are certainly not limited to specific systems."
And yet people with the knowledge and inclination to perform those attacks represent an extremely tiny segment of the population. Most IT professionals wouldn't know how to perform the attacks you are referring to.
"choosing a bad password, choosing the wrong protocol or letting your neighbours near your WiFi router, choosing the wrong brand of router"
The chance of your neighbors consistently being able to figure out why their desktop won't power up when the strip is turned off are slim let alone knowing what to do with access to your router or knowing what a router is. Hell in most cases your neighbor probably has a router and doesn't know what a router is.
"But wireless network encryption is only capable of protecting against someone who doesn't already have physical access anyway."
Your son/daughter/roommate is far more than likely to be trying to get into your wifi than any sort of thief and physical access doesn't help the layman one bit.
Even if they were technically capable to take advantage of physical access (you do realize that an extremely small fraction of the population has that level of technical capability right?) that isn't going to do them much good unless they don't mind you knowing your network was tampered with.
Wifi doesn't exist for corporate networks that need to prevent access to data. The bulk of wifi networks aren't protecting sensitive data at all, the protection is to prevent people from accessing your internet connection.
Only here in geekland do we even consider obscure scenerios like actual thought out and sophisticated attacks.
"Without that, you never know if you're being middled, and therefore you might as well not have encryption."
In theory yes. In practice no. Yes theoretically speaking I can't trust anything because I can't trust the cert authority.
I used to admin mail servers. There is nothing stopping bob the night backup babysitter from reading your plain text mail all night long. Its a hell of a lot harder for him to manage to compromise a mail with a man in the middle attack.
In practice I am far more secure because joe was joe yesterday and had the same cert. Because I can call joe and ask if his cert changed if he got the mail.
And because my ISP and carnivore can't simply sniff the traffic going over the wires and read the mail I'm sending. They actually have to successfully manage a man-in-the-middle attack that tricks me.
"I use it and simply to keep neighbors off my budget connection."
That is what security pros ignore with their ridiculous security recommendations and stringent long and complicated passwords you have to write down.
Aside from your neighbors the most likely person to try to hack your wifi is your 12 year old girl trying to sext when she should in bed. Neither of them is likely to be able to gain access without having to account for it if you use a simple password. The minute you have to stick something to the router though, baby girl is sending me jail bait titty shots.
The mere existence of superior technology does not make the older technology obsolete.
Ultimately there is a critical mass of usage that must be achieved by the new technology to call it obsolete. For instance, cars are not obsolete despite the existence of rocket packs, leer jets, and flying cars.
Without question the vast majority of wifi networks are NOT running WPA2/AES. I don't know how many 9's you need to throw in but its definitely less than one percent. I would be surprised if half the wifi networks out there could even run WPA2/AES without at least replacing hardware (and that isn't even counting the fact the no shortage of equipment rated for WPA2/AES doesn't actually reliably run in that mode).
Nobody replaces their router/wireless nics just because there are a couple hundred or even a thousand people in the country with the tech savy and equipment to potentially mooch their wifi. Those who have the new equipment are just incidental to other upgrades.
Even among those who have the hardware and would be willing they aren't willing to have wifi that won't work when their niece comes to visit with a two or three year old laptop.
WDS can support WEP, WPA, and WPA2 Personal (TKIP+AES) as of v24sp1 (and earlier?). [tested with wrt150n1.1 AP + wrt54g6 R] (TODO: test other hardware). R/RB [repeater/repeater bridge] support all encryption algorithms.
The most likely people to try to break into the router are people who have physical access and little technical knowledge. Like kids and employees.
The second most likely group is random people who are close by and want to mooch some free internet but they on average are less tech savy than your kids and any form of security will keep them out.
People lose perspective in computer security. Big time.
"And I would recommend to write down the password and put it in a drawer. Chances are that you only need to type it in after your system went fubar, and if that happens, you may have lost your password. Drawers are also very difficult to hack from the internet."
and yet, they are ever so easy to hack from inside the room.
