So now they'll have to run point to point links to every VA and Social Security office to the closest gateway. At the cost of fiber these days, that'll be an amazingly high cost, when they could get much much less expensive internet through local suppliers.
If they want to standardize their security, there are other ways to do this. They could decide on one line of router/firewall and remotely update the configurations.
Sadly, management tools and report generation are hard, and they require a level of coordination that's easier to achieve in one company and much harder in an open source environment.
A brilliant programmer can come up with some really solid and innovative code (ex. reiserfs), but to make a nontrivial management tool you need a combination of programmers, designers, and yes, managers, working in tight concert.
I personally am okay with paying for front ends when they're needed, so we can get kickass scalable solid database backends for free.
Contrary to what one might assume, the in-memory clustering is generally slower than storing the files on disk.. Are you sure this holds up for mysql's shared nothing architecture? Most other DBMSes use a shared block device (a SAN) for clustered databases, which is a whole other perforamnce profile.
So, could you offer a bonded "secure" computer repair service through attorneys? No. In order for attorney client privilege to apply, the lawyer has to be rendering legal services.
Basically, if it were the Chinese government behind it, they would find machines in the US and Europe to zombify, and launch their attacks on government computers from those machines. They would use so many layers of net access that it'd be exceedingly difficult to track it back to hacker.gov.cn. If there was a coordinated attack by the Chinese government, and the US managed to track it back to them, the NSA would probably keep quiet about it so that they don't give away their capabilities and so that they'd have a method to feed China misinformation.
This is most likely a coordinated attack by someone who wants US information (could be any country/organization in the world) and developed a botnet which happens to mostly reside in China, since China's computers tend not to get frequent security updates. The fact that the IP addresses are originating from China indicates that it's probably anyone but China.
However...China-bashing does score political points right now.
The aircraft is private property run by a private company, and as such can refuse business to any individual they wish for any reason they wish.
The airline is required by security directives to refuse service in certain situations (i.e. the person's name matches one in the no-fly list). This isn't about the airlines voluntarily refusing service, it's about the government saying they have to.
a host of musicians (over 4,500 of them, including poor, starving stars such as U2, Paul McCartney and Peter Gabriel) have taken out a big ad in the FT to back the call for an extension to copyright in the UK. Allegedly, that's what the British public wants
I guess the starving musicians have to spend big money take out an ad to tell the public what they already want.
I also wonder in yousendit.com can handle a slashdotting. I guess we'll know soon!
On the link I posted below I've only gotten 50 hits, and yousendit has a limit of 100 downloads for unpaid accounts, so I don't think too many people are clicking on the article.
[quote]Stuff like UI-isms (paperclips, ribbons, hiding the file menu, etc) isnt "innovation".... Stuff like Dtrace , TCP/IP, xml,.. THAT is innovation. Lets have MS give us some real innovation, you know - stuff that wont just change "the way things are done" inside of thier own software ecosystem.[/quote]
Tell that to your grandma. Computers are made to be used, not to be repositories for acronyms.
These phones look pretty stylish, but I'd stay away from any touchpad style phone if you ever plan on wearing gloves. Touchpads only work with direct skin contact, so that rules out using them outside in cold weather.
Cell phones could go a long way, but I think that something like this limits the environment too much.
the complaint asks AOL to notify all users affected by the data disclosure via certified mail
Unless I'm being sued or in immediate legal danger, I don't want to get any certified mail.
When I do, I have to interrupt my work day and drive 10 miles over questionable roads to the post office.
The fact that some of my searches may have been leaked without my name on them is not a reason to send a
certified letter, however an insert in my next bill would be completely reasonable.
The EFF has good intentions, but in this case they are going overboard.
Yes, AOL made a mistake by releasing that information. They've admitted to the mistake, apologized, and I doubt anyone will try to do this again.
On the other hand, one needs to recognize that they didn't release the information for the purposes of making money, or defrauding the customers, or anything else. They collected the data in order to help a researcher write an extremely informative paper[pdf] about human behavior as it relates to searches. That researcher decided that other's might benefit from the information, and convinced AOL to make it publically available. It turns out that that was a huge lapse in judgement, but nonetheless, intentions are also important and while criticizing AOL, we should also complement them for their effort to interface with the academic community.
AOL has been punished enough in the press. Given the circumstances I don't think that any legal action is necessary.
Guess what? A proxy server passes on an HTTP connections request headers, including the cookies. All that it Masq's is the IP address. This way you can use a proxy server with websites that require session tracking.
This issue isn't about google keeping track of your IP address, it's them tying all your queries to one identifier, using longterm cookies. To protect yourself from this, configure your browser so that it clears your cookies each time you reload it.
Actually, TPM is a perfect solution for the military. For several years the military has been having issues that there's no standard hardware encryption, so in some cases they use hardware that goes on the SCSI bus and hard drive and encrypts everything that goes onto the hard drive (obviously not useful for a laptop). In other cases, they use software for encrypting specific files, but have no way of guaranteeing that the user encrypts everything that's sensitive. In other words, the military is still relying on physical security, and that's not good enough anymore as laptops are way to easy to steal or "lose"
I'll bet that the military is a big enough customer that they could get their own public key put on the TPM chips. This would allow them to encrypt every one of their hard drives so it doesn't work on non-military hardware (remember the USB drives showing up at bazaars? Imagine if they were only readable on military-issued laptops or in conjunction with RSA fobs) They could also make it so that people can only install software on their computers which is certified for the classification level of that computer, as another problem that they have is people installing programs that could contain spyware on their secret and top-secret laptops.
