The way they proved the Four Color Theorem was to first delineate ALL possible ways the theorem needed to be proven true and then use computer to prove true each combination. They basically generated a lot of graphs and then demonstrated that each graph had a four color partite set. The graphs they chose were representative of all possible graphs that could exist.
So 1) it required a lot of thinking as to what graphs were representative of all possible graphs and 2) the computer just did the grunt work of performing the coloring of each graph 3) A human could have done this, but it would have taken a long, long time 4) who cares about four color theorem anyways?
Whoever created the Wikipedia article is a moron. If they were going to expand out the IUPUC form for some protein (a molecule which has its own nomenclature btw) then they should have chosen Dystrophin.
The Dystrophin exon (coding sequence) is over 2.4 MILLION bases or 800,000 amino acids long.
Using the moron's system of naming proteins, Dystrophin's name would be ~3.5 MILLION characters long.
Wow and this made it past the Slashdot editors. Good job guys! Maybe it's because the editors have no clue about most science. Maybe they need to hire someone who does.
I'll second the part about Intel having a kickass Fortran95 compiler. Using interprocedural optimization speeds up programs by like 2x compared to not using it.
And the Linux version is FREE (as in beer) for non-commercial usage.
I've programmed in other languages, but I find I get more done in less time using Fortran95. Why? I only code for scientific computing and I can give a rats ass about 'features' such as garbage collection, strange usage of pointers, etc etc that Fortran95 doesn't have.
In the end, Fortran95 is fantastic for scientific computing and the only other language that comes close is C. C is just as fast, but (omfg) you can make a lot of mistakes that will just suck the time from you. No thanks.
Sitting right next to the online version of the paper could be... dum dum dum... an advertisement!
Even better, most people download the PDF version, print it out, read it, and stick it in their cabinet for future reference. Sticking in an ad midway through the paper would not only cause the person to read the ad (even fleetingly), but also to SAVE the ad where others might pick it up and read it.
What sort of ads would be relevant? Well, that depends on the paper, of course. Experimental papers using culturing could feature ads on media supplies, incubators, etc etc. Applied math papers could feature ads on Matlab, Mathematica, stats programs, etc etc.
Or, even better, have Google scan the paper and dynamically place ads in it based on its content.
Why didn't anyone suggest this yet? (Or did the daily rant on editorial copy staff cloud the first 100 comments??)
What if I ate 2 tons of the selenium plant and died?
What if I ate 2 tons of beas who ate the selenium plant and died?
What if I got stung by a bea who ate the selenium plant and died from an acute allergy to beas?
What if I got hit by a car whose driver was a lab tech whose company developed the plant which absorbs selenium?
What if I ate two tons of dirt which contained so much selenium, that would otherwise be absorbed by genetically engineered planted, that I died?
What if I ate two tons of dirt? Eeww.
Ok, enough. There are so many irrelevant "what if" questions. The main questions are: What are the benefits of this plant and how do they compare to the risks of this plant _relative_ to the risk of exististing on this planet (including us, other organisms, and the planet itself). Most genetically engineered (and commercially viable) plants have so much benefit that their risks are wildly outweighed. Even with the "what if"'s.
The Gates Foundation just donated $46 million to a research group / company in Berkeley so that they could research the genetic engineering of a malaria drug and produce enough to make the drug super cheap.
The Slashdot post makes it seem like the people at MIT invented the idea of synthetic biology. Well, I'm sure the good guys over at MIT would agree that the hallmark papers that started the craze didn't originate from MIT...they came from Princeton & Berkeley and there's plenty of other institutions who are making major contributions (some greater than MIT's), especially on the science end.
That being said, their idea of Biobricks is very innovative and they did host the first conference on the topic. So the popular press can be easily misled.
Don't feel bad about being rejected by Slashdot. It's like a monkey throwing away caviar. The editors sometime overlook obviously interesting stories (to us) that, in due time, become obviously interesting to them.
