It is trivial for anybody with the resources to build a faster than light drive, too - for some definitions of "with the resources".
It would depend upon whom this guy wishes to protect his conversations against - J. Random Carbonunit or Special Agent TLA.
If the former, than the encryption used in GSM is enough - few people have the gear to modulate and demodulate a GSM signal with proper time slotting, time of flight correction, etc. Making a GSM signal is HARD - I build gear that does it.
If the latter, then they won't screw around picking the signal off the air - they will throw a CALEA intercept on his phone when it hits the PTSN. Then the only thing that can protect him would be VERY strong encryption seperate from the phone - which as I said in my first posting is difficult due to the nature of digital phones.
Lastly, if he is trying to protect himself from Special Agent TLA, encrypting his signal like this won't help - it will just raise a big red flag saying "Look At Me! I Am Hiding SomeThing!". He would be far better served making an innocuous word code and using that.
GSM (and PCS) phones encrypt the traffic anyway (at least they do outside the USA).
That is one of the big advantages of digital cellular modes over older, analog cellular modes - the ease of adding encryption.
However, if you want to throw another layer on top of this, it gets more difficult - since digital phones take the audio signal and vocode it, you cannot just scramble your voice and feed it in - the vocoder won't know what to do with it and won't encode it properly. You would have to inject your signal after the vocoder but before the Viterbi/Trellis coding.
Hmmmm. Given the amount of bandwidth Windows Update takes, I wonder how much of AOL's bandwidth it takes.
Hmmm...
On a related note: I haven't looked recently, but it used to be that Windows clients were TERRIBLE about DNS lookups - they would not cache anything, and were always making DNS requests on every little thing. I was helping a FOF set up his DSL, and his DNS lookups were taking 3-5 seconds, because his ISP's name servers (swbell) were overloaded. We finally set up his own internal name server, and set it to do the name lookups itself - time went from 5 seconds to <.1 second. (Yes, this puts some more work on the root servers, but not much, as his name server will cache the locations of the TLD servers).
I wonder if MS is setting the proxy:nocache header in the HTTP reply, or if their client is always doing a reload rather than an if-modified-since.
Perhaps allowing Squid to be configured to ignore proxy:nocache and to convert reload into IMS based on an ACL would allow a site admin to tweak around this without breaking other sites>
Not filtering. Disconnection.
on
AOL Sues Spammers
·
· Score: 2, Insightful
Most people who want the backbone providers to "do something" about spam don't want the backbone providers to filter.
They want the backbone providers to pull the plug on the mainsleaze spammers directly connected to them.
They want the backbone providers to insist that the Tier-(N+1:N>=1) providers to enforce their TOS. Failing that, they want the backbone providers to pull the plug on those who support spamming.
Do me a favor - read the March Linux Journal (which contains my letter to the editor).
Then write LJ in response to their response to my response to Rackspace's ad.
AOL is suing a Norton spammer
on
AOL Sues Spammers
·
· Score: 4, Interesting
One of the defendants would appear to be one of the myriad pirated Norton/Symantec spammers (George Moore, Maryland Internet Marketing of Maryland, and 14 of their advertising affiliates. Spam Content: software products (www.getnortonhere.net))
Question: could/would Symantec join in this suit, or better still bring copyright violation and (ahem)piracy charges against this fool?
I have long held the belief that Symantec does not more aggressively crack down on all the Norton spammers because once somebody has purchased an unauthorized copy of Norton, they will have to pay Symantec for updates. Thus, Symantec makes money on the subscription fees and doesn't have to mess around with actually making a disk, printing a manual, etc.
Upon reflection I thought I would add this: in my own experience, I have found:
1) If you have a decent amount of bandwidth between server and client, X does better than VNC in terms of update rate. For example, over a 10 or 100MBit LAN connection X runs faster than VNC.
2) If the apps are older apps that do more of the work server-side, then X does better than VNC.
3) If the apps are more recent apps that do a lot of XRender stuff, then VNC does better over low bandwidth connections. This seems to be more because VNC can "skip" updates, and wait until the screen stablizes, while X will push all the operations across the pipe.
Hold those who host spammers responsible
on
AOL Sues Spammers
·
· Score: 4, Interesting
Read my journal - the most recent entry as of this writing is about my writing to Linux Journal and raising the point that Rackspace (who has been taking out full page ads in LJ) are very spam friendly.
In my journal, one person responded about her experiences as a Rackspace customer.
One thing we can do is to make it VERY public that places like Rackspace, Verio, UUNET etc. are unwilling to do anything to enforce their own Terms Of Service against spam. Granted, if you follow the various anti-spamming news groups you will know this, but most PHBs don't follow the anti-spamming newsgroups.
