2008/8 - Users begin switching from encryption to sending lots of plain text. 2008/9 - When it becomes a felony to use any encryption that does not have a back door for the NSA (or RIAA... whichever comes first).
Do we all have a right (by DMCA in US) or otherwise to the encryption we put on our data?
I presume that such a right would be unenumerated and reserved to the people by the Ninth Amendment. There is no current law forbidding the use of encryption as far as I know and any such would be quickly challenged on multiple constitutional grounds.
Does it take a court order or other legal instrument to lawfully break encryption?
No. If they already have legally seized the data or communication through a warrant or other process then they do not require anything else to attempt to break the encryption. Recovering a key on the other hand could be very difficult if the suspect was prudent.
Here is a good legal discussion about encryption and privacy including a link to a paper written by Professor Orin Kerr:
Abstract:
Does encrypting Internet communications create a reasonable expectation of privacy in their contents, triggering Fourth Amendment protection? At first blush, it seems that the answer must be yes: A reasonable person would surely expect that encrypted communications will remain private. In this paper, Professor Kerr explains why this intuitive answer is entirely wrong: Encrypting communications cannot create a reasonable expectation of privacy. The reason is that the Fourth Amendment regulates access, not understanding: no matter how unlikely it is that the government will successfully decrypt ciphertext, the Fourth Amendment offers no protection if it succeeds. As a result, the government does not need a search warrant to decrypt encrypted communications. This surprising result is consistent with Fourth Amendment caselaw: it matches how courts have resolved cases involving the reassembly of shredded documents, recovery of deleted files, and the translation of foreign languages. The Fourth Amendment may regulate government access to ciphertext, but it does not regulate government efforts to translate ciphertext into plaintext.
There are some issues with battery use and design that make your suggestion difficult to implement in practice.
The cells have to be wired in series to allow the minimum wire cross section in the wiring harness and connectors. Even given that, the current demands are high enough that simple pressure contacts between cells will not be sufficient to generate a low enough resistance to prevent thermal heating at the contact and possibly catastrophic damage. In the past when I have built my own high power packs using cylindrical potassium hydroxide based cells (NiCd and NiMH), I etched the cell terminals and soldered directly to them for this very reason.
There is also an issue with using large single cells with a high volume to surface area. The internal resistance can generate considerable heat and the low surface area makes cooling difficult. It is not uncommon for multiple smaller cell stacks to be connected in parallel for this very reason in high current applications.
Cell stack voltages are typically 170 volts or higher in electric cars and there is no real way to turn the cells off during replacement which presents another hazard. High power solar cell installations can have this same issue as well which is why they often keep the voltages lower then optimum from a wiring standpoint.
5: Invent/produce technology that allows other companies to do their business and increase profit w/o having to buy carbon credits. 6: Profit!!!
Would that be such a bad thing?
The problem as I understand it is that carbon trading provides endless opportunity for rent seeking so in that respect the profit comes at an even larger overall cost. Why crush the competition or be profitable through productivity when you can control the legislative process? Most of our regulation schemes suffer from that anyway so maybe on a larger scale it is an insignificant problem.
I would have gone with a straight carbon tax despite how taxes are often abused but I am not knowledgeable enough to know where the collected funds would need to go for the best results. The obvious place would be to support conservation and less polluting energy generation. As a bonus, it would have created the structure to add a Middle-East importation tax (would the WTO have a cow over that if it was to fight carbon pollution?) to pay for our military expenditures there and encourage other energy and petroleum sources although given the fungible nature of oil I am not sure how well that would work in practice.
There is a significant environmental difference when you deal with powered circuitry in a vacuum verses an electrically insulating atmosphere. The space station not only travels through a good vacuum but any gases present will tend to be an an ionized state and conduct so it is not necessarily a matter of avoiding physical contact from the panel to a conductive area of the space suit.
I suspect the best defense would be a conductive suit similar to that worn by linemen who work on power lines. I do not know if the space suits have a conductive layer for protection but if they do it is possible it was not intended for an ionized space environment that includes exposed high power 160 volt circuits.
Which Flash memory manufacturers were you thinking of?
