For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
No finite polynomial with rational coefficient can take on only and every prime value. The proof of this is elementary. Rather surprisingly, there is a real number C such that C^3^n is always prime for all n [cf Mills' Theorem]. Unfortunately, the only known way to calculate C is working backward from a list of primes...
I see that the phrase "releasing under the GPL" has combined with "genetic programming" to create a comment that will inevitably be moderated up. It's really too bad that the proposal is most probably nonsense:
Genetic programming is in reality a novel, but not revolutionary, set of techniques for the finding the extrema of a fitness condition. For example, the fitness function for reverse engineering algoritms could be f_actual(x)-f_test(x). Genetic programming, and indeed most extrema finding algorithms, have the weekness of falling into local minimum wells instead of finding the global extrema.
Morever, it relies on the search surface of the fitness function being adaquately "rough" so that the algorithm can proceed systematically in some direction. Certainly, a GP algorithm that does polynomial or rational interpolation is doable, however, this is an entirely different task than cryptanalysis of a block cipher or the integer factorization problem.
Cryptanalysis of a block cipher- this is unlikely to work for two main reasons, one empirical and the other a theoretical argument based on computational complexity theory. A well designed block cipher's result surface is too "smooth" (ie. the Hamming distance between f(correct_key) and f(correct_key + 1) will essentially be the block size/2 when averaged over all inputs). As a result, I don't see with a GP would have to work on. Furthermore, the general case of determining the output of a Boolean function in the presence of noise, the Parity Assumption, is conjuctured (proven?) to be NP-complete. Solving this would likely imply that P=NP.
Integer factorization problem- Firstly, the author displays a misunderstanding of the IFP. The best algorithms for extracting the prime factors of a large composite number are heuristically asymptopic to a sub-exponential function, not fully exponential. The existence of an algorithm that can do this in *linear* time is almost certainly impossible. If this were ever to occur, there would need to be a revolutionary change in our understanding of number theory.
>Presumably a gravity wave will hit all the >detectors at the same time, since gravity moves >as fast as it wants to and doesnt have to be >held back by the speed of light rule.. NO!!! Gravity moves at the speed of light. The importance of the seperation is that local interference will presumably not be the same at both points, allowing a statistical cancellation of the error effects. (BTW, I left my account logged on in a library computer and some idiot has been doing those second-first-post messages. A -31 karma's pretty bad...maybe I should just cancel this account.)
Slashdot Mirror
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
by kuro5hin
For a long time now, those who want to use and improve the slashdot code have been wondering, and waiting, and hoping for the much promised 0.4 tarball. Many of them have in fact become quite irate about the lag between code releases, the lack of a CVS server, and the overall appearance that the slashdot gang doesn't practice what it preaches ("release early, release often"). How would you respond to these criticisms, and do you intend to change the development practices in any way in the future?
CmdrTaco:
I get a nice flamey email about once a week from some ass who calls me a hypocrite and slams me for not getting out a new release. My usual response is to tell them that I delay the release by 24 hours each time someone asks me when a new Slash tarball will be out.
Seriously, there are only 3 people who really know how much work a source release for this is: CowboyNeal, Patrick and Me. And the three of us have been working on a lot of stuff. As I write this, we are bugfixing and documenting and preparing for a source release. There is a private CVS server that one day soon will be publicly read only.
This isn't like other projects: it has been custom fit to our hardware and to our needs. It doesn't have install scripts or help or even comments in the code. We're just too busy to play tech support helping dozens of people compile mod_perl and tune Apache. We've decided to squash the bugs and make a clean release rather than rush it.
It's really easy for someone to complain that I didn't release a new version of the source code every week. Its also easy to forget that in the last 6 months we've doubled in traffic and we've had to optimize our code and hardware to handle that. A new source release is secondary: Our job is running Slashdot. We want to release new versions of Slash, but it is a definite second priority to keeping Slashdot moving.
Finally, it's coming soon. It'll be out when its finished. And if you ask me again I'll postpone it again.
_
/\ /\| |/\
/\| | | ||\ / / / /
|\_/|
|---|
| |
| |
_|=-=|_
_
| | | | |\>
| | | | | \
|- - - -|) )
|
\
\
\
http://www-csag.cs.uiuc.edu/individual/pakin/compl aint-info/new-improved.html
Same as you: too much time on my hands. Isn't it obvious?
Yes it was. Totally uncalled for and very mean.
No finite polynomial with rational coefficient can take on only and every prime value. The proof of this is elementary. Rather surprisingly, there is a real number C such that C^3^n is always prime for all n [cf Mills' Theorem]. Unfortunately, the only known way to calculate C is working backward from a list of primes...
I see that the phrase "releasing under the GPL" has combined with "genetic programming" to create a comment that will inevitably be moderated up. It's really too bad that the proposal is most probably nonsense:
Genetic programming is in reality a novel, but not revolutionary, set of techniques for the finding the extrema of a fitness condition. For example, the fitness function for reverse engineering algoritms could be f_actual(x)-f_test(x). Genetic programming, and indeed most extrema finding algorithms, have the weekness of falling into local minimum wells instead of finding the global extrema.
Morever, it relies on the search surface of the fitness function being adaquately "rough" so that the algorithm can proceed systematically in some direction. Certainly, a GP algorithm that does polynomial or rational interpolation is doable, however, this is an entirely different task than cryptanalysis of a block cipher or the integer factorization problem.
Cryptanalysis of a block cipher- this is unlikely to work for two main reasons, one empirical and the other a theoretical argument based on computational complexity theory. A well designed block cipher's result surface is too "smooth" (ie. the Hamming distance between f(correct_key) and f(correct_key + 1) will essentially be the block size/2 when averaged over all inputs). As a result, I don't see with a GP would have to work on. Furthermore, the general case of determining the output of a Boolean function in the presence of noise, the Parity Assumption, is conjuctured (proven?) to be NP-complete. Solving this would likely imply that P=NP.
Integer factorization problem- Firstly, the author displays a misunderstanding of the IFP. The best algorithms for extracting the prime factors of a large composite number are heuristically asymptopic to a sub-exponential function, not fully exponential. The existence of an algorithm that can do this in *linear* time is almost certainly impossible. If this were ever to occur, there would need to be a revolutionary change in our understanding of number theory.
>Presumably a gravity wave will hit all the >detectors at the same time, since gravity moves >as fast as it wants to and doesnt have to be >held back by the speed of light rule.. NO!!! Gravity moves at the speed of light. The importance of the seperation is that local interference will presumably not be the same at both points, allowing a statistical cancellation of the error effects. (BTW, I left my account logged on in a library computer and some idiot has been doing those second-first-post messages. A -31 karma's pretty bad...maybe I should just cancel this account.)
not that close to first post!!!!
not that close to first post!!!!
not that close to first post!!!!
not that close to first post!!!!
not that close to first post!!!!
not that close to first post!!!!
second first post!?!?!
FIRST POSTxyz11!!!
second first post!?!?!
second first post!?!?!