You probably already know it, but here are a couple of other collections of open-source projects in your parent company. They can probably provide advice about such topics as which person in your legal departments has expertise and advice on the subject and other institution specific headaches and shortcuts.
You need to incorporate the product of the following conditions:
Patient's certainty: Uncertain Certain and correct Certain and incorrect
Getting the drug: Yes No
This would leave us with the following groups: Not sure and recieving drug Not sure and not recieving drug Certain of recieving drug and recieving drug Certain of not recieving drug and not recieving drug Certain of not recieving drug and recieving drug Certain of recieving drug and not recieving drug
Then you need many replicates, include all the interactions in your ANOVA (i.e. do it the simple, correct way with none of the monkeying around that bad statisticians will prescribe), and report the results that pass Ficher's LSD (the most powerful detector of significant difference), and possibly also include results passing more stringent significance tests.
Then we will have the answer. Wait 4 years for people to do it with other drugs and make more complicated expirements with more degrees of freedom and it will be canon.
And yes, you will have to LIE to and DECIEVE your patients. This is considered unethical, so this simple basic expirement will never be done in the "developed" world. There can be no waiver of "you may or may not recieve medication" because if introduced it would place everyone in the group "Uncertain." If the patients have a bias towards believing that a medical experiment does not medicate as stated then the patients must not know that they are participating in the experiment.
Don't know if this is new or not, but a streaming peer-to-peer protocol like bittorrent would be pretty cool. It could be used to inexpensively broadcast audio or video almost live, potentially making news reporting available to a wider selection of journalists. Checksums on data would obviously be a problem here and malicious nodes in the network would have an easier time of disrupting communications. This mechanism needs to be independent of media type, and rely on being used in combination with file formats that can be picked up and played from any small chunk. The client could decide which portions of the stream it would rather get, sacrificing liveness to get as much as possible, trying to pick up the nearest blocks in the future first to stay as smooth as possible, or minimizing buffer size and going after the most recent blocks to stay as live as possible.
I tried this against my girlfriend's computer with a wireless network connection. She has XP home SP2. With the firewall ON it caused 100% CPU usage during the attack, and loss of connection. Connection was restored after sending ^C to hping2. With the firewall OFF it caused 100% CPU usage during the attack, and loss of connection. CPU usage and connection were not restored for about a minute after the end of the attack. The included firewall with default settings is NOT a solution to this problem on XP Home SP2.
You're right. Spyware or adware (these would probably benefit from blocking windows updates more then viruses) can emulate whatever fingerprints of Wine Microsoft uses for countermeasures to prevent updates. Wine doesn't need to enter the arms race against Microsoft - more nefarious groups with financial incentive will do it for them.
Konqueror 3.3.1 is vulnerable if Global JavaScript Policies - Open New Windows is set to Allow and you use the no popup blocker link. With Open New Windows set to Smart neither link will work.
Sorry if my typing is terrible. I can't feel my finger tips as they are covered in super glue. I just finished my landscape architecture final project.
Re:Solution - no conditionals or templates!
on
Programming Puzzles
·
· Score: 1
You are missing the point which is to use an indexed array and pointers to switch paths instead of the other features of the language.
If these compile to a branch its because the compiler made them conditional and it was the compiler's choice to be that way. It could have compiled them differently. I included the == just 'cause, so we can remove it and switch the indices back. The >= is a little bit harder.
First we subtract the numbers a la:
num - max
Now we have either a positive number, a negative number, or 0. Assuming int is a 16 bit number then the most significant bit will be 0 if this result is 0 or positive, and 1 if its negative. Let's make this the least significant bit:
(num - max) >> 15
Now we have either a 0 if num >= max or a 1 if num < max. We can flip these around easily enough with a bitwise not to flip all the bits then and with one to get rid of 1s in the leading bits:
(~((num - max) >> 15 )) & 1
We can rewite the whole program as (this is not debugged - I'm in Windows for a change):
int ge(int a, int b) { return (~((a - b) >> 15 )) & 1; }
Solution - no conditionals or templates!
on
Programming Puzzles
·
· Score: 1
This C solution uses no control structures, loops, templates, conditionals, ? operator, or shortcut evaluation of logical operators to accomplish this. I could use just one printf if I wanted, but this is supposed to be fairly clear. It works if you reverse the array and remove the == operator, but I think switchptr is more robust as written.
That's some pretty impressive edge detection, thanks for pointing it out. A related problem is to identify areas in an image that are the same thing. The best algorithm I know of for segmenting images is Leo Grady's brand new iso parametric graph partitioning method. His work is at http://cns.bu.edu/~lgrady/. His PhD thesis is probably the best place to start.