The Flash memory integrators who are using Flash chips for storage devices are using the same notation that the hard drive manufacturers are using. The chip manufacturers themselves however are using the notation that the semiconductor memory manufacturers are using except they do not count the area reserved for error correcting data in the actual storage capacity.
In some respects this is a little odd because the actual number of bits stored in the array is not a power of 2 when you count the extra error correcting bits (64 bytes per 2048 byte block) but that is only at the lowest level and the pages and blocks themselves are addressed in powers of 2. It is just the page size itself which is not a power of 2.
I guess the lesson here is that it is the interface that matters. This is especially the case for low pin count parallel interfaces (DRAM and SRAM) where initially the chip organization was driven more by the cost of the package pins then chip layout considerations. If your interface relies on powers of 2, then a kword (whatever the word size happens to be) is 1024 of something. Maybe the hard drive manufacturers should have switched to binary coded decimal.
A typical bulk NAND Flash device is specified like this: (copied from a Micron datasheet)
So for this device not counting the page level error correcting bytes:
4 Gb = 4096 blocks x 64 pages/block x 2048 bytes/page = 4294967296 bytes
You might also note that the block size of 128K is equal to 131072 bytes.
I remember when the hard drive manufacturers switched to base 10 notation and I do not remember anybody who thought it was done for anything except marketing reasons. It made their own drives look larger then the drives of their competitor's who had not switched yet.
My oscilloscope dithers the clock rate so even though it's 350 MHz, it acts cleaner because it moves back and forth across a signal, to evade aliasing artifacts.
Your oscilloscope uses random repetitive sampling to reconstruct repetitive waveforms which general audio is not. There is a trigger circuit which serves to align the sampling clock with the presumed start of the waveform so that new samples are placed into the correct place. This technique allows full reconstruction of a high bandwidth repetitive signal using a relatively slow sampling system. Oscilloscope sampling rates however have gotten very high (2 gigasamples per second and higher) using interleaved fast converters which allows single shot sampling at rates previously only available through repetitive or sequential sampling.
. . . what forbids comcast, your ISP from just changing a legitimate TCP packet to a simple RST (instead of injecting extra packets) . ..
Nothing except that it requires the hardware to be inline to the connection where currently the hardware is connected off of a monitor port or similar. It is cheaper from a cost and performance perspective to be outside of the routed path.
P2P protocol designers are pretty agile and clever. In the face of regular faked TCP RST bits on a connection, they'll evolve the protocol to make shorter connections, and to make repeated attempts to reconnect when an unexpected RST is received. Expect tuning "knobs" in clients very soon now, on how resilient to make the connections or how many bytes to transfer before tearing down and rebuilding the connection.
Since presumably traffic analysis without inspection is being used to identify which TCP connections to kill, tunneling over a any TCP connection will not work because the attack works against TCP itself and short of a customized IP stack or an external firewall, the forged RST packets can not be ignored. From what I can tell, IPSEC authenticated header mode will not work either because the state control flags are considered mutable not to mention the address translation issues. IPSEC is resistant to inspection and IP/port forging but not RST based DoS attacks. Either tunneling over UDP or straight UDP would work however because RST has no meaning for a stateless protocol. SCTP might work but I do not know the low level details well enough to be sure.
What is really needed is something like TCP with the appropriate flags authenticated to prevent
Magnetic bearings have their place but it is not in this sort of application. They excel where the physical contact of normal bearings would cause low lifetime at high speed but in other applications normal bearing are just so much easier to build, use, and replace that their limited lifetimes are not significant.
The space station failure is probably either related to a temperature coefficient mismatch between two parts that are now rubbing or physical damage. A magnetic bearing would not specifically solve either problem.
This is nothing new at Intel or any comparable semiconductor manufacturer who owns their own fabrication plants. Bob Colwell discussed this in a presentation he gave at Stanford. Intel has separate design teams to handle new designs and refinements to existing designs. The later teams are often linked with process technology or fabrication plants because it is very very expensive to have a new process become available for use in production while having nothing available to take advantage of it.
When I write to the trusted friend and gently try to explain the pitfalls of mass address sharing or how to use BCC, they invariably respond with a "huh?", or get all offended and never speak again.
