Actually, I believe that most universities and their students can view the Windows source code, after agreeing to a non-disclosure agreement.
Yeah right; you're running an educational operation, so you tell your students to use something that requires an NDA. But this rather misses the whole point of an educational institution.
Funny thing: If you look into the history of unix and its clones, you find that the most successful clones (e.g., minix and linux) were started largely because of licensing restrictions for student access to the source code. Yes, you could get student access licenses from AT&T, with a lot of paperwork including signatures n NDAs, plus paying license fees. But Andy and Linus decided, quite reasonably, that this wasn't reasonable, and it would be better to build their own from scratch.
Both have commented that this was not just a better educational approach; it was also a lot more fun for everyone involved. And both projects, started for personal and educational use, have given us useful, unencumbered systems that are much more useful to a wide variety of applications.
For an interesting non-unix example, look into the Japanese ITRON system. This was also started in an academic environment, as a tool for teaching real-time computing without the legal hassles involved in showing the code to students. It's now one of the major embedded real-time kernels. Much of this is because you don't need to sign an NDA to study the code.
Heck, science disproves itself every year, changing some previous thought out "solution" to a new one.For example, for years science said that ulcers were caused by stress and/or increased stomach acid. Then one day (I believe about 6 months ago), a group of scientists found that they are actually caused by bacteria!
Actually, the stress theory was never scientific. It was a medical myth, repeated in medical texts but never tested in any way. It was disproved by a scientific researcher who saw some suspicious evidence and decided to investigate. After he showed that Helicobacter causes ulcers, people went back to the medical literature and said "Hey, whaddaya know; there is no published study showing a causative relation between stress and ulcers. I guess that theory was never scientific at all."
Medicine is full of myths like this, mostly because people demand that doctors be all-knowing, and doctors want their patients to feel good, so they go along with the myths even if they know that they don't really know.
Most new scientific theories don't replace previous scientific theories. Rather, they replace beliefs that had never been properly tested. Typically the scientific answer had been "Further research is needed", i.e., "We don't really know".
One of the main differences between scientists and religions is that scientists are expected to admit that they don't know. Instead of just guessing and then insisting that their guess is right, scientists are supposed to say "Further research is needed" and apply for grants to figure out what's really going on.
Actually, it did get quite a bit of coverage, both in the scientific press and the mass media, when it was first figured out - about 20 years ago.
There used to be a joke about how scientists had shown that bumblebees can't fly, because there's no way they could develop enough aerodynamic lift with those tiny wings to keep their mass in the air. Of course, it turns out that they don't use aerodynamic lift. And that was pretty obvious all along, since their wings are flat (and high-speed photos showed that they didn't deform their wings into airfoils when flying).
Part of the joke was that someone was supposed to say that it was a good thing that bumblebees weren't smart enough to understand the scientific explanation of why they couldn't fly.
Well, it was funny decades ago, before we knew how they developed the needed lift.
In principle, I'd strongly agree, and so would most scientists. But in the real world, people often don't have that luxury. We don't always have the resources or the time. No matter what we wish, the fact that X has published a string of 20 papers over 10 years that have all been checked and found perfectly correct will have an effect on even the most skeptical. So X's latest results will be accepted at face value, at least for a while.
That's partly why we really need skeptical, mostly younger researchers out to make a name for themselves by poking holes in some established star's results.
Of course, if the star researcher really did make a mistake, it will be discovered eventually. And in the meantime, people will be basing their work on the mistake. So it's too bad we don't always have the ability to triple-check everything in independent labs.
But even with this admittedly human failing, the scientific demand for reproducibiity has a much better record than any legal investigations, both in discovering incorrect results and in teaching, punishing or rehabilitating the errant researcher.
There's really no need to try for legal punishments for scientific fraud. Having one's results shown irreproducible is enough of a punishment. And, unlike legal sanctions, it has a good chance of persuading the errant researchers to mend their ways and contribute more usefult results in the future.
Not to be prejudicial but from now on anything that comes out of SK's academia will be taken with a "grain of salt."
We might add that the scientific community as a whole has a long history of this approach. It's called "reproducibility", and standard procedure is to apply it to all results from all labs.
I'd suggest that we do such in this case, and dispense with the legal and political attacks. Either Dr Hwang's results are reproducible and thus credible, or they aren't reproducible and should be dismissed as erroneous. Not just for this lab, but for any.
In some cases, you find that a particular lab's results have always been easily reproducible, so you start accepting them before they've been independently tested. But that should be the special case. The default should be that everything is considered tentative until independently verified.
Science has historically had little but grief from attempts to control fraud by legal or political means. This rarely leads to good scientific results. But verification by reproduction has a long history of success.
This approach has the advantage of not attributing fraud when the problem was poor methodology or writing. If a result can't be reproduced because the published reports were misunderstood, this tends to come out quickly, and the original researchers publish a "clarification" that better explains what they really did.
But the legal and political systems tend to jump quickly to a "fraud" conclusion, from which there is little escape even when it was just a mistake. Then we lose good research and good researchers.
The fact that some of Dr Hwang's earlier results have been verified implies that we should be cautious about jumping to conclusions. Even if there was intentional fraud, we don't know just who did it, or what sort of pressure they were under.
And if this was done because of outside pressure, which is highly likely in this case, the usual scientific approach should quickly get across the idea that you can't get away with forcing your researchers to publish invalid results.
For that matter, we in the US could benefit from getting this message across to some of our grant agencies, especially the corporate ones.
The real problem here is that cloning has become a hot political topic in some countries. This usually leads to bad science. We can't fight it politically; the politicians have all the clout. But we can easily fight it with scientific methods.
8.4 -> 10.6 is what, 25% error margin? seems a bit far off from an exact science to me
The Hubble constant is not well understood and is very hard to measure, hence the error margin.