Re:Sounds comprehensive
on
Knoppix Hacks
·
· Score: 2, Interesting
I made this guide, which is sadly in need of an update*. One of the most frequent email questions I get is, "Help - I lost my partition table, how do I get it back"? There's a great utility included in knoppix called gpart that searches the hard drive for partitions and constructs a new partition table. It also makes backups of master boot records. Its amazing all of the stuff they've thought to include.
* If anyone wants to mail me a copy of the newest knoppix cd I'll update it. I hate downloading ISO images. My address is 888 E 18th Ave, Apt 8. Eugene, OR 97401.
This tells us more about what we don't know about honeybees than it tells us about the cataclysmic event of 65 million years ago. And its not much of a mystery anyway - many types of bees hibernate, and can be kept for years in a freezer for pollinating orchards.
It was a mass extinction, not a total extinction. If nothing had survived, we would have started over again 65 million years ago at a few species near ocean bottom vents. Many, many, many land plants and creatures survived. A much more interesting question would be, "How did Cretotrigona prisca or their close ancestors survive the mass extinction event about 65 million years ago"?
Marginal cost is the cost of producing one more item, and has nothing to do with R&D and initial costs. Please mod parent up, and the grandparent down.
real 0m0.060s user 0m0.020s sys 0m0.010s cedric@cedric:~$ time gcc fibo.c -O3 -o fibo.gcc
real 0m0.441s user 0m0.150s sys 0m0.030s cedric@cedric:~$ time./fibo.gcc 1
real 0m0.074s user 0m0.000s sys 0m0.000s cedric@cedric:~$ time./fibo.tcc 1
real 0m0.072s user 0m0.000s sys 0m0.000s cedric@cedric:~$ time./fibo.gcc 1
real 0m0.071s user 0m0.000s sys 0m0.000s cedric@cedric:~$ time./fibo.tcc 1
real 0m0.074s user 0m0.000s sys 0m0.000s cedric@cedric:~$ time./fibo.gcc 1
real 0m0.071s user 0m0.000s sys 0m0.010s cedric@cedric:~$ time./fibo.tcc 1
real 0m0.070s user 0m0.000s sys 0m0.000s
I can't believe this is what GCLS uses as a benchmark. The running time is so short it's probably all start-up and shutdown. Anyway, the numbers are pretty fair for a compiler compiling code that was tweaked to get the other compiler to be as fast as possible.
I've always wanted an input device that had a handful of sliders and knobs for adjusting things like color balance, 3d position and rotation in modeling programs, etc. I've envisioned a little device with about 3 sliders and knobs, and maybe a couple of buttons. The joystick, mouse, and even keyboard arrow input should all be abstracted to an axis input which would also work for this kind of device. Forget touchpads - the power in a device like this would be the tactile sensation, the ability to remember how far you pushed the slider, turned the knob, etc. Also people can easily turn knobs and move sliders very quickly for corse adjustments and very slowly and delicately for fine adjustments.
In real life that would be the sort of thing to do. But in this silly competetion, since we can't get to the compiler, I was trying to do it through one incorrect malloc like this:
unsigned long *Tally; *Tally = (unsigned long) malloc(sizeof(unsigned long)*256);
to get the array to hold the counts, and trying to manipulate the code so that when the compiler compiles it, the unchanged pointer Tally ends up being in another portion of the program that would be interpereted as some totally wacko initial counts, without segfaulting. Of course this wouldn't pass inspections before the election nor get the correct counts, but it would have code that was symmetrical across candidates, while favoring one with a specific version of the compiler (fortunantly I'm working on debian too). Unfortunantly I havn't found one yet that even compiles with tally pointing into the program to begin with. It would be easier if I changed their code around a bit to put the favored candidate in the 0th index of the array.
The first amendment guarantees the right to hold stupid, idiotic political opinions. If you don't like it, there are other countries with different constitutions, feel free to emigrate. Personally, I like the Bill of Rights just fine, thank you.
The first amendment guarantees the right to hold stupid, idiotic political opinions. If you don't like it, there are other countries with different constitutions, feel free to emigrate. Personally, I like the Bill of Rights just fine, thank you.
An administrative law judge over-ruled an administrative decision Friday. The 15 counties that use touch-screen voting systems must be able to perform manual recounts in extremely close elections.
Or ask the Wright brothers. They sunk their company by investing all their time in litigation against competitors instead of development and innovation.
Slashdot Mirror
http://www.intel.com/software/products/opensource/
http://www.intel.com/cd/ids/developer/asmo-na/eng/ 52779.htm
You need to incorporate the product of the following conditions:
Patient's certainty:
Uncertain
Certain and correct
Certain and incorrect
Getting the drug:
Yes
No
This would leave us with the following groups:
Not sure and recieving drug
Not sure and not recieving drug
Certain of recieving drug and recieving drug
Certain of not recieving drug and not recieving drug
Certain of not recieving drug and recieving drug
Certain of recieving drug and not recieving drug
Then you need many replicates, include all the interactions in your ANOVA (i.e. do it the simple, correct way with none of the monkeying around that bad statisticians will prescribe), and report the results that pass Ficher's LSD (the most powerful detector of significant difference), and possibly also include results passing more stringent significance tests.