Where do you get this "before anything useful is allowed to happen" crap? That's not what they're doing. They are snuffing out long-lived transactions. If you've transferred twenty or thirty megs from a single host, that's hardly "before anything useful is allowed to happen". If you can cite stories to the contrary, then do so.
My own measurements show Comcast forging the RST packet to force the TCP connection to terminate about 30 seconds after the connection is initiated. It can easily happen quickly enough that no meaningful data is exchanged and robust clients will continuously reconnect only to be reset 15 to 30 seconds later each time. If the torrent is only seeded from clients through Comcast, it is effectively dead despite multiple seeders. This does not happen on all torrents but the pattern of what is blocked and what is not is difficult to discern.
Unfortunately, most current secure transport schemes were only designed to protect data from eavesdropping, not to protect against denial of service attacks against the connection. For example, SSL and TLS both need to be layered above a reliable transport layer (usually TCP), and it is TCP itself that Comcast is attacking.
In examining various ways to ameliorate the problem Comcast is causing I looked into a number of techniques including IPSEC which supports header authentication separate from encryption when used in transport mode but the 8 TCP flag bits are considered mutable for connection management reasons and may be altered maliciously.
Tunneling works of course because the entire IP packet including the header is encrypted and/or authenticated but the tunnel itself is subject to attack in the same way that the SSL/TLS tunnel is as you point out.
Actually, they're not blocking an entire protocol.
They're actively resetting ANY TCP connection that involves uploading significant amounts of data for more than a few seconds.
I enjoyed the statement in the article about Comcast technically not blocking anything:
When the BitTorrent clients receive the false reset packets, they themselves terminate the connection, as they think the other host has told them to go away. Thus, through sneaky techniques and network-level false statements, Comcast is able to trick users' software into terminating their own transfers.
I worry that if done wrong, legislation will be passed that even forbids QoS, which will make things really bad for both users and ISPs. The legislation would have to have wording that QoS is OK as long as the "bottom of the barrel" protocols are able to use full bandwidth when no one else is using the network.
I anticipate the same problems. The legislation will either be too broad and overreaching or so narrow that it will cause as many problems as it attempts to solve. Legislate effectively concerning technology? Ha! The confusion between QoS and Net Neutrality already exists.
Technological solutions like filtering RST packets, packet authentication (Hmmm. This might actually work as another post pointed out but it can still interfere with QoS.), and packet encryption will have the effect of making legitimate QoS more difficult and less effective.
Slashdot Mirror
2008/8 - Users begin switching from encryption to sending lots of plain text.
2008/9 - When it becomes a felony to use any encryption that does not have a back door for the NSA (or RIAA... whichever comes first).
I presume that such a right would be unenumerated and reserved to the people by the Ninth Amendment. There is no current law forbidding the use of encryption as far as I know and any such would be quickly challenged on multiple constitutional grounds.
Does it take a court order or other legal instrument to lawfully break encryption?
No. If they already have legally seized the data or communication through a warrant or other process then they do not require anything else to attempt to break the encryption. Recovering a key on the other hand could be very difficult if the suspect was prudent.
Here is a good legal discussion about encryption and privacy including a link to a paper written by Professor Orin Kerr:
Abstract:
http://volokh.com/posts/1157133639.shtml
There are some issues with battery use and design that make your suggestion difficult to implement in practice.
The cells have to be wired in series to allow the minimum wire cross section in the wiring harness and connectors. Even given that, the current demands are high enough that simple pressure contacts between cells will not be sufficient to generate a low enough resistance to prevent thermal heating at the contact and possibly catastrophic damage. In the past when I have built my own high power packs using cylindrical potassium hydroxide based cells (NiCd and NiMH), I etched the cell terminals and soldered directly to them for this very reason.
There is also an issue with using large single cells with a high volume to surface area. The internal resistance can generate considerable heat and the low surface area makes cooling difficult. It is not uncommon for multiple smaller cell stacks to be connected in parallel for this very reason in high current applications.
Cell stack voltages are typically 170 volts or higher in electric cars and there is no real way to turn the cells off during replacement which presents another hazard. High power solar cell installations can have this same issue as well which is why they often keep the voltages lower then optimum from a wiring standpoint.