Um, the topic is the shape of our galaxy. The Hubble Constant is utterly irrelevant to this topic. The very fact that it appeared here is a sign of total cluelessness.
It's true that the galaxy's size and shape was mostly determined by measuring distances and relative motions of various things within the galaxy. But these motions are unrelated to the universal expansion that the Hubble Constant is a measure of. The galaxy isn't itself expanding.
If I have 120 dollars and I donate 20, I get taxed on the remaining 100 dollars...
Ah, but you're not doing it like Bill. As others have pointed out, most of his donations have been in the form of software licenses.
It's like you have 120 dollars, so you "donate" a piece of software that you've listed (but rarely sold) for $20, which costs you maybe $1 for the media. You take a $20 deduction, but you still have $119 of your original $120, while you pay taxes on only $100.
Sounds like a good deal to me. Especially if the software you donated won't work unless the recipient upgrades their OS, which by some chance you are happy to sell to them.
And 5 more of these $20 donations will wipe out the entire tax on your $120, while leaving you with $115. Of course, in Bill's case, you can add 5 or 6 zeroes to all those numbers.
In that case EVERY corporation on earth has your money and the only way to fight this evil is to sell everythign and start a commune in south america. Just don't drink the cool aid.
Well, I have three computers that are running linux, but the older two of them had MS windows when they were delivered. I overwrote Windows with linux, but that doesn't change the fact that part of what I paid for those machines went to Microsoft. And even worse: Microsoft gets to list me as a "satisfied customer".
OTOH, when I bought the Guinness and Negra Modelo that's in our refrigerator, none of the purchase price went to the maker of Kool Aid[TM]. And there are a couple of bags of coffee beans, whose price didn't include anything that went to Starbucks. So Kool Aid isn't a good metaphor here.;-)
[S]oftware used to always be open source for the most part. If you were buying something for your fucking expensive mainframe, you better be able to tinker around with the software running on it.
This is something that seems to be unknown to most of the comuter biz these days. Back in the 70's, and early 80's, I worked at a number of places where the main computer was a big IBM monster. The admins running VM always had complete source, and the people in charge of the other OSs that ran on VM usually had the source, too.
Funny story: one of my first experiences with unix was being "elected" the system manager for Amdahl's unix (UTS) which ran on top of VM. When we asked the Amdahl reps about source, their answer was "Source isn't an option; you get it whether you want it or not." They were happy to see that we had C programmers that could read their code; that made their support job a lot easier.
Within a month of getting their system and learning to use it, I had already sent them a bug fix that immediately went into their code base. This was funny, too: As the clock went into the DST half of the year, all sorts of timestamps were off by an hour. When I investigated, I found that unix got its time from VM, which used local time internally. It seems that the guy who ran VM thought that the way to handle this was to change the VM clock. Rather than let him (and his bosses) know what an idiot he was, I added code to the unix clock routine that tracked the error in the VM clock. The Amdahl guys thought that this was hilarious, and agreed that they had other customers that didn't understand time, so my match would be generally useful.
On my side, my boss was impressed when my patch worked correctly the first time, and was again imressed when Amdahl thanked me for the patch. He didn't hear about their amusement.
Anyway, the loss of open source in the 80's was a real setback to the computer industry and all our users. This was one of many ways that Bill Gates & Company hurt us all. It's really disappointing that he got away with it as easily as he did.
Of course, back then we didn't have a term like "open source", because we didn't need a term for it. Maybe that's part of why MS got away with their secrecy as easily as they did.
Bill just says, 'You've got to take the long view--history will get this right.'
If history is going to get things right, it won't be during his lifetime.
It's entirely possible that "history" (i.e., historians) will get it right, but society as a whole won't.
There's a lot of precedent for this. One of my favorite examples, which runs a risk of invoking Godwin's Law, is the explanation back in the 1930's by Adolf Hitler for why the rest of the world wouldn't care what they did to the Jews. He mentioned two precedents: the slaughter of the Armenians by the Turks two decades earlier, and the slaughter of the Plains Indians by the Americans in the late 1800s.
The interesting thing here is that all three of these have been thoroughly documented by various historians, American and otherwise. But try finding any mention (much less serious analysis) of any of them in history textbooks anywhere. To a very great degree, Adolf was right in his prognosis. He didn't lose because of what he did to Jews; he lost because of the Blitz and the mistake of sinking American ships (and thinking he could take Russia in the winter;-).
Similarly, there have been some fairly thorough analyses of what was going on inside the US government during the last weeks of WWII and most of the Vietnamese war. None of this makes it into the public understanding of these events, despite the fact that hard-hitting histories have already been written. The road to war in Iraq has already been quite analyzed by historians, with no visible effect on the politics of it at all.
So Bill might better say that "history" will indeed get it right, but it won't matter, because society will go with its mythology rather than what the historians say actually happened. And he's the one paying the mythmakers (who are paid a lot better than any historians).
And how can anyone not see the mediocrity in other software in Office like Powerpoint?
I asked you to name the software suite which is so great that you can factually state that Microsoft Office is mediocre in comparison.
Um; this case has little if anything to do with competition. "PowerPoint" has become the joke word in much of the business (and political) world. Even amount people who would never even look at anything computerish unless it says IBM or Microsoft on the outside, "PowerPoint" is used the same way we geeks use "PHB". Outside top management, it's widely considered the ultimately sign of an empty-headed waste of time.
Actually, there are a number of other packages for building presentations. But they'll never be a competitor for PowerPoint, no matter how good they might be. Unless they say "IBM" or "Microsoft" on the label, they won't even be noticed by management.
It's great that he's giving some of this money to charity.
Um, well, maybe.
