Openly admitting your security is based on obscurity sounds a little strange IMHO.
Instead of using a secret protocol, plainly give out the necessary certifiates only via email and kill them off after abuse. Especially since everybody can use the Skpe API to spam if he wants.
We have no idea what you are doing and what the office is doing. If no central database or file access is required, then a server for backup seems fine to me.
You mean, lets say, any Muslim in traditional outfit can enter a plane and because he is afraid of flying pray "Allah is great", without any fearing to be removed from the plane? You mean, one can wear a t-shirt with "fuck [name of local governor]" on it and police will threat him like anybody else? You mean one can not be imprisoned without seeing a lawyer for saying something which the police *considers* to be a terrorist threat.
I am glad to hear that.
There are no political forces who want to have a "kill switch" on the Internet?
I appreciate that the USA and the western world in general is very free right now, but i urge you to be careful when making comparisons, or take this for granted. Good chess players watch their own defense.
Its more complicated than "the Chinese" "censoring the net" for "political reasons". Its that han-Chinese in some regions are not tolerated so well in the local population - to say it nicely. Its sometimes less of a political conflict, but more a race problem. Sometimes external organizations give support for political reasons, but it ends up for stimulating demonstrations - which are close to racial nationalism - during which people have been killed, not by the police, but by the protesters, telling that these protests where *not* peaceful sit-ins on the streets, but violent unrest. Stimulating unrest *is* a crime in many European states, and i have no doubt that it could happen that your web page may be put off line.
If you want to have a confirmation that feeling suppressed by a central government does not make people nicer or more understanding for human rights, look to the Baltic states. Some of these try to put the considerable Russian speaking population (people who also had no say in going there) into a serious disadvantage, up to forbidding to speak the Russian language in public places.
So while i hope the Chinese government finds a way to deal with this in a constructive manner, i have to say: maybe the time to solve this complicated problem has just not yet come. Maybe they figured that if they use this "kill switch" deaths can be prevented. While we may dispute about it (and i would), this idea is not *so* far from some things i have heard in Europe/the US.
Disclaimer: i worked on QC, but not for dwave (although one of my former employers had dealing with them):
dwave does not aim to build a machine to crack codes. They build a machine which can do what can be done using the technology available now.
The normal ideas about how to make a QC work, even for lets say factoring 128 bit numbers require many more logical qubits than available. The logical qubits should be composed from physical qubits, and for all coding schemes besides some quite new ones the minimum error rate of the phycical qubits needs to 100 times lower than what its normally right now, and even then you need 10s to 1000s of physical qubits PER logical qubit. this means, we need 1000s to 100000s of physical qubits on the chip with error rates of 10^-4 - 10^-3 per operation. Right now most approaches at QC are at 1-10 qubits with.1%-10% error rate.
What does dwave do? They plainly select the things which their qc should do very clever, and skip everything for which real quantum coherence is required. In some sense they build an analog computer for a certain class of problems. The way in which they construct it uses quantum mechanics to overcome the performance of a comparable purely classical machine.
However this machine is quite a different QC from the other QCs proposed. It wont provide exponential speedup on breaking codes
How to try co catch a foothold in a communication based market by giving the message to the world that anybody using anything else than windows should see if he can live with google talk or competitors. yes, thats you, android and iphone users; better switch now and send the skype contacts your new address.
> We have lived without such technology for generations.
Wow. Thats your argument?
We did not have mobile phones for generations. We did not have cars for generations. We were not even able to sustain cities so dense that a lot of people collide for generations. We were not able to support a population dense enough that transport is needed for generations.
How does this fit with your argument?
The cost of a single person killed or heavily injured in a car accident is very high, even if you dont consider that the person may suffer pain. If we build a technology to solve a certain problem (here: sustain a high transport density at low costs), then we should do our best to optimize that technology.
To the sociopaths who think that the highways are free in the sense of doing a sport there, i say: if i want to play golf i go to the golf yard, If i want to do sport shooting, i go to the shooting lane; if i want to swim i go to the swimming pool. So if you want to test how fast you can drive, the go to the next racetrack. Most of these are used only a few times fer year and under the week they are quite cheap (or acceptable at least) to rent for a few rounds.
Yes, and that may happen. They can plainly ask who drove this truck before, and ask where that guy is (he is *obviously not* in the sleeper now.). If the truckers are unlucky the police will have 2 posts (actually i was told by an informed source it is done like that) in 200km distance
this is something the judges in the court have long asked for. After an accident its difficult to establish howfar the accident was caused by the behavior of driver or circumstances - e.g. - did the driver go 150km/h for 6h without a break or did he feel compelled by somebody driving 1m to his rear end to go faster just at the location of the accident? I also think its ok to confiscate the record if the driver was caught speeding or stopped in a control because its suspected that he drove to long (a mechanical recorder to prevent speeding/going without a break is mandatory for trucks in Germany, and in general the experiences seem to be quite good).
