As I wrote, it still does not solve the mass of DNA required to enumerate all solution. It should be however noted, that a good solution of this particular problem is important, because as it's NP-complete, all other NP problems could be solved with cca the same complexity.
So this is a good problem, because it can be easily represented in DNA sequences and if solved, any NP algorithm could be solved using the machine solving this problem.
This is not that important as it sounds. It would not be able to solve NP-complete problems for large inputs, because it enumerates all possibilities in DNA base-pair combinations. This has actually been done before with pure DNA and their manipulation (now they use bacteria for color-marking and thus selection of the right solution's DNA sequence).
Anyways, this does not scale well, having only a few hundred cities would require DNA, that would weight cca the mass of our earth.
So this result is nice as a genetic manipulation excercise, I like that they contribute to the "standard biological components database", but it has no implication for computational complexity.
And it won't solve complex problems better than sillicon, because instead of time, you need mass. There's simply not enough material and space on this small planet for DNA solution of hamiltonian path for >500 cities.
There are lots of interesting algorithms, that are probabilistic. The definition is, that if you run it long enough, the solution will be found.
Reliability is also backed up by the times you run experiment. You can easily discard pink colours, if you get yellow, you can check the result (remember, it's NP, so checking, if the answer is okay is deterministic polynomial, i. e. fast). So filtering out the wrong solutions is not a problem.
The principial problem is, that it does not scale well. You need to enumerate all combinations in DNA. Only for a few hundred cities, you get enormous mass of DNA (earth-like weights and more).
In our country, people usually do both at once: masters and work experience. Anyways, masters is not something you would do later, that's something, you should do now. After going to work, I guarantee you, you are not going back to the university.
So go for it, the worst that could happen is that you don't finish it.
Maybe try abroad combined with some work experience.
This is bullshit. First -- one time pads are _universally unbreakable_ without knowing the key. This was mathematically proved by Claude Shannon in Communication Theory of Secrecy Systems in 1949. Contrast this with other ciphers, which we believe are not easily breakable.
One of the assumptions (or maybe the most important assumption) is, that plain text does not hold more information than key. Which is not true for quasars -- as stated in other comments, you would only need to know the source, frequency and start of encoding, which is by all means at most the same amount of information (see information enthropy, also defined by Shannon), but never more.
Actually, I've been looking for replication in a Free database for months. The things I don't get with the various "proxy" solutions:
if I do insert delayed to a database (which is replicated to two databases), or a simple insert. I have an autoincrement field. Who guarantees, that they will have the same value. If insert delayed is performed, how does the "proxy" guarantee they are actually issued in the same order. i don't care what's the order, but I want the order to be the same on both databases.
if a database falls down, how do I get instant resynchronization? I don't want to copy the full *GB database back and forth, while having a read lock. i want instant resynchronization from the point where the database fell down.
I want to write to both databases and the changes replicated to the others. I want peer to peer database, not master-server, because that involves always knowing who is the master on the application level (okay, I can grab an IP address, but much easier and nicer solution is having a cloud of servers and these things solved in the cluster, not by my hacks).
Another thing -- did anyone had a look at SAPdb and Interbase? They are Free too and there's not much talk about them. Are they useable? Do they provide replication?
The shockwave site fails to detect I have realplayer installed, I don't have windows media player nor quicktime installed (I use linux and don't have crossover or some stuff like that).
Or you can send some URL mplayer can play (I believe they can stream windows media, I have binary plugins installed).
If you have real player installed correctly, could you just play it and paste me an URL it plays?
Hello work. I'm from slovakia and I have been working from home for two years now.
Personally, not needing to pay attention to political sagas is an advantage for me, since I don't have to care about that. I have my work done and I don't have to bother with unrelated stuff.
The disadvantage is, that you are not seen and people don't think you are commited. They will eventually see the hard work, when something bad happens and they see how promptly you can solve issues.