But as others have pointed out, the great mjority of those billions of dollars he has supposedly given away were in the form of software licenses. He didn't pay for those licenses. Not even in the form of "lost sales", because such "gifts" have mostly been to organizations that wouldn't have paid for the licenses anyway.
The real motive for such gifts is to tie the recipients to Windows. Especially when the software given won't run on older machines or Windows (or DOS) releases, and require that the recipient pay for an upgrade.
Gates has given some actual money, true. But the publicly-stated amounts are nearly an order of magnitude greater than what Gates actually paid. Most is just a funny-money estimate that functions primarily as a useful tax deduction.
Bill Gates didn't pioneer this sort of business giving, of course. It has been part of IBM's marketing strategy for decades, and is a good part of the reason for their eventual takover of the mainframe "market". Bill just extended this approach into the small-computer realm where IBM wasn't succeeding.
That's an interesting letter to read in the light of Microsoft subsequently paying programmers to build Internet Explorer and then give it out free, thus bankrupting Netscape. Of course, after that, Bill & Steve have let us know that the Free (in both senses) Software people are nothing but communists trying to undermine the American economy.
There's room in there for a lot of cynicism.
I have personally had a bit of fun with this on occasion, by arguing with a straight face that the FOSS crowd should publicly praise Bill Gates as their mentor. Try it yourself sometime; it's a lot of fun. The IE-vs-Netscape story is a nice example of the effects of giving out your software free. Now we just need to "do a Netscape" on Microsoft.
These six principles are not in themselves difficult to apply, and may even appear obvious. However, in the authors' experience, many software development projects fail to deliver against many, if any, of these principles.
I'd argue that this is true, and the main reason that so many software products are so buggy is that most companies reward their developers for making buggy software, and punish them for making quality software.
The latter point is easy to show. I've worked on any number of products where I produced something fairly quickly that worked, and nobody could find any bugs. How was I rewarded? In nearly every case, I was laid off. I'd done the job that I was hired for, they didn't have another project that "matched my talents", and the software was so well documented that if any bugs did appear, others could fix them. So why should they continue to pay me to be nonproductive?
The lesson to developers is obvious: If you want to keep working here, you'll have to make sure that there is still work for you to do. This is not difficult. We all know the results.
It's especially easy to do when you look at the illogical demands that most companies put on programmers. I've been denied all sorts of information about the workings of the tools that I'm required to use, on the grounds that what I'm asking about is proprietary. So I have tools whose behavior isn't fully knowable, and I spend a lot of time looking at output and asking "What could that thing be doing to give such bizarre results?"
Of course, the funniest part of the article was the first principle, of "using a sound, formal notation for all deliverables". In most companies, this means a PowerPoint file.
My general conclusion is that lots of people say that they want quality software, but they intentionally prevent developers from producing it, and they punish developers who do so despite all the barriers. The result shouldn't be at all surprising.
Probably, because the classical definition is nonsensical. It has no obvious relation to the meanings of the three words. Well, ok, "the" has the right meaning. But there's no obvious sense of begging in the definition, and there's usually no obvious question.
There's nothing wrong with such nonsense, of course. It's clearly an idiom, i.e., a phrase whose meaning can't be derived from the literal meanings of its words. We use idioms all the time. Many are like this one, an obsolete usage that lingers on.
The modern meaning is an obvious, literal interpretation of the phrase. As such, it should be approved by us geeks and nerds, since one of our regular sources of humor is taking an idiom or metaphor in its literal sense.
So I'd expect this dispute to come up nearly every time someone uses "beg(ging) the question". It will take a lot of education to reinstall the older definition in the minds of people accustomed to the modern usage.
While we're at it, maybe we can work on getting people to use "thou", "thy" and "thee" correctly. I see these misused all the time.;-)
I'm arguing against those who say that you should only vote for major party candidates, because otherwise you are "wasting your vote".
I've found that a good reply to this is "No, voting for someone other than the best candidate is wasting your vote."
If you vote for a major-party candidate because your preferred third-party candidate "can't win", then you have truly been tricked into wasting your vote. And in the worst possible way: You not only gave your vote to a candidate you didn't want to win; you also added data to the idea that third-party candidates can't win.
What specifications in digital video do Linux and Microsoft Windows fail to meet?
Dunno if there are any that are relevant to the kernel. Like any unixoid system, linux does support streaming data. Though I'm not a Windows expert, I'd guess that it does, too. The work doesn't belong in the kernel; it should be in an app, so it's not hogging space when you're not looking at a video.
The only real OS question is how much overhead there is in getting the data between the input card and the app, and then getting the scan lines to the screen. But this doesn't really have anything to do with any video spec; it's just a bit-moving question.
Vague complaints about an "old" OS not being suitable for any specific stream of bits strikes me as a bit clueless. I'll believe it when I read something specific about an OS interfering with the task.
Actually, I'd expect it to be more of a programming-language issue than an OS issue.
Anyone got a 2-line perl video codec?;-)
(Yes, I do have the perl 2-line RSA codec t-shirt.)
What We Believe But Cannot Prove: Today's Leading Thinkers on Science in the Age of Certainty
One of the oft-repeated important points about scientific methodology is that science very rarely, if ever, actually proves anything. Rather, science mostly gives us results in the form of a double negative: We've tried, and found that we can't disprove a hypothesis. After sufficiently many such tries, such an "undisprovable" hypothesis graduates to the class of theory, which really just means that it's tentatively accepted as the best exlanation so far, until something better comes along.
I'd expect that any purported Leading Thinkers on Science would happily agree that they can't really prove anything that they believe. Proof is for mathematicians, not scientists.
But some of their comments are well worth reading and thinking about.
Wow. Only a psychologist would come up with an idea like this.
