Other scientists shouldn't be using the same software to reproduce the results. That's pretty obvious.
What we're talking about here is an added layer of assurance that the initial results are as good as they can be. Having an independent auditing of code means that the methodology described in a paper is the one actually used in the software. It fixes simple errors, and ensures that there's no fraud. The programmers who would do this are not being taken away from creating new scientific research, because they're not scientists, nor are they competing for the same research grants. What's more, it's free QA that keeps the science more scientific.
You have a whole lot of coders looking at the code and making sure it's a perfect implementation of the model or algorithms described by the scientists, and it won't matter if somebody uses the same code to reproduce the results, because you'll know the software is working perfectly. All that's left then is for other scientists to criticize the methodology and assumptions of the research as part of the normal peer review process. It turns the code into just another tool at a scientist's disposal.
We're talking about science here. The whole point of science is challenging something repeatedly and thoroughly, and if it stands up, you know it's true. Adding more challenges is the most scientific thing you can do.
Truth is a defense against libel. If you disclose all your code, and problems with your code are found and fixed, it's actually harder to be sued over your results.
The scientific community can actually run the code and verify that it produces the same output on another machine in another lab. What we're talking about here is programmers making sure the code is a perfect representation of the scientific formulas described by the research paper. Basically, did the researchers follow the methodology they claimed to? Is data being lost or altered due to limitations of the software? This doesn't take any special knowledge of the science involved, just the calculations involved.
I'm quite sure there would be exceptions to state secrets. I'm also sure the folks who built F-15s did a heck of a lot more testing and verification than your typical scientific research at a university.
There's going to be people criticizing code for improper documentation, but I feel that's something that should be a concern in general for scientific research. If you don't document your code properly, you aren't properly documenting your methodology. Code that works, though, regardless of style will probably not get too much criticism. I can't imagine some suggestions for efficiency would hurt though. If you can do your calculations more quickly, or with less computing resources required, you can accomplish more science in less time, for cheaper.
Really what we're talking about here is verifying that your desired algorithms are what actually ended up in the code. Having some extra eyes on your code can only be a good thing for that reason. Programmers will leave the science behind the algorithms up to the scientists.
Verifying that the code used by the researchers is consistent with the methods described, without accidental errors, is basically what we're talking about here, and is essentially equivalent to writing your own code. Eliminate researcher errors in their coding, and you're left with one less possible flaw in an experiment.
The release of raw data along with the results is a good thing, I agree. But somebody checking to make sure the scientists' algorithms are implemented correctly in an existing program is easier, and more likely to happen, than somebody independently coding their own implementation, particularly in scientific research where the computing power necessary is prohibitive.
Whatever he wants. Since when was having information a bad thing? I'm sure a lot of scientists would be a lot happier if there was a greater understanding and curiosity about science in the world. Ultimately it will translate to more money for research, but in the shorter term, a more scientifically-conversant society can only be a good thing. Yes, there will be stuff that most of the population won't understand. But you'll use that as a basis for depriving everyone of that knowledge? There will be people who do understand it, and if it's only a single person understanding a single paper out of thousands, the entire exercise was still worth it.
If you're the only one looking at your code, and it's so complex, how do you know it's working as you intended it to? This is the question that is at the very heart of the issue we're discussing here. It really doesn't matter what the problem is you're working on, I assume the algorithms can be described in some logical/mathematical way that somebody can turn into code without understanding what it's all about. After all, there's got to be something at the heart of your project you can share with those 8 other scientists so they know your methodology.
What would be better, keeping your code to yourself and not knowing if there are even tiny bugs in the implementation of your algorithms, or releasing it and having interested parties examine it and tell you if your algorithms are actually converted properly to code or not? It doesn't matter if anybody can run it, or what the data is you're number-crunching on, I would think you'd want somebody to make sure you didn't type 1+2 instead of 2+3.
As a (wo)man of science, I would expect your primary concern would be in the truth. You cannot know you're getting the truth if your code is never seen by another set of eyes. Nobody's going to be challenging your algorithms outside of the scientific community. What we're talking about is equivalent to a lab tech making sure your scales are calibrated. Trust me, no matter whatever special new language you created, there's more than a few computer scientists out there who can figure it out without you spending a month going back and documenting everything.
That said, if scientists were shamed enough into writing good code to begin with, and documenting everything, that wouldn't be a bad thing either, for science or programing.