Don't they have internal compensation for that?
This is just delayed crosstalk which his research helps to prevent or at least accommodate.
5: Invent/produce technology that allows other companies to do their business and increase profit w/o having to buy carbon credits.
6: Profit!!!
Would that be such a bad thing?
The problem as I understand it is that carbon trading provides endless opportunity for rent seeking so in that respect the profit comes at an even larger overall cost. Why crush the competition or be profitable through productivity when you can control the legislative process? Most of our regulation schemes suffer from that anyway so maybe on a larger scale it is an insignificant problem.
I would have gone with a straight carbon tax despite how taxes are often abused but I am not knowledgeable enough to know where the collected funds would need to go for the best results. The obvious place would be to support conservation and less polluting energy generation. As a bonus, it would have created the structure to add a Middle-East importation tax (would the WTO have a cow over that if it was to fight carbon pollution?) to pay for our military expenditures there and encourage other energy and petroleum sources although given the fungible nature of oil I am not sure how well that would work in practice.
. . . it will be called Skynet, Borg, or the Matrix, depending upon which type of nerd you are.
Is your list inclusive? I would have picked Colossus or maybe Proteus myself.
There is a significant environmental difference when you deal with powered circuitry in a vacuum verses an electrically insulating atmosphere. The space station not only travels through a good vacuum but any gases present will tend to be an an ionized state and conduct so it is not necessarily a matter of avoiding physical contact from the panel to a conductive area of the space suit.
I suspect the best defense would be a conductive suit similar to that worn by linemen who work on power lines. I do not know if the space suits have a conductive layer for protection but if they do it is possible it was not intended for an ionized space environment that includes exposed high power 160 volt circuits.
I am not clear how serious you are being so I will sum up my argument in three words:
Ornithology recapitulates physiology.
I made a small mistake:
4 Gb = 4096 blocks x 64 pages/block x 2048 bytes/page = 4294967296 bytes
should have been:
4 Gb = 4096 blocks x 64 pages/block x 2048 bytes/page x 8 bits/byte = 4294967296 bits
Which Flash memory manufacturers were you thinking of?
The Flash memory integrators who are using Flash chips for storage devices are using the same notation that the hard drive manufacturers are using. The chip manufacturers themselves however are using the notation that the semiconductor memory manufacturers are using except they do not count the area reserved for error correcting data in the actual storage capacity.
In some respects this is a little odd because the actual number of bits stored in the array is not a power of 2 when you count the extra error correcting bits (64 bytes per 2048 byte block) but that is only at the lowest level and the pages and blocks themselves are addressed in powers of 2. It is just the page size itself which is not a power of 2.
I guess the lesson here is that it is the interface that matters. This is especially the case for low pin count parallel interfaces (DRAM and SRAM) where initially the chip organization was driven more by the cost of the package pins then chip layout considerations. If your interface relies on powers of 2, then a kword (whatever the word size happens to be) is 1024 of something. Maybe the hard drive manufacturers should have switched to binary coded decimal.
A typical bulk NAND Flash device is specified like this: (copied from a Micron datasheet)
Page size x8: 2,112 bytes (2,048 + 64 bytes)
Block size: 64 pages (128K + 4K bytes)
Plane size: 2,048 blocks
Device size: 4 Gb: 4096 blocks; 8 Gb: 8,192 blocks; 16 Gb: 16,384 blocks
So for this device not counting the page level error correcting bytes:
4 Gb = 4096 blocks x 64 pages/block x 2048 bytes/page = 4294967296 bytes
You might also note that the block size of 128K is equal to 131072 bytes.
I remember when the hard drive manufacturers switched to base 10 notation and I do not remember anybody who thought it was done for anything except marketing reasons. It made their own drives look larger then the drives of their competitor's who had not switched yet.
My oscilloscope dithers the clock rate so even though it's 350 MHz, it acts cleaner because it moves back and forth across a signal, to evade aliasing artifacts.
Your oscilloscope uses random repetitive sampling to reconstruct repetitive waveforms which general audio is not. There is a trigger circuit which serves to align the sampling clock with the presumed start of the waveform so that new samples are placed into the correct place. This technique allows full reconstruction of a high bandwidth repetitive signal using a relatively slow sampling system. Oscilloscope sampling rates however have gotten very high (2 gigasamples per second and higher) using interleaved fast converters which allows single shot sampling at rates previously only available through repetitive or sequential sampling.
