I became so enthusiastic after having lzip compress my NT partition that I assigned 2 of my PhD students to the task of investigating the possibility to enhance the lunzip algorithm with a random data generation method that I developed when I was an software engineer at Microsoft (R). At that time I used to generate software, but now I am brainstorming about using it for filelunziping. If my contribution is accepted by the lunzip team, this will have effect on the type of licence. It'll probably no longer be shareware and will cost future users a minor fee of 450$ (a small price for state-of-the-art technology)
The Netherlands is a good place to work.
I've been working in IT in Holland a few years.
I started for a new company in amsterdam 1 month ago. One of the advantages is thatalmost every
dutch man speaks english quite well.
I've worked with dutch and french colleages and
noticed that their english is quite bad.
Besides dutch people are known for their tolerance.
The second best choice in europe IMO would be the UK.
One can never garantuee that the GA will not end up in a local minimum. In my opinion it's a big mistake to start with a beginpopulation that is not completely random (he includes the 'baseline', a solution that is known to be good). In this way you loose the global character of the search and the robustness of the method although it may save some time. The population will probably converge to something that is near to this baseline.
Software is made up of 0's and 1's, so a computer program is a representation of an integer number. I don't think anyone should be able to become the owner of an integer number.
I tried to use octave. I never succeeded in porting my Matlab scripts, I think the help function is bad. I guess I'll have to boot NT and run matlab if I want to use the scripts
Slashdot Mirror
I became so enthusiastic after having lzip compress my NT partition that I assigned 2 of my PhD students to the task of investigating the possibility to enhance the lunzip algorithm with a random data generation method that I developed when I was an software engineer at Microsoft (R). At that time I used to generate software, but now I am brainstorming about using it for filelunziping. If my contribution is accepted by the lunzip team, this will have effect on the type of licence. It'll probably no longer be shareware and will cost future users a minor fee of 450$ (a small price for state-of-the-art technology)
The Netherlands is a good place to work. I've been working in IT in Holland a few years. I started for a new company in amsterdam 1 month ago. One of the advantages is thatalmost every dutch man speaks english quite well. I've worked with dutch and french colleages and noticed that their english is quite bad. Besides dutch people are known for their tolerance. The second best choice in europe IMO would be the UK.
One can never garantuee that the GA will not end up in a local minimum. In my opinion it's a big mistake to start with a beginpopulation that is not completely random (he includes the 'baseline', a solution that is known to be good). In this way you loose the global character of the search and the robustness of the method although it may save some time. The population will probably converge to something that is near to this baseline.
I kind of agree with you about the uselessness of my point, I just wanted to make a ridiculous statement.
Software is made up of 0's and 1's, so a computer program is a representation of an integer number. I don't think anyone should be able to become the owner of an integer number.
I tried to use octave. I never succeeded in porting my Matlab scripts, I think the help function is bad. I guess I'll have to boot NT and run matlab if I want to use the scripts