You are dramatically more secure using a dictionary word password of less than ten characters, without mixed case, numbers or special characters than using a long and complex password that you have to write down somewhere to remember.
The people who are most likely to try to break into your internet are people you know and especially people you live and/or work with.
As a security professional you may encounter people being attacked by wardrivers and dictionary attacks all the time but as an individual you are probably more likely to be struck by lightning than have someone more sophisticated than your average random idiot looking for an open wifi link to check his email try to get in.
As a freelance network technician I've encountered ONE count them, ONE server running any operating system that has actually been deliberately hacked in ten years on the job. Despite having proof on the box my co-workers were skeptical because despite each having between 20 and 40 years of experience administering small business systems none of them had ever seen a hacked box.
That of course doesn't count worms, spyware, virus, and other automated attacks; open relays; or the kids/underling employees figuring out how to get around the proxy server.
The only thing that will make e-mail encryption take hold is an advance in the technology or at least the clients.
When using e-mail encryption is as simple as checking a box (or better yet, enabled by default) and the key generation, registration, retrieval, etc are completely automatic it might catch on.
So far I've yet to see a client that does this in a successful and consistent way.
Limiting division by telomere length kills off healthy and undamaged cells preventing them from dividing. Removing this "feature" while introducing a replacement mechanism to protect from cancer would leave only damaged cells unable to reproduce.
All else being equal that should increase the ratio of undamaged cells to damaged cells.
I really don't see how anyone could deny that would be a good thing.
"Once the telomeres are eliminated, the ends of chromosomes are treated as double strand breaks by genetic repair machinery which more often than not, results in apoptosis, chromosome joining and ultimately cancer in some cases."
So the way in which p16 is believed to protect cells from cancer does not conflict with turning on telomerase and turning on telomerase would prevent this kind of damage causing at least some impact on aging and cell damage.
"Much of aging has to do with genetic damage"
Wouldn't immortalizing healthy undamaged cells counteract this to some extent?
If nothing else it would seem as if this may at least be part of the puzzle along with increased antioxidants, sunscreen, and viagra.
"If we were really patient, we could knock out p16 in these moles and see if they get cancer. That would pretty well establish whether or not it was just p16 that was responsible for the relative resistance to cancer."
It would establish if p16 is is an essential part of their cancer resistance but if p16 is working in combination with something else your experiment wouldn't reveal it.
The next step would be to then splice this gene into mice and activate it and see if the gene alone is enough to make it cancer resistant. That step will conclusively establish if something else is responsible (or an additional something) or if the gene alone is enough.
Slashdot Mirror
With no third party software/drivers/etc?
Third parties drivers aren't microsoft windows hardware support they are third party support of windows despite microsoft. You won't see microsoft taking responsibility if those drivers don't work will you? Of course not, if they don't take the responsibility they certainly aren't entitled to the credit if the drivers do work.
"When I said it can't be both - I meant that both of the above can't be true. You can buy any PC - even one preloaded with Linux and there is zero doubt in my mind that Windows will be able to run on that hardware."
Both can be true. I've never seen a non-preloaded windows system where windows supported all the hardware. In every case full hardware support required downloading third party drivers. Ubuntu may or may not support the hardware but if it is going to work at all, it most likely worked out of the box with no additional configuration or third party downloads required. In the few cases where they are needed the system uses detects it and prompts you to download them.
The difference might not be especially troublesome for you today but it will be when that hardware is a few years old. For instance I guarantee when many windows users "upgrade" to vista aka windows 7 their perfectly functional printers/scanners/multi-functions/digital cameras/web cams that are a few years old will have to be replaced to accommodate the upgrade. Ubuntu will continue to support nearly every piece of hardware it supported with the last release on into the future until some compelling TECHNICAL reason makes it infeasible.
If they don't want their data to be an ISP's hostage they do. And if they don't want their inbox searched by the FBI without warrants.
Of course the anti-spam effort have pretty much outlawed private mail servers so downloading/deleting the mail as fast as you can is about all you have left.
"but you need to enter the address of your mail server into a mail client and end users seem to be able to manage that..."
