The full 40% coverage is needed only for solar neutrinos, and to some extent possible super nova neutrinos, both of whos interactions produce far less light than other classes of neutrinos. For higher energy atmospheric and K2K neutrinos 20% coverage will translate into only slightly reduced ability to reconstruct the events and do physics.
It is a good compromise to do the most physics with the resources at hand.
SK is/was the far detector for the K2K long baseline neutrino experiment which is/was still running. With out SK, there isn't much point. SK was also planned to be used in another future lbl experiment, JHF-SK, which could pin down some of the mixing parameters (theta_13) and possibly put a limit on CP violation in the neutrino system.
So, this short sighted crass statement by this anonymous physicist really annoys me.
You can do OO style programming in Lisp and Scheme. Lisp has Common Lisp Object System (CLOS). In Guile Scheme, there is GOOPS which closely resembles CLOS.
This is (almost) what is done in ROOT (root.cern.ch). Instead of #including the system compiler's source they have something which is called ACLiC. With ACLiC you can compile some C/C++ code to a shared object (.so) using your systems compiler via a fork/exec and link it in to the running code. This is done with a single simple command.
Using the same command, minus an option, one can instead just interpret the code with CINT. This lets you trade off run time speed with compile time delay.
For a "real" C/C++ interpreter take a look at CINT, used extensively in ROOT:
http://root.cern.ch/root/Cint.html. It doesn't cover 100% of C/C++ but it covers a lot.
I agree that C/C++ is not the best choice for a scripting language in general, but it does allow one to prototype fast.
I need some more convincing that neutrinos are changing on their way to Earth.
A good way to give more credence to this idea would be to place both types of
detectors at several points...
Speak-freely is a venerable program and there is both a Linux and a windows version which can intercomunicate, but its sound quality is pretty poor. `ohphone' gives much better sound quality and is easier to set up.
Coincidently I was just starting to mess with H.323 this weekend. With a direct net connection, one can use `ohphone' with out needing any other programs (except another one to talk to, of course). It's pretty simple to set up and use. If you are behind a NAT router things are trickier (I haven't gotten it to work yet). There are Debian packages for `ohphone' and a few other H.323 packages.
I was also able to use openmcu to provide rudimentary group conference services. This is also packaged for Debian.
BTW, with ohphone you can talk with windows Netmeeting users.
After finishing, Nasubi was asked what he most wanted. He said kimchi and yakiniku (Korean spicy cabbage and Korean BBQ). So the producers flew him to Korea, gave him a big urn of kimchi, a huge yakiniku meal and..... A chance to do it all over again, but this time in Korea where he doesn't speak the language. The lunatic accepted.
Some of the funny things about that show I remember was the time he finally won some clothes. He won a pair of shorts and he was so happy at first but then he realized that he just liked dancing around naked too much to wear them, so they ended up on the wall. Speaking of dancing, the producers always added a huge eggplant (which is ``Nasubi'' in Japanese) instead of the more western fig leaf, always cracked me up.
The ``exposure'' has done him good as he is now one of the official 12 famous TV people on Jp TV shows.
Granted, I never spent the same effort to learn VMS as I have for UNIX, but I never felt comfortable in the VMS shell. In the case of changing directory, one had to use a clumsy (opinion!) path statements with too many brackets and colons. For removing directories there was always some dificulty with permissions so that by default, I could not delete my own directories.
I know there are many people who swear by VMS so I am sure it isn't a terrible OS, but i just can't see it ``kicking the crud'' out of UNIX.
Anyways, I was certainly not asserting that having a CD command makes a OS better, but rather the level of comfort that a VMS shell gives me is minimal in comparison to UNIX shells (even plain sh). In my mind, DOS is only slightly more comfortable than VMS as an environment. (This is very likely due to just familiarity).
One thing I do remember fondly about that year of VMS is the help system was quite usefull. In comparison, UNIX's man system suffers as it is focused more as a reference guide than a guiding reference.
You are possibly correct about banks using VMS, but walking into a bank I see a lot of Windows machines.
