Seems to me the best way of doing it would be to put two 911 buttons side by side on these "disposable" phones, so you have to press both at the same time for it to dial.
It's also interesting to note that while Sony Ericsson phones allow you to leave bluetooth active, but not receive messages - apparently there's nothing in the Nokia phones to do the same thing (heh heh heh).
Unfortunately bluetooth phones are still very rare in north america, which is sort of unfortunate given my P800 and the excellent application "SMAN":)
But if nobody is getting the spam anyway, would it make much of a difference?
Having a few hundred thousand (million?) machines chipping away at your bandwidth by requesting the page, without the spammer selling any product because their messages are getting filtered, would get pretty expensive, pretty fast.
I've seen a few messages with the "unrelated text" trick attempted. Unfortunately it doesn't work.
Part of the joy of bayesian filtering is that it doesn't categorize the entire message - it calculates probability factors for each word in the message, but it just takes the top 10 and bottom 10 (or whatever value you set) words, and calculates probability from that.
Even if the spammer pasted an entire page out of a dictionary, all it would take is one word like "viagra" or "mortgage" to tip the scale so far towards spam that there's no way a huge chunk of randomly chosen paragraphs would help at all.
To be sure, a few messages do make it past filtering, but they're few and far between. So far between that I just nuke my "spam" box without reading it most of the time.
Spamassassin is great for ISPs and other companies that need rule-based spam checkers that are sort of "generic".
For personal filtering, nothing beats a good bayesian filter. I use POPFile myself and it's approaching 99% accuracy and I _LOVE_ it.
Spam very, very rarely makes it past, and if it does, it's the generic "check out this site" type message with no other information. Even spammers trying this technique aren't having much success as I'm seeing less and less of it (maybe 1 or 2 message a month make it past the filters).
The next step in anti-spam evolution will be spam-scanning software that automatically follows links back to webpages and looks for "spammy" content and tags the message as spam in the email system.
For those out there that havn't tried a bayesian form of filtering yet, give POPFile a try: (http://popfile.sourceforge.net/). Just be sure to read the instructions.
I'm waiting for it to hit the fan over Adobe's product activation that's required on their new "Creative Suite" products (ie: Photoshop and Friends).
Just like XP, you have to let your machine either contact Adobe over the internet, or phone their customer service number to get the activation code that's locked to your individual computer.
Oh yes, and apparently you can only activate twice over the internet, then you HAVE to phone their CSRs to explain why you're not a pirate giving copies to your friends to justify additional installs. Guilty until proven innocent I suppose.
I'm willing to bet there's going to be a backlash against them similar to the Intuit tax activation fiasco.
Sorry, I didn't explain well I suppose - the 256 codes only applies to the specific remote in question (that would be disabled). The receiver itself would never be disabled.
Depending on the size of the code (say, 256 bits), hacking it would be next to impossible unless you wanted to wait hours, if not days to try and brute-force a light that will trigger in a minute or so anyway.
I use an infrared remote to access my condo - it uses an infrared remote system that's fairly simple, yet effective:
-The remote (much like garage door openers) uses a rolling pseudo-random number sequence. The remote generates a code based on the next number in sequence from a random number generator seeded with a known key for that particular remote (the main controller needs to be "paired" with the remote before use so it knows what seed each remote is using).
-The controller keeps track of all of the remotes for the building and pre-computes the next 256 valid codes that each remote will generate based on the seed exchanged when the remote is paired.
When a remote triggers, it sends the random number code (NOT the key) by way of infrared to the controller receiver. The controller checks to see if the number the remote transmitted is in the next 256 valid numbers for that particular remote, if so, you get in. If not, you don't and the attempt is logged.
If you press the button more than 256 times (playing with the remote button for example) when you're not around the sensor, none of the precompute codes will match the next time the remote is used and it will be useless until re-paired.
Even if you capture the code being sent from the remote, you won't know the key that the random number generator is using in that particular remote to generate the number sequence, or any of the subsequent numbers that the remote would generate. You'd only capture the code that was sent, and once that was used, it wouldn't work again anyway.
If a remote key is compromised, it's simple to simply deactivate that particular remote key. If the system is brute-force attacked, it can either deactivate the sensor that's being attacked, or just call security to the appropriate location.
They can't put it on the subtitle/angle tracks or they'd just get removed during the rip. Which only leaves actual messing with the MPEG encoded video.
I suppose someone could create a MPG "tweaker" to just decompress/watermark/recompress select blocks on the encoded files to serialize them, but it would be expensive, and would mean DVD-R'ing every disc (compatibility issues, far more expensive production)
Interestingly enough, I saw "Kill Bill" last night, and although I really enjoyed the movie, Miramax's watermarking of the image (red-dot patterns appearing for a frame or two during the movie) was very evident.