But if LJ gets flooded with people calling RackedWaste to task, then it is possible that it might catch the eye of potential SpamSpace customers. Who knows? It might even catch the eye of the marketing group at SpamWaste and they might, just might, start pushing to enforce their TOS.
I have also read the myriad of other articles examine the likelyhood of a stranglet forming. Since collision events many orders of magnitude greater in energy occur all the time in the upper atmosphere, were the strangelet runaway scenario likely we would already be a hyperdense sphere 100 meters in diameter.
Also, the odds that we will be able to create better nanobots than nature has is equally remote - as any nanobot that could "grey goo" the world would have to work against quite an entropy gradient, and were that possible some microbe would have stumbled across it by now.
"Stumbled across it" - that's a very good phrase to keep in mind when reading this sort of thing. Do you really think that just because N scientists don't follow a line of research that all scientists won't?
I would far rather have responsible scientists examining the world, finding the "bad things", and allowing mankind to deal with them, than to allow one scientist working for the only people who will fund him (this would be the "naughty people") to come across the idea first.
As I stated before - all too often the people who most preach "self control" are really more interested in "other control" - after all, were they interested in self control they would control themselves.
I think that in this case, either he a) believes that this won't effect him personnaly, or b) he doesn't care. And that sickens me, because the truely great astronomers, physicists, geneticists, and other scientists believed that while there are evil men, man can be trusted.
And look at it like this: if we cannot trust man, then we are screwed because that's what we are.
I find it interesting that this man is an astronomer. I guess he figures that his particular branch of science will never be considered "dangerous" and need to be "limited", unlike those other blighters in physics.
One trick around this is to call the war rooms "Break rooms".
However, the best approach is to convince management of the need for these rooms, so that they will support you.
I suggest the use of hidden cameras, prostitutes, and extortion. Also effectatious are blunt force trauma, cattle prods, and capsacin coated toilet paper in the executive bathroom.
You want "war rooms" - a room with a whiteboard, a door, perhaps a water cooler, and a network drop or wireless LAN.
You need several. Don't allow them to be "reserved" - no sign up sheets for these. These are not "conference rooms". These are places your people can go to hash things out on an ad-hoc basis.
You need an absolute ban on speaker phones.
You should discourage anybody from using speakers on their computer - encourage headphone use (at a reasonable volume level).
It still will suck. I went from an office with a door that I could close to a cube farm, and it gets very hard to concentrate. The only benefit cubes have over offices is that management can change things around whenever they feel bored.
That would depend upon the nature of the app you are running - if you are running simple text apps X protocol is MUCH faster because all that gets sent across the link is "Draw this text in this font at this location", not a bunch of pixels.
Granted, if you have some app that is doing XRender on the client side then VNC might be faster, but that is as much the app's fault as the protocol.
Run a tcpdump (or better still use Ethereal) and watch what your favorite apps do.
Reading the lengths to which you must go to get a remote display on your Windows machine amazes me.
Give me the same basic hardware, but rip WinCE out and put a lightweight X server into it, and I could remote the display on my workstation without any software changes on it at all (except perhaps for adding a line to my X0.hosts file).
AND if the table spoke SSH, I wouldn't even have to do that.
AND the fact that I could also redirect the displays of my SGI, my other server, my service monitor, and anything else that spoke X Windows system protocol.
For all you naysayers who poop-poo the need for network transparency in your GUI, I say:
I've a friend who works for Coleman (he's one of their webmonkeys, let's give him a heart attack by putting some "referrer=http://slashdot.org"'s in his logs).
He gets plenty of fun stuff from work. One of the things they have is a little crank powered flashlight that uses a Brinkman style little incand.
It will run for a fair length of time on the batteries, but I have to wonder how much longer it would run with an LED.
Slashdot Mirror
And if I just had a cup of hot tea, or coffee, or am nervous, or if my traveling companion just said something horribly embarrassing, or....
I was not saying it was trivial - the person to whom I was responding said it was.
I was saying quite the contrary - it WASN'T trivial.
You may notice that I compared "on their heads" vs "IN their nose"
ON vs IN are two very different things.
As George Carlin said, "You can get ON the plane, I'M going to get IN the plane!"
It is trivial for anybody with the resources to build a faster than light drive, too - for some definitions of "with the resources".
It would depend upon whom this guy wishes to protect his conversations against - J. Random Carbonunit or Special Agent TLA.
If the former, than the encryption used in GSM is enough - few people have the gear to modulate and demodulate a GSM signal with proper time slotting, time of flight correction, etc. Making a GSM signal is HARD - I build gear that does it.