Now I'm also a contractor for other companies, also working from home mostly. I don't have to be anywhere during working hours, I just need to have my laptop and cellphone connection with me in case something bad happens. And of course I need to have my work done. Pretty nice, I wouldn't change.
First, I must say anyone who pays for hosting based on bandwidth they get is stupid. Anyone can do ping -f IP_address for a night (for example a competitor). I refuse any offer for hosting, if it's counted by amount of bandwidth. I get billed by kbps (or mbps).
The second thing is what I believe is how a good contract should look. In contract I give to my customers, there's a clausule about such things. It means, client does not get billed excessively if such a disaster takes place, but I'm not responsible for the service not working 100% in this case. I believe it's fair to both sides.
You are not true. As you probably know, if you have read these books, One Time Pad is _provably_ unbreakable. If it has a mathematical proof, as he claims, no test of time is needed. It's proved, period. (the question is, if the proof is okay and each step would survive, but if it is, as he claims -- which _can_ be checked, it's the invention right here right now).
There are lots of people claiming they have unbreakable encryption, but if they have correct mathematical proof, man, this would be invention!
Hello, is there some way to detect the hole (as opposed to check the version number)? I have to recheck bunch of computers (not necessarily Linux or Intel) that they're safe. But as you stated, lots of vendors backported the fix, so I can't rely on version number. Has nessus plugin, that checks for the actual bug (not version number)?
Maybe it's time for moving away from IRC. There's a nice project and protocol called SILC, which tries to prevent lots of problems of IRC (mainly the lack of enciphering and signing). It has also better network structure (you don't need 10 bots just to guard your channel from takeovers, since this is protected by cryptography).
So the question should not be, to which IRC network we want to move, but where we want to move.
Ah, for all you standard weenies, SILC has been submitted to IETF as a next-generation chatting protocol draft. Really check it out, this is the future of chatting (and it's free and open source).
Re:Linux for desktop, *BSD for servers? - why not.
on
FreeBSD 4.6
·
· Score: 1
I agree -- do what you're comfortable with. If you have to use Linux on server and are comfortable with BSD, try Slackware.
Anyways, Debian is great on servers. Don't take me wrong, I'm a consultant/administrator for many companies and I admin various Linuxes (Debian, Slack, Mandrake, even RedHat) and various BSDs and even Solaris. I don't see a great difference here. There are differences however. *BSD and Debian-stable are very very very stable. If you need raw computing power and have multiprocessor system, don't use BSD.
But I'm not such liberal on desktop. I bought IBM Thinkpad and installed FreeBSD 4.5. It just sucked completely. No national keyboard support because of old XFree (this is gone in 4.6), very bad support for hardware (Linmodem, soundcard). IBM has great support for Linux and I'm happy with Debian here yet. BSD just is not for desktop (yet).
OK, what about accepting their rules -- you will read their docs, put the knowledge into the code under some kind of X11 or any other GPL compatible license. The code won't run without the original samba, but will not be licensed under GPL and will meet the criteria, but since it requires GPLed code to operate, it will be effectively infected by GPL. The possible uses of this code would be:
Use the code as a whole, thus be bound be GPL
Use only the otherwise (X11 or so) licensed part -- do we really care?
I think the code without all the other samba code (licensed under GPL) will be pretty useless. Yes, this certainly is a workaround, but it will allow the samba developers to actually use the docs (thus they don't need to reverse-engineer) and also effectively force people to accept the GPL, without breaking rules.
Hello, has anyone of you seen the virtual keyboard? Have you tried it? How does it work? How much will it cost and when it will it be available? Some URLs?
Has someone of you used this to port Delphi app to
Linux? I had a nice free LGPL-covered application, that
I wanted to compile using kylix open edition. But a lot of things are different.I see a lot of units, like QDialog, QForm, etc. under Linux, but they're counterparts in Windows are Dialog, Form, etc. So is there any sourcecode compatibility? Is there a tool for doing this?