They better be careful, if they tamper with my moviegoing experience, I will not be seeing their new releases in the theatre. Putting red dots all over a movie that I've paid $10 or more to see practically guarantees that I won't be in the audience.
I own a P800, but I don't plan on upgrading to a P900 - the improvements just aren't that significant in my book to warrant the hundreds of dollars of money, not to mention the time to sell-off the old unit and upgrade.
1 - the screen resolution is the same, but looks larger because the flip is smaller. I have my flip removed anyway, so no big deal.
2 - landscape mode is nice, though there's only a few apps that I need that use it.
3 - firmware can be uploaded into the P800 as well, but it must be done at a dealer. I havn't heard anywhere yet that the P900 is user-upgradable without a trip to the SE service center though that would be very, very nice.
4 through 11 - Nice, but nothing that I'd pay hundreds of dollars to obtain.
I'm going to skip the P900 and probably pick up whatever they're releasing next summer/fall.
I'm supremely dissapointed that they didn't put a better camera into the P900 because the model from the 800 (and which is in the 900) is just barely acceptable in terms of quality. Even a one-megapixel camera would've been very, very nice.
The odd thing is that region control was originally designed mainly to stop "foreign regions" (ie: regions outside the US/Canada) from viewing all of the juicy MPAA wares before they were released in their domestic language.
The reality is that nearly ALL of the regions outside the US/Canada have region coding disabled on their players by default (nobody will buy players in those countries unless the player is region-disabled, so the companies have learned pretty damn quick to give consumers what they want).
So, the net result? Nearly the only place in the world with region-coding enabled on DVD players is... You guessed it, the US/Canada.
Just out of interest, was there info that ATI gave out the Doom 3 alpha?
There's a big difference between "leaked" (ie: gave out), and someone who "stole" it from them by popping a disk into a drive or hacked into a server or something...
Part of the windows problem seems to be that startup applications and services all try and execute in parallel, thus thrashing the hell out of the drive and slowing the machine a whole lot.
Why not force (or allow the user to set) a limit on the number of startup applications that execute simultaneously to avoid thrashing the drive and making the startup sluggish.
I rather think the Republicans aren't all that worried about a "Florida happening again". After all, it did get a Republican into the oval office didn't it...
It's odd though, speaking as a Canadian who has always though that although not perfect, the US electoral system had a fair number of checks and balances, it absolutely blows my mind that this sort of un-checked corporate crap isn't being stopped in it's tracks.
It's like 9/11 gave the politicians and big business license to do whatever the hell they want to with your entire country and the economy, and they're screwing it up at a simply astounding rate. "Patriot" take-away-your rights acts, a court denying a "do-not call list" that 50 MILLION people want for the benefit of a few telemarketing lobbyists, big companies trying to patent even the most trivial of ideas... Where does it end?
I mean, this latest info about a company making machines to support democratic elections that has no "unalterable record", easy bypassing (or complete lack) of database passwords, and executives talking about just printing "system check" on the screen without any actual checking being done because the electoral regulations require a full system check before the system begins recording votes.
For what it's worth, they're not having all that much luck.
My ISP's DNS servers have been updated with the "delegation-only" patch so they don't return any wildcarded results. I've also updated my own DNS servers at work with the same result.
The more ISPs that do this, the less money verisign makes until they just drop the idea alltogether.
It's worth noting that recent "broadband" releases in Japan are in the 50-100megabit range to your home. Yes, that means transferring anywhere from 6,000 to 12,000kb per second.
By contrast, my ADSL (Canada) downloads at around 350kb per second, and it's considered fast...
On the upstream side of things, it's around 50kb per second.
We have a looooong way to go in north america, folks...
Slashdot Mirror
Seems like the way to help prevent this would be local or federal regulations requiring a $25 refundable deposit on disposable technology like this.
Dump the phone in a bin and you're out $25. Bring it back and you get a refund.
That would be enough to ensure that most phones are recycled properly.
Seems to me the best way of doing it would be to put two 911 buttons side by side on these "disposable" phones, so you have to press both at the same time for it to dial.
That's a good idea, I should patent it! (lol)
I've got a Sony Ericsson P800, and it allows you to use any WAV file as a ring tone. Very, very nice.
The P900 is even newer (and slightly nicer), and allows you to use MP3s as ringtones.
http://www.sonyericsson.com/P900/main.htm
I honestly don't care.
Any time I go looking for networking gear, I'm making it a point of avoiding belkin hardware.
They screwed up, I'll never buy their stuff. Let this be a lesson to other companies that are considering a plan like this...
Serves 'em right for supporting this kind of stupidity.
