Science should be defended by peer review and by thesis defence, not by challenging it in a court of law, with the possibility of a legal punishment for being wrong, or for producing a politically inconvenient result.
I agree. I am a scientist myself. However, i am also an employee. Instead of letting idiotic morons who believe the earth is 6000years old and relativity is bullshit because its to complicated for them (see Andrew Schlafly) throw mud on me in public, i would rather prefer that they go to court. Because then there is a good chances it hurts them.
Well. did you write to these matrices? I guess you are aware that matlab has something like copy on write - which can lead to funny behaviours. I also find the lack of pass by reference a little disturebing to say the least, but i guess the idea behind is to keeep i teasy to debug
a) The data types in matlab have absolutely nothing to do with the processor you are running on - and that is good - all my ml programs run on all my machines
b) already the 8087 supported 64 bit integer - so no i dont know why matlab does not have it since a long time
c) Usually the mathworks people care more about compatibility to the toolboxes. As long as a new feature collides with one of the important toolboxes, they rather don`t introduce it. E.g. if the signal processing toolbox fucks up when being fed 64 bit data
a) The same story was about a year ago about embassies using tor and being sniffed on
b) All anonymizing techniques rely on a sufficiently high ration of suitable "good" to "bad" nodes. Nowadays, injecting 1000 bad nodes is not costly. I suppose many secret services have 1000s of machines (or virtual machines) in the Tor network
c) If your endpoint needs to keep a stateful connection for your machine, he will be able to sniff the total connection. At least he will be able to extract metadata, and unless you use encryption, also get the whole content of the stream
d) P2P is inherently not-encrypted, at least no the content level. In the contrary, it is indexed by content, which reduces the effort to analyse what somebody wants to download (which is why P2P works).
e) Your ISP is subject to a contract with you and privacy laws. An arbitrary guy being a tor node is not
f) If you dont trust you ISP, then use a VPN you pay for with a clearly defined policy on when to hand over you records (the one i use specifically states only when forced by a judge). You anyway should have one of these for going online abroad.
It has been discussed over and over. Two difficult persons collide in a job, something goes wrong, it escalates beyond the point of repairing the damage, some formal rules and orders are pulled out and thrown in the court - from there on its pure luck what happens. Court decisions about duties in the job etc. are always very uncertain, not only for admins.
Not if the consulting team messes up. If you are the one who called for them and something is fucked up in the eye of the upper morons (who may have hired the student doing the crap back then), it does not matter if the consulting team did their job right.
Once the crappy internal web applications for managing some forms have been duct-taped together by a student worker, nodody dares to touch a single thing. You can only get burned.
The head of the biggest institution protecting child abusers with strange bending of the own laws tells something about moral realtivism? Yes, moral relativism is bad. especially if you use it to argue with strange arguments, why its not so bad that your Bishops protected the Child molesters. Get busy with to analyze the bigotry inside the church instead of mourning that now everybody can communicate - and even faster than the church.
Select the twenty people most important to you and save their names/telephone numbers/e-mail addresses in your cellphone. Thats it. dont copy people who you have not talked with for three years to you new cellphone. Call the ones who are important. Send them a new-year/christmas/easter/whatever e-mail. Normally 2 e-mails per year are enough to follow up.
Use social media with very incomplete identities to meet new people.
i mention it when its on-topic. That was to illustrate the absurdity of designing an rf modulator instead of simulating the picture of an analog tv on a monitor.
You can also emulate HW with a decent effort nowadays. Take an CPLD with few 10000 gates, throw 2 DACs in the output and buy an signal source and an IQ modulator. I am pretty sure that iff you have a CRT with an RF input (e.g. a television - which I btw dont own for 11years) and you are into it (which i am not)then it will be not so difficult to simulate the whole signal path up to the output (given anybody attached it to a Network Analyser).
Yes. the out-of the box configurability of the WPS was *great*. once you educated the PEBKACs to just use the template file to create a new text document of a certain type, this text document could get a context menu like "Copy to Floppy", "Send to x", etc. I liked the possibilities which the consistent implementation and use of the EAs created.
Moreover: DeScribe for OS2 is still the most reactive word processor i have seen. Even nowadays MS Office (not to speak about OO) makes more breaks in which it does react to user input. Programs really written for OS2 made use of threading like on no other system i saw.