I personally use SILC. I think this is quite understandable, because of the nature of this project. It should be very secure, so it's quite difficult to accept binary package from someone, the developers and user community hardly heard from.
The second reason could be, that the compilation in this phase is a way of debugging, i.e. the developers would like people to compile it to test for most of the compilation errors. Please stay ahead with silc, they do a good job.
I'm not very active contributor, but I've been playing with several project, mainly posting internationalization patches.
PHPNuke project (a slashdot-like weblog) makes commiting changes as difficult as possible. Nukelite mainly does not accept patches, which is quite sad, when you do the work to improve something and he reimplements your changes his way in the next version or even forgets it.
The FreeBSD project also makes contribution quite difficult, but not as PHPNuke.
I've made several project forks because of this only to find out, that I don't have the time to maintain my versions and making the fork was good for me, but mainly didn't help the others.
I also think, this is because of copyright. Many people write a GPL covered library, make it very used and then wait, if someone needs it under different license for a fee. This would be very difficult, if the code was from several different authors. So the author won't accept your patch, because it would contamine the code and he wouldn't be the only copyright holder (and so he can't sell the software or library under different license, which would f.e. allow linking with proprietary code).
Yep, exactly. Don't get me wrong, I _lived_ in Czechoslovakia, I know what communism is. I was too young to learn the propaganda, I know how bad it is.
But I really got that feeling when reading FSF pages _AND_ don't think anything wrong about it. Communism NOT as a political force, freedom-taker or anything else, but as a philosophy.
Hey guy, this is so funny. Yes, there are drugs, that are addictive: heroine, pervitine, crack,...
A lot of them are _NOT_ addictive (marijuanna, LSD, ecstasy). The _MOST_ risk comes from the fact, that they're illegal. The dealers add amphetamines and addictive parts to it, so the drug becomes addictive and even dangerous. So those dangers come from the drug war.
Anyway, here in Europe, we have more freedom, I think. We always could export cryptography. We can take some drugs, some are tolerated. In Netherlands, homosexuals can get married, there's a new eutanasia law. Our country (Slovak republic) wanted to accept the,,Contract with Vatican'', so there would be no possibility for interruption in woman's pregancy.
What I see here is freedom. I mean, I'm not a homosexual, I don't have to be married with a male. I don't like interruption and eutanasia, so I won't take it. I know ecstasy is not secure for my liver, so I won't use it. The fun of all is, that I actually can, but it doesn't mean I have to do it. People who take drugs are everywhere, you won't stop them by some stupid laws. They're just not working. If you want to help them and make them safe, make it legal, so the stuff is clean and secure (because is made in a controlled environment). They will pay for the drug the same price (the actual cost of manufacturing is 5%, the 95% of the price would be tax for propagation). Isn't this better?
Slashdot Mirror
As I wrote, it still does not solve the mass of DNA required to enumerate all solution. It should be however noted, that a good solution of this particular problem is important, because as it's NP-complete, all other NP problems could be solved with cca the same complexity.
So this is a good problem, because it can be easily represented in DNA sequences and if solved, any NP algorithm could be solved using the machine solving this problem.
This is not that important as it sounds. It would not be able to solve NP-complete problems for large inputs, because it enumerates all possibilities in DNA base-pair combinations. This has actually been done before with pure DNA and their manipulation (now they use bacteria for color-marking and thus selection of the right solution's DNA sequence).
Anyways, this does not scale well, having only a few hundred cities would require DNA, that would weight cca the mass of our earth.
So this result is nice as a genetic manipulation excercise, I like that they contribute to the "standard biological components database", but it has no implication for computational complexity.
And it won't solve complex problems better than sillicon, because instead of time, you need mass. There's simply not enough material and space on this small planet for DNA solution of hamiltonian path for >500 cities.
There are lots of interesting algorithms, that are probabilistic. The definition is, that if you run it long enough, the solution will be found.
Reliability is also backed up by the times you run experiment. You can easily discard pink colours, if you get yellow, you can check the result (remember, it's NP, so checking, if the answer is okay is deterministic polynomial, i. e. fast). So filtering out the wrong solutions is not a problem.