And then I sometimes think they just randomly select stories. Shakespeare via 1 million submitters.
I find that most journals charge per page and charge per color figure. Typical costs can run to $2000 easily.
Ever try to cram results into a readable black and white figure? It's really hard! I think journals should offer free online color figures. Anyways, that's just about the only way I retrieve articles anymore.
Fyi, my advisor wants me to publish in PLOS Comp. Bio., but (if I have a choice) I'd rather send papers to PNAS. Another open access journal (6 months after publication). I like the idea of PLOS though and I think it'll make a good contribution to the field.
If you consider the citation network as a directed graph, the degree of the nodes follow a power law. See authors Strogatz, Albert, and Barabasi for the details.
One of the familiar results of power law networks is that, out of all papers, only 20% of them are cited 80% of the time. The famous get more famous and the obscure stay obscure.
Not to sound like an arrogant American, but you do realize that the U.S is the first modern democracy in the world. I'm excluding Athens, Rome, etc.
When the U.S broke off from England, the founding fathers extensively analyzed the ancient governments of Greece & Rome to create our national constitution. Shortly thereafter, France declared itself a republic and modeled their constitution off of ours. The UK came later.
The U.S inherited common law from England, but the U.S was most definately the pioneer in modern democracy and subsequently became the model for many future democracies.
And, remember, the only reason why Europe mostly contains democratic governments is because the U.S held back the Soviet Union and indirectly caused its downfall. At the very least, the Warsaw Pact countries would still be communist.
How would you reform the current departments of Justice, Treasury, Homeland Security, and Defense so that intelligence gathering and analysis is faster and more thorough? Specifically, what do you suggest would be the optimal agency structure to provide the fastest response to actionable intelligence and better collection of information for long term analysis?
The 9/11 Commission Report details the enormous failings of our intelligence establishment. The system was truly "Blinking Red" and yet neither the FBI, CIA, INS, etc could find the link in the pattern. What would you specifically do to force coordination between different agencies and encourage more intelligence sharing?
The Information Technology used by the FBI, CIA, and other agencies are woefully inadequate for current requirements. We live in the most technologically advanced country in the world and yet our government agencies do not possess this technology to protect us. Will you spend the money to give it to them?
Notice how I said a "better question". Why? Because ANY proposed voting system that lacks a trusted and verifiable way to collect and tally votes is truly a WASTE OF TIME.
The argument against the electoral college has been going on for 50 years. It's not new. Electronic voting is new. Paper-less voting is new. Let's not rehash old boring questions with the same old stodgy answers. The future of accountable democracy is in the hands of a computer that we supposedly must just 'trust' to give the right answer.
Any computer programmer (or any Slashdotter, for that matter) should know that computers only give you the answer the human operator programs it to do. There is no inherent security to it. There is no trust without validation.
We're messing with our democracy and people are still ogling the LCD screen.
What happens if democracy finally flourishes in Iraq and/or Afghanistan and the government they select is one composed primarily of clerics, mullahs, etc? While democratic choices may not be the right ones, do we, as a fellow democracy, respect the choice of another people even though the result may hinder the national security of our nation?
I have no doubt that a democracy that choses to be a theocracy with little pluralism or tolerance will quickly become a failed state. Ideas supposedly mandated by a higher power are not subject to compromise. And yet compromise is the essence of a successful democracy.
Do we intercede in a democratically made, but unwise, choice? Or will doing so actually exacerbate the problem and further implicate America as an imperialist power?
I'd like an answer now, because it's going to happen in a couple of years.
A better electorate question might ask the candidates how they feel about the usage of paper-less electronic voting machines which have proven vulnerabilities.
There are many ways to tally the votes (electoral college, proportional electorate by state, etc), but if the votes themselves are vulnerable to fraud then democracy of any type is in peril.
Intel has a set of optimized mathematical libraries for all sorts of applications (linear algebra, image processing, random number generation, FFT's, etc). Not only are they optimized for Intel systems, but they save you the time of coding it yourself.
