NTFS and FAT checksum blocks in files, I'd assume EVERY filesystem would do this. In the days of dos floppies, thats what "CRC Error" meant - the checksum didn't match the data. I believe modern ATA drives also detect & correct bit errors at a level not normally visible to the OS - you can see these error counts via SMART counters, and if they are rising, your drive is on the way out. I'd imagine this would be a symptom of exactly what TFA is describing.
And RAID CAN detect and repair these errors. What you said might be true for RAID 1 (mirroring) except for the filesystem checksums would reveal which volume was corrupt and which was correct. RAID 5 spreads the parity over three volumes in a way that means that if one drive flips a bit (or even completely dies) the information on that drive can be reconstructed from the parity information on the other two drives.
Yes, ZFS is cool (the best filesystem out there IMHO) but its not really using any revolutionary new techniques, just combining all the good ideas from the last 50 years in a complete package.
I thought that TCP hijacking was possible because of vulnerbilities in Sequence number generation in some TCP implementations, which have now been fixed?
Do you have big signs up warning users not to use internet banking or online shopping from these PCs, as you are recording their online banking passwords and credit card numbers?
How well do you secure your systems and backups? Do you ever worry someone will grab those HTTPs logs and get access to a whole lot of very sensitive information?
I'm talking about what happens on the client end - say for example you connect to your bank's online banking website (https)
They have a cert that is signed by verisign or someone.
Now would you feel happy if your ISP was decrypting, logging and potentially altering (to insert ads etc) that traffic to/from you your bank - which would include your balances, your CC numbers, and your username and password.
To have the ability to do this your ISP would need to decrypt the traffic and re-sign it with a cert of theirs that was trusted by your PC - or else you browser would say the site's cert isn't trusted, and/or the site's cert doesn't match the site URL.
Things like feeding you a bogus SSL cert while making it appear perfectly legit and decrypting your traffic,
Fortunately they can't do that without your browser screaming the name on the cert doesn't match the hostname.
Of course, a large % of clueless users will ignore the strongly worded warning and click ok.
Only way they could do that is if they had their own trusted root certification authority - then they could make up a new cert for the website you asked for on the fly, and your browser would trust it.
I beleive in china you must have a root cert the govt has issued in your root certs store. That would let them evesdrop on HTTPs sessions without triggering any obvious alerts on the client site (although if you checked the certification path you could see that the site's cert was issued by the chinese govt, not verisign or similar)
Re:Well, Linus is an ass, what's new.
on
Linus on GIT and SCM
·
· Score: 3, Funny
wow, you aren't just an amzing ass with a huge ego, you are modest about it too!:D
Late model Xboxes have a 10gb drive. An xbox drive is split into 5 partitions (boot, 3 swap, & data) and if you have a 10gb drive, 2gb is simply unpartitioned. If you put a modchip in you can create a new partion, and chuck more stuff in there. Unfortunately you can't make any of the stock partitions any bigger.
If you put a large HD in your xbox then the extra space is used in a partition unmodded xboxes don't have (and is unuseable without the modchip enabled)
The international acedemy of science awarded Bussard & team the "Outstanding technology of the year award" last year (linky)
According to that page, Bussard's reactor could be on the market in 6-10 years.
Interestingly the design isn't a "steam kettle" system, like all existing thermal power plants - coal, natural gas or nuclear, which all use a heat source to boil water to spin a steam turbine.
Bussard's Pollywell design generates high-energy alpha particles, which can be used to directly produce an electrical current.
It looks like Bussard is finally getting the attention he deserves, rather than the incredibly expensive magnetic confinement systems like ITER, which has so far spent billions of dollars and needs billions more before anyone can even say for sure if it will work or not...
If Bussard pulls this off, this could be an incredibly disruptive technology. Clean, cheap power... what the nuclear age has so long promised but failed to deliver.
Yes it works 100% with NTFS. It doesn't care in the slightest what filesystem the drive hosting the volume is using, or what the filesystem inside the encrypted volume is.
This is typical, get used to it - or get a job where this stuff is left to specialists, business analysts.
Although I beleive you should go through the pain of requirements gathering at least once, it will make you a better developer.
I reccommend workshops. Get some users (and preferably also a manager or team leader who can give a different perspective) in a quiet room with a whiteboard for two or three hours at a time, and get them to walk you through the process. Draw diagrams, get them to explain things. Getting what they actually want out of them can be like pulling teeth. They will assume you understand their problems... assume nothing.
