My 2cents police radio use is not for point 2 point communications but broadcast communication so that everyone on the team maintains an image of whats going on. Police have always had alternate methods of communicating sensitive information off the radio even if that was only cell phones.
In this context my concern is not that encrypting broadcast is a bad thing but that encryption will be seen as an excuse for being lazy and not using point to point communication systems to convey operationally sensitive information.
Even if the encryption were 100% perfect and you had perfect operational security there are "alleged" bad guys routinely being escorted to station in the back seats of these vechicles.
TFA is an embodiement of everything wrong with public global warming debate. There are two generally separate issues.
1. What is the problem? 2. What should be done about it?
You are entitled to have an "opinion" on what should be done about it.
You are not entitled to an opinion with regards to objective reality. You may disagree with models, datasets, methodologies..etc these disagreements must be specific and substantive.
"In fact, a large and growing number of distinguished scientists and engineers do not agree that drastic actions on global warming are needed"
Scientists have no business making such statements in the first place. It is the job of politicians and the public to answer what if anything should be done with the projections and predictions provided by scientists.
This is an issue on both sides of the coin... IPCC et al need to refrain from in any way stating action is needed and just focus on making predictions of possible futures with their confidence in those predictions. Let the reader figure out what is needed based on the evidence provided by scientists.
Injecting political opinion into scientific issues hurts everyone. It makes people mistrust scientists and lowers the threshold for politicians ignoring scientific advice they receive when it is inconvenient for them.
Global warming is not the only issue under the "climate change" umbrella which TFA and most other political opinion pieces attacking global warming summarily ignore.
If you don't want to worry about global warming think of the oceans which >1billion people depend for food... real troubling things going on right now... shelled creatures melting away, 25% of reefs gone, issues with plankton at the very bottom of food chain gone. Evidence directly linking acidification to human produced carbon.
IANAL. Conspiracy to commit a felony can be punished pretty severely as is evidenced by this situation. Some people will argue that this tramples rights because you cannot even read something without risk of going to jail.
Why is downloading information any more an indicator of intent than purchasing a couple 30lb bags of fertilizer from your local garden store?
Wouldn't you want them arrested BEFORE they killed you?
Head-of-line blocking may happen at different levels.
In the case of pipelined HTTP, if the first request in the pipeline is slow, it will block all the others, because answers have to be delivered in order.
From what I understand they are partially re-inventing TCP in order to mux and demux transactions within a single TCP session.
In my opinion all that really matters is that pipelines are kept busy transmitting data 100% of the time and never stalled on ack windows or waiting for RTTs involved with the browser to figure out what to ask for next. Stuffing new requests while existing ones are still executing to mask RTT delays is good enough if your smart about what your asking for.
Multiple TCP sessions effectivly work around head-of-line at *all* layers as well as mitigate ordering dependancies. You don't need to proritize as much as you need hurestics in the browser to ensure it selects content that will keep the pipeline busy masking RTTs.
It would be interesting to see not just ideal marketing benchmarks of SPDY but how it compares to alternatives.
If we are going to go for a new protocol why not go big with a new IP protocol. All most people ever really want was a fast multi-stream (optionally) reliable message passing protocol anyway. Yea it would take a while to catch on but if it was good enough (without limitations of SCTP) eventually 20 years from now it will happen.
Another thing worth mentioning HTTP is not really the problem with delay nowadays it is the ad networks... and javascript that goes as far as checking to make sure ad content is loaded before content the user cares about is displayed.
No, HTTP suffers from head-of-line blocking. There is no way around that. HTTP also doesn't have header compression (which matters quite a bit on low-bandwidth pipes) nor prioritization to ensure that your HTTP, javascript and CSS load before the lolkittens:)
Please tell me your joking.
SPDY is layered on top of TCP... **TCP** suffers from head-of-line blocking and therefore so does SPDY.
By collapsing everything into a single connection you induce *more* latency on low speed, high latency connections because the chance of idling the link when waiting for an ACK/response with one TCP session or waiting for the next request increases vs multiple concurrent TCP sessions.
You don't need to invent a new protocol for header compression if you intend to always use TLS anyway. That is stupid... Just enable compression within TLS.