The reason that we dislike TPM is because it gives the administrator complete control over the platform. That's exactly what the military needs.
Write and call your senators letting them know exactly why this is a bad idea. If enough people make their opposition clear, we may still have a chance of getting it stopped there.
Ain't it great that open source tools can now benefit from the same meaningless marketing drivel
which has consistently been a strong feature of proprietary software?
If something can be read, and written - it can be copied.
Copying takes time. Any system like this would be carefully timed with an atomic clock on both sides so that a latency change would immediately be discovered. It has always been possible in a fiber system to detect an active man-in-the-middle by monitoring the latency.
This prevents passive listening, where a portion of the beam is split off and monitored. If you're only sending a single photon, there isn't a portion to split.
It would be _stupid_ to put something so critical on a shared band.
And I'm sure that someone wishing to disrupt communications system would obey the FCC regulations by staying out of government bands. In radio, the entire spectrum is a shared band.
Slashdot Mirror
She meant free up developer time.
How long will it take for that list to get published as a list of places to go?
So now they'll have to run point to point links to every VA and Social Security office to the closest gateway. At the cost of fiber these days, that'll be an amazingly high cost, when they could get much much less expensive internet through local suppliers. If they want to standardize their security, there are other ways to do this. They could decide on one line of router/firewall and remotely update the configurations.
A brilliant programmer can come up with some really solid and innovative code (ex. reiserfs), but to make a nontrivial management tool you need a combination of programmers, designers, and yes, managers, working in tight concert.
I personally am okay with paying for front ends when they're needed, so we can get kickass scalable solid database backends for free.
Have a look at this user's comment history. Quite the one joke horse. http://slashdot.org/~New+Here
Wrong. The syslog protocol only allows you to write to send data.
I guess someone will have to add him to the List of Premature obituaries
This is most likely a coordinated attack by someone who wants US information (could be any country/organization in the world) and developed a botnet which happens to mostly reside in China, since China's computers tend not to get frequent security updates. The fact that the IP addresses are originating from China indicates that it's probably anyone but China.
However...China-bashing does score political points right now.
Interesting stuff. Here's a link to the full text:
d f
http://130.58.240.179:8080/~erek/minorityreport.p
[quote]Stuff like UI-isms (paperclips, ribbons, hiding the file menu, etc) isnt "innovation" .... Stuff like Dtrace , TCP/IP, xml, .. THAT is innovation. Lets have MS give us some real innovation, you know - stuff that wont just change "the way things are done" inside of thier own software ecosystem.[/quote]
Tell that to your grandma. Computers are made to be used, not to be repositories for acronyms.
Cell phones could go a long way, but I think that something like this limits the environment too much.
The EFF has good intentions, but in this case they are going overboard.
Yes, AOL made a mistake by releasing that information. They've admitted to the mistake, apologized, and I doubt anyone will try to do this again.
On the other hand, one needs to recognize that they didn't release the information for the purposes of making money, or defrauding the customers, or anything else. They collected the data in order to help a researcher write an extremely informative paper[pdf] about human behavior as it relates to searches. That researcher decided that other's might benefit from the information, and convinced AOL to make it publically available. It turns out that that was a huge lapse in judgement, but nonetheless, intentions are also important and while criticizing AOL, we should also complement them for their effort to interface with the academic community.
AOL has been punished enough in the press. Given the circumstances I don't think that any legal action is necessary.
Guess what? A proxy server passes on an HTTP connections request headers, including the cookies. All that it Masq's is the IP address. This way you can use a proxy server with websites that require session tracking. This issue isn't about google keeping track of your IP address, it's them tying all your queries to one identifier, using longterm cookies. To protect yourself from this, configure your browser so that it clears your cookies each time you reload it.
Actually, TPM is a perfect solution for the military. For several years the military has been having issues that there's no standard hardware encryption, so in some cases they use hardware that goes on the SCSI bus and hard drive and encrypts everything that goes onto the hard drive (obviously not useful for a laptop). In other cases, they use software for encrypting specific files, but have no way of guaranteeing that the user encrypts everything that's sensitive. In other words, the military is still relying on physical security, and that's not good enough anymore as laptops are way to easy to steal or "lose"
I'll bet that the military is a big enough customer that they could get their own public key put on the TPM chips. This would allow them to encrypt every one of their hard drives so it doesn't work on non-military hardware (remember the USB drives showing up at bazaars? Imagine if they were only readable on military-issued laptops or in conjunction with RSA fobs) They could also make it so that people can only install software on their computers which is certified for the classification level of that computer, as another problem that they have is people installing programs that could contain spyware on their secret and top-secret laptops.
The reason that we dislike TPM is because it gives the administrator complete control over the platform. That's exactly what the military needs.
Write and call your senators letting them know exactly why this is a bad idea. If enough people
make their opposition clear, we may still have a chance of getting it stopped there.
From the company website:
Copying takes time. Any system like this would be carefully timed with an atomic clock on both sides so that a latency change would immediately be discovered. It has always been possible in a fiber system to detect an active man-in-the-middle by monitoring the latency.
This prevents passive listening, where a portion of the beam is split off and monitored. If you're only sending a single photon, there isn't a portion to split.