Then we will have the answer. Wait 4 years for people to do it with other drugs and make more complicated expirements with more degrees of freedom and it will be canon.
And yes, you will have to LIE to and DECIEVE your patients. This is considered unethical, so this simple basic expirement will never be done in the "developed" world. There can be no waiver of "you may or may not recieve medication" because if introduced it would place everyone in the group "Uncertain." If the patients have a bias towards believing that a medical experiment does not medicate as stated then the patients must not know that they are participating in the experiment.
Yep. Nice University projects. No source code for either, as far as I can tell.
Don't know if this is new or not, but a streaming peer-to-peer protocol like bittorrent would be pretty cool. It could be used to inexpensively broadcast audio or video almost live, potentially making news reporting available to a wider selection of journalists. Checksums on data would obviously be a problem here and malicious nodes in the network would have an easier time of disrupting communications. This mechanism needs to be independent of media type, and rely on being used in combination with file formats that can be picked up and played from any small chunk. The client could decide which portions of the stream it would rather get, sacrificing liveness to get as much as possible, trying to pick up the nearest blocks in the future first to stay as smooth as possible, or minimizing buffer size and going after the most recent blocks to stay as live as possible.
I tried this against my girlfriend's computer with a wireless network connection. She has XP home SP2. With the firewall ON it caused 100% CPU usage during the attack, and loss of connection. Connection was restored after sending ^C to hping2. With the firewall OFF it caused 100% CPU usage during the attack, and loss of connection. CPU usage and connection were not restored for about a minute after the end of the attack. The included firewall with default settings is NOT a solution to this problem on XP Home SP2.
You're right. Spyware or adware (these would probably benefit from blocking windows updates more then viruses) can emulate whatever fingerprints of Wine Microsoft uses for countermeasures to prevent updates. Wine doesn't need to enter the arms race against Microsoft - more nefarious groups with financial incentive will do it for them.
Kudos, that was a very good idea!
$10 per month is way too much to pay for any computer game.
People are using knoppix for this all the time; I can tell by the amount of email I deal with on the subject.
Konqueror 3.3.1 is vulnerable if Global JavaScript Policies - Open New Windows is set to Allow and you use the no popup blocker link. With Open New Windows set to Smart neither link will work.
Sorry if my typing is terrible. I can't feel my finger tips as they are covered in super glue. I just finished my landscape architecture final project.
You are missing the point which is to use an indexed array and pointers to switch paths instead of the other features of the language.
If these compile to a branch its because the compiler made them conditional and it was the compiler's choice to be that way. It could have compiled them differently. I included the == just 'cause, so we can remove it and switch the indices back. The >= is a little bit harder.
First we subtract the numbers a la:
num - max
Now we have either a positive number, a negative number, or 0. Assuming int is a 16 bit number then the most significant bit will be 0 if this result is 0 or positive, and 1 if its negative. Let's make this the least significant bit:
(num - max) >> 15
Now we have either a 0 if num >= max or a 1 if num < max. We can flip these around easily enough with a bitwise not to flip all the bits then and with one to get rid of 1s in the leading bits:
(~((num - max) >> 15 )) & 1
We can rewite the whole program as (this is not debugged - I'm in Windows for a change):
int ge(int a, int b) {
return (~((a - b) >> 15 )) & 1;
}
void * switchptr(int condition, void * trueptr, void * falseptr) {
void * ptrarray[2];
ptrarray[1] = trueptr;
ptrarray[0] = falseptr;
return ptrarray[condition];
}
int stopnums(int num, int max) {
return 0;
}
int printnums(int num, int max) {
int (*nextfun) (int, int);
printf("%i\n", num);
nextfun = switchptr( ge(num,max), &stopnums, &printnums);
nextfun(num + 1, max);
printf("%i\n", num);
return 0;
}
int main () {
return printnums(1, 100);
}
This C solution uses no control structures, loops, templates, conditionals, ? operator, or shortcut evaluation of logical operators to accomplish this. I could use just one printf if I wanted, but this is supposed to be fairly clear. It works if you reverse the array and remove the == operator, but I think switchptr is more robust as written.
void * switchptr(int condition, void * trueptr, void * falseptr) {
void * ptrarray[2];
ptrarray[0] = trueptr;
ptrarray[1] = falseptr;
return ptrarray[condition == 0];
}
int stopnums(int num, int max) {
return 0;
}
int printnums(int num, int max) {
int (*nextfun) (int, int);
printf("%i\n", num);
nextfun = switchptr( num >= max, &stopnums, &printnums);
nextfun(num + 1, max);
printf("%i\n", num);
return 0;
}
int main () {
return printnums(1, 100);
}
That's some pretty impressive edge detection, thanks for pointing it out. A related problem is to identify areas in an image that are the same thing. The best algorithm I know of for segmenting images is Leo Grady's brand new iso parametric graph partitioning method. His work is at http://cns.bu.edu/~lgrady/. His PhD thesis is probably the best place to start.