I would love to see a prioritzation scheme that works but the cold hard truth is you won't get much better than simple hurestics that don't require prioritization anyway. Dependancy graph is discovered late and simply isn't knowable ahead of time. You can do something like PGO to figure it out but this only works with static content.
If we are going to bother with a new protocol it needs to be a new *IP* layer protocol or at least layered on UDP... Anything less is akin to purchasing "Internet accelerator software" from an infomercial.
Just to restate: You don't care if a diagnostic tool is less deterministic and specific as long as you don't have to learn anything new to deal with new situations (like dualstack). Alright, thing I've got it now.
If you didn't catch the hint from my previous message your 100% correct I simply don't want to know or care whether a site is IPv6, IPv4 or some combination of both... Why should I?
I want a diagnostic tool that follows RFC 3484 not some crappy artifact from ancient history when IPv6 support in the linux kernel was experimental and 3484 did not exist. I also want a tool that allows me to explicitly define the address family of my choice should I ever have a reason to care.
"you don't have to learn anything new to deal with new situations "
Ding Ding Ding... this is the whole fucking point of RFC 3484.
I remember in previous incidents the TSA has fined people people who fail to complete screening and decide to go the other way and leave the airport instead. Will they do the same to Rand Paul? If not why?
Not only did MegaUpload not delete the actual files when sent DMCA notices (but did when sent abuse letters about illegal content like child porn)
This is not necessary. If you read the DMCA it is enough to simply remove *access* to the content.
This was almost all the times used for spreading copyright infringing material and MegaUpload was notoriously known for being good site for such use.
The Internet is notoriously known for being a good method of transporting such material. What is your point? I've used megaupload many times over the years but never to download movies or cracked software.
As the internal emails show they were also fully aware of this fact.
This is problematic...
not only did MegaUpload staff know about this activity and try to get around DMCA notices and laws, they did copyright infringement themselves
Actually the number of possible states of a single qubit is uncountably infinite (according to quantum mechanics). Of course the number of results of any given complete measurement of that qubit is exactly two, but then, the number of measurements you can do on it is uncountably infinite. In theory, at least. In practice the number may be very high, but certainly finite, because you cannot make your measurement apparatus settings infinitely precise.
Thank you, my mistake seems to be the variation in ways the question can be prepared has no bearing on the count of possible output states.
Is this a better analogy?
I want to add 1 + 1 without revealing to the cloud I don't know what 1 + 1 is fearing I will be rained on if it found out.
My question is prepared with two parameters (1 + x) and (1 + y) where x and y are random numbers I want added.
The random numbers are known only to me and can be as large as ones own imagination almost uncountably infinite.
The cloud returns an answer. I subtract x and y from the clouds answer to get my 2.
Is this system just a useful version of the scheme above?
Religion is the most dangerous thing facing our population, not overpopulation. They all claim to be peaceful, but criticize them - and you'll see their true nature.
Religion is just one of many tools men who seek to impose their agendas on others use to get their way. If not religion it will be something else to fill the vacuume.
I think the word "perfect" is too strong. A system is only as good as its weakest link.
Quantum systems are not able to provide any guarantees WRT to *what* the system is entangled with.
You still need "classical" source of trust to bind the quantum system to do anything useful. (See MITM..quantum proxy server..)
Questions:
What prevents the replacement of the quantum cloud service with an attacker who intentionally provides wrong answers?
Or simply ignores a request pretending they did not get it in a bid to gain additional knowledge about the question? (Thereby revealing distribution)
How do you prevent information leakage WRT state density and count of qubits needed to perform a computation?
Anyway I'm still a little fuzzy on what they mean by "The user prepares qubits â" the fundamental units of quantum computers â" in a state known only to himself and sends these qubits to the quantum computer"
It seems like they are tweaking the distribution, transfering the system to the quantum processor and reading back an outcome. What is hidden to the computer is the initial probability distribution which is needed to understand outcomes in a useful context? Does this sound right or am I missing something?
If it is right how is this really any different than asking a million questions, collecting a million answers and only choosing the one you intended to ask in the first place?