As for VMS itself, have you ever used it? I used it the first year of grad school and I am very happy that my mind has purged everything it could about this OS. Uhg, what a pain in ass to use.
When VMS can do simple things simply (like, cd, and rmdir) on VMS, then let's talk about it ``kicking the crud out of everything else''.
VMS was the target of some of the first crackers. But now, since it is extreamly rare to see VMS used anymore it enjoys some security through obscurity: with out the scripts, there can be no script kiddies.
One more mark against proprietary software.
on
Microsoft Cracked
·
· Score: 2
This developement is a bad thing considering how much mission critical stuff, for good or ill, depends on MS software. Just think of all those stock market, insurance company, bank, and government computers running MS software. Now think of all that data under the control of an outside untrusted agent. It's enough to cause a bank run!
However, as bad as this is, it is good for free software as highlights the benefit of having access to the source and the drawback of proprietary software. It should be strongly stressed that this break in and possible insertion of back doors in literally millions of computers via MS software just underlines things we all already know: When the source is not open, the consumer has *no* way to prove its level of security.
In the past MS and others have used the ``argument'' that having the source available to black hat hackers makes free / open source less secure. This (false) argument rested on the assumption that Uncle Bill kept MS source under lock and key. Today this argument is now double false.
Proven and observed are not the same thing. It is possible to build a system which has no connection to observable reality, (for example, starting with false premises).
From reading the NYT article I don't really see what the big deal is. As far as I can tell from the info there, all that is being done is to amplify the leading edge of the wave packet while at the same time cancelling out the remaining body of the packet.
It seems to me that the ``speed'' of the wave packet is being measured using the time the peak enters the chamber and the time the amplified tail exists the chamber.
If this is correct you could tune the wave packet to give you any ``speed'' greater than c you wanted by changing the width of the wave packet. (assuming you could still get this trick of cancelling the peak while amplifying the leading tail).
If I missed something, I am sure I'll be corrected.
I would say that bug fixing is a statistical process. This means that the time a bug exists is distributed by some (probably complicated) distribution which is a function of several parameters, of which the number of eyes looking at the code is one.
I think most (honestly critical) people would say that the higher the number of eyes checking a body of code the lower the mean bug lifetime will be. But, as with any distributed value, one can always get a measurement way out on the tail of the distribution. This is why anecdotal evidence is not very useful for trying to find out how some quantity is distributed.
Think of how dumb the average person is. By definition, half of the population is dumber than that...
This is not true ``by definition'' at all. Half the population is dumber than the average person only if the median of the dumbness distribution is equal to the mean.
Slashdot Mirror
The full 40% coverage is needed only for solar neutrinos, and to some extent possible super nova neutrinos, both of whos interactions produce far less light than other classes of neutrinos. For higher energy atmospheric and K2K neutrinos 20% coverage will translate into only slightly reduced ability to reconstruct the events and do physics.
It is a good compromise to do the most physics with the resources at hand.
SK is/was the far detector for the K2K long baseline neutrino experiment which is/was still running. With out SK, there isn't much point. SK was also planned to be used in another future lbl experiment, JHF-SK, which could pin down some of the mixing parameters (theta_13) and possibly put a limit on CP violation in the neutrino system.
So, this short sighted crass statement by this anonymous physicist really annoys me.
SK water is (was) about as ph balanced as you can get.
You can do OO style programming in Lisp and Scheme. Lisp has Common Lisp Object System (CLOS). In Guile Scheme, there is GOOPS which closely resembles CLOS.
It does make menus and some other GTK fonts look significantly nicer (at least on this 133 DPI LCD).
However, the three things I stare at most are xemacs, gnome-terminal and galeon, none of which are changed by this hack.
Ha! A Hampton Inn I stay at semi freqently near Chicago is exactly the first thing that came to mind when I read this article!
This is (almost) what is done in ROOT (root.cern.ch). Instead of #including the system compiler's source they have something which is called ACLiC. With ACLiC you can compile some C/C++ code to a shared object (.so) using your systems compiler via a fork/exec and link it in to the running code. This is done with a single simple command.