Intel also provides the VTune Performance Analyzer, which allows you to trace the path through your programs and determine where the bottlenecks are.
I've used the Intel Linux Fortran compiler and I am very happy with it. Code that runs fine on my Sun workstation (950 Mhz, 6 gig RAM) at school works 4-5x faster on my home PC (2.8 Ghz, 1 gig RAM). It's got all the fancy optimization options, but a simple -O3 -ipo will get you 90% there.
Their computer system went down for at least *3* hours in Minneapolis, shutting down the entire terminal. They couldn't check flight plans, ticket information, scheduling, logistics, etc. No planes in, no planes out.
You'd think they'd have redundancy and backups, but they probably don't. That requires some planning beyond the immediate need of the company and, even if it's more profitable to invest in backups, long term planning simply isn't considered as much.
This happens to my University all the time. The power goes out in one building for a few hours and services across the entire University are disrupted completely. This building happens to house most of the license servers for important software, but no one would _ever_ think of putting a backup license server in another building _just in case_. No, that'd be thinking ahead.
True in most respects, but the relationship between an electric field and a magnetic field is mathematically similar to the idea of space and time being relative. It's a mathematical shift.
I should warn that I'm not a physicist (I've just taken quite a few physics classes) and so don't ask me what exactly the shift is, but I remember having known it at one point.;)
Slashdot Mirror
The way they proved the Four Color Theorem was to first delineate ALL possible ways the theorem needed to be proven true and then use computer to prove true each combination. They basically generated a lot of graphs and then demonstrated that each graph had a four color partite set. The graphs they chose were representative of all possible graphs that could exist.
So
1) it required a lot of thinking as to what graphs were representative of all possible graphs and
2) the computer just did the grunt work of performing the coloring of each graph
3) A human could have done this, but it would have taken a long, long time
4) who cares about four color theorem anyways?
Hey, I'm in a bad mood and the icon is a giant foot.
And it's not really that funny. I mean, if I told the joke to all of my engineer friends, there'd be crickets. And they're engineers. That bad.
Whoever created the Wikipedia article is a moron. If they were going to expand out the IUPUC form for some protein (a molecule which has its own nomenclature btw) then they should have chosen Dystrophin.
The Dystrophin exon (coding sequence) is over 2.4 MILLION bases or 800,000 amino acids long.
Using the moron's system of naming proteins, Dystrophin's name would be ~3.5 MILLION characters long.
Wow and this made it past the Slashdot editors. Good job guys! Maybe it's because the editors have no clue about most science. Maybe they need to hire someone who does.
Well, thank you very much.
I've read through the --help for xlf about four times without finding that flag.
Except that the IBM's xlf Fortran95 compiler only accepts .f files.
.f90 files to .f).
Why, why, oh why?! It is really, really annoying (and I've seen no flag to auto-convert
I'll second the part about Intel having a kickass Fortran95 compiler. Using interprocedural optimization speeds up programs by like 2x compared to not using it.
And the Linux version is FREE (as in beer) for non-commercial usage.
I've programmed in other languages, but I find I get more done in less time using Fortran95. Why? I only code for scientific computing and I can give a rats ass about 'features' such as garbage collection, strange usage of pointers, etc etc that Fortran95 doesn't have.
In the end, Fortran95 is fantastic for scientific computing and the only other language that comes close is C. C is just as fast, but (omfg) you can make a lot of mistakes that will just suck the time from you. No thanks.
Sitting right next to the online version of the paper could be ... dum dum dum ... an advertisement!
Even better, most people download the PDF version, print it out, read it, and stick it in their cabinet for future reference. Sticking in an ad midway through the paper would not only cause the person to read the ad (even fleetingly), but also to SAVE the ad where others might pick it up and read it.
What sort of ads would be relevant? Well, that depends on the paper, of course. Experimental papers using culturing could feature ads on media supplies, incubators, etc etc. Applied math papers could feature ads on Matlab, Mathematica, stats programs, etc etc.