Exactly. Let the scientific community evaluate the algorithms, and let the computer science community evaluate its implementation as code. The end result is better code and better science. How is that a bad thing? Any true scientist can't possibly object to something that will result in ensuring the validity of experiment results. As I understand it, scientists are after the truth, not merely research grants and political attention, right? Right??
I really don't think anybody actually expects a few million people challenging the basis of the algorithms here. Programmers will just be challenging the implementation and making sure it's correct.
There's probably other variations out there that can provide an energy source, things much harder, and thus never took off on Earth where energy was easy. I fully expect we'll end up finding life a lot more frequently than we expect at the moment. It'll mostly be stuff we can't have conversations with though.
There's also the whole thing about asteroids carrying life from one place to another. Just because it's hard for life to arise in a particular place in a solar system doesn't mean it can't get transported elsewhere and evolve to thrive, or at least survive.
They're not going to build their own space shuttle if they get their hands on our space shuttle plans. They'll learn from our designs and build something else like a missile, or use the material science for some new fighter jet. It's better they start from scratch than get a leg-up from our designs. The shuttle may be obsolete, but what do you think we use to learn from and improve on?
The idea that they're all criminals is kind of silly. Most spies operating abroad actually work with diplomatic immunity, gathering relatively public information and doing analysis. There are spies, working with that immunity, who go about trying to recruit assets like Mr. Chung... somebody disaffected, or somebody who has a reason to feel a stronger tie towards another country, or merely somebody who can be bought. They're the criminals, but they're a fairly small portion of the intelligence community as a whole. They're the ones who take the biggest risks, and the ones who get the biggest rewards for their work, if they do it well.
Unfortunately, we live in a world that's not all that friendly. Without spies, there'd be a lot more chaos and death in the world. Spies are the ones that allow leaders to go into a meeting with another leader and tell them "we know you're doing x, so cut it out" and lets military powers stay balanced enough that nobody gains a massive advantage and goes to war. Basically, spies allow for diplomacy to flourish, and they prevent conflicts. China is fully justified in wanting to spy on us and build up technologically and militarily in order to ensure their interests are satisfied. We're equally as justified in wanting to stop them from doing that so our own interests are satisfied. As a patriot, I want my own nation to have the best spies and maintain or achieve superiority in all things. As a rational human being, I'm not going to begrudge any other nation the same desire.
I'm against the death penalty on moral grounds, but treason/spying cases are pretty much the only time when I hesitate to rule it out entirely. The cost of spying against my nation should be prohibitive, and I wouldn't be terribly upset if other nations followed suit. That's the risk assets take.
I would *hope* that we weren't letting foreign nationals have access to all our secrets. Now, giving clearance levels to naturalized citizens is another debate. Hopefully background checks get a little better after something like this, anyway.
As far as I'm concerned, it's rocket science, and that's worth keeping a lid on. Any kind of rocket science, unless you can buy a kit in the mail or a hobby store, has pretty serious potential applications. China getting their hands on space shuttle plans doesn't mean they'll build their own shuttle, it means they take some tiny piece of it and improve some military application like a missile or airplane.
In any case, China is trying to get every scrap of technology they can get their hands on, regardless of its potential applications or obsolescence level. Some stuff may be useful immediately, others may be stepping stones to other technologies. Shuttle tech is stuff we improve on to develop new technologies. It's better, from our national security standpoint, for them to be stuck starting from scratch than from getting an idea of where our technology is headed and what we're starting with.
There's always the so-called "military shuttle" rumors for just such a purpose during the Cold War. How likely there is that one exists or ever existed is debatable, of course, and it's pretty unlikely there's ever been an actual launch of one without somebody noticing. We do know that there's been other military/spy missions going on with the NASA shuttle program though, for satellites and such.
And you'll still be able to use VLC, there just won't be any new versions for a while. Chances are, as grandparent said, somebody else will come along soon enough. I'm glad this article is here because it might generate some interest, and maybe next week we'll get an article saying "VLC for Mac Saved!"
Yeah, VLC isn't perfect, but I much prefer it to MPlayer. Still, I'm not terribly worried. It's not like they're going to come to your house and uninstall whatever version you have now. And their big announcement is surely a way of getting some new developers to stand up and help out. Some new eyes on the code could be a good thing, at any rate. Presumably whoever does take over will have plenty of enthusiasm and a lot will get done.
It's free, and you probably have it installed already. Give it a whirl and see if it works for you. If not, pencil and paper works great for math notes....