If the latter, then they won't screw around picking the signal off the air - they will throw a CALEA intercept on his phone when it hits the PTSN. Then the only thing that can protect him would be VERY strong encryption seperate from the phone - which as I said in my first posting is difficult due to the nature of digital phones.
Lastly, if he is trying to protect himself from Special Agent TLA, encrypting his signal like this won't help - it will just raise a big red flag saying "Look At Me! I Am Hiding SomeThing!". He would be far better served making an innocuous word code and using that.
GSM (and PCS) phones encrypt the traffic anyway (at least they do outside the USA).
That is one of the big advantages of digital cellular modes over older, analog cellular modes - the ease of adding encryption.
However, if you want to throw another layer on top of this, it gets more difficult - since digital phones take the audio signal and vocode it, you cannot just scramble your voice and feed it in - the vocoder won't know what to do with it and won't encode it properly. You would have to inject your signal after the vocoder but before the Viterbi/Trellis coding.
You don't put the laser on the robot dogs head.
You put the laser in the dog's nose.
Hmmmm. Given the amount of bandwidth Windows Update takes, I wonder how much of AOL's bandwidth it takes.
.1 second. (Yes, this puts some more work on the root servers, but not much, as his name server will cache the locations of the TLD servers).
Hmmm...
On a related note: I haven't looked recently, but it used to be that Windows clients were TERRIBLE about DNS lookups - they would not cache anything, and were always making DNS requests on every little thing. I was helping a FOF set up his DSL, and his DNS lookups were taking 3-5 seconds, because his ISP's name servers (swbell) were overloaded. We finally set up his own internal name server, and set it to do the name lookups itself - time went from 5 seconds to <
I wonder if MS is setting the proxy:nocache header in the HTTP reply, or if their client is always doing a reload rather than an if-modified-since.
Perhaps allowing Squid to be configured to ignore proxy:nocache and to convert reload into IMS based on an ACL would allow a site admin to tweak around this without breaking other sites>
Most people who want the backbone providers to "do something" about spam don't want the backbone providers to filter.
They want the backbone providers to pull the plug on the mainsleaze spammers directly connected to them.
They want the backbone providers to insist that the Tier-(N+1:N>=1) providers to enforce their TOS. Failing that, they want the backbone providers to pull the plug on those who support spamming.
Great - with both AOL and Symantec suing this moron, maybe the number spams I get for Symantec will go down!
....
Now, nuke all the elargement spams, credit spams,
Sigh. Well, there went that moment of elation.
Do me a favor - read the March Linux Journal (which contains my letter to the editor).
Then write LJ in response to their response to my response to Rackspace's ad.
One of the defendants would appear to be one of the myriad pirated Norton/Symantec spammers (George Moore, Maryland Internet Marketing of Maryland, and 14 of their advertising affiliates. Spam Content: software products (www.getnortonhere.net))
Question: could/would Symantec join in this suit, or better still bring copyright violation and (ahem)piracy charges against this fool?
I have long held the belief that Symantec does not more aggressively crack down on all the Norton spammers because once somebody has purchased an unauthorized copy of Norton, they will have to pay Symantec for updates. Thus, Symantec makes money on the subscription fees and doesn't have to mess around with actually making a disk, printing a manual, etc.
Upon reflection I thought I would add this: in my own experience, I have found:
1) If you have a decent amount of bandwidth between server and client, X does better than VNC in terms of update rate. For example, over a 10 or 100MBit LAN connection X runs faster than VNC.
2) If the apps are older apps that do more of the work server-side, then X does better than VNC.
3) If the apps are more recent apps that do a lot of XRender stuff, then VNC does better over low bandwidth connections. This seems to be more because VNC can "skip" updates, and wait until the screen stablizes, while X will push all the operations across the pipe.
Read my journal - the most recent entry as of this writing is about my writing to Linux Journal and raising the point that Rackspace (who has been taking out full page ads in LJ) are very spam friendly.
In my journal, one person responded about her experiences as a Rackspace customer.
One thing we can do is to make it VERY public that places like Rackspace, Verio, UUNET etc. are unwilling to do anything to enforce their own Terms Of Service against spam. Granted, if you follow the various anti-spamming news groups you will know this, but most PHBs don't follow the anti-spamming newsgroups.
But if LJ gets flooded with people calling RackedWaste to task, then it is possible that it might catch the eye of potential SpamSpace customers. Who knows? It might even catch the eye of the marketing group at SpamWaste and they might, just might, start pushing to enforce their TOS.