What would *not* be ok would be any function where the police can ask "list all drivers who did this or that". There is no way to prevent this from being used to track people, e.g. by setting up a 50cm long speed regulated zone in the database to get all driver passing this point.
The mandatory things would be:
*encryption, where the keys are stored in a way that they can be only recomposed either from the owners/driver (the driver can e.g. insert an electronic license) keyring (to defend himself), or from several institutions agreeing and providing the key for a specific case.
*a legal framework which highly discourages institutions from even trying to abuse this data
*no network connection of the device. The memory should be a removable part, which is secured by a normal lock/seal. Implementations which do not contain the data only in this removable part should be forbidden.
*the only normally accessible interface should be a port used to set the currently used license (this is, downloading a public key from it).
I think actually, some technologies should fall under export control laws. Like weapons. But as long as its legal to sell it there is little point to this.
Yes and no. If your capable it staff figured out that something specific which is needed and should work really does not work and that it is a problem of the preinstalled software/hardware then it if a difference if you are put trough to the "did you check it is plugged in" customer phone support, which will file a case which will rot in the depths of their database unless 10000 other customers have the same problem, or if you actually have a support contract.
The logics behind that is simple: each effort you make has positive and negative consequences. All levels on the orders of some mSv/year are *safe* in the sense that if you apply that radiation to 10000 people, you probably wont be able to see the effects - even statistically. This means if you have a population of 100000 where you have to decide to move them - or not - you have to take into account that the adverse effects may impose a bigger problem than the radiation. The psychological stress and possible loss of the workplace will probably cause an rate of death via suicide or alcoholism which far outweighs the effects of increasing the radiation level from 2mSv/year to 4mSv/year.
However it may make sense to keep the rule not to try to expose 30000000 people to such a radiation levels. And this is exactly why i think that the persons responsible for not venting the reactors/the building immediately should go to prison.
if you know somebody who organizes a lab course in physics, in a university you can ask if you can take the geiger counter there and compare it to their calibrated samples. Typically there is a box of sealed test samples (well locked away), which have well defined radiation doses in different gamma-ranges, so you can test the sensitivity. However, you will have to take an safety instruction to even touch the box. So if you know somebody there well, he may help you. He may even tell you how to calibrate the device correctly using that sample. Another way, which is less technically challenging and will not give you a quantitative calibration is to use one of the typical stones which radiate stronger. Refer to any standard textbook which these are in you region. Look e.g. for granite on wikipedia and follow to the original sources. However none of these means will provide you with any information about the sensitivity of the counter.
As for your friend trying to measure food: More than a quantitative comparison "this radiates stronger than that" will not be possible. The data will be problematically low for the prescribed doses if the counter has no good integrator/long term counter and is stable. Any quantitative measurement of contamination with isotopes is completely unrealistic outside the lab and with an inexperienced operator, especially if the device has no energy resolution. A simple workaround around the latter would be insert materials with different absorption coefficients into the path and compare the measurements, but i cant tell how well that works. Moreover 100-1000Bq/kg is not much. I doubt you manage to get more than a count rate of 1-10clicks per second from a sample of acceptable size. which means that in order to get a 10Percent resolution you may have to integrate over 100seconds or more. That means that the dark count rate should be acceptably stable.
If your friend does this to protect the own health, i recommend the following: don't do it. There are two possibilities: either the food in monitored professionally and marked correctly (which i believe is normally the case in Japan) or its not. If its monitored professionally then there will be no long-term contamination which is undetected. The effect of a spurious peak in one meal to ten or even hundred times of the allowed level wont kill you or have any adverse effects, and reliably i think you will be only able to detect starting from about 10-100 times of the allowed dose. If the food which is not monitored professionally *and* comes from within 50-100km around the reactor then don't eat it, if you have the choice, until the situation stabilized (that is, when any kind of containment, even by a simple plastic foil is reestablished and then after a few months, look at the ieae website). If you believe you must support the farmers there, then donate money, don't buy the food.
An non-reading can also provide you with a false sense of safety, and that is true for all uncontrolled foods. There is no way for a layman to establish safety of a food which comes from within the problematic range around the reactor.