I think they should take the oil out and leave it there as long as possible until it sinks below the surface. Then salvage, turn into a reef or divers amusement park.
In the good ole days car wrecks were prominently displayed on side of public roads to serve as a warning.
I say not only keep it but spend time and money to secure it in place as long as possible.
uh yeah.. that little part about the kernel? that's critical.. can't have development on an open source os if the hardware requires a signed kernel to boot
Every package a modern linux distribution downloads nowadays is "signed". Whats the big deal? Even if you compile your own kernel from source it is one extra thing make does at the end.
The core issue is having the ability to define certs your computer is willing to trust. If you can do that as TFA infers you can do anything you want including develop an intermediate handler which loads an untrusted operating system.
They are just completely fucked, and I actually feel kind of bad for them. I mean, they theoretically "own" this nice slice of wireless spectrum, which they bought at considerable expense, and they want to do something pretty cool with it, but they're not allowed to because some other industry has been illegally bleeding into their spectrum for years and now it's too late to fix it.
Here we go again... and again...and again. A grant to spectrum comes with fine print dictating how it can be used. You simply do not have the right to do whatever you want as long as it remains in-band..
Their existing allocation was provided under the ATC integrated services rule which explicitly prohibits proliferation of ground stations.
What they did was apply for an exemption to the rule they had purchased their spectrum under which has thus far not been granted due to interference concerns.
"considerable expense" is peanuts compared to what an AT&T or a verizon would pay at auction for the same bandwidth nationally. Why does LS deserve a loophole?
I've been known to piss on requirements in specifications from time to time because they subvert my interests or they have effects I believe to be more harmful than helpful.
All secure boot does is give the computer some assurance whatever it is handing off control to can be trusted.
There is no technical way for UEFI or anything else to enforce signed drivers in the form of modules loaded dynamically at runtime. If the kernel is blessed by the computer these "requirements" are simply empty words on a page that can and will be ignored with impunity.
Any action under SOPA/PIPA requires a Court Order. Which you won't get by pressing a button and filling out a form.
You can't just read a bill and assume the words of it mean anything. Bills apply to reality. You must account for all aspects of human nature and chains of possible effects when determining whether a bill is good or bad.
Section 104 and 105 give immunity to people who act to block sites, funds, products, transactions...etc they themselves think are infringing with no due process consequences be damned. No way that will ever be abused by competitors or provide legal cover for actions against outfits like wikileaks.
Actually, you won't have to worry about it in any case, unless your site is based outside the USA, and you're not a US resident. In either of those cases, current law allows legal relief, and SOPA/PIPA don't deal with you at all....
Section 201 makes you a criminal for performing "happy birthday" if the cost the copyright owner sets per ear exceeds 1k. If you record someone singing the whole thing good luck with getting a court of law to deem this anything other than willfull infringement.
I want Rossi to produce working e-cats because he looks like a crackpot, has the historical baggage of a crackpot, makes extraordinary claims and does not appear to give two shits about the theoretical basis for his work.
Everyone who fell over each other dismissing Pons and Fleischmans sloppy work with sloppy reproductions of their own deserve to be strung up and publically ridiculed by a character like Rossi even if he turns out to be the crackpot he seems to be.
Except that ping is not an app, it is a diagnostic tool. If you're in a dual-stack situation, you want to know whether a host is reachable over IPv4 or over IPv6 independently. You could do this with a -4 and -6 flag to ping, but then you'd need to type two more characters for the v6 version and three more for the v4 version.
No I don't. I don't give a flying rats ass who is dualstacked. All I want to know is why x is not working. By ping being the only app...err "diagnostic tool" which does not follow the same addressing standard everything else on my system uses to access network resources my job is harder not easier.
In France here, a new mobile operator appeared, giving a lot more of technical informations than its competitor. It explained it could not accept more than ~100 open TCP connection per user, because it only has 8000 IPv4 addresses for its mobile network and expects 3 millions users. If users use more than 100 ports, he will have an exhaustion of IP/port combinations. We finally reach the point where IPv4, even using NAT techniques, is becoming impractical. The switch should happen soon.