Make sure you do a thourough job, and get them to sign off on the requirements documentation you come up with in the end. If you don't and then end up building something that doesn't meet their needs then its difficult and expensive to change, and you will get the blame.
I thought thats how all stream cyphers worked? By generating a pseudo-random stream, and then XORing this with the source stream?
Thats certainly how RC4 works (SSL, HTTPs...). I wrote an implementation of RC4 once when I was between jobs, it was a very interesting exercise and it taught me a lot about cryptography.
The 255 char limit isn't the filesystem, its explorer (and the api it uses.)
You can use subst, reparse points or just map a share to get to the files too deep to get straight too.
I've seen quite a few network drives organised into a very clever heirachy that is 10+ levels deep hit this problem. You have to make a new share, deeper down.
Best multiplayer game I ever played was Tribes (the first one) but maybe that was because I played with a bunch of regulars, and we actually did the whole teamplay thing, rather than all being lone wolves.
Don't count on your visual basic skills you learned "over a decade ago" helping you out much. VB.Net syntax is similar to old-school vb5-6, but writing apps for the.net framework is a completely different kettle of fish. The difference between C# and vb.net is tiny compared with the difference between vb6 and vb7 (vb.net)
You can still hack something ugly together with vb.net (or c#) in a hurry, but if you want to do it right you have to take a different approach.
The problem is where does this leave tools like daemon tools, which require a device driver? They are screwed, unless they use hacks like the article describes. free/open source apps won't be able to afford a cert for their drivers, and MS may not give them one anyway.
Yeah I haven't read the paper either, and all I know about special relativity are the general principles but all frames of reference are equal is certainly one of the most important points.
But now I think about it, another principle of GR is a constant acceleration is indistinguishable from a gravity field - so if this thing can supply a constant force against gravity, shouldn't it feel the same force and so accelerate when not in a gravity field (or in orbit etc?)
In essence, the Emdrive is a resonating bottle full of microwaves. Because microwaves are a low frequency form of light, their behaviour is governed by Einstein's Theory of Special Relativity. And while microwaves and other forms of electromagnetic radiation may be thought of as very fast moving particles, they also have to be thought of as waves. At the same time that the constituent particles are moving at light speed, or their phase velocity, energy is transferred by the wave aspect travelling at group velocity. Group velocity is the result of waves of different wavelengths interacting with each other. While, according to Einstein, the phase velocity of electromagnetic waves is the speed of light in the appropriate medium whatever happens, and in whatever moving frame of reference the observer happens to be, group velocity varies. Group velocity can be any speed from stationary to light speed (a few physicists suggest the additional possibility of faster than light), and this varies the amount of momentum striking an impenetrable barrier, and thus the force exerted on it. Hence, it is possible to have a bottle full of electromagnetic waves exerting more force on one end than the other, whereas this is not possible for anything else that an engineer would normally be expected to encounter.
IANAP, but this might explain what he means by this effect not being useful for continuous acceleration - If the "bottle" accelerates, won't this change the phase velocity of the microwaves, and then reduce the thrust effect?
I submitted this story a week ago, and was rejected:( (but I didn't have a link to the new scientist article, so not surprising)
Not if you are using a defragger that uses MS's defrag API on an NTFS volume.
On NTFS moving a cluster is an atomic operation. The data is duplicated to an unused space, then the metadata (file table) is updated, which is a journaled operation.
More info (including a free defragger) here: http://www.kessels.com/defrag/index.html. Scroll down to "How safe is it" and it even mentions that "heavy use of the harddisk may trigger a hardware fault", which is most likely what happened to the parent poster.
Its a hardware issue.. most likely a bad disk. Probably brought on by a high temp, I bet the disk got hot while it was defragging. You really should run a free app like HDDHealth that can monitor your disk temp and the SMART counters.
Defragging a disk using the built-in utility is very safe, however its not that effective because it doesn't normally move the defragged files to a very smart location, so you have to run it multiple times to make much of a difference. I reccommend this one instead - http://www.whitneyfamily.org/Hacks/?item=Defrag. It uses the same windows defrag API, but its smarter.
FYI, the defrag API doesn't let you defrag in-use files, and each file is moved in an atomic transaction - the file contents are copied to the new location, then the file table is updated (which is a journaled operation). You can power off at any time, and you won't damage the filesystem, and it will most definitely not cause a crash.