As I understand it, It's illegal to put nukes in space (reactors are fine, I'm talking about things that intentionally go boom). It's the same international treaty or treaties that stopped us from putting nuclear-lasers and other Star Wars types of satellites in orbit that involved nuclear explosions. It's hard to keep MAD going if somebody has nuclear missiles in orbit pointed straight down at your cities. A successful first strike is way too easy with orbital bombardment. A lot harder when they can watch you launch from the ground with ICBMs.
If you play a video game about breaking somebody's arm, you're not actually breaking somebody's arm. You're showing restraint by not doing it in real life, instead choosing to do it virtually. So you're not doing lasting physical harm, your pleasure from it is irrelevant.
Dogs and cats are predatory animals. They evolved to be lethal and efficient killers. We domesticated them to help us kill animals for us (dogs for hunting food for us, cats to kill pests). Humans evolved to be lethal and efficient killers as well. None of us alive can't find many, many, many ancestors in our lineage who hunted animals for food, went to war, or did stuff like murdering children or committing rapes. If they didn't do those things, frankly you wouldn't exist. So these things are in our nature. Because we survived as a species and developed increased intelligence and more complex social structures, most of us can avoid doing nasty, despicable actions (usually by sending other people to do them in our place like nearly every war in the last few hundred years), but it's still part of us.
If I had a cat that played a video game involving biting a human arm, I wouldn't be worried in the slightest, so long as the cat never did it to a real human in real life. Cats don't play video games, however, so play bites are fine too. Play-murder is fine for humans as well. It used to be done in stuff like "cowboys and indians", or just with sticks. The fact that we have more advanced technology to do it makes no difference whatsoever. One outlet is as good as another, as far as I'm concerned.
I would love to see Journeyman Project get a reboot. I still load up my three Buried in Time CDs from time to time with an emulator. Never got to play the others in the series, but I gathered they were pretty good too. Great story, really interesting puzzles, and it was probably one of the best mystery-type games I've ever played.
Well, ignoring your exaggeration for a moment, the better point you could have made is that OS X comes out with a new major version a lot more often than windows does, so people wanting the latest and greatest have to buy more OS X versions in the same time a Windows user upgrades.
Slashdot Mirror
Other scientists shouldn't be using the same software to reproduce the results. That's pretty obvious.
What we're talking about here is an added layer of assurance that the initial results are as good as they can be. Having an independent auditing of code means that the methodology described in a paper is the one actually used in the software. It fixes simple errors, and ensures that there's no fraud. The programmers who would do this are not being taken away from creating new scientific research, because they're not scientists, nor are they competing for the same research grants. What's more, it's free QA that keeps the science more scientific.
You have a whole lot of coders looking at the code and making sure it's a perfect implementation of the model or algorithms described by the scientists, and it won't matter if somebody uses the same code to reproduce the results, because you'll know the software is working perfectly. All that's left then is for other scientists to criticize the methodology and assumptions of the research as part of the normal peer review process. It turns the code into just another tool at a scientist's disposal.
We're talking about science here. The whole point of science is challenging something repeatedly and thoroughly, and if it stands up, you know it's true. Adding more challenges is the most scientific thing you can do.
Truth is a defense against libel. If you disclose all your code, and problems with your code are found and fixed, it's actually harder to be sued over your results.
The scientific community can actually run the code and verify that it produces the same output on another machine in another lab. What we're talking about here is programmers making sure the code is a perfect representation of the scientific formulas described by the research paper. Basically, did the researchers follow the methodology they claimed to? Is data being lost or altered due to limitations of the software? This doesn't take any special knowledge of the science involved, just the calculations involved.
I'm quite sure there would be exceptions to state secrets. I'm also sure the folks who built F-15s did a heck of a lot more testing and verification than your typical scientific research at a university.
There's going to be people criticizing code for improper documentation, but I feel that's something that should be a concern in general for scientific research. If you don't document your code properly, you aren't properly documenting your methodology. Code that works, though, regardless of style will probably not get too much criticism. I can't imagine some suggestions for efficiency would hurt though. If you can do your calculations more quickly, or with less computing resources required, you can accomplish more science in less time, for cheaper.
Really what we're talking about here is verifying that your desired algorithms are what actually ended up in the code. Having some extra eyes on your code can only be a good thing for that reason. Programmers will leave the science behind the algorithms up to the scientists.
Verifying that the code used by the researchers is consistent with the methods described, without accidental errors, is basically what we're talking about here, and is essentially equivalent to writing your own code. Eliminate researcher errors in their coding, and you're left with one less possible flaw in an experiment.