While I want IPv4 gone yesterday this is not true. The touple used by NATSs are unique combinations of src ip, dst ip, src port and dst port. You can reuse the same port on the same source IP as long as the destination address is different.
The reason providers are doing port allocation schemes is so that if you have port information in a trace it will tie back to the end user for liability purposes. However even here the same idea applies it is not 100 open TCP connections it is 100 open TCP connections to the *same* destination address.
What the IPv6-people just refuse to understand is that there is zero benefit for running IPv6 now.
What the IPv6 naysayers just refuse to understand is that we have no choice. NAT works great for you because you have at least one public IPv4 address that you control.
The problem with this thinking is there are real consequences to running out of IPv4 addresses.
When you push NAT out to the carrier and that IP address is serving hundreds of customers then what? If you think setting up DNS or using torrent software or skype that does not bounce content through strangers systems was hard just wait till you want to publish anything through said carrier NAT.
I think most IPv6 people are quite happy to move on without you. Comcast is deploying to millions. All major ISPs have active trials. Asia is going crazy you should see all the crap being pushed through softwires at the moment... IPv6 only content coming soon to a theatre near you...like it or not it is happening with or without you.
RFC1112 muticast space needs L2/L3 multicast mapping traceroute utilities which has been around longer than I would like to admit... grrr.. i'm old...
I think we were talking about different things. I was talking about the difference between ping and traceroute that appear in Linux. Neither utility has any ability to ping an L2 address. Neither utility has any multicast support unless pinging the broadcast/multicast address counts as multicast support and all of the L2 details are abstracted by the kernel.
I'm sure your fancy "l3 switch" can trace through STP or whatever holds your L2 together but none of the above is relevent to ping and traceroute on linux.
Like I said before all traceroute amounts to is a bunch of ping incrementing the TTL each time. It is not rocket science.
Slashdot Mirror
My 2cents police radio use is not for point 2 point communications but broadcast communication so that everyone on the team maintains an image of whats going on. Police have always had alternate methods of communicating sensitive information off the radio even if that was only cell phones.
In this context my concern is not that encrypting broadcast is a bad thing but that encryption will be seen as an excuse for being lazy and not using point to point communication systems to convey operationally sensitive information.
Even if the encryption were 100% perfect and you had perfect operational security there are "alleged" bad guys routinely being escorted to station in the back seats of these vechicles.
It is disingenuous to claim US does not have the ordinance to destroy Iranian underground facilities. It clearly does.
The most important question is not "how" but "why".
TFA is an embodiement of everything wrong with public global warming debate. There are two generally separate issues.
1. What is the problem?
2. What should be done about it?
You are entitled to have an "opinion" on what should be done about it.
You are not entitled to an opinion with regards to objective reality. You may disagree with models, datasets, methodologies..etc these disagreements must be specific and substantive.
"In fact, a large and growing number of distinguished scientists and engineers do not agree that drastic actions on global warming are needed"
Scientists have no business making such statements in the first place. It is the job of politicians and the public to answer what if anything should be done with the projections and predictions provided by scientists.
This is an issue on both sides of the coin... IPCC et al need to refrain from in any way stating action is needed and just focus on making predictions of possible futures with their confidence in those predictions. Let the reader figure out what is needed based on the evidence provided by scientists.
Injecting political opinion into scientific issues hurts everyone. It makes people mistrust scientists and lowers the threshold for politicians ignoring scientific advice they receive when it is inconvenient for them.
Global warming is not the only issue under the "climate change" umbrella which TFA and most other political opinion pieces attacking global warming summarily ignore.
If you don't want to worry about global warming think of the oceans which >1billion people depend for food... real troubling things going on right now... shelled creatures melting away, 25% of reefs gone, issues with plankton at the very bottom of food chain gone. Evidence directly linking acidification to human produced carbon.
Al-Zaida Terror1sts develop tricobalt salted antimatter bomb. Antardic ice sheet vaporized. Millions of penguins now homeless.
IANAL. Conspiracy to commit a felony can be punished pretty severely as is evidenced by this situation. Some people will argue that this tramples rights because you cannot even read something without risk of going to jail.
Why is downloading information any more an indicator of intent than purchasing a couple 30lb bags of fertilizer from your local garden store?