Or, even better, have Google scan the paper and dynamically place ads in it based on its content.
Why didn't anyone suggest this yet? (Or did the daily rant on editorial copy staff cloud the first 100 comments??)
The world would implode.
What if I ate 2 tons of the selenium plant and died?
What if I ate 2 tons of beas who ate the selenium plant and died?
What if I got stung by a bea who ate the selenium
plant and died from an acute allergy to beas?
What if I got hit by a car whose driver was a lab tech whose company developed the plant which absorbs selenium?
What if I ate two tons of dirt which contained so much selenium, that would otherwise be absorbed by genetically engineered planted, that I died?
What if I ate two tons of dirt? Eeww.
Ok, enough. There are so many irrelevant "what if" questions. The main questions are: What are the benefits of this plant and how do they compare to the risks of this plant _relative_ to the risk of exististing on this planet (including us, other organisms, and the planet itself). Most genetically engineered (and commercially viable) plants have so much benefit that their risks are wildly outweighed. Even with the "what if"'s.
My two cents.
The Gates Foundation just donated $46 million to a research group / company in Berkeley so that they could research the genetic engineering of a malaria drug and produce enough to make the drug super cheap.
I think civilization is doing just fine.
The Slashdot post makes it seem like the people at MIT invented the idea of synthetic biology. Well, I'm sure the good guys over at MIT would agree that the hallmark papers that started the craze didn't originate from MIT...they came from Princeton & Berkeley and there's plenty of other institutions who are making major contributions (some greater than MIT's), especially on the science end.
That being said, their idea of Biobricks is very innovative and they did host the first conference on the topic. So the popular press can be easily misled.
Don't feel bad about being rejected by Slashdot. It's like a monkey throwing away caviar. The editors sometime overlook obviously interesting stories (to us) that, in due time, become obviously interesting to them.
And then I sometimes think they just randomly select stories. Shakespeare via 1 million submitters.
I find that most journals charge per page and charge per color figure. Typical costs can run to $2000 easily.
Ever try to cram results into a readable black and white figure? It's really hard! I think journals should offer free online color figures. Anyways, that's just about the only way I retrieve articles anymore.
Fyi, my advisor wants me to publish in PLOS Comp. Bio., but (if I have a choice) I'd rather send papers to PNAS. Another open access journal (6 months after publication). I like the idea of PLOS though and I think it'll make a good contribution to the field.
You might find this interesting:
If you consider the citation network as a directed graph, the degree of the nodes follow a power law. See authors Strogatz, Albert, and Barabasi for the details.
One of the familiar results of power law networks is that, out of all papers, only 20% of them are cited 80% of the time. The famous get more famous and the obscure stay obscure.
IANAGT - I Am Not a Graph Theoretician
Thank God!
there is nothign to fear
Except bad spelling!
Not to sound like an arrogant American, but you do realize that the U.S is the first modern democracy in the world. I'm excluding Athens, Rome, etc.
When the U.S broke off from England, the founding fathers extensively analyzed the ancient governments of Greece & Rome to create our national constitution. Shortly thereafter, France declared itself a republic and modeled their constitution off of ours. The UK came later.
The U.S inherited common law from England, but the U.S was most definately the pioneer in modern democracy and subsequently became the model for many future democracies.
And, remember, the only reason why Europe mostly contains democratic governments is because the U.S held back the Soviet Union and indirectly caused its downfall. At the very least, the Warsaw Pact countries would still be communist.
Dear Sirs,
How would you reform the current departments of Justice, Treasury, Homeland Security, and Defense so that intelligence gathering and analysis is faster and more thorough? Specifically, what do you suggest would be the optimal agency structure to provide the fastest response to actionable intelligence and better collection of information for long term analysis?
The 9/11 Commission Report details the enormous failings of our intelligence establishment. The system was truly "Blinking Red" and yet neither the FBI, CIA, INS, etc could find the link in the pattern. What would you specifically do to force coordination between different agencies and encourage more intelligence sharing?