* If anyone wants to mail me a copy of the newest knoppix cd I'll update it. I hate downloading ISO images. My address is 888 E 18th Ave, Apt 8. Eugene, OR 97401.
This tells us more about what we don't know about honeybees than it tells us about the cataclysmic event of 65 million years ago. And its not much of a mystery anyway - many types of bees hibernate, and can be kept for years in a freezer for pollinating orchards.
It was a mass extinction, not a total extinction. If nothing had survived, we would have started over again 65 million years ago at a few species near ocean bottom vents. Many, many, many land plants and creatures survived. A much more interesting question would be, "How did Cretotrigona prisca or their close ancestors survive the mass extinction event about 65 million years ago"?
Marginal cost is the cost of producing one more item, and has nothing to do with R&D and initial costs. Please mod parent up, and the grandparent down.
Thank you
--Cedric
fibo.c from GCLS:
./fibo.gcc
./fibo.tcc
./fibo.gcc
./fibo.tcc
./fibo.gcc
./fibo.tcc
cedric@cedric:~$ time tcc fibo.c -o fibo.tcc
real 0m0.060s
user 0m0.020s
sys 0m0.010s
cedric@cedric:~$ time gcc fibo.c -O3 -o fibo.gcc
real 0m0.441s
user 0m0.150s
sys 0m0.030s
cedric@cedric:~$ time
1
real 0m0.074s
user 0m0.000s
sys 0m0.000s
cedric@cedric:~$ time
1
real 0m0.072s
user 0m0.000s
sys 0m0.000s
cedric@cedric:~$ time
1
real 0m0.071s
user 0m0.000s
sys 0m0.000s
cedric@cedric:~$ time
1
real 0m0.074s
user 0m0.000s
sys 0m0.000s
cedric@cedric:~$ time
1
real 0m0.071s
user 0m0.000s
sys 0m0.010s
cedric@cedric:~$ time
1
real 0m0.070s
user 0m0.000s
sys 0m0.000s
I can't believe this is what GCLS uses as a benchmark. The running time is so short it's probably all start-up and shutdown. Anyway, the numbers are pretty fair for a compiler compiling code that was tweaked to get the other compiler to be as fast as possible.
I've always wanted an input device that had a handful of sliders and knobs for adjusting things like color balance, 3d position and rotation in modeling programs, etc. I've envisioned a little device with about 3 sliders and knobs, and maybe a couple of buttons. The joystick, mouse, and even keyboard arrow input should all be abstracted to an axis input which would also work for this kind of device. Forget touchpads - the power in a device like this would be the tactile sensation, the ability to remember how far you pushed the slider, turned the knob, etc. Also people can easily turn knobs and move sliders very quickly for corse adjustments and very slowly and delicately for fine adjustments.
In real life that would be the sort of thing to do. But in this silly competetion, since we can't get to the compiler, I was trying to do it through one incorrect malloc like this:
;
unsigned long *Tally;
*Tally = (unsigned long) malloc(sizeof(unsigned long)*256)
to get the array to hold the counts, and trying to manipulate the code so that when the compiler compiles it, the unchanged pointer Tally ends up being in another portion of the program that would be interpereted as some totally wacko initial counts, without segfaulting. Of course this wouldn't pass inspections before the election nor get the correct counts, but it would have code that was symmetrical across candidates, while favoring one with a specific version of the compiler (fortunantly I'm working on debian too). Unfortunantly I havn't found one yet that even compiles with tally pointing into the program to begin with. It would be easier if I changed their code around a bit to put the favored candidate in the 0th index of the array.
The first amendment guarantees the right to hold stupid, idiotic political opinions. If you don't like it, there are other countries with different constitutions, feel free to emigrate. Personally, I like the Bill of Rights just fine, thank you.
My eyes thank you in advance.
Current phone numbers
Massachusetts:
617-861-9507
New Hampshire:
603-413-3623
Rhode Island:
401-648-7950
New York:
212-660-4245
Washington:
206-888-5400
Florida:
321-485-0033
772-974-0041
786-743-0033
An amusing thought: Imagine the next person to get such a number after a service like this closes.
Cedric Shock
888 E 18th Ave Apt 8
Eugene, Oregon 97401
(541) 343-6640
An administrative law judge over-ruled an administrative decision Friday. The 15 counties that use touch-screen voting systems must be able to perform manual recounts in extremely close elections.
Or ask the Wright brothers. They sunk their company by investing all their time in litigation against competitors instead of development and innovation.