Really? That is news to me. Last I checked end users need a disc from the ISP to do that, or they call their tech, or in a real pinch their teenager.
"Wardrivers and dictionary attacks are certainly not limited to specific systems."
And yet people with the knowledge and inclination to perform those attacks represent an extremely tiny segment of the population. Most IT professionals wouldn't know how to perform the attacks you are referring to.
"choosing a bad password, choosing the wrong protocol or letting your neighbours near your WiFi router, choosing the wrong brand of router"
The chance of your neighbors consistently being able to figure out why their desktop won't power up when the strip is turned off are slim let alone knowing what to do with access to your router or knowing what a router is. Hell in most cases your neighbor probably has a router and doesn't know what a router is.
"But wireless network encryption is only capable of protecting against someone who doesn't already have physical access anyway."
Your son/daughter/roommate is far more than likely to be trying to get into your wifi than any sort of thief and physical access doesn't help the layman one bit.
Even if they were technically capable to take advantage of physical access (you do realize that an extremely small fraction of the population has that level of technical capability right?) that isn't going to do them much good unless they don't mind you knowing your network was tampered with.
Wifi doesn't exist for corporate networks that need to prevent access to data. The bulk of wifi networks aren't protecting sensitive data at all, the protection is to prevent people from accessing your internet connection.
Only here in geekland do we even consider obscure scenerios like actual thought out and sophisticated attacks.
"Without that, you never know if you're being middled, and therefore you might as well not have encryption."
In theory yes. In practice no. Yes theoretically speaking I can't trust anything because I can't trust the cert authority.
I used to admin mail servers. There is nothing stopping bob the night backup babysitter from reading your plain text mail all night long. Its a hell of a lot harder for him to manage to compromise a mail with a man in the middle attack.
In practice I am far more secure because joe was joe yesterday and had the same cert. Because I can call joe and ask if his cert changed if he got the mail.
And because my ISP and carnivore can't simply sniff the traffic going over the wires and read the mail I'm sending. They actually have to successfully manage a man-in-the-middle attack that tricks me.
"I use it and simply to keep neighbors off my budget connection."
That is what security pros ignore with their ridiculous security recommendations and stringent long and complicated passwords you have to write down.
Aside from your neighbors the most likely person to try to hack your wifi is your 12 year old girl trying to sext when she should in bed. Neither of them is likely to be able to gain access without having to account for it if you use a simple password. The minute you have to stick something to the router though, baby girl is sending me jail bait titty shots.
The mere existence of superior technology does not make the older technology obsolete.
Ultimately there is a critical mass of usage that must be achieved by the new technology to call it obsolete. For instance, cars are not obsolete despite the existence of rocket packs, leer jets, and flying cars.
Without question the vast majority of wifi networks are NOT running WPA2/AES. I don't know how many 9's you need to throw in but its definitely less than one percent. I would be surprised if half the wifi networks out there could even run WPA2/AES without at least replacing hardware (and that isn't even counting the fact the no shortage of equipment rated for WPA2/AES doesn't actually reliably run in that mode).
Nobody replaces their router/wireless nics just because there are a couple hundred or even a thousand people in the country with the tech savy and equipment to potentially mooch their wifi. Those who have the new equipment are just incidental to other upgrades.
Even among those who have the hardware and would be willing they aren't willing to have wifi that won't work when their niece comes to visit with a two or three year old laptop.
Actually, according to the dd-wrt wiki you can use WDS with the TKIP+AES setting and then just configure your clients to use AES.
http://www.dd-wrt.com/wiki/index.php/Repeating_Mode_Comparisons
Specifically:
WDS can support WEP, WPA, and WPA2 Personal (TKIP+AES) as of v24sp1 (and earlier?). [tested with wrt150n1.1 AP + wrt54g6 R] (TODO: test other hardware). R/RB [repeater/repeater bridge] support all encryption algorithms.
Couldn't you just configure one to operate as a repeater or repeater bridge?
It isn't even a good practice. In my experience few card/router combinations are able to connect reliably when the SSID isn't broadcast.