The release of raw data along with the results is a good thing, I agree. But somebody checking to make sure the scientists' algorithms are implemented correctly in an existing program is easier, and more likely to happen, than somebody independently coding their own implementation, particularly in scientific research where the computing power necessary is prohibitive.
Whatever he wants. Since when was having information a bad thing? I'm sure a lot of scientists would be a lot happier if there was a greater understanding and curiosity about science in the world. Ultimately it will translate to more money for research, but in the shorter term, a more scientifically-conversant society can only be a good thing. Yes, there will be stuff that most of the population won't understand. But you'll use that as a basis for depriving everyone of that knowledge? There will be people who do understand it, and if it's only a single person understanding a single paper out of thousands, the entire exercise was still worth it.
If you're the only one looking at your code, and it's so complex, how do you know it's working as you intended it to? This is the question that is at the very heart of the issue we're discussing here. It really doesn't matter what the problem is you're working on, I assume the algorithms can be described in some logical/mathematical way that somebody can turn into code without understanding what it's all about. After all, there's got to be something at the heart of your project you can share with those 8 other scientists so they know your methodology.
What would be better, keeping your code to yourself and not knowing if there are even tiny bugs in the implementation of your algorithms, or releasing it and having interested parties examine it and tell you if your algorithms are actually converted properly to code or not? It doesn't matter if anybody can run it, or what the data is you're number-crunching on, I would think you'd want somebody to make sure you didn't type 1+2 instead of 2+3.
As a (wo)man of science, I would expect your primary concern would be in the truth. You cannot know you're getting the truth if your code is never seen by another set of eyes. Nobody's going to be challenging your algorithms outside of the scientific community. What we're talking about is equivalent to a lab tech making sure your scales are calibrated. Trust me, no matter whatever special new language you created, there's more than a few computer scientists out there who can figure it out without you spending a month going back and documenting everything.
That said, if scientists were shamed enough into writing good code to begin with, and documenting everything, that wouldn't be a bad thing either, for science or programing.
Exactly. Let the scientific community evaluate the algorithms, and let the computer science community evaluate its implementation as code. The end result is better code and better science. How is that a bad thing? Any true scientist can't possibly object to something that will result in ensuring the validity of experiment results. As I understand it, scientists are after the truth, not merely research grants and political attention, right? Right??
I really don't think anybody actually expects a few million people challenging the basis of the algorithms here. Programmers will just be challenging the implementation and making sure it's correct.
If climate scientists release their code, people will be able to control the weather!
There's probably other variations out there that can provide an energy source, things much harder, and thus never took off on Earth where energy was easy. I fully expect we'll end up finding life a lot more frequently than we expect at the moment. It'll mostly be stuff we can't have conversations with though.
There's also the whole thing about asteroids carrying life from one place to another. Just because it's hard for life to arise in a particular place in a solar system doesn't mean it can't get transported elsewhere and evolve to thrive, or at least survive.
They're not going to build their own space shuttle if they get their hands on our space shuttle plans. They'll learn from our designs and build something else like a missile, or use the material science for some new fighter jet. It's better they start from scratch than get a leg-up from our designs. The shuttle may be obsolete, but what do you think we use to learn from and improve on?
Yeeeeaaaahhhh.....
The idea that they're all criminals is kind of silly. Most spies operating abroad actually work with diplomatic immunity, gathering relatively public information and doing analysis. There are spies, working with that immunity, who go about trying to recruit assets like Mr. Chung... somebody disaffected, or somebody who has a reason to feel a stronger tie towards another country, or merely somebody who can be bought. They're the criminals, but they're a fairly small portion of the intelligence community as a whole. They're the ones who take the biggest risks, and the ones who get the biggest rewards for their work, if they do it well.
Unfortunately, we live in a world that's not all that friendly. Without spies, there'd be a lot more chaos and death in the world. Spies are the ones that allow leaders to go into a meeting with another leader and tell them "we know you're doing x, so cut it out" and lets military powers stay balanced enough that nobody gains a massive advantage and goes to war. Basically, spies allow for diplomacy to flourish, and they prevent conflicts. China is fully justified in wanting to spy on us and build up technologically and militarily in order to ensure their interests are satisfied. We're equally as justified in wanting to stop them from doing that so our own interests are satisfied. As a patriot, I want my own nation to have the best spies and maintain or achieve superiority in all things. As a rational human being, I'm not going to begrudge any other nation the same desire.