The principial problem is, that it does not scale well. You need to enumerate all combinations in DNA. Only for a few hundred cities, you get enormous mass of DNA (earth-like weights and more).
In our country, people usually do both at once: masters and work experience. Anyways, masters is not something you would do later, that's something, you should do now. After going to work, I guarantee you, you are not going back to the university.
So go for it, the worst that could happen is that you don't finish it.
Maybe try abroad combined with some work experience.
I wrote a similiar article back in 2003 for 2600 magazine.
I copied it here to my blog right now (so you don't need to find it in paper edition of 2600 magazine):
http://jooray.soup.io/post/10105517/State-of-art-certificates-Whom-do-you
Seems nothing has changed in five years.
This is bullshit. First -- one time pads are _universally unbreakable_ without knowing the key. This was mathematically proved by Claude Shannon in Communication Theory of Secrecy Systems in 1949. Contrast this with other ciphers, which we believe are not easily breakable.
One of the assumptions (or maybe the most important assumption) is, that plain text does not hold more information than key. Which is not true for quasars -- as stated in other comments, you would only need to know the source, frequency and start of encoding, which is by all means at most the same amount of information (see information enthropy, also defined by Shannon), but never more.
So in short -- this is worse than one time pads.
Another thing -- did anyone had a look at SAPdb and Interbase? They are Free too and there's not much talk about them. Are they useable? Do they provide replication?
ok, I found it out myself:
n g_line_wmp_300.asf'
n g_line_wmp_300.asf'
mplayer 'http://205.252.48.142:80/wm.atomfilms/full/waiti
if you do
mplayer -dumpstream 'http://205.252.48.142:80/wm.atomfilms/full/waiti
you get a direct local copy of the file, which can be then recoded (with mencoder) to something more sane than Windows Media.
Is there a direct url for the zero 7 video?
The shockwave site fails to detect I have realplayer installed, I don't have windows media player nor quicktime installed (I use linux and don't have crossover or some stuff like that).
Or you can send some URL mplayer can play (I believe they can stream windows media, I have binary plugins installed).
If you have real player installed correctly, could you just play it and paste me an URL it plays?
Thanks!
Personally, not needing to pay attention to political sagas is an advantage for me, since I don't have to care about that. I have my work done and I don't have to bother with unrelated stuff.
The disadvantage is, that you are not seen and people don't think you are commited. They will eventually see the hard work, when something bad happens and they see how promptly you can solve issues.
Now I'm also a contractor for other companies, also working from home mostly. I don't have to be anywhere during working hours, I just need to have my laptop and cellphone connection with me in case something bad happens. And of course I need to have my work done. Pretty nice, I wouldn't change.
First, I must say anyone who pays for hosting based on bandwidth they get is stupid. Anyone can do ping -f IP_address for a night (for example a competitor). I refuse any offer for hosting, if it's counted by amount of bandwidth. I get billed by kbps (or mbps).
The second thing is what I believe is how a good contract should look. In contract I give to my customers, there's a clausule about such things. It means, client does not get billed excessively if such a disaster takes place, but I'm not responsible for the service not working 100% in this case. I believe it's fair to both sides.
There are lots of people claiming they have unbreakable encryption, but if they have correct mathematical proof, man, this would be invention!
Here is a user-friendly description of what should be
done to prevent the worm getting to your site and how to minimize the impact on huge network.
Hello, is there some way to detect the hole (as opposed to check the version number)? I have to recheck bunch of computers (not necessarily Linux or Intel) that they're safe. But as you stated, lots of vendors backported the fix, so I can't rely on version number. Has nessus plugin, that checks for the actual bug (not version number)?
How many false positives does it have? Is the use of geoip legal for this matter? (Won't they start complaining about it?).
So the question should not be, to which IRC network we want to move, but where we want to move.