My personal feeling is that *in Japan, which has high food quality in general* an inexperienced operator of a Geiger counter trying to measure his own food will have higher stress due to mis/unclear readings and the constant (lets remember, this may have to be done for 20years if you take it seriously) reminder of the danger just before eating. The adverse health effects of this and possible associated psychological effects (stress before eating) will outweigh the negative effects of getting a higher dose from time to time. If you take the 30min-1h per day which you need to check the food *seriously* for such low doses of radiation, then there are other thing you can do in this hour (go jogging, ride a bike etc.) which will help the body more to develop the immune system.
No, the CEO gets also companies IT experts time to prepare and set up his personal iphone with everything required to comply to - on the management level there is no big difference between personal and private any more; everything which can be done to save his time is good for the company.
If there is a possible security implication by the person with access to all the data will be using an device which is not secured properly, this has big possible implications. Give the CEO a second iphone, if necessary, any other device, or a personal assistent carrying a laptop behind him 24x7.
All of this is better than to explain the shareholders and the stock market why somebody could have access to overall sales numbers which only the board level could have known. The possible damage times multiplied by the risk of an uncontrolled device being hacked rank high enough that your CEO will listen - he will just say "make it work". Otherwise its a good idea if you covered your ass but a written order.
and i am alienated by todays user interfaces. What alienates me most is that showing the keybinding seem to be a thing of the past and pure text menus are not possible to turn on.I like a simple alphabetically sorted list to start apps, which would take less space and not be as weird as having 9 screenful of badly designed, stupidly copied or sometime identical icons. And sorry on the most alarm clocks on smartphones you could instead of the plus easily write "add/set alarm" - no lack of space there.
I dont think that monitoring the data use precisely is a big problem that it would be a big increase. In Germany one company has a 100MB for 2Euro/month package and their estimation (after that: throtteling) did not differ significantly from what i measured on my computer.
If it would so expensive to meter it exacty that offer would not work out.
Slashdot Mirror
Openly admitting your security is based on obscurity sounds a little strange IMHO.
Instead of using a secret protocol, plainly give out the necessary certifiates only via email and kill them off after abuse. Especially since everybody can use the Skpe API to spam if he wants.
We have no idea what you are doing and what the office is doing. If no central database or file access is required, then a server for backup seems fine to me.
So you want to say, half of the conversations gave the police occasion to bully the persons at will?
You mean, lets say, any Muslim in traditional outfit can enter a plane and because he is afraid of flying pray "Allah is great", without any fearing to be removed from the plane? You mean, one can wear a t-shirt with "fuck [name of local governor]" on it and police will threat him like anybody else? You mean one can not be imprisoned without seeing a lawyer for saying something which the police *considers* to be a terrorist threat.
I am glad to hear that.
There are no political forces who want to have a "kill switch" on the Internet?
I appreciate that the USA and the western world in general is very free right now, but i urge you to be careful when making comparisons, or take this for granted. Good chess players watch their own defense.
Its more complicated than "the Chinese" "censoring the net" for "political reasons". Its that han-Chinese in some regions are not tolerated so well in the local population - to say it nicely. Its sometimes less of a political conflict, but more a race problem. Sometimes external organizations give support for political reasons, but it ends up for stimulating demonstrations - which are close to racial nationalism - during which people have been killed, not by the police, but by the protesters, telling that these protests where *not* peaceful sit-ins on the streets, but violent unrest. Stimulating unrest *is* a crime in many European states, and i have no doubt that it could happen that your web page may be put off line.
If you want to have a confirmation that feeling suppressed by a central government does not make people nicer or more understanding for human rights, look to the Baltic states. Some of these try to put the considerable Russian speaking population (people who also had no say in going there) into a serious disadvantage, up to forbidding to speak the Russian language in public places.
So while i hope the Chinese government finds a way to deal with this in a constructive manner, i have to say: maybe the time to solve this complicated problem has just not yet come. Maybe they figured that if they use this "kill switch" deaths can be prevented. While we may dispute about it (and i would), this idea is not *so* far from some things i have heard in Europe/the US.
Did he make sure his emulators use firmware in the right way?
They wanted to write: we found the 3.5 inch fd and its still readable.
Disclaimer: i worked on QC, but not for dwave (although one of my former employers had dealing with them):
dwave does not aim to build a machine to crack codes. They build a machine which can do what can be done using the technology available now.