Wouldn't you want them arrested BEFORE they killed you?
No.
Head-of-line blocking may happen at different levels.
In the case of pipelined HTTP, if the first request in the pipeline is slow, it will block all the others, because answers have to be delivered in order.
From what I understand they are partially re-inventing TCP in order to mux and demux transactions within a single TCP session.
In my opinion all that really matters is that pipelines are kept busy transmitting data 100% of the time and never stalled on ack windows or waiting for RTTs involved with the browser to figure out what to ask for next. Stuffing new requests while existing ones are still executing to mask RTT delays is good enough if your smart about what your asking for.
Multiple TCP sessions effectivly work around head-of-line at *all* layers as well as mitigate ordering dependancies. You don't need to proritize as much as you need hurestics in the browser to ensure it selects content that will keep the pipeline busy masking RTTs.
It would be interesting to see not just ideal marketing benchmarks of SPDY but how it compares to alternatives.
If we are going to go for a new protocol why not go big with a new IP protocol. All most people ever really want was a fast multi-stream (optionally) reliable message passing protocol anyway. Yea it would take a while to catch on but if it was good enough (without limitations of SCTP) eventually 20 years from now it will happen.
Another thing worth mentioning HTTP is not really the problem with delay nowadays it is the ad networks... and javascript that goes as far as checking to make sure ad content is loaded before content the user cares about is displayed.
As one of the creators of SPDY..
No, HTTP suffers from head-of-line blocking. There is no way around that. :)
HTTP also doesn't have header compression (which matters quite a bit on low-bandwidth pipes) nor prioritization to ensure that your HTTP, javascript and CSS load before the lolkittens
Please tell me your joking.
SPDY is layered on top of TCP... **TCP** suffers from head-of-line blocking and therefore so does SPDY.
By collapsing everything into a single connection you induce *more* latency on low speed, high latency connections because the chance of idling the link when waiting for an ACK/response with one TCP session or waiting for the next request increases vs multiple concurrent TCP sessions.
You don't need to invent a new protocol for header compression if you intend to always use TLS anyway. That is stupid... Just enable compression within TLS.
I would love to see a prioritzation scheme that works but the cold hard truth is you won't get much better than simple hurestics that don't require prioritization anyway. Dependancy graph is discovered late and simply isn't knowable ahead of time. You can do something like PGO to figure it out but this only works with static content.
If we are going to bother with a new protocol it needs to be a new *IP* layer protocol or at least layered on UDP... Anything less is akin to purchasing "Internet accelerator software" from an infomercial.
Just to restate: You don't care if a diagnostic tool is less deterministic and specific as long as you don't have to learn anything new to deal with new situations (like dualstack). Alright, thing I've got it now.
If you didn't catch the hint from my previous message your 100% correct I simply don't want to know or care whether a site is IPv6, IPv4 or some combination of both... Why should I?
I want a diagnostic tool that follows RFC 3484 not some crappy artifact from ancient history when IPv6 support in the linux kernel was experimental and 3484 did not exist. I also want a tool that allows me to explicitly define the address family of my choice should I ever have a reason to care.
"you don't have to learn anything new to deal with new situations "
Ding Ding Ding... this is the whole fucking point of RFC 3484.
I remember in previous incidents the TSA has fined people people who fail to complete screening and decide to go the other way and leave the airport instead. Will they do the same to Rand Paul? If not why?
Not only did MegaUpload not delete the actual files when sent DMCA notices (but did when sent abuse letters about illegal content like child porn)
This is not necessary. If you read the DMCA it is enough to simply remove *access* to the content.
This was almost all the times used for spreading copyright infringing material and MegaUpload was notoriously known for being good site for such use.
The Internet is notoriously known for being a good method of transporting such material. What is your point? I've used megaupload many times over the years but never to download movies or cracked software.
As the internal emails show they were also fully aware of this fact.
This is problematic...
not only did MegaUpload staff know about this activity and try to get around DMCA notices and laws, they did copyright infringement themselves
Very problematic...