Not true at all. Consider the famous quote:
Not only is the universe stranger than we imagine, it is stranger than we can imagine.
This was from Sir Arthur Eddington (1882-1944). He was an astronomer.
You also see variants of this quote attributed to assorted others, including J.B.S.Haldane and Arthur C.Clarke, neither of whom is/was a psychologist. It's a common conjecture among many kinds of scientists.
A quantified version of this has popped up in the computer field. The idea is that your mind is produced by your brain, which has a large but finite number of atoms (or elementary particles if you prefer). Thus, there is a finite limit to what you are capable of holding in your mind at any given time. If understanding the universe requires more bits than this, then you are not capable of understanding the universe.
Expressed this way, it even seems ultimately testable. But not soon; we're still a long way from knowing how many bits a human brain contains. We don't even know the physical representation of information in the brain. So maybe it's possible for a human to understand the universe; maybe not. Maybe some day we'll know. Or maybe we'll find a way of adding plug-in memory to our brain.
How can you create functionality for something that doesn't exist yet?
Huh? What's the problem? It's called "implementing to spec". I've done it on a lot of projects.
Of course, there are always problems when some of the other components aren't yet available for testing. So you discuss the interfaces with the other implementers, and write small emulators that handle the test cases that you need. You usually do this anyway, because real-world tests often refuse to give you some of the infrequent cases that you'd like to test against.
And, needless to say, this never works perfectly. So you don't just package up the components and ship them to customers; you re-test after you think you've got all the components together.
But saying that you can't create functionality for something that doesn't exist yet - that's just showing ignorance of standard engineering practice. If every component had to wait until all the other interacting component existed, we'd never build anything new.
I'd say that the original unix guys at Bell Labs did a pretty good job of designing for things that didn't exist yet. But if you dig up their writings on the subject, you won't find this surprising.
Actually, it's more an examlple of the fact that the English doesn't have a legal authority with the power to enforce any particular definition of a word. So there often isn't a "proper" definition of many words.
If you consult one dictionary, you might think you know the proper definition of words like "atheist" and "agnostic". But if you consult different dictionaries, which is easy now that they're online, you find that people routinely use both words with all sorts of meanings, and even interchange them.
A bit of investigation quickly shows that there's no way to win this one. No matter which definition you use, you'll be attacked by people who insist that you're wrong. And many of them have emotional attachments to the words, so there's little chance of any agreement.
Part of the problem is the old fallacy of the "excluded middle". Many people routinely refuse to accept that one might not have a commitment to something as emotional as their God, so you must either believe as they do or you must believe that their God doesn't exist. You aren't allowed a skeptical attitude; you must commit yourself. Attempting to opt out will be taken as disbelief.
So, although "agnostic" was originally coined from the Greek roots ("a-" = "without" + "gnosis" = "knowledge"), and meant belief that we don't know, the excluded-middle approach immediately interpreted this as disbelief in the existence of God. Many people
You really can't win in an argument like this. It's best to not get involved. In any public forum, such arguments can never be anything other than a flame fest.
(And yes, I do realize that I'm making the same sort of meta-argument that I'm saying can't be won.;-)
[J]ust as easy to make are verb number agreement in comlpex sentences like one I wrote just a few minutes ago: "Meta information like comments and 'track changes' is lost or corrupt."
Hey, that should be "Meta information likes comments, and 'track changes' are lost or corrupt.";-)
I have no idea what that means. But, as a human emulating a grammar checker, I don't have to understand it; I just have to find a parsing and point out the errors.
[Written as an illustration of why grammar checkers are so often worthless.]
... they want to expand into other ventures- such as print versions for third world countries.
Hmmm... I'd think a more practical aproach would be to back the Negroponte "laptop for every child" project. That, and pushing for wireless coverage everywhere, would benefit the poorer parts of the world a lot more than centuries of printing books has done.
Here's another idea: Hint to third-world governments that it could help their people a lot if wikipedia (and other such efforts) had more material in their native language(s). It's not reasonable to expect Americans to do the translation, of course; the thing to do is to subsidize locals to do the translation. And donate to wikipedia, because they have to buy disks and bandwidth for the effort.
An "It's for your children" campaign could shake loose some funds, especially for something that's relatively so cheap.
Perhaps it's just me, but I find the attempts by librarians to stay 'relevant', while understandable, to be just a little bit pathetic.
It's not you; that definitely comes across. But "librarian" isn't a monolithic borg-like population. Some of them have taken to the internet as a great new extension of their traditional role. The city library a mile from where I live has a room full of internet terminals; the "card catalog" has been electronic for 20 years and online for 10 years. And so on.
When's the last time you asked a librarian a question?
Probably not long. The problem here is: On the Internet, nobody can tell you're a librarian.
Actually, I have knowingly communicated with a number of librarians, via email, in the past couple of months. This was as part of several ongoing efforts to turn local archives of paper into electronic forms.
One of these is the Village Music Project, which has a collection of personal notebooks of local working musicians' in the past three centuries or so, converting them to a form that's easily usable (by mostly free, open-source software) by musicians who are interested in that sort of thing. The director of the project has complained about being terminally busy, partly due to such things as an effort to preserve and make available online the Vaughan Williams Memorial Library, a similar effort based on Cecil Sharpe's life work.
There are a zillion efforts like this going on right now, and librarians are deeply involved in many of them. This shouldn't be a surprise, since librarian training does include a lot of the techniques needed for such efforts. Only the medium has changed, not the information. And our new medium has the ability to organize stuff in ways that weren't possible before, while not invalidating any of the older organizing techniques.
Slashdot Mirror
Actually, I believe that most universities and their students can view the Windows source code, after agreeing to a non-disclosure agreement.
Yeah right; you're running an educational operation, so you tell your students to use something that requires an NDA. But this rather misses the whole point of an educational institution.