Ah, for all you standard weenies, SILC has been submitted to IETF as a next-generation chatting protocol draft. Really check it out, this is the future of chatting (and it's free and open source).
Anyways, Debian is great on servers. Don't take me wrong, I'm a consultant/administrator for many companies and I admin various Linuxes (Debian, Slack, Mandrake, even RedHat) and various BSDs and even Solaris. I don't see a great difference here. There are differences however. *BSD and Debian-stable are very very very stable. If you need raw computing power and have multiprocessor system, don't use BSD.
But I'm not such liberal on desktop. I bought IBM Thinkpad and installed FreeBSD 4.5. It just sucked completely. No national keyboard support because of old XFree (this is gone in 4.6), very bad support for hardware (Linmodem, soundcard). IBM has great support for Linux and I'm happy with Debian here yet. BSD just is not for desktop (yet).
I think the code without all the other samba code (licensed under GPL) will be pretty useless. Yes, this certainly is a workaround, but it will allow the samba developers to actually use the docs (thus they don't need to reverse-engineer) and also effectively force people to accept the GPL, without breaking rules.
the release (Linux binaries, Solver tree and sources)?
I'm trying to download it and provide a mirror, but it's impossible yet.
Hello, has anyone of you seen the virtual keyboard? Have you tried it? How does it work? How much will it cost and when it will it be available? Some URLs?
Has someone of you used this to port Delphi app to
Linux? I had a nice free LGPL-covered application, that
I wanted to compile using kylix open edition. But a lot of things are different.I see a lot of units, like QDialog, QForm, etc. under Linux, but they're counterparts in Windows are Dialog, Form, etc. So is there any sourcecode compatibility? Is there a tool for doing this?
The second reason could be, that the compilation in this phase is a way of debugging, i.e. the developers would like people to compile it to test for most of the compilation errors. Please stay ahead with silc, they do a good job.
PHPNuke project (a slashdot-like weblog) makes commiting changes as difficult as possible. Nukelite mainly does not accept patches, which is quite sad, when you do the work to improve something and he reimplements your changes his way in the next version or even forgets it.
The FreeBSD project also makes contribution quite difficult, but not as PHPNuke.
I've made several project forks because of this only to find out, that I don't have the time to maintain my versions and making the fork was good for me, but mainly didn't help the others.
I also think, this is because of copyright. Many people write a GPL covered library, make it very used and then wait, if someone needs it under different license for a fee. This would be very difficult, if the code was from several different authors. So the author won't accept your patch, because it would contamine the code and he wouldn't be the only copyright holder (and so he can't sell the software or library under different license, which would f.e. allow linking with proprietary code).
But I really got that feeling when reading FSF pages _AND_ don't think anything wrong about it. Communism NOT as a political force, freedom-taker or anything else, but as a philosophy.
A lot of them are _NOT_ addictive (marijuanna, LSD, ecstasy). The _MOST_ risk comes from the fact, that they're illegal. The dealers add amphetamines and addictive parts to it, so the drug becomes addictive and even dangerous. So those dangers come from the drug war.
Anyway, here in Europe, we have more freedom, I think. We always could export cryptography. We can take some drugs, some are tolerated. In Netherlands, homosexuals can get married, there's a new eutanasia law. Our country (Slovak republic) wanted to accept the ,,Contract with Vatican'', so there would be no possibility for interruption in woman's pregancy.
What I see here is freedom. I mean, I'm not a homosexual, I don't have to be married with a male. I don't like interruption and eutanasia, so I won't take it. I know ecstasy is not secure for my liver, so I won't use it. The fun of all is, that I actually can, but it doesn't mean I have to do it. People who take drugs are everywhere, you won't stop them by some stupid laws. They're just not working. If you want to help them and make them safe, make it legal, so the stuff is clean and secure (because is made in a controlled environment). They will pay for the drug the same price (the actual cost of manufacturing is 5%, the 95% of the price would be tax for propagation). Isn't this better?
War on drugs is propaganda, it is not working.