Whatever happened to your disk was probably caused by the stress of defragging, but it wasn't the defraggers fault. I've seen plenty of corupt NTFS volumes, and every one of them was caused by bad hardware.
Slashdot Mirror
Almost completely wrong.
NTFS and FAT checksum blocks in files, I'd assume EVERY filesystem would do this. In the days of dos floppies, thats what "CRC Error" meant - the checksum didn't match the data. I believe modern ATA drives also detect & correct bit errors at a level not normally visible to the OS - you can see these error counts via SMART counters, and if they are rising, your drive is on the way out. I'd imagine this would be a symptom of exactly what TFA is describing.
And RAID CAN detect and repair these errors. What you said might be true for RAID 1 (mirroring) except for the filesystem checksums would reveal which volume was corrupt and which was correct. RAID 5 spreads the parity over three volumes in a way that means that if one drive flips a bit (or even completely dies) the information on that drive can be reconstructed from the parity information on the other two drives.
Yes, ZFS is cool (the best filesystem out there IMHO) but its not really using any revolutionary new techniques, just combining all the good ideas from the last 50 years in a complete package.
Umm so whats special about Duchovny's hands?
I thought that TCP hijacking was possible because of vulnerbilities in Sequence number generation in some TCP implementations, which have now been fixed?
Thats exactly what has happened. What do you think the british classification board classified? Some paper sketches? :)
Its not just illegal, its impossible.
At the end of development game must be signed with microsoft's private key. Only dev kits will run a game that isn't signed.
A modchip bypasses this check. However, there are no modchips for the 360 or ps3 yet... I beleive there are some for the Wii
Having a matt black container with a reflective cover above it is probably the best configuration as black is a much better emitter of heat than white
Do you have big signs up warning users not to use internet banking or online shopping from these PCs, as you are recording their online banking passwords and credit card numbers?
How well do you secure your systems and backups? Do you ever worry someone will grab those HTTPs logs and get access to a whole lot of very sensitive information?
I'm talking about what happens on the client end - say for example you connect to your bank's online banking website (https)
/or the site's cert doesn't match the site URL.
They have a cert that is signed by verisign or someone.
Now would you feel happy if your ISP was decrypting, logging and potentially altering (to insert ads etc) that traffic to/from you your bank - which would include your balances, your CC numbers, and your username and password.
To have the ability to do this your ISP would need to decrypt the traffic and re-sign it with a cert of theirs that was trusted by your PC - or else you browser would say the site's cert isn't trusted, and
Fortunately they can't do that without your browser screaming the name on the cert doesn't match the hostname.
Of course, a large % of clueless users will ignore the strongly worded warning and click ok.
Only way they could do that is if they had their own trusted root certification authority - then they could make up a new cert for the website you asked for on the fly, and your browser would trust it.
I beleive in china you must have a root cert the govt has issued in your root certs store. That would let them evesdrop on HTTPs sessions without triggering any obvious alerts on the client site (although if you checked the certification path you could see that the site's cert was issued by the chinese govt, not verisign or similar)
wow, you aren't just an amzing ass with a huge ego, you are modest about it too! :D
Ok, If this drive has 5 platters, then why not 6? There must be 6 platters in there, why is one not used?
They actualy did do exactly that...
Late model Xboxes have a 10gb drive. An xbox drive is split into 5 partitions (boot, 3 swap, & data) and if you have a 10gb drive, 2gb is simply unpartitioned. If you put a modchip in you can create a new partion, and chuck more stuff in there. Unfortunately you can't make any of the stock partitions any bigger.
If you put a large HD in your xbox then the extra space is used in a partition unmodded xboxes don't have (and is unuseable without the modchip enabled)
The international acedemy of science awarded Bussard & team the "Outstanding technology of the year award" last year (linky)
According to that page, Bussard's reactor could be on the market in 6-10 years.
Interestingly the design isn't a "steam kettle" system, like all existing thermal power plants - coal, natural gas or nuclear, which all use a heat source to boil water to spin a steam turbine.
Bussard's Pollywell design generates high-energy alpha particles, which can be used to directly produce an electrical current.
It looks like Bussard is finally getting the attention he deserves, rather than the incredibly expensive magnetic confinement systems like ITER, which has so far spent billions of dollars and needs billions more before anyone can even say for sure if it will work or not...