The normal ideas about how to make a QC work, even for lets say factoring 128 bit numbers require many more logical qubits than available. The logical qubits should be composed from physical qubits, and for all coding schemes besides some quite new ones the minimum error rate of the phycical qubits needs to 100 times lower than what its normally right now, and even then you need 10s to 1000s of physical qubits PER logical qubit. this means, we need 1000s to 100000s of physical qubits on the chip with error rates of 10^-4 - 10^-3 per operation. Right now most approaches at QC are at 1-10 qubits with .1%-10% error rate.
What does dwave do? They plainly select the things which their qc should do very clever, and skip everything for which real quantum coherence is required. In some sense they build an analog computer for a certain class of problems. The way in which they construct it uses quantum mechanics to overcome the performance of a comparable purely classical machine.
However this machine is quite a different QC from the other QCs proposed. It wont provide exponential speedup on breaking codes
Is there anything fundamental which prevents automating the orientation of the package? Or sticking barcodes on several sides of the luggage?
They work, they are cheap, you can stick them on anything,the limits are well known, and if the printer fails you can even send the tag by fax.
while i like USB, i would appreciate a form factor which includes screws, like the sub-d connectors.
do they choose it based on the assumption that there will be no big natural desasters close to this place?
How to try co catch a foothold in a communication based market by giving the message to the world that anybody using anything else than windows should see if he can live with google talk or competitors. yes, thats you, android and iphone users; better switch now and send the skype contacts your new address.
Commodore 64 Porn arrives in high-res
> We have lived without such technology for generations.
Wow. Thats your argument?
We did not have mobile phones for generations. We did not have cars for generations. We were not even able to sustain cities so dense that a lot of people collide for generations. We were not able to support a population dense enough that transport is needed for generations.
How does this fit with your argument?
The cost of a single person killed or heavily injured in a car accident is very high, even if you dont consider that the person may suffer pain. If we build a technology to solve a certain problem (here: sustain a high transport density at low costs), then we should do our best to optimize that technology.
To the sociopaths who think that the highways are free in the sense of doing a sport there, i say: if i want to play golf i go to the golf yard, If i want to do sport shooting, i go to the shooting lane; if i want to swim i go to the swimming pool. So if you want to test how fast you can drive, the go to the next racetrack. Most of these are used only a few times fer year and under the week they are quite cheap (or acceptable at least) to rent for a few rounds.
Yes, and that may happen. They can plainly ask who drove this truck before, and ask where that guy is (he is *obviously not* in the sleeper now.). If the truckers are unlucky the police will have 2 posts (actually i was told by an informed source it is done like that) in 200km distance
You are misinformed, the driver has to write his name onto the card. If you are caught with another drivers card, you are in trouble.
this is something the judges in the court have long asked for. After an accident its difficult to establish howfar the accident was caused by the behavior of driver or circumstances - e.g. - did the driver go 150km/h for 6h without a break or did he feel compelled by somebody driving 1m to his rear end to go faster just at the location of the accident? I also think its ok to confiscate the record if the driver was caught speeding or stopped in a control because its suspected that he drove to long (a mechanical recorder to prevent speeding/going without a break is mandatory for trucks in Germany, and in general the experiences seem to be quite good).
What would *not* be ok would be any function where the police can ask "list all drivers who did this or that". There is no way to prevent this from being used to track people, e.g. by setting up a 50cm long speed regulated zone in the database to get all driver passing this point.
The mandatory things would be:
*encryption, where the keys are stored in a way that they can be only recomposed either from the owners/driver (the driver can e.g. insert an electronic license) keyring (to defend himself), or from several institutions agreeing and providing the key for a specific case.
*a legal framework which highly discourages institutions from even trying to abuse this data
*no network connection of the device. The memory should be a removable part, which is secured by a normal lock/seal. Implementations which do not contain the data only in this removable part should be forbidden.
*the only normally accessible interface should be a port used to set the currently used license (this is, downloading a public key from it).
I think actually, some technologies should fall under export control laws. Like weapons. But as long as its legal to sell it there is little point to this.
I rather think it would be more realistic and appropriate to negotiate for variable pay, depending on the companies income and your performance.
Yes and no. If your capable it staff figured out that something specific which is needed and should work really does not work and that it is a problem of the preinstalled software/hardware then it if a difference if you are put trough to the "did you check it is plugged in" customer phone support, which will file a case which will rot in the depths of their database unless 10000 other customers have the same problem, or if you actually have a support contract.
The logics behind that is simple: each effort you make has positive and negative consequences. All levels on the orders of some mSv/year are *safe* in the sense that if you apply that radiation to 10000 people, you probably wont be able to see the effects - even statistically. This means if you have a population of 100000 where you have to decide to move them - or not - you have to take into account that the adverse effects may impose a bigger problem than the radiation. The psychological stress and possible loss of the workplace will probably cause an rate of death via suicide or alcoholism which far outweighs the effects of increasing the radiation level from 2mSv/year to 4mSv/year.