Funny thing: If you look into the history of unix and its clones, you find that the most successful clones (e.g., minix and linux) were started largely because of licensing restrictions for student access to the source code. Yes, you could get student access licenses from AT&T, with a lot of paperwork including signatures n NDAs, plus paying license fees. But Andy and Linus decided, quite reasonably, that this wasn't reasonable, and it would be better to build their own from scratch.
Both have commented that this was not just a better educational approach; it was also a lot more fun for everyone involved. And both projects, started for personal and educational use, have given us useful, unencumbered systems that are much more useful to a wide variety of applications.
For an interesting non-unix example, look into the Japanese ITRON system. This was also started in an academic environment, as a tool for teaching real-time computing without the legal hassles involved in showing the code to students. It's now one of the major embedded real-time kernels. Much of this is because you don't need to sign an NDA to study the code.
Heck, science disproves itself every year, changing some previous thought out "solution" to a new one.For example, for years science said that ulcers were caused by stress and/or increased stomach acid. Then one day (I believe about 6 months ago), a group of scientists found that they are actually caused by bacteria!
Actually, the stress theory was never scientific. It was a medical myth, repeated in medical texts but never tested in any way. It was disproved by a scientific researcher who saw some suspicious evidence and decided to investigate. After he showed that Helicobacter causes ulcers, people went back to the medical literature and said "Hey, whaddaya know; there is no published study showing a causative relation between stress and ulcers. I guess that theory was never scientific at all."
Medicine is full of myths like this, mostly because people demand that doctors be all-knowing, and doctors want their patients to feel good, so they go along with the myths even if they know that they don't really know.
Most new scientific theories don't replace previous scientific theories. Rather, they replace beliefs that had never been properly tested. Typically the scientific answer had been "Further research is needed", i.e., "We don't really know".
One of the main differences between scientists and religions is that scientists are expected to admit that they don't know. Instead of just guessing and then insisting that their guess is right, scientists are supposed to say "Further research is needed" and apply for grants to figure out what's really going on.
Actually, it did get quite a bit of coverage, both in the scientific press and the mass media, when it was first figured out - about 20 years ago.
There used to be a joke about how scientists had shown that bumblebees can't fly, because there's no way they could develop enough aerodynamic lift with those tiny wings to keep their mass in the air. Of course, it turns out that they don't use aerodynamic lift. And that was pretty obvious all along, since their wings are flat (and high-speed photos showed that they didn't deform their wings into airfoils when flying).
Part of the joke was that someone was supposed to say that it was a good thing that bumblebees weren't smart enough to understand the scientific explanation of why they couldn't fly.
Well, it was funny decades ago, before we knew how they developed the needed lift.
In principle, I'd strongly agree, and so would most scientists. But in the real world, people often don't have that luxury. We don't always have the resources or the time. No matter what we wish, the fact that X has published a string of 20 papers over 10 years that have all been checked and found perfectly correct will have an effect on even the most skeptical. So X's latest results will be accepted at face value, at least for a while.
That's partly why we really need skeptical, mostly younger researchers out to make a name for themselves by poking holes in some established star's results.
Of course, if the star researcher really did make a mistake, it will be discovered eventually. And in the meantime, people will be basing their work on the mistake. So it's too bad we don't always have the ability to triple-check everything in independent labs.
But even with this admittedly human failing, the scientific demand for reproducibiity has a much better record than any legal investigations, both in discovering incorrect results and in teaching, punishing or rehabilitating the errant researcher.
There's really no need to try for legal punishments for scientific fraud. Having one's results shown irreproducible is enough of a punishment. And, unlike legal sanctions, it has a good chance of persuading the errant researchers to mend their ways and contribute more usefult results in the future.
Not to be prejudicial but from now on anything that comes out of SK's academia will be taken with a "grain of salt."
We might add that the scientific community as a whole has a long history of this approach. It's called "reproducibility", and standard procedure is to apply it to all results from all labs.
I'd suggest that we do such in this case, and dispense with the legal and political attacks. Either Dr Hwang's results are reproducible and thus credible, or they aren't reproducible and should be dismissed as erroneous. Not just for this lab, but for any.
In some cases, you find that a particular lab's results have always been easily reproducible, so you start accepting them before they've been independently tested. But that should be the special case. The default should be that everything is considered tentative until independently verified.
Science has historically had little but grief from attempts to control fraud by legal or political means. This rarely leads to good scientific results. But verification by reproduction has a long history of success.
This approach has the advantage of not attributing fraud when the problem was poor methodology or writing. If a result can't be reproduced because the published reports were misunderstood, this tends to come out quickly, and the original researchers publish a "clarification" that better explains what they really did.
But the legal and political systems tend to jump quickly to a "fraud" conclusion, from which there is little escape even when it was just a mistake. Then we lose good research and good researchers.
The fact that some of Dr Hwang's earlier results have been verified implies that we should be cautious about jumping to conclusions. Even if there was intentional fraud, we don't know just who did it, or what sort of pressure they were under.
And if this was done because of outside pressure, which is highly likely in this case, the usual scientific approach should quickly get across the idea that you can't get away with forcing your researchers to publish invalid results.
For that matter, we in the US could benefit from getting this message across to some of our grant agencies, especially the corporate ones.
The real problem here is that cloning has become a hot political topic in some countries. This usually leads to bad science. We can't fight it politically; the politicians have all the clout. But we can easily fight it with scientific methods.
8.4 -> 10.6 is what, 25% error margin? seems a bit far off from an exact science to me
The Hubble constant is not well understood and is very hard to measure, hence the error margin.
Um, the topic is the shape of our galaxy. The Hubble Constant is utterly irrelevant to this topic. The very fact that it appeared here is a sign of total cluelessness.