Actually the number of possible states of a single qubit is uncountably infinite (according to quantum mechanics). Of course the number of results of any given complete measurement of that qubit is exactly two, but then, the number of measurements you can do on it is uncountably infinite. In theory, at least. In practice the number may be very high, but certainly finite, because you cannot make your measurement apparatus settings infinitely precise.
Thank you, my mistake seems to be the variation in ways the question can be prepared has no bearing on the count of possible output states.
Is this a better analogy?
I want to add 1 + 1 without revealing to the cloud I don't know what 1 + 1 is fearing I will be rained on if it found out.
My question is prepared with two parameters (1 + x) and (1 + y) where x and y are random numbers I want added.
The random numbers are known only to me and can be as large as ones own imagination almost uncountably infinite.
The cloud returns an answer. I subtract x and y from the clouds answer to get my 2.
Is this system just a useful version of the scheme above?
Religion is the most dangerous thing facing our population, not overpopulation. They all claim to be peaceful, but criticize them - and you'll see their true nature.
Religion is just one of many tools men who seek to impose their agendas on others use to get their way. If not religion it will be something else to fill the vacuume.
Your close, it's not a million though, it is probably more like uncountably infinite.
How do you figure? I would have thought given the state of todays useless "quantum" computers a million is more than generous.
I think the word "perfect" is too strong. A system is only as good as its weakest link.
Quantum systems are not able to provide any guarantees WRT to *what* the system is entangled with.
You still need "classical" source of trust to bind the quantum system to do anything useful. (See MITM..quantum proxy server..)
Questions:
What prevents the replacement of the quantum cloud service with an attacker who intentionally provides wrong answers?
Or simply ignores a request pretending they did not get it in a bid to gain additional knowledge about the question? (Thereby revealing distribution)
How do you prevent information leakage WRT state density and count of qubits needed to perform a computation?
Anyway I'm still a little fuzzy on what they mean by "The user prepares qubits â" the fundamental units of quantum computers â" in a state known only to himself and sends these qubits to the quantum computer"
It seems like they are tweaking the distribution, transfering the system to the quantum processor and reading back an outcome. What is hidden to the computer is the initial probability distribution which is needed to understand outcomes in a useful context? Does this sound right or am I missing something?
If it is right how is this really any different than asking a million questions, collecting a million answers and only choosing the one you intended to ask in the first place?
This stupid temperature debate sucks up all the oxygen in the room while all of the really important environmental issues are summarily ignored.
I think they should take the oil out and leave it there as long as possible until it sinks below the surface. Then salvage, turn into a reef or divers amusement park.
In the good ole days car wrecks were prominently displayed on side of public roads to serve as a warning.
I say not only keep it but spend time and money to secure it in place as long as possible.
uh yeah.. that little part about the kernel? that's critical.. can't have development on an open source os if the hardware requires a signed kernel to boot
Every package a modern linux distribution downloads nowadays is "signed". Whats the big deal? Even if you compile your own kernel from source it is one extra thing make does at the end.
The core issue is having the ability to define certs your computer is willing to trust. If you can do that as TFA infers you can do anything you want including develop an intermediate handler which loads an untrusted operating system.
They are just completely fucked, and I actually feel kind of bad for them. I mean, they theoretically "own" this nice slice of wireless spectrum, which they bought at considerable expense, and they want to do something pretty cool with it, but they're not allowed to because some other industry has been illegally bleeding into their spectrum for years and now it's too late to fix it.
Here we go again... and again...and again. A grant to spectrum comes with fine print dictating how it can be used. You simply do not have the right to do whatever you want as long as it remains in-band..
Their existing allocation was provided under the ATC integrated services rule which explicitly prohibits proliferation of ground stations.
What they did was apply for an exemption to the rule they had purchased their spectrum under which has thus far not been granted due to interference concerns.
"considerable expense" is peanuts compared to what an AT&T or a verizon would pay at auction for the same bandwidth nationally. Why does LS deserve a loophole?
I've been known to piss on requirements in specifications from time to time because they subvert my interests or they have effects I believe to be more harmful than helpful.
All secure boot does is give the computer some assurance whatever it is handing off control to can be trusted.