If Bussard pulls this off, this could be an incredibly disruptive technology. Clean, cheap power... what the nuclear age has so long promised but failed to deliver.
Yes it works 100% with NTFS. It doesn't care in the slightest what filesystem the drive hosting the volume is using, or what the filesystem inside the encrypted volume is.
I second Greg Bear... Although Greg Egan is also good.
I ate up books like Eon, forge of god, eternity, moving mars while I was at high school.
This is typical, get used to it - or get a job where this stuff is left to specialists, business analysts.
Although I beleive you should go through the pain of requirements gathering at least once, it will make you a better developer.
I reccommend workshops. Get some users (and preferably also a manager or team leader who can give a different perspective) in a quiet room with a whiteboard for two or three hours at a time, and get them to walk you through the process. Draw diagrams, get them to explain things. Getting what they actually want out of them can be like pulling teeth. They will assume you understand their problems... assume nothing.
Make sure you do a thourough job, and get them to sign off on the requirements documentation you come up with in the end. If you don't and then end up building something that doesn't meet their needs then its difficult and expensive to change, and you will get the blame.
I thought thats how all stream cyphers worked? By generating a pseudo-random stream, and then XORing this with the source stream?
Thats certainly how RC4 works (SSL, HTTPs...). I wrote an implementation of RC4 once when I was between jobs, it was a very interesting exercise and it taught me a lot about cryptography.
The 255 char limit isn't the filesystem, its explorer (and the api it uses.)
You can use subst, reparse points or just map a share to get to the files too deep to get straight too.
I've seen quite a few network drives organised into a very clever heirachy that is 10+ levels deep hit this problem. You have to make a new share, deeper down.
Best multiplayer game I ever played was Tribes (the first one) but maybe that was because I played with a bunch of regulars, and we actually did the whole teamplay thing, rather than all being lone wolves.
Don't count on your visual basic skills you learned "over a decade ago" helping you out much. VB.Net syntax is similar to old-school vb5-6, but writing apps for the .net framework is a completely different kettle of fish. The difference between C# and vb.net is tiny compared with the difference between vb6 and vb7 (vb.net)
You can still hack something ugly together with vb.net (or c#) in a hurry, but if you want to do it right you have to take a different approach.
The problem is where does this leave tools like daemon tools, which require a device driver? They are screwed, unless they use hacks like the article describes. free/open source apps won't be able to afford a cert for their drivers, and MS may not give them one anyway.
Yeah I haven't read the paper either, and all I know about special relativity are the general principles but all frames of reference are equal is certainly one of the most important points.
But now I think about it, another principle of GR is a constant acceleration is indistinguishable from a gravity field - so if this thing can supply a constant force against gravity, shouldn't it feel the same force and so accelerate when not in a gravity field (or in orbit etc?)
From that link:
IANAP, but this might explain what he means by this effect not being useful for continuous acceleration - If the "bottle" accelerates, won't this change the phase velocity of the microwaves, and then reduce the thrust effect?
I submitted this story a week ago, and was rejected
Not if you are using a defragger that uses MS's defrag API on an NTFS volume.
On NTFS moving a cluster is an atomic operation. The data is duplicated to an unused space, then the metadata (file table) is updated, which is a journaled operation.
More info (including a free defragger) here: http://www.kessels.com/defrag/index.html. Scroll down to "How safe is it" and it even mentions that "heavy use of the harddisk may trigger a hardware fault", which is most likely what happened to the parent poster.
Its a hardware issue.. most likely a bad disk. Probably brought on by a high temp, I bet the disk got hot while it was defragging. You really should run a free app like HDDHealth that can monitor your disk temp and the SMART counters.
Defragging a disk using the built-in utility is very safe, however its not that effective because it doesn't normally move the defragged files to a very smart location, so you have to run it multiple times to make much of a difference. I reccommend this one instead - http://www.whitneyfamily.org/Hacks/?item=Defrag. It uses the same windows defrag API, but its smarter.
FYI, the defrag API doesn't let you defrag in-use files, and each file is moved in an atomic transaction - the file contents are copied to the new location, then the file table is updated (which is a journaled operation). You can power off at any time, and you won't damage the filesystem, and it will most definitely not cause a crash.
Whatever happened to your disk was probably caused by the stress of defragging, but it wasn't the defraggers fault. I've seen plenty of corupt NTFS volumes, and every one of them was caused by bad hardware.