It's true that the galaxy's size and shape was mostly determined by measuring distances and relative motions of various things within the galaxy. But these motions are unrelated to the universal expansion that the Hubble Constant is a measure of. The galaxy isn't itself expanding.
If I have 120 dollars and I donate 20, I get taxed on the remaining 100 dollars ...
Ah, but you're not doing it like Bill. As others have pointed out, most of his donations have been in the form of software licenses.
It's like you have 120 dollars, so you "donate" a piece of software that you've listed (but rarely sold) for $20, which costs you maybe $1 for the media. You take a $20 deduction, but you still have $119 of your original $120, while you pay taxes on only $100.
Sounds like a good deal to me. Especially if the software you donated won't work unless the recipient upgrades their OS, which by some chance you are happy to sell to them.
And 5 more of these $20 donations will wipe out the entire tax on your $120, while leaving you with $115. Of course, in Bill's case, you can add 5 or 6 zeroes to all those numbers.
In that case EVERY corporation on earth has your money and the only way to fight this evil is to sell everythign and start a commune in south america. Just don't drink the cool aid.
;-)
Well, I have three computers that are running linux, but the older two of them had MS windows when they were delivered. I overwrote Windows with linux, but that doesn't change the fact that part of what I paid for those machines went to Microsoft. And even worse: Microsoft gets to list me as a "satisfied customer".
OTOH, when I bought the Guinness and Negra Modelo that's in our refrigerator, none of the purchase price went to the maker of Kool Aid[TM]. And there are a couple of bags of coffee beans, whose price didn't include anything that went to Starbucks. So Kool Aid isn't a good metaphor here.
[S]oftware used to always be open source for the most part. If you were buying something for your fucking expensive mainframe, you better be able to tinker around with the software running on it.
This is something that seems to be unknown to most of the comuter biz these days. Back in the 70's, and early 80's, I worked at a number of places where the main computer was a big IBM monster. The admins running VM always had complete source, and the people in charge of the other OSs that ran on VM usually had the source, too.
Funny story: one of my first experiences with unix was being "elected" the system manager for Amdahl's unix (UTS) which ran on top of VM. When we asked the Amdahl reps about source, their answer was "Source isn't an option; you get it whether you want it or not." They were happy to see that we had C programmers that could read their code; that made their support job a lot easier.
Within a month of getting their system and learning to use it, I had already sent them a bug fix that immediately went into their code base. This was funny, too: As the clock went into the DST half of the year, all sorts of timestamps were off by an hour. When I investigated, I found that unix got its time from VM, which used local time internally. It seems that the guy who ran VM thought that the way to handle this was to change the VM clock. Rather than let him (and his bosses) know what an idiot he was, I added code to the unix clock routine that tracked the error in the VM clock. The Amdahl guys thought that this was hilarious, and agreed that they had other customers that didn't understand time, so my match would be generally useful.
On my side, my boss was impressed when my patch worked correctly the first time, and was again imressed when Amdahl thanked me for the patch. He didn't hear about their amusement.
Anyway, the loss of open source in the 80's was a real setback to the computer industry and all our users. This was one of many ways that Bill Gates & Company hurt us all. It's really disappointing that he got away with it as easily as he did.
Of course, back then we didn't have a term like "open source", because we didn't need a term for it. Maybe that's part of why MS got away with their secrecy as easily as they did.
Bill just says, 'You've got to take the long view--history will get this right.'
;-).
If history is going to get things right, it won't be during his lifetime.
It's entirely possible that "history" (i.e., historians) will get it right, but society as a whole won't.
There's a lot of precedent for this. One of my favorite examples, which runs a risk of invoking Godwin's Law, is the explanation back in the 1930's by Adolf Hitler for why the rest of the world wouldn't care what they did to the Jews. He mentioned two precedents: the slaughter of the Armenians by the Turks two decades earlier, and the slaughter of the Plains Indians by the Americans in the late 1800s.
The interesting thing here is that all three of these have been thoroughly documented by various historians, American and otherwise. But try finding any mention (much less serious analysis) of any of them in history textbooks anywhere. To a very great degree, Adolf was right in his prognosis. He didn't lose because of what he did to Jews; he lost because of the Blitz and the mistake of sinking American ships (and thinking he could take Russia in the winter
Similarly, there have been some fairly thorough analyses of what was going on inside the US government during the last weeks of WWII and most of the Vietnamese war. None of this makes it into the public understanding of these events, despite the fact that hard-hitting histories have already been written. The road to war in Iraq has already been quite analyzed by historians, with no visible effect on the politics of it at all.
So Bill might better say that "history" will indeed get it right, but it won't matter, because society will go with its mythology rather than what the historians say actually happened. And he's the one paying the mythmakers (who are paid a lot better than any historians).
And how can anyone not see the mediocrity in other software in Office like Powerpoint?
I asked you to name the software suite which is so great that you can factually state that Microsoft Office is mediocre in comparison.
Um; this case has little if anything to do with competition. "PowerPoint" has become the joke word in much of the business (and political) world. Even amount people who would never even look at anything computerish unless it says IBM or Microsoft on the outside, "PowerPoint" is used the same way we geeks use "PHB". Outside top management, it's widely considered the ultimately sign of an empty-headed waste of time.
Actually, there are a number of other packages for building presentations. But they'll never be a competitor for PowerPoint, no matter how good they might be. Unless they say "IBM" or "Microsoft" on the label, they won't even be noticed by management.
It's great that he's giving some of this money to charity.
Um, well, maybe.
But as others have pointed out, the great mjority of those billions of dollars he has supposedly given away were in the form of software licenses. He didn't pay for those licenses. Not even in the form of "lost sales", because such "gifts" have mostly been to organizations that wouldn't have paid for the licenses anyway.