As I said here:
http://mobile.slashdot.org/comments.pl?sid=1423971&cid=29919403
The most likely people to try to break into the router are people who have physical access and little technical knowledge. Like kids and employees.
The second most likely group is random people who are close by and want to mooch some free internet but they on average are less tech savy than your kids and any form of security will keep them out.
People lose perspective in computer security. Big time.
"And I would recommend to write down the password and put it in a drawer. Chances are that you only need to type it in after your system went fubar, and if that happens, you may have lost your password. Drawers are also very difficult to hack from the internet."
and yet, they are ever so easy to hack from inside the room.
You are dramatically more secure using a dictionary word password of less than ten characters, without mixed case, numbers or special characters than using a long and complex password that you have to write down somewhere to remember.
The people who are most likely to try to break into your internet are people you know and especially people you live and/or work with.
As a security professional you may encounter people being attacked by wardrivers and dictionary attacks all the time but as an individual you are probably more likely to be struck by lightning than have someone more sophisticated than your average random idiot looking for an open wifi link to check his email try to get in.
As a freelance network technician I've encountered ONE count them, ONE server running any operating system that has actually been deliberately hacked in ten years on the job. Despite having proof on the box my co-workers were skeptical because despite each having between 20 and 40 years of experience administering small business systems none of them had ever seen a hacked box.
That of course doesn't count worms, spyware, virus, and other automated attacks; open relays; or the kids/underling employees figuring out how to get around the proxy server.
"I get it, it all started when Adam dropped his fig leaf. Eve's first scream. I think I have a jpeg of that. Stupid fig leaf..."
I think it started when that pic was first transferred. We all know the internet is for pr0n.
You listed windows 7 before linux you sick fsck. Get the fsck back in your hole and don't come back.
We don't tolerate your ki...
lo, I for one welcome ou...
"Celebrating the day the Internet was *conceived*... well, that seems a bit weird."
Not if you are looking at it from dad's perspective. From the dad's perspective it is all down hill from the moment of conception.
"Maybe because you aren't a network of computers."
What an odd assumption.
The only thing that will make e-mail encryption take hold is an advance in the technology or at least the clients.
When using e-mail encryption is as simple as checking a box (or better yet, enabled by default) and the key generation, registration, retrieval, etc are completely automatic it might catch on.
So far I've yet to see a client that does this in a successful and consistent way.
"smaller than a planck length (which is the assumed smallest possible distance that something measurable can happen in classical physics)"
Something seems to be amiss here. If the variation is smaller than a planck doesn't that debunk aforementioned assumption?
If you do buy into the inefficiency thing then go old school and send an email that begins...
"Dear Uncle bin laden, what is your new address again?"
Limiting division by telomere length kills off healthy and undamaged cells preventing them from dividing. Removing this "feature" while introducing a replacement mechanism to protect from cancer would leave only damaged cells unable to reproduce.
All else being equal that should increase the ratio of undamaged cells to damaged cells.
I really don't see how anyone could deny that would be a good thing.
"Once the telomeres are eliminated, the ends of chromosomes are treated as double strand breaks by genetic repair machinery which more often than not, results in apoptosis, chromosome joining and ultimately cancer in some cases."
So the way in which p16 is believed to protect cells from cancer does not conflict with turning on telomerase and turning on telomerase would prevent this kind of damage causing at least some impact on aging and cell damage.
"Much of aging has to do with genetic damage"
Wouldn't immortalizing healthy undamaged cells counteract this to some extent?
If nothing else it would seem as if this may at least be part of the puzzle along with increased antioxidants, sunscreen, and viagra.
"If we were really patient, we could knock out p16 in these moles and see if they get cancer. That would pretty well establish whether or not it was just p16 that was responsible for the relative resistance to cancer."
It would establish if p16 is is an essential part of their cancer resistance but if p16 is working in combination with something else your experiment wouldn't reveal it.
The next step would be to then splice this gene into mice and activate it and see if the gene alone is enough to make it cancer resistant. That step will conclusively establish if something else is responsible (or an additional something) or if the gene alone is enough.