I am waiting since over 10 years that IBM takes the best out of OS/2 and puts it on the top of Linux or another Unix. Gnome, KDE, and Windows are still missing some features available in OS/2. The only thing rivaling the consistency of the GUI may be the Apple desktop.
Planes are one of the safest means of transportation. This is reached by systematically evaluating all risks. The exact effect of vulcanic ash on the various types of engines seem not to be known. Normally engine failures are not correlated on a signle plane. However there have been examples of planes loosing several (or all) engines at onces when flying trough volcanic ash. This means that this (unknown) risk does not enter in the usual power law for several redundant systems. Moreover its known that in influences sensors of the plane. A loss of sensors caused the crash of the Air France flight some time ago. If several engines fail at once of the sensors fail in a fatal way, people may die.
The logics for this must be: "Do we for sure that a plane can operate as designed under these conditions?" instead of "do we know fore sure its dangerous?"
Look. Its a simple equation. The power plants hae to be dimensioned to the maximum consumption. The difference between maximum and average over the time the technology have to be available at a certain level is unused and has to be paid by the customers. If you can spread the 2h peak of air conditioning to lets say a 6h peak of one third the peak height, you may have reduce the unused power capacity by a factor of three.
Sounds good for the customer, doesn't it? Prices which are lower off-season or at times of low demands are usual in many capitalistic countries and companies. Airlines, Phone companies, Hotels, even restaurants/bars will try to make the peak-to-average ratio lower by giving the customers a financial incentive not to choose the time to use the service all at the same time.
Slashdot Mirror
Let not forget which party made this their slogan.
Science should be defended by peer review and by thesis defence, not by challenging it in a court of law, with the possibility of a legal punishment for being wrong, or for producing a politically inconvenient result.
I agree. I am a scientist myself. However, i am also an employee. Instead of letting idiotic morons who believe the earth is 6000years old and relativity is bullshit because its to complicated for them (see Andrew Schlafly) throw mud on me in public, i would rather prefer that they go to court. Because then there is a good chances it hurts them.
Well. did you write to these matrices? I guess you are aware that matlab has something like copy on write - which can lead to funny behaviours. I also find the lack of pass by reference a little disturebing to say the least, but i guess the idea behind is to keeep i teasy to debug
a) The data types in matlab have absolutely nothing to do with the processor you are running on - and that is good - all my ml programs run on all my machines b) already the 8087 supported 64 bit integer - so no i dont know why matlab does not have it since a long time c) Usually the mathworks people care more about compatibility to the toolboxes. As long as a new feature collides with one of the important toolboxes, they rather don`t introduce it. E.g. if the signal processing toolbox fucks up when being fed 64 bit data
Yes. On that aspect there is not much change. But if the process is lost it will be much easier to sue for defamation.
Its better to test these claims in front of a court than to listen to the defamation of sciene much longer. Much easier to defend yourself there.
Yes. if a system is important to you, then backup with a bare-metal restore is the only way to be sure.
Any professional who tells different is directly responsible for loss of data.
Highly suprised? Not at all.
a) The same story was about a year ago about embassies using tor and being sniffed on
b) All anonymizing techniques rely on a sufficiently high ration of suitable "good" to "bad" nodes. Nowadays, injecting 1000 bad nodes is not costly. I suppose many secret services have 1000s of machines (or virtual machines) in the Tor network
c) If your endpoint needs to keep a stateful connection for your machine, he will be able to sniff the total connection. At least he will be able to extract metadata, and unless you use encryption, also get the whole content of the stream
d) P2P is inherently not-encrypted, at least no the content level. In the contrary, it is indexed by content, which reduces the effort to analyse what somebody wants to download (which is why P2P works).
e) Your ISP is subject to a contract with you and privacy laws. An arbitrary guy being a tor node is not
f) If you dont trust you ISP, then use a VPN you pay for with a clearly defined policy on when to hand over you records (the one i use specifically states only when forced by a judge). You anyway should have one of these for going online abroad.
Basically what Jobs says is: dont use things which dont work well, even if you kill off some things which work.