The real motive for such gifts is to tie the recipients to Windows. Especially when the software given won't run on older machines or Windows (or DOS) releases, and require that the recipient pay for an upgrade.
Gates has given some actual money, true. But the publicly-stated amounts are nearly an order of magnitude greater than what Gates actually paid. Most is just a funny-money estimate that functions primarily as a useful tax deduction.
Bill Gates didn't pioneer this sort of business giving, of course. It has been part of IBM's marketing strategy for decades, and is a good part of the reason for their eventual takover of the mainframe "market". Bill just extended this approach into the small-computer realm where IBM wasn't succeeding.
That's an interesting letter to read in the light of Microsoft subsequently paying programmers to build Internet Explorer and then give it out free, thus bankrupting Netscape. Of course, after that, Bill & Steve have let us know that the Free (in both senses) Software people are nothing but communists trying to undermine the American economy.
There's room in there for a lot of cynicism.
I have personally had a bit of fun with this on occasion, by arguing with a straight face that the FOSS crowd should publicly praise Bill Gates as their mentor. Try it yourself sometime; it's a lot of fun. The IE-vs-Netscape story is a nice example of the effects of giving out your software free. Now we just need to "do a Netscape" on Microsoft.
The article comments:
These six principles are not in themselves difficult to apply, and may even appear obvious. However, in the authors' experience, many software development projects fail to deliver against many, if any, of these principles.
I'd argue that this is true, and the main reason that so many software products are so buggy is that most companies reward their developers for making buggy software, and punish them for making quality software.
The latter point is easy to show. I've worked on any number of products where I produced something fairly quickly that worked, and nobody could find any bugs. How was I rewarded? In nearly every case, I was laid off. I'd done the job that I was hired for, they didn't have another project that "matched my talents", and the software was so well documented that if any bugs did appear, others could fix them. So why should they continue to pay me to be nonproductive?
The lesson to developers is obvious: If you want to keep working here, you'll have to make sure that there is still work for you to do. This is not difficult. We all know the results.
It's especially easy to do when you look at the illogical demands that most companies put on programmers. I've been denied all sorts of information about the workings of the tools that I'm required to use, on the grounds that what I'm asking about is proprietary. So I have tools whose behavior isn't fully knowable, and I spend a lot of time looking at output and asking "What could that thing be doing to give such bizarre results?"
Of course, the funniest part of the article was the first principle, of "using a sound, formal notation for all deliverables". In most companies, this means a PowerPoint file.
My general conclusion is that lots of people say that they want quality software, but they intentionally prevent developers from producing it, and they punish developers who do so despite all the barriers. The result shouldn't be at all surprising.
Do we have to go through this EVERY time?
;-)
Probably, because the classical definition is nonsensical. It has no obvious relation to the meanings of the three words. Well, ok, "the" has the right meaning. But there's no obvious sense of begging in the definition, and there's usually no obvious question.
There's nothing wrong with such nonsense, of course. It's clearly an idiom, i.e., a phrase whose meaning can't be derived from the literal meanings of its words. We use idioms all the time. Many are like this one, an obsolete usage that lingers on.
The modern meaning is an obvious, literal interpretation of the phrase. As such, it should be approved by us geeks and nerds, since one of our regular sources of humor is taking an idiom or metaphor in its literal sense.
So I'd expect this dispute to come up nearly every time someone uses "beg(ging) the question". It will take a lot of education to reinstall the older definition in the minds of people accustomed to the modern usage.
While we're at it, maybe we can work on getting people to use "thou", "thy" and "thee" correctly. I see these misused all the time.
I'm arguing against those who say that you should only vote for major party candidates, because otherwise you are "wasting your vote".
I've found that a good reply to this is "No, voting for someone other than the best candidate is wasting your vote."
If you vote for a major-party candidate because your preferred third-party candidate "can't win", then you have truly been tricked into wasting your vote. And in the worst possible way: You not only gave your vote to a candidate you didn't want to win; you also added data to the idea that third-party candidates can't win.
What specifications in digital video do Linux and Microsoft Windows fail to meet?
;-)
Dunno if there are any that are relevant to the kernel. Like any unixoid system, linux does support streaming data. Though I'm not a Windows expert, I'd guess that it does, too. The work doesn't belong in the kernel; it should be in an app, so it's not hogging space when you're not looking at a video.
The only real OS question is how much overhead there is in getting the data between the input card and the app, and then getting the scan lines to the screen. But this doesn't really have anything to do with any video spec; it's just a bit-moving question.
Vague complaints about an "old" OS not being suitable for any specific stream of bits strikes me as a bit clueless. I'll believe it when I read something specific about an OS interfering with the task.
Actually, I'd expect it to be more of a programming-language issue than an OS issue.
Anyone got a 2-line perl video codec?
(Yes, I do have the perl 2-line RSA codec t-shirt.)
TFA has the title:
What We Believe But Cannot Prove: Today's Leading Thinkers on Science in the Age of Certainty
One of the oft-repeated important points about scientific methodology is that science very rarely, if ever, actually proves anything. Rather, science mostly gives us results in the form of a double negative: We've tried, and found that we can't disprove a hypothesis. After sufficiently many such tries, such an "undisprovable" hypothesis graduates to the class of theory, which really just means that it's tentatively accepted as the best exlanation so far, until something better comes along.
I'd expect that any purported Leading Thinkers on Science would happily agree that they can't really prove anything that they believe. Proof is for mathematicians, not scientists.
But some of their comments are well worth reading and thinking about.
Wow. Only a psychologist would come up with an idea like this.
Not true at all. Consider the famous quote:
Not only is the universe stranger than we imagine, it is stranger than we can imagine.