The Information Technology used by the FBI, CIA, and other agencies are woefully inadequate for current requirements. We live in the most technologically advanced country in the world and yet our government agencies do not possess this technology to protect us. Will you spend the money to give it to them?
Such trolling.
But, here's my rebuttal. I'm only going to reply once.
The Cray X1 supercomputer (156th fastest supercomputer in the world) has only two supported compilers:
C
Fortran95/2k
I can do all my scientific computing in Fortran whereas it would take me twice as long to code it in C (I do know both).
So I get to do twice as much. And for those who know higher level mathematics, Fortran's vector syntax is much more comfortable.
I'm not going to get into further argument. This is my last reply on the topic.
Notice how I said a "better question". Why? Because ANY proposed voting system that lacks a trusted and verifiable way to collect and tally votes is truly a WASTE OF TIME.
The argument against the electoral college has been going on for 50 years. It's not new. Electronic voting is new. Paper-less voting is new. Let's not rehash old boring questions with the same old stodgy answers. The future of accountable democracy is in the hands of a computer that we supposedly must just 'trust' to give the right answer.
Any computer programmer (or any Slashdotter, for that matter) should know that computers only give you the answer the human operator programs it to do. There is no inherent security to it. There is no trust without validation.
We're messing with our democracy and people are still ogling the LCD screen.
This is a very good question.
What happens if democracy finally flourishes in Iraq and/or Afghanistan and the government they select is one composed primarily of clerics, mullahs, etc? While democratic choices may not be the right ones, do we, as a fellow democracy, respect the choice of another people even though the result may hinder the national security of our nation?
I have no doubt that a democracy that choses to be a theocracy with little pluralism or tolerance will quickly become a failed state. Ideas supposedly mandated by a higher power are not subject to compromise. And yet compromise is the essence of a successful democracy.
Do we intercede in a democratically made, but unwise, choice? Or will doing so actually exacerbate the problem and further implicate America as an imperialist power?
I'd like an answer now, because it's going to happen in a couple of years.
A better electorate question might ask the candidates how they feel about the usage of paper-less electronic voting machines which have proven vulnerabilities.
There are many ways to tally the votes (electoral college, proportional electorate by state, etc), but if the votes themselves are vulnerable to fraud then democracy of any type is in peril.
Intel has a set of optimized mathematical libraries for all sorts of applications (linear algebra, image processing, random number generation, FFT's, etc). Not only are they optimized for Intel systems, but they save you the time of coding it yourself.
Intel also provides the VTune Performance Analyzer, which allows you to trace the path through your programs and determine where the bottlenecks are.
I've used the Intel Linux Fortran compiler and I am very happy with it. Code that runs fine on my Sun workstation (950 Mhz, 6 gig RAM) at school works 4-5x faster on my home PC (2.8 Ghz, 1 gig RAM). It's got all the fancy optimization options, but a simple -O3 -ipo will get you 90% there.
My two bits.
Their computer system went down for at least *3* hours in Minneapolis, shutting down the entire terminal. They couldn't check flight plans, ticket information, scheduling, logistics, etc. No planes in, no planes out.
You'd think they'd have redundancy and backups, but they probably don't. That requires some planning beyond the immediate need of the company and, even if it's more profitable to invest in backups, long term planning simply isn't considered as much.
This happens to my University all the time. The power goes out in one building for a few hours and services across the entire University are disrupted completely. This building happens to house most of the license servers for important software, but no one would _ever_ think of putting a backup license server in another building _just in case_. No, that'd be thinking ahead.
True in most respects, but the relationship between an electric field and a magnetic field is mathematically similar to the idea of space and time being relative. It's a mathematical shift.
;)
I should warn that I'm not a physicist (I've just taken quite a few physics classes) and so don't ask me what exactly the shift is, but I remember having known it at one point.