Using the same command, minus an option, one can instead just interpret the code with CINT. This lets you trade off run time speed with compile time delay.
I agree that C/C++ is not the best choice for a scripting language in general, but it does allow one to prototype fast.
http://k2k.physics.sunysb.edu/k2k/
http://www.hep.anl.gov/ndk/hypertext/numi.html
Speak-freely is a venerable program and there is both a Linux and a windows version which can intercomunicate, but its sound quality is pretty poor. `ohphone' gives much better sound quality and is easier to set up.
I was also able to use openmcu to provide rudimentary group conference services. This is also packaged for Debian.
BTW, with ohphone you can talk with windows Netmeeting users.
I have such a display (Inspiron 5000e) and I must say that I don't know how people slog around in anything less.
After finishing, Nasubi was asked what he most wanted. He said kimchi and yakiniku (Korean spicy cabbage and Korean BBQ). So the producers flew him to Korea, gave him a big urn of kimchi, a huge yakiniku meal and..... A chance to do it all over again, but this time in Korea where he doesn't speak the language. The lunatic accepted.
Some of the funny things about that show I remember was the time he finally won some clothes. He won a pair of shorts and he was so happy at first but then he realized that he just liked dancing around naked too much to wear them, so they ended up on the wall. Speaking of dancing, the producers always added a huge eggplant (which is ``Nasubi'' in Japanese) instead of the more western fig leaf, always cracked me up.
The ``exposure'' has done him good as he is now one of the official 12 famous TV people on Jp TV shows.
VB is just Fortran with GUI callbacks.
I know there are many people who swear by VMS so I am sure it isn't a terrible OS, but i just can't see it ``kicking the crud'' out of UNIX.
Anyways, I was certainly not asserting that having a CD command makes a OS better, but rather the level of comfort that a VMS shell gives me is minimal in comparison to UNIX shells (even plain sh). In my mind, DOS is only slightly more comfortable than VMS as an environment. (This is very likely due to just familiarity).
One thing I do remember fondly about that year of VMS is the help system was quite usefull. In comparison, UNIX's man system suffers as it is focused more as a reference guide than a guiding reference.
As for VMS itself, have you ever used it? I used it the first year of grad school and I am very happy that my mind has purged everything it could about this OS. Uhg, what a pain in ass to use. When VMS can do simple things simply (like, cd, and rmdir) on VMS, then let's talk about it ``kicking the crud out of everything else''.
VMS was the target of some of the first crackers. But now, since it is extreamly rare to see VMS used anymore it enjoys some security through obscurity: with out the scripts, there can be no script kiddies.
However, as bad as this is, it is good for free software as highlights the benefit of having access to the source and the drawback of proprietary software. It should be strongly stressed that this break in and possible insertion of back doors in literally millions of computers via MS software just underlines things we all already know: When the source is not open, the consumer has *no* way to prove its level of security.
In the past MS and others have used the ``argument'' that having the source available to black hat hackers makes free / open source less secure. This (false) argument rested on the assumption that Uncle Bill kept MS source under lock and key. Today this argument is now double false.
Remains? Since when has there been any integrity to MS code?
Proven and observed are not the same thing. It is possible to build a system which has no connection to observable reality, (for example, starting with false premises).
Rational.
My wife took some pics yesterday at the Macworld show. They are at http://superk.physics.sunysb .edu/~yoko/Mac/World/thumb.
It seems to me that the ``speed'' of the wave packet is being measured using the time the peak enters the chamber and the time the amplified tail exists the chamber.
If this is correct you could tune the wave packet to give you any ``speed'' greater than c you wanted by changing the width of the wave packet. (assuming you could still get this trick of cancelling the peak while amplifying the leading tail).
If I missed something, I am sure I'll be corrected.
I think most (honestly critical) people would say that the higher the number of eyes checking a body of code the lower the mean bug lifetime will be. But, as with any distributed value, one can always get a measurement way out on the tail of the distribution. This is why anecdotal evidence is not very useful for trying to find out how some quantity is distributed.
This is not true ``by definition'' at all. Half the population is dumber than the average person only if the median of the dumbness distribution is equal to the mean.