Its the counterexample off microsoft: Never kill things which work, even if some things dont work right.
It has been discussed over and over. Two difficult persons collide in a job, something goes wrong, it escalates beyond the point of repairing the damage, some formal rules and orders are pulled out and thrown in the court - from there on its pure luck what happens. Court decisions about duties in the job etc. are always very uncertain, not only for admins.
Use a VPN provider of your choice.
in the lab:
oscilloscopes, network analysers, pulse generators etc.
Thanks for illustrating my original point. The way you have chosen is.... Interesting... to say the least...
Not if the consulting team messes up. If you are the one who called for them and something is fucked up in the eye of the upper morons (who may have hired the student doing the crap back then), it does not matter if the consulting team did their job right.
Once the crappy internal web applications for managing some forms have been duct-taped together by a student worker, nodody dares to touch a single thing. You can only get burned.
The head of the biggest institution protecting child abusers with strange bending of the own laws tells something about moral realtivism? Yes, moral relativism is bad. especially if you use it to argue with strange arguments, why its not so bad that your Bishops protected the Child molesters. Get busy with to analyze the bigotry inside the church instead of mourning that now everybody can communicate - and even faster than the church.
Ok i forgot: i mainly use my phone for email and IM
i can only say that harddrives are not tracked well enough that not relying on encryption for sensitive data is idiotic.
Select the twenty people most important to you and save their names/telephone numbers/e-mail addresses in your cellphone. Thats it. dont copy people who you have not talked with for three years to you new cellphone. Call the ones who are important. Send them a new-year/christmas/easter/whatever e-mail. Normally 2 e-mails per year are enough to follow up.
Use social media with very incomplete identities to meet new people.
i mention it when its on-topic. That was to illustrate the absurdity of designing an rf modulator instead of simulating the picture of an analog tv on a monitor.
You can also emulate HW with a decent effort nowadays. Take an CPLD with few 10000 gates, throw 2 DACs in the output and buy an signal source and an IQ modulator. I am pretty sure that iff you have a CRT with an RF input (e.g. a television - which I btw dont own for 11years) and you are into it (which i am not)then it will be not so difficult to simulate the whole signal path up to the output (given anybody attached it to a Network Analyser).
Yes. the out-of the box configurability of the WPS was *great*. once you educated the PEBKACs to just use the template file to create a new text document of a certain type, this text document could get a context menu like "Copy to Floppy", "Send to x", etc. I liked the possibilities which the consistent implementation and use of the EAs created.
Moreover: DeScribe for OS2 is still the most reactive word processor i have seen. Even nowadays MS Office (not to speak about OO) makes more breaks in which it does react to user input. Programs really written for OS2 made use of threading like on no other system i saw.
I am waiting since over 10 years that IBM takes the best out of OS/2 and puts it on the top of Linux or another Unix. Gnome, KDE, and Windows are still missing some features available in OS/2. The only thing rivaling the consistency of the GUI may be the Apple desktop.
Planes are one of the safest means of transportation. This is reached by systematically evaluating all risks. The exact effect of vulcanic ash on the various types of engines seem not to be known. Normally engine failures are not correlated on a signle plane. However there have been examples of planes loosing several (or all) engines at onces when flying trough volcanic ash. This means that this (unknown) risk does not enter in the usual power law for several redundant systems. Moreover its known that in influences sensors of the plane. A loss of sensors caused the crash of the Air France flight some time ago. If several engines fail at once of the sensors fail in a fatal way, people may die.
The logics for this must be: "Do we for sure that a plane can operate as designed under these conditions?" instead of "do we know fore sure its dangerous?"
Look. Its a simple equation. The power plants hae to be dimensioned to the maximum consumption. The difference between maximum and average over the time the technology have to be available at a certain level is unused and has to be paid by the customers. If you can spread the 2h peak of air conditioning to lets say a 6h peak of one third the peak height, you may have reduce the unused power capacity by a factor of three.
Sounds good for the customer, doesn't it? Prices which are lower off-season or at times of low demands are usual in many capitalistic countries and companies. Airlines, Phone companies, Hotels, even restaurants/bars will try to make the peak-to-average ratio lower by giving the customers a financial incentive not to choose the time to use the service all at the same time.