I'm against the death penalty on moral grounds, but treason/spying cases are pretty much the only time when I hesitate to rule it out entirely. The cost of spying against my nation should be prohibitive, and I wouldn't be terribly upset if other nations followed suit. That's the risk assets take.
I would *hope* that we weren't letting foreign nationals have access to all our secrets. Now, giving clearance levels to naturalized citizens is another debate. Hopefully background checks get a little better after something like this, anyway.
As far as I'm concerned, it's rocket science, and that's worth keeping a lid on. Any kind of rocket science, unless you can buy a kit in the mail or a hobby store, has pretty serious potential applications. China getting their hands on space shuttle plans doesn't mean they'll build their own shuttle, it means they take some tiny piece of it and improve some military application like a missile or airplane.
In any case, China is trying to get every scrap of technology they can get their hands on, regardless of its potential applications or obsolescence level. Some stuff may be useful immediately, others may be stepping stones to other technologies. Shuttle tech is stuff we improve on to develop new technologies. It's better, from our national security standpoint, for them to be stuck starting from scratch than from getting an idea of where our technology is headed and what we're starting with.
There's always the so-called "military shuttle" rumors for just such a purpose during the Cold War. How likely there is that one exists or ever existed is debatable, of course, and it's pretty unlikely there's ever been an actual launch of one without somebody noticing. We do know that there's been other military/spy missions going on with the NASA shuttle program though, for satellites and such.
Maybe you could elaborate, so we know what you're talking about?
And you'll still be able to use VLC, there just won't be any new versions for a while. Chances are, as grandparent said, somebody else will come along soon enough. I'm glad this article is here because it might generate some interest, and maybe next week we'll get an article saying "VLC for Mac Saved!"
Yeah, VLC isn't perfect, but I much prefer it to MPlayer. Still, I'm not terribly worried. It's not like they're going to come to your house and uninstall whatever version you have now. And their big announcement is surely a way of getting some new developers to stand up and help out. Some new eyes on the code could be a good thing, at any rate. Presumably whoever does take over will have plenty of enthusiasm and a lot will get done.
It's free, and you probably have it installed already. Give it a whirl and see if it works for you. If not, pencil and paper works great for math notes....
As I understand it, It's illegal to put nukes in space (reactors are fine, I'm talking about things that intentionally go boom). It's the same international treaty or treaties that stopped us from putting nuclear-lasers and other Star Wars types of satellites in orbit that involved nuclear explosions. It's hard to keep MAD going if somebody has nuclear missiles in orbit pointed straight down at your cities. A successful first strike is way too easy with orbital bombardment. A lot harder when they can watch you launch from the ground with ICBMs.
If you play a video game about breaking somebody's arm, you're not actually breaking somebody's arm. You're showing restraint by not doing it in real life, instead choosing to do it virtually. So you're not doing lasting physical harm, your pleasure from it is irrelevant.
Dogs and cats are predatory animals. They evolved to be lethal and efficient killers. We domesticated them to help us kill animals for us (dogs for hunting food for us, cats to kill pests). Humans evolved to be lethal and efficient killers as well. None of us alive can't find many, many, many ancestors in our lineage who hunted animals for food, went to war, or did stuff like murdering children or committing rapes. If they didn't do those things, frankly you wouldn't exist. So these things are in our nature. Because we survived as a species and developed increased intelligence and more complex social structures, most of us can avoid doing nasty, despicable actions (usually by sending other people to do them in our place like nearly every war in the last few hundred years), but it's still part of us.
If I had a cat that played a video game involving biting a human arm, I wouldn't be worried in the slightest, so long as the cat never did it to a real human in real life. Cats don't play video games, however, so play bites are fine too. Play-murder is fine for humans as well. It used to be done in stuff like "cowboys and indians", or just with sticks. The fact that we have more advanced technology to do it makes no difference whatsoever. One outlet is as good as another, as far as I'm concerned.
I would love to see Journeyman Project get a reboot. I still load up my three Buried in Time CDs from time to time with an emulator. Never got to play the others in the series, but I gathered they were pretty good too. Great story, really interesting puzzles, and it was probably one of the best mystery-type games I've ever played.
Well, ignoring your exaggeration for a moment, the better point you could have made is that OS X comes out with a new major version a lot more often than windows does, so people wanting the latest and greatest have to buy more OS X versions in the same time a Windows user upgrades.