I have also read the myriad of other articles examine the likelyhood of a stranglet forming. Since collision events many orders of magnitude greater in energy occur all the time in the upper atmosphere, were the strangelet runaway scenario likely we would already be a hyperdense sphere 100 meters in diameter.
Also, the odds that we will be able to create better nanobots than nature has is equally remote - as any nanobot that could "grey goo" the world would have to work against quite an entropy gradient, and were that possible some microbe would have stumbled across it by now.
"Stumbled across it" - that's a very good phrase to keep in mind when reading this sort of thing. Do you really think that just because N scientists don't follow a line of research that all scientists won't?
I would far rather have responsible scientists examining the world, finding the "bad things", and allowing mankind to deal with them, than to allow one scientist working for the only people who will fund him (this would be the "naughty people") to come across the idea first.
As I stated before - all too often the people who most preach "self control" are really more interested in "other control" - after all, were they interested in self control they would control themselves.
I think that in this case, either he a) believes that this won't effect him personnaly, or b) he doesn't care. And that sickens me, because the truely great astronomers, physicists, geneticists, and other scientists believed that while there are evil men, man can be trusted.
And look at it like this: if we cannot trust man, then we are screwed because that's what we are.
I didn't say that his branch of science wouldn't be affected - I said that I thought that he thought that it would not be, which is a different thing.
All science is interconnected - anybody who doubts that should watch Connections.
And anybody who thinks that research in any given area won't happen "because it could be dangerous" is foolish.
I am just saying that this particular instance of Class Scientist might not think he would be affected by such "self-restraint".
Funny, how those how most preach "self-restraint" usually mean "other-restraint".
I find it interesting that this man is an astronomer. I guess he figures that his particular branch of science will never be considered "dangerous" and need to be "limited", unlike those other blighters in physics.
One trick around this is to call the war rooms "Break rooms".
However, the best approach is to convince management of the need for these rooms, so that they will support you.
I suggest the use of hidden cameras, prostitutes, and extortion. Also effectatious are blunt force trauma, cattle prods, and capsacin coated toilet paper in the executive bathroom.
You want "war rooms" - a room with a whiteboard, a door, perhaps a water cooler, and a network drop or wireless LAN.
You need several. Don't allow them to be "reserved" - no sign up sheets for these. These are not "conference rooms". These are places your people can go to hash things out on an ad-hoc basis.
You need an absolute ban on speaker phones.
You should discourage anybody from using speakers on their computer - encourage headphone use (at a reasonable volume level).
It still will suck. I went from an office with a door that I could close to a cube farm, and it gets very hard to concentrate. The only benefit cubes have over offices is that management can change things around whenever they feel bored.
That would depend upon the nature of the app you are running - if you are running simple text apps X protocol is MUCH faster because all that gets sent across the link is "Draw this text in this font at this location", not a bunch of pixels.
Granted, if you have some app that is doing XRender on the client side then VNC might be faster, but that is as much the app's fault as the protocol.
Run a tcpdump (or better still use Ethereal) and watch what your favorite apps do.
Reading the lengths to which you must go to get a remote display on your Windows machine amazes me.
Give me the same basic hardware, but rip WinCE out and put a lightweight X server into it, and I could remote the display on my workstation without any software changes on it at all (except perhaps for adding a line to my X0.hosts file).
AND if the table spoke SSH, I wouldn't even have to do that.
AND the fact that I could also redirect the displays of my SGI, my other server, my service monitor, and anything else that spoke X Windows system protocol.
For all you naysayers who poop-poo the need for network transparency in your GUI, I say:
BEHOLD
A segment override is a lot cheaper than a seperate instruction on every test.
And protecting the code from modification is much better done by protecting the pages themselves in the page table than by using segementation tricks.
What if AOL were to go to the Tier-1 ISPs that fail to enforce their terms of service against spammers, and say,
<voice character="ED-209">Your customers are in violation of your terms of service. You will terminate them. You have 15 hours to comply.</voice>
And should they fail to comply, null-route those Tier-1's at AOL's border routers.
What do you think Exodus, Verio, and UUNet would do when they faced the very real possiblity of being blocked from AOL?
And Mozilla itself shall be renamed "Mercury Marauder" - it is big, heavy, powerful, and not quite as expensive as the competition.
Meanwhile, rumors from Redmond indicate they have begun on a project codenamed "Ricer" - no additional data found.
I've a friend who works for Coleman (he's one of their webmonkeys, let's give him a heart attack by putting some "referrer=http://slashdot.org"'s in his logs).
He gets plenty of fun stuff from work. One of the things they have is a little crank powered flashlight that uses a Brinkman style little incand.
It will run for a fair length of time on the batteries, but I have to wonder how much longer it would run with an LED.
Perhaps next generation....