"You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License."
"It's so beautiful, I may just cry openly."
IANAL, but if don't accept GPL then the text in the license doesn't have any legal effects. Default laws etc. (of an user's country) should be used instead.
I don't exactly if how this noise-cancelling system work but article mentiones that it's most effective with repeative sounds (drills and also music with some degree).
If the sound is repeative/has much correlation with itself (when time changes) then the future sound can be predicted quite well and noise cancelling probably works even with high frequencies.
Standard statistical signal processing methods and neural networks could probably do quite a good job if the sound is predictable enough. If you have many microphones in a room you could probably train NN to give good noise cancelling response (I have only limited understanding of NNs) that guesses what kind the not yet arrived sound is and adapts to the "sound response of the room" which is probably non-linear.
> Why complain about these harmless newbies. They'll learn from their mistakes > and start more reasonable projects later.
Yes and usually it's the case that the ones that should read these rants generally speaking don't do it. However, hopefully even someone reads it so the effort isn't wasted.
Yes. But (afaik) one can still get out of black hole by having very much (easily usable) energy. 'Just' use energy to create enough antimatter that annihilates enough mass and then you're free.
So does this mean that if one goes fast enough (near c) towards black hole and inside event horison manages to slow down to get enough energy and has a way to create antimatter then with enough speed one can escape from black hole?
I don't really understand this attitude. If one puts his code under BSD license, then he is giving it away free. It's like selling your house away for 100 dollars and then being later pissed off for 'being screwed' (assuming bargain was done legally and morally correct).
So they'll always choose the Microsoft(tm) option. That's just the way it is in the real business world. Even if there's no possibility of recovering losses from the vendor, at least there will always (?) be that vendor to blame/approach. And in the off (heh) chance that there are others with the same problems, the likelihood of finding a solution will be greater.
But that just not how it works (in certain cases):
Why this is on slashdot?
This is just another case of experts speaking about subject area that is outside they expertise.
Interesting reading (with some (understandable flaws), but nothing really new if one has ever read a bit about genetic algorithms and/or other system analysis related topics.
Yes, DOS was/is "stable", but only because requirements for the modern OS were much lower than nowadays.
a) if bad behaving applications can cause crash with modern OSes, OS is considered to be unstable.
b) Modern OSes must support lot more different hardware and any combinations of different hardware.
c) Modern OS must do multitasking, multiple, good memory management, handle different priviledge levels, support multiple users
d) Modern OS (kernel+core libs) must support lot of different APIs, executable formats, abstract away direct hardware accesses etc..
It's _relatively_ easy to code 'dos'-size program to be efficient and (mostly) always working when the requirements aren't very demanding. When the program size / number of features grow the number ways things can go wrong increases dramatically (O(n!) interacting parts (in theory)).
You're right, but only partly IMHO. Learning new languages (syntax) is easy but learning all the names of library calls etc. can take considerable amount of time because you simply have to learn N (N = big) new symbols and how to use them. And if you don't know them by heart it can take very much time to find the right function from documentation.
It also takes time to get a deep understanding of a new language (how it's implemented, what is fast, what is not, how to combine it with other languages, portability, differences between different compilers/compiler versions etc.)
Slashdot Mirror
Have you ever heared about CVS?
(in this case it should be automized)
"You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License."
"It's so beautiful, I may just cry openly."
IANAL, but if don't accept GPL then the text in the license doesn't have any legal effects. Default laws etc. (of an user's country) should be used instead.I don't exactly if how this noise-cancelling system work but article mentiones that it's most effective with repeative sounds (drills and
also music with some degree).
If the sound is repeative/has much correlation with itself (when time changes) then the future sound can be predicted quite well and noise cancelling probably works even with high frequencies.
Standard statistical signal processing methods and neural networks could probably do quite a good job if the sound is predictable enough.
If you have many microphones in a room you could probably train NN to give good noise cancelling response (I have only limited understanding of NNs) that guesses what kind the not yet arrived sound is and adapts to the "sound response of the room" which is probably non-linear.
> Why complain about these harmless newbies. They'll learn from their mistakes
> and start more reasonable projects later.
Yes and usually it's the case that the ones that should read these rants generally speaking don't do it. However, hopefully even someone reads it so the effort isn't wasted.
Site's seems to be slashdotted already. Does anyone know any mirrors?
Yes. But (afaik) one can still get out of black hole by having very much (easily usable) energy. 'Just' use energy to create enough antimatter that annihilates enough mass and then you're free.
So does this mean that if one goes fast enough (near c) towards black hole and inside event horison manages to slow down to get enough energy and has a way to create antimatter then with enough speed one can escape from black hole?
> BSD code, on the other...
I don't really understand this attitude. If one puts his code under BSD license, then he is giving it away free.
It's like selling your house away for 100 dollars and then being later pissed off for 'being screwed' (assuming bargain was done legally and morally correct).
So they'll always choose the Microsoft(tm) option. That's just the way it is in the real business world. Even if there's no possibility of recovering losses from the vendor, at least there will always (?) be that vendor to blame/approach. And in the off (heh) chance that there are others with the same problems, the likelihood of finding a solution will be greater.
But that just not how it works (in certain cases):
Apache ~ 60%, Microsoft ~ 30% [netcraft.com]
Why this is on slashdot?
This is just another case of experts speaking about subject area that is outside they expertise.
Interesting reading (with some (understandable flaws), but nothing really new if one has ever read a bit about genetic algorithms and/or other system analysis related topics.
Yes, DOS was/is "stable", but only because requirements for the modern OS were much lower than nowadays.
a) if bad behaving applications can cause crash with modern OSes, OS is considered to be unstable.
b) Modern OSes must support lot more different hardware and any combinations of different hardware.
c) Modern OS must do multitasking, multiple, good memory management, handle different priviledge levels, support multiple users
d) Modern OS (kernel+core libs) must support lot of different APIs, executable formats, abstract away direct hardware accesses etc..
It's _relatively_ easy to code 'dos'-size program to be efficient and (mostly) always working when the requirements aren't very demanding. When the program size / number of features grow the number ways things can go wrong increases dramatically (O(n!) interacting parts (in theory)).
You're right, but only partly IMHO. Learning new languages (syntax) is easy but learning all the names of library calls etc. can take considerable amount of time because you simply have to learn N (N = big) new symbols and how to use them. And if you don't know them by heart it can take very much time to find the right function from documentation.
It also takes time to get a deep understanding of a new language (how it's implemented, what is fast, what is not, how to combine it with other languages, portability, differences between different compilers/compiler versions etc.)