/. would be a good place for potential investors in technology to at least do a little "reading between the lines"...
People who actually did some research (and therefore decided not to buy stock in this company) would then be safe.
You'd think that
N.
2 weeks then? :)
N.
It's also interesting to note that while Sony Ericsson phones allow you to leave bluetooth active, but not receive messages - apparently there's nothing in the Nokia phones to do the same thing (heh heh heh).
:)
Unfortunately bluetooth phones are still very rare in north america, which is sort of unfortunate given my P800 and the excellent application "SMAN"
N.
But if nobody is getting the spam anyway, would it make much of a difference?
Having a few hundred thousand (million?) machines chipping away at your bandwidth by requesting the page, without the spammer selling any product because their messages are getting filtered, would get pretty expensive, pretty fast.
It's not a perfect solution, but every bit helps.
I've seen a few messages with the "unrelated text" trick attempted. Unfortunately it doesn't work.
Part of the joy of bayesian filtering is that it doesn't categorize the entire message - it calculates probability factors for each word in the message, but it just takes the top 10 and bottom 10 (or whatever value you set) words, and calculates probability from that.
Even if the spammer pasted an entire page out of a dictionary, all it would take is one word like "viagra" or "mortgage" to tip the scale so far towards spam that there's no way a huge chunk of randomly chosen paragraphs would help at all.
To be sure, a few messages do make it past filtering, but they're few and far between. So far between that I just nuke my "spam" box without reading it most of the time.
N.
Spamassassin is great for ISPs and other companies that need rule-based spam checkers that are sort of "generic".
For personal filtering, nothing beats a good bayesian filter. I use POPFile myself and it's approaching 99% accuracy and I _LOVE_ it.
Spam very, very rarely makes it past, and if it does, it's the generic "check out this site" type message with no other information. Even spammers trying this technique aren't having much success as I'm seeing less and less of it (maybe 1 or 2 message a month make it past the filters).
The next step in anti-spam evolution will be spam-scanning software that automatically follows links back to webpages and looks for "spammy" content and tags the message as spam in the email system.
For those out there that havn't tried a bayesian form of filtering yet, give POPFile a try: (http://popfile.sourceforge.net/). Just be sure to read the instructions.
Won't do them any good here. Local bayesian filter. Approaching 99% classification accuracy after 6 months.
:)
Spam doesn't stand a chance
N.
I'm waiting for it to hit the fan over Adobe's product activation that's required on their new "Creative Suite" products (ie: Photoshop and Friends).
Just like XP, you have to let your machine either contact Adobe over the internet, or phone their customer service number to get the activation code that's locked to your individual computer.
Oh yes, and apparently you can only activate twice over the internet, then you HAVE to phone their CSRs to explain why you're not a pirate giving copies to your friends to justify additional installs. Guilty until proven innocent I suppose.
I'm willing to bet there's going to be a backlash against them similar to the Intuit tax activation fiasco.
N.
Sorry, I didn't explain well I suppose - the 256 codes only applies to the specific remote in question (that would be disabled). The receiver itself would never be disabled.
Depending on the size of the code (say, 256 bits), hacking it would be next to impossible unless you wanted to wait hours, if not days to try and brute-force a light that will trigger in a minute or so anyway.
Wouldn't work if it was done right.
I use an infrared remote to access my condo - it uses an infrared remote system that's fairly simple, yet effective:
-The remote (much like garage door openers) uses a rolling pseudo-random number sequence. The remote generates a code based on the next number in sequence from a random number generator seeded with a known key for that particular remote (the main controller needs to be "paired" with the remote before use so it knows what seed each remote is using).
-The controller keeps track of all of the remotes for the building and pre-computes the next 256 valid codes that each remote will generate based on the seed exchanged when the remote is paired.
When a remote triggers, it sends the random number code (NOT the key) by way of infrared to the controller receiver. The controller checks to see if the number the remote transmitted is in the next 256 valid numbers for that particular remote, if so, you get in. If not, you don't and the attempt is logged.
If you press the button more than 256 times (playing with the remote button for example) when you're not around the sensor, none of the precompute codes will match the next time the remote is used and it will be useless until re-paired.
Even if you capture the code being sent from the remote, you won't know the key that the random number generator is using in that particular remote to generate the number sequence, or any of the subsequent numbers that the remote would generate. You'd only capture the code that was sent, and once that was used, it wouldn't work again anyway.
If a remote key is compromised, it's simple to simply deactivate that particular remote key. If the system is brute-force attacked, it can either deactivate the sensor that's being attacked, or just call security to the appropriate location.
N.
Doesn't go deep enough. Want something that can go up to around 500m for any kind of real exploring fun.
N.