However it may make sense to keep the rule not to try to expose 30000000 people to such a radiation levels. And this is exactly why i think that the persons responsible for not venting the reactors/the building immediately should go to prison.
if you know somebody who organizes a lab course in physics, in a university you can ask if you can take the geiger counter there and compare it to their calibrated samples. Typically there is a box of sealed test samples (well locked away), which have well defined radiation doses in different gamma-ranges, so you can test the sensitivity. However, you will have to take an safety instruction to even touch the box. So if you know somebody there well, he may help you. He may even tell you how to calibrate the device correctly using that sample. Another way, which is less technically challenging and will not give you a quantitative calibration is to use one of the typical stones which radiate stronger. Refer to any standard textbook which these are in you region. Look e.g. for granite on wikipedia and follow to the original sources. However none of these means will provide you with any information about the sensitivity of the counter.
As for your friend trying to measure food: More than a quantitative comparison "this radiates stronger than that" will not be possible. The data will be problematically low for the prescribed doses if the counter has no good integrator/long term counter and is stable. Any quantitative measurement of contamination with isotopes is completely unrealistic outside the lab and with an inexperienced operator, especially if the device has no energy resolution. A simple workaround around the latter would be insert materials with different absorption coefficients into the path and compare the measurements, but i cant tell how well that works. Moreover 100-1000Bq/kg is not much. I doubt you manage to get more than a count rate of 1-10clicks per second from a sample of acceptable size. which means that in order to get a 10Percent resolution you may have to integrate over 100seconds or more. That means that the dark count rate should be acceptably stable.
If your friend does this to protect the own health, i recommend the following: don't do it. There are two possibilities: either the food in monitored professionally and marked correctly (which i believe is normally the case in Japan) or its not. If its monitored professionally then there will be no long-term contamination which is undetected. The effect of a spurious peak in one meal to ten or even hundred times of the allowed level wont kill you or have any adverse effects, and reliably i think you will be only able to detect starting from about 10-100 times of the allowed dose. If the food which is not monitored professionally *and* comes from within 50-100km around the reactor then don't eat it, if you have the choice, until the situation stabilized (that is, when any kind of containment, even by a simple plastic foil is reestablished and then after a few months, look at the ieae website). If you believe you must support the farmers there, then donate money, don't buy the food.
An non-reading can also provide you with a false sense of safety, and that is true for all uncontrolled foods. There is no way for a layman to establish safety of a food which comes from within the problematic range around the reactor.
My personal feeling is that *in Japan, which has high food quality in general* an inexperienced operator of a Geiger counter trying to measure his own food will have higher stress due to mis/unclear readings and the constant (lets remember, this may have to be done for 20years if you take it seriously) reminder of the danger just before eating. The adverse health effects of this and possible associated psychological effects (stress before eating) will outweigh the negative effects of getting a higher dose from time to time. If you take the 30min-1h per day which you need to check the food *seriously* for such low doses of radiation, then there are other thing you can do in this hour (go jogging, ride a bike etc.) which will help the body more to develop the immune system.
No, the CEO gets also companies IT experts time to prepare and set up his personal iphone with everything required to comply to - on the management level there is no big difference between personal and private any more; everything which can be done to save his time is good for the company.
If there is a possible security implication by the person with access to all the data will be using an device which is not secured properly, this has big possible implications. Give the CEO a second iphone, if necessary, any other device, or a personal assistent carrying a laptop behind him 24x7.
All of this is better than to explain the shareholders and the stock market why somebody could have access to overall sales numbers which only the board level could have known. The possible damage times multiplied by the risk of an uncontrolled device being hacked rank high enough that your CEO will listen - he will just say "make it work". Otherwise its a good idea if you covered your ass but a written order.
and i am alienated by todays user interfaces. What alienates me most is that showing the keybinding seem to be a thing of the past and pure text menus are not possible to turn on.I like a simple alphabetically sorted list to start apps, which would take less space and not be as weird as having 9 screenful of badly designed, stupidly copied or sometime identical icons. And sorry on the most alarm clocks on smartphones you could instead of the plus easily write "add/set alarm" - no lack of space there.
I dont think that monitoring the data use precisely is a big problem that it would be a big increase. In Germany one company has a 100MB for 2Euro/month package and their estimation (after that: throtteling) did not differ significantly from what i measured on my computer.
If it would so expensive to meter it exacty that offer would not work out.