So they are repeating what they have been doing to gun dealers who had a box of fender washers sitting in their garage. Great. Now what?
Nothing except that it requires the hardware to be inline to the connection where currently the hardware is connected off of a monitor port or similar. It is cheaper from a cost and performance perspective to be outside of the routed path.
P2P protocol designers are pretty agile and clever. In the face of regular faked TCP RST bits on a connection, they'll evolve the protocol to make shorter connections, and to make repeated attempts to reconnect when an unexpected RST is received. Expect tuning "knobs" in clients very soon now, on how resilient to make the connections or how many bytes to transfer before tearing down and rebuilding the connection.
Since presumably traffic analysis without inspection is being used to identify which TCP connections to kill, tunneling over a any TCP connection will not work because the attack works against TCP itself and short of a customized IP stack or an external firewall, the forged RST packets can not be ignored. From what I can tell, IPSEC authenticated header mode will not work either because the state control flags are considered mutable not to mention the address translation issues. IPSEC is resistant to inspection and IP/port forging but not RST based DoS attacks. Either tunneling over UDP or straight UDP would work however because RST has no meaning for a stateless protocol. SCTP might work but I do not know the low level details well enough to be sure.
What is really needed is something like TCP with the appropriate flags authenticated to prevent
If you want a real thrill ask them to open you DVDs or even better anything in those vacuum formed plastic containers.
So you are an identity as well as a hard drive thief! That is my name and address!
Err, wait. You have my apologies. I misread my ID which says Jules Vern.
Magnetic bearings have their place but it is not in this sort of application. They excel where the physical contact of normal bearings would cause low lifetime at high speed but in other applications normal bearing are just so much easier to build, use, and replace that their limited lifetimes are not significant.
The space station failure is probably either related to a temperature coefficient mismatch between two parts that are now rubbing or physical damage. A magnetic bearing would not specifically solve either problem.
This is nothing new at Intel or any comparable semiconductor manufacturer who owns their own fabrication plants. Bob Colwell discussed this in a presentation he gave at Stanford. Intel has separate design teams to handle new designs and refinements to existing designs. The later teams are often linked with process technology or fabrication plants because it is very very expensive to have a new process become available for use in production while having nothing available to take advantage of it.
I do not have any trusted friends left either.
2008: Allow us more choices.
Vote for competence and well-reasoned judgment in '08.
But . . . but . . . what if I want to vote?
My own measurements show Comcast forging the RST packet to force the TCP connection to terminate about 30 seconds after the connection is initiated. It can easily happen quickly enough that no meaningful data is exchanged and robust clients will continuously reconnect only to be reset 15 to 30 seconds later each time. If the torrent is only seeded from clients through Comcast, it is effectively dead despite multiple seeders. This does not happen on all torrents but the pattern of what is blocked and what is not is difficult to discern.
In examining various ways to ameliorate the problem Comcast is causing I looked into a number of techniques including IPSEC which supports header authentication separate from encryption when used in transport mode but the 8 TCP flag bits are considered mutable for connection management reasons and may be altered maliciously.
Tunneling works of course because the entire IP packet including the header is encrypted and/or authenticated but the tunnel itself is subject to attack in the same way that the SSL/TLS tunnel is as you point out.
I enjoyed the statement in the article about Comcast technically not blocking anything:
When the BitTorrent clients receive the false reset packets, they themselves terminate the connection, as they think the other host has told them to go away. Thus, through sneaky techniques and network-level false statements, Comcast is able to trick users' software into terminating their own transfers.
I anticipate the same problems. The legislation will either be too broad and overreaching or so narrow that it will cause as many problems as it attempts to solve. Legislate effectively concerning technology? Ha! The confusion between QoS and Net Neutrality already exists.
Technological solutions like filtering RST packets, packet authentication (Hmmm. This might actually work as another post pointed out but it can still interfere with QoS.), and packet encryption will have the effect of making legitimate QoS more difficult and less effective.