They can't put it on the subtitle/angle tracks or they'd just get removed during the rip. Which only leaves actual messing with the MPEG encoded video.
I suppose someone could create a MPG "tweaker" to just decompress/watermark/recompress select blocks on the encoded files to serialize them, but it would be expensive, and would mean DVD-R'ing every disc (compatibility issues, far more expensive production)
Interestingly enough, I saw "Kill Bill" last night, and although I really enjoyed the movie, Miramax's watermarking of the image (red-dot patterns appearing for a frame or two during the movie) was very evident.
They better be careful, if they tamper with my moviegoing experience, I will not be seeing their new releases in the theatre. Putting red dots all over a movie that I've paid $10 or more to see practically guarantees that I won't be in the audience.
I'll either wait to see it on DVD, or...
N.
I own a P800, but I don't plan on upgrading to a P900 - the improvements just aren't that significant in my book to warrant the hundreds of dollars of money, not to mention the time to sell-off the old unit and upgrade.
1 - the screen resolution is the same, but looks larger because the flip is smaller. I have my flip removed anyway, so no big deal.
2 - landscape mode is nice, though there's only a few apps that I need that use it.
3 - firmware can be uploaded into the P800 as well, but it must be done at a dealer. I havn't heard anywhere yet that the P900 is user-upgradable without a trip to the SE service center though that would be very, very nice.
4 through 11 - Nice, but nothing that I'd pay hundreds of dollars to obtain.
I'm going to skip the P900 and probably pick up whatever they're releasing next summer/fall.
I'm supremely dissapointed that they didn't put a better camera into the P900 because the model from the 800 (and which is in the 900) is just barely acceptable in terms of quality. Even a one-megapixel camera would've been very, very nice.
N.
The odd thing is that region control was originally designed mainly to stop "foreign regions" (ie: regions outside the US/Canada) from viewing all of the juicy MPAA wares before they were released in their domestic language.
The reality is that nearly ALL of the regions outside the US/Canada have region coding disabled on their players by default (nobody will buy players in those countries unless the player is region-disabled, so the companies have learned pretty damn quick to give consumers what they want).
So, the net result? Nearly the only place in the world with region-coding enabled on DVD players is... You guessed it, the US/Canada.
Pathetic.
N.
Don't be an ass. The line about "nuking people" was made by moderators on Shacknews.
Gabe didn't say that at all.
How about verifying your roumers before posting them as facts?
N.
Just out of interest, was there info that ATI gave out the Doom 3 alpha?
There's a big difference between "leaked" (ie: gave out), and someone who "stole" it from them by popping a disk into a drive or hacked into a server or something...
N.
A good point...
I still find it odd how the RIAA and MPAA folks persist on calling it "stealing".
People are not "stealing" anything, they're "infringing a copyright".
N.
Part of the windows problem seems to be that startup applications and services all try and execute in parallel, thus thrashing the hell out of the drive and slowing the machine a whole lot.
Why not force (or allow the user to set) a limit on the number of startup applications that execute simultaneously to avoid thrashing the drive and making the startup sluggish.
On the contrary...
I rather think the Republicans aren't all that worried about a "Florida happening again". After all, it did get a Republican into the oval office didn't it...
It's odd though, speaking as a Canadian who has always though that although not perfect, the US electoral system had a fair number of checks and balances, it absolutely blows my mind that this sort of un-checked corporate crap isn't being stopped in it's tracks.
It's like 9/11 gave the politicians and big business license to do whatever the hell they want to with your entire country and the economy, and they're screwing it up at a simply astounding rate. "Patriot" take-away-your rights acts, a court denying a "do-not call list" that 50 MILLION people want for the benefit of a few telemarketing lobbyists, big companies trying to patent even the most trivial of ideas... Where does it end?
I mean, this latest info about a company making machines to support democratic elections that has no "unalterable record", easy bypassing (or complete lack) of database passwords, and executives talking about just printing "system check" on the screen without any actual checking being done because the electoral regulations require a full system check before the system begins recording votes.
Frightening, absolutely frightening...
N.
For what it's worth, they're not having all that much luck.
My ISP's DNS servers have been updated with the "delegation-only" patch so they don't return any wildcarded results. I've also updated my own DNS servers at work with the same result.
The more ISPs that do this, the less money verisign makes until they just drop the idea alltogether.
N.
It's worth noting that recent "broadband" releases in Japan are in the 50-100megabit range to your home. Yes, that means transferring anywhere from 6,000 to 12,000kb per second.
By contrast, my ADSL (Canada) downloads at around 350kb per second, and it's considered fast...
On the upstream side of things, it's around 50kb per second.
We have a looooong way to go in north america, folks...
N.