There is no technical way for UEFI or anything else to enforce signed drivers in the form of modules loaded dynamically at runtime. If the kernel is blessed by the computer these "requirements" are simply empty words on a page that can and will be ignored with impunity.
Any action under SOPA/PIPA requires a Court Order. Which you won't get by pressing a button and filling out a form.
You can't just read a bill and assume the words of it mean anything. Bills apply to reality. You must account for all aspects of human nature and chains of possible effects when determining whether a bill is good or bad.
Section 104 and 105 give immunity to people who act to block sites, funds, products, transactions...etc they themselves think are infringing with no due process consequences be damned. No way that will ever be abused by competitors or provide legal cover for actions against outfits like wikileaks.
Actually, you won't have to worry about it in any case, unless your site is based outside the USA, and you're not a US resident. In either of those cases, current law allows legal relief, and SOPA/PIPA don't deal with you at all....
Section 201 makes you a criminal for performing "happy birthday" if the cost the copyright owner sets per ear exceeds 1k. If you record someone singing the whole thing good luck with getting a court of law to deem this anything other than willfull infringement.
I want Rossi to produce working e-cats because he looks like a crackpot, has the historical baggage of a crackpot, makes extraordinary claims and does not appear to give two shits about the theoretical basis for his work.
Everyone who fell over each other dismissing Pons and Fleischmans sloppy work with sloppy reproductions of their own deserve to be strung up and publically ridiculed by a character like Rossi even if he turns out to be the crackpot he seems to be.
Except that ping is not an app, it is a diagnostic tool. If you're in a dual-stack situation, you want to know whether a host is reachable over IPv4 or over IPv6 independently. You could do this with a -4 and -6 flag to ping, but then you'd need to type two more characters for the v6 version and three more for the v4 version.
No I don't. I don't give a flying rats ass who is dualstacked. All I want to know is why x is not working. By ping being the only app...err "diagnostic tool" which does not follow the same addressing standard everything else on my system uses to access network resources my job is harder not easier.
In France here, a new mobile operator appeared, giving a lot more of technical informations than its competitor. It explained it could not accept more than ~100 open TCP connection per user, because it only has 8000 IPv4 addresses for its mobile network and expects 3 millions users. If users use more than 100 ports, he will have an exhaustion of IP/port combinations. We finally reach the point where IPv4, even using NAT techniques, is becoming impractical. The switch should happen soon.
While I want IPv4 gone yesterday this is not true. The touple used by NATSs are unique combinations of src ip, dst ip, src port and dst port. You can reuse the same port on the same source IP as long as the destination address is different.
The reason providers are doing port allocation schemes is so that if you have port information in a trace it will tie back to the end user for liability purposes. However even here the same idea applies it is not 100 open TCP connections it is 100 open TCP connections to the *same* destination address.
What the IPv6-people just refuse to understand is that there is zero benefit for running IPv6 now.
What the IPv6 naysayers just refuse to understand is that we have no choice. NAT works great for you because you have at least one public IPv4 address that you control.
The problem with this thinking is there are real consequences to running out of IPv4 addresses.
When you push NAT out to the carrier and that IP address is serving hundreds of customers then what? If you think setting up DNS or using torrent software or skype that does not bounce content through strangers systems was hard just wait till you want to publish anything through said carrier NAT.
I think most IPv6 people are quite happy to move on without you. Comcast is deploying to millions. All major ISPs have active trials. Asia is going crazy you should see all the crap being pushed through softwires at the moment... IPv6 only content coming soon to a theatre near you...like it or not it is happening with or without you.
RFC1112 muticast space needs L2/L3 multicast mapping traceroute utilities which has been around longer than I would like to admit... grrr.. i'm old...
I think we were talking about different things. I was talking about the difference between ping and traceroute that appear in Linux. Neither utility has any ability to ping an L2 address. Neither utility has any multicast support unless pinging the broadcast/multicast address counts as multicast support and all of the L2 details are abstracted by the kernel.
I'm sure your fancy "l3 switch" can trace through STP or whatever holds your L2 together but none of the above is relevent to ping and traceroute on linux.
Like I said before all traceroute amounts to is a bunch of ping incrementing the TTL each time. It is not rocket science.