This was from Sir Arthur Eddington (1882-1944). He was an astronomer.
You also see variants of this quote attributed to assorted others, including J.B.S.Haldane and Arthur C.Clarke, neither of whom is/was a psychologist. It's a common conjecture among many kinds of scientists.
A quantified version of this has popped up in the computer field. The idea is that your mind is produced by your brain, which has a large but finite number of atoms (or elementary particles if you prefer). Thus, there is a finite limit to what you are capable of holding in your mind at any given time. If understanding the universe requires more bits than this, then you are not capable of understanding the universe.
Expressed this way, it even seems ultimately testable. But not soon; we're still a long way from knowing how many bits a human brain contains. We don't even know the physical representation of information in the brain. So maybe it's possible for a human to understand the universe; maybe not. Maybe some day we'll know. Or maybe we'll find a way of adding plug-in memory to our brain.
How can you create functionality for something that doesn't exist yet?
Huh? What's the problem? It's called "implementing to spec". I've done it on a lot of projects.
Of course, there are always problems when some of the other components aren't yet available for testing. So you discuss the interfaces with the other implementers, and write small emulators that handle the test cases that you need. You usually do this anyway, because real-world tests often refuse to give you some of the infrequent cases that you'd like to test against.
And, needless to say, this never works perfectly. So you don't just package up the components and ship them to customers; you re-test after you think you've got all the components together.
But saying that you can't create functionality for something that doesn't exist yet - that's just showing ignorance of standard engineering practice. If every component had to wait until all the other interacting component existed, we'd never build anything new.
I'd say that the original unix guys at Bell Labs did a pretty good job of designing for things that didn't exist yet. But if you dig up their writings on the subject, you won't find this surprising.
Actually, it's more an examlple of the fact that the English doesn't have a legal authority with the power to enforce any particular definition of a word. So there often isn't a "proper" definition of many words.
;-)
If you consult one dictionary, you might think you know the proper definition of words like "atheist" and "agnostic". But if you consult different dictionaries, which is easy now that they're online, you find that people routinely use both words with all sorts of meanings, and even interchange them.
A bit of investigation quickly shows that there's no way to win this one. No matter which definition you use, you'll be attacked by people who insist that you're wrong. And many of them have emotional attachments to the words, so there's little chance of any agreement.
Part of the problem is the old fallacy of the "excluded middle". Many people routinely refuse to accept that one might not have a commitment to something as emotional as their God, so you must either believe as they do or you must believe that their God doesn't exist. You aren't allowed a skeptical attitude; you must commit yourself. Attempting to opt out will be taken as disbelief.
So, although "agnostic" was originally coined from the Greek roots ("a-" = "without" + "gnosis" = "knowledge"), and meant belief that we don't know, the excluded-middle approach immediately interpreted this as disbelief in the existence of God. Many people
You really can't win in an argument like this. It's best to not get involved. In any public forum, such arguments can never be anything other than a flame fest.
(And yes, I do realize that I'm making the same sort of meta-argument that I'm saying can't be won.
[J]ust as easy to make are verb number agreement in comlpex sentences like one I wrote just a few minutes ago: "Meta information like comments and 'track changes' is lost or corrupt."
;-)
Hey, that should be "Meta information likes comments, and 'track changes' are lost or corrupt."
I have no idea what that means. But, as a human emulating a grammar checker, I don't have to understand it; I just have to find a parsing and point out the errors.
[Written as an illustration of why grammar checkers are so often worthless.]
... they want to expand into other ventures- such as print versions for third world countries.
... I'd think a more practical aproach would be to back the Negroponte "laptop for every child" project. That, and pushing for wireless coverage everywhere, would benefit the poorer parts of the world a lot more than centuries of printing books has done.
Hmmm
Here's another idea: Hint to third-world governments that it could help their people a lot if wikipedia (and other such efforts) had more material in their native language(s). It's not reasonable to expect Americans to do the translation, of course; the thing to do is to subsidize locals to do the translation. And donate to wikipedia, because they have to buy disks and bandwidth for the effort.
An "It's for your children" campaign could shake loose some funds, especially for something that's relatively so cheap.
Here's my fearless prediction:
...
The Google Box will run a custom Linux distro, GoogLinux. It will probably be largely based on either Ubuntu or Mepis
It'll have a hand crank on the side. And it'll sell for $100.
But google will hand them out for free in third-world countries.
Admit it; you want one.
Perhaps it's just me, but I find the attempts by librarians to stay 'relevant', while understandable, to be just a little bit pathetic.
It's not you; that definitely comes across. But "librarian" isn't a monolithic borg-like population. Some of them have taken to the internet as a great new extension of their traditional role. The city library a mile from where I live has a room full of internet terminals; the "card catalog" has been electronic for 20 years and online for 10 years. And so on.
When's the last time you asked a librarian a question?
Probably not long. The problem here is: On the Internet, nobody can tell you're a librarian.
Actually, I have knowingly communicated with a number of librarians, via email, in the past couple of months. This was as part of several ongoing efforts to turn local archives of paper into electronic forms.
One of these is the Village Music Project, which has a collection of personal notebooks of local working musicians' in the past three centuries or so, converting them to a form that's easily usable (by mostly free, open-source software) by musicians who are interested in that sort of thing. The director of the project has complained about being terminally busy, partly due to such things as an effort to preserve and make available online the Vaughan Williams Memorial Library, a similar effort based on Cecil Sharpe's life work.
There are a zillion efforts like this going on right now, and librarians are deeply involved in many of them. This shouldn't be a surprise, since librarian training does include a lot of the techniques needed for such efforts. Only the medium has changed, not the information. And our new medium has the ability to organize stuff in ways that weren't possible before, while not invalidating any of the older organizing techniques.