I see a fundamental issue with the central storyline, which is that drug companies are seeing a stronger placebo response in drug trials. But drug trials are not designed to measure the placebo response; they are designed to measure the drug against the placebo. It would be like comparing 100 different scales for accuracy, and then going back into the data set to try to discover any differences in the standard weights that were used. A placebo can either be a control or an effect; you can't run one experiment and then treat it both ways. Based on your article it sounds like this is what the drug companies are trying to do though.
The specific test of a double-blind drug trial is to measure the effectiveness of the drug under study. The point of using a placebo for control is that it mirrors the drug--same size and color for the pill, same presentation and language from the doctor, etc. In fact the doctor does not themselves know whether they are administering the drug or placebo. This controls for all those extraneous factors that might affect "normality". Since all the details are the same, any "pleasing effect" (no matter how strong) will confer upon the drug as much as on the placebo, creating a common baseline.
The only controlled--and thus meaningful--result of a drug test is how the drug performs vs. the placebo. If the drug-takers do not show significant improvement compared to the placebo-takers, all that tells you is that the active ingredient in the drug does not have much effect--thus answering the trial.
If both the drug-takers and placebo-takers show significant improvements, that is an interesting result but it is of questionable scientific value if it is not properly controlled itself. To study WHY they all improved would take another, different test. The Wired article gets into the aspects of really testing the placebo effect scientifically.
Basically because the controls are so different, you can run a good study of the effectiveness of a drug, or a good study of the strength of the placebo effect, but not both at the same time. You can't use a placebo as both the independent and dependent variable simultaneously and expect to get a good result. It's got to be one or the other.
I work for myself. I use my boss and my employer to provide money and health care to myself, giving them some of my talent, time*, and effort in return. I'm an equal partner in my working conditions; if I don't like them I have great freedom to go get another job or start my own company.
I have 2 computers at work and 3 at home. I have a Blackberry. Plus I travel and sometimes do not bring any of these with me.
I can pick my e-mail from any of these computers, with perfect sync between all of them, because I use Gmail. Which is a cloud service.
I remember the days of having a dedicated e-mail application on my home computer that reached out to my ISP's mail server to download my e-mail. But, I don't remember it well because I switched to Hotmail just about as fast as I could when it came out. It's incredibly inconvenient to have my communications tied to one particular device sitting on a desk somewhere.
People like cloud services for the same reason most people like cell phones better than landlines: convenience and portability.
Every month with them its some new "thing" with its own name and brand. "The Long Tail." "Free." "Good Enough." It's not insightful! It's an exercise in meme-making and marketing.
"Good enough" has always been, um, good enough. For a given need, people always take something over nothing and then competitive markets walk the quality up and the cost down. A CD player has far, far better fidelity and portability than a phonograph did. But a phonograph was better than no music, so it sold well when it came out. MP3s players are already walking the quality curve--iTunes is now the #1 music retailer selling AAC (better than MP3), recently bumped to 256 kbps with "iTunes Plus."
Look at TV, which became ubiquitous using high-contrast, grainy, black and white screens. Now you can buy a 32" 1080p resolution HDTV for like $400.
Off the top of my head, these are three fields where small smart teams can still do research and produce game-changing ideas.
Game-changers are sometimes only visible after they've changed the game. I bet we'll look back at this time in 30 years and the early signs of the next "big thing" will be obvious. The hard part is spotting them now.
CD-RW is more archival than CD-R because it uses re-meltable metal instead of dye to hold the data. It's a good middle ground in terms of archival quality, and should last 16 years in dark cool storage pretty easily.
If someone defaces it or uses it as a means to alter someone's reputation (for good or ill), it will lose credibility.
Can we please let this argument die? It has no factual support. There have in fact been a number of very high-profile instances of defacing or altering entries and they have had no measurable effect on the growth of the content or the traffic. It's a meme that continues to live on solely by repetition.
Wikipedia is never going to be Encyclopedia Britannica. It's never going to be thought of that way. That's ok because in many ways it is much better than Britannica, despite the greater potential for abuse. Where was Britannica's entry on the London Underground bombs one hour after they happened, for instance. (But this new system could easily take that value away!)
So can we please let Wikipedia be Wikipedia and stop comparing it to printed encyclopedias? Give people some credit--they know it's not Britannica.
This requirement first appeared in politically controversial articles. Then it spread to most articles on serious subjects. Now it's applied even to fancruft. ("What do you mean I can't write about 'Zords in Power Rangers: Jungle Fury' because they weren't mentioned in a Journal of Popular Culture article?") The detailed fancruft is gradually moving to Wikia, which has lower standards.
I believe this is a conscious commercial strategy designed to drive more and more content to Wikia, which is a for-profit company founded by Jimmy Wales--who also happens to be the leader of the "inner circle" at Wikipedia.
Command systems only do good if the commanding authority is good. If the command authority is compromised, the entire system is compromised.
A better, more flexible system is the wisdom of the crowds and the marketplace of ideas, which naturally tempers extremist viewpoints. See: Federalist #10.
I cannot believe I am having to make this point in a thread about Wikipedia.
Wikipedia grew to where it is by allowing anyone to edit it. That is what makes (made) Wikipedia different from other Web sites and sources of information. Yes there has been a lot of criticism, but it is criticism of the core concept...a concept that has proven itself in spectacular fashion! So why would anyone listen to the criticism? Yet over time a small group of people has altered and altered Wikipedia in response to the criticism. This is yet one more step down that losing path.
How should one measure the success of Wikipedia? In "criticism" or pageviews? I would submit that pageviews is the right answer. If it is useful it will be used, criticism be damned.
Wikipedia is not in any less need of updating now than the day it started. In fact because it is so much bigger, it is in exponentially MORE need of updating than the day it started. It would be a huge, huge mistake to think that it is at all "finished" and ready to be put into maintenance mode. If anything the Foundation ought to be finding ways to make it easier and easier to update and correct pages.
Steve Jobs on multiple occasions has said that Apple is at heart a software company, then quotes Alan Kay saying that "people who are really serious about software do their own hardware." The implication (right or wrong, you decide) is that they can't do the best software unless they control the hardware too.
Sorry, I once worked on a site, where we got 16 million visits *a day*! And that's only for the top country.
Yeah, we've all "worked on" big sites. It's quite another thing to be financially accountable for a big site. You'll find it's a little less easy to cast away 10% of your users overnight when your profit margin is only 1% to 3%.
The proper way to build a big site is to build to standards and then add exception handling for any significant user bases. Over time some of these will shrink below the "who cares" limit and you can get rid of that exception. Obviously those limits will vary per site and audience.
Your rant about leadership is fine, but you do not lead your customers, you serve them. You lead your employees, and then, yes, you sometimes make decisions that are necessary but not popular.
Big difference. The "rubber sheet" is a simplified model of Einstein's approach to gravity, which is descriptive. In other words general relativity doesn't tell us "why" gravity works, it just provides the maths for modeling the "shape" of a gravitational field and predicting what will happen to objects moving within it. So the GP is exactly right--general relativity doesn't tell us why things roll around on the rubber sheet. But it does tell us with great accuracy where they will go. This practical approach goes all the way back to Newton.
The F-35 is being built by Lockheed as well, and much of what they learned on the F-22 project has gone into the F-35 project. The planes even look fairly similar.
Remember that these "greedy schmucks" are the ones lobbying and influencing the law. You, I, the Slashdot community, we do not. But we are tax paying constituents. The only time we influence this is when we vote--and let's face it, it's not a voting issue.
Speak for yourself. I call or write my elected representatives all the time about issues. I donate money to nonprofits who have professional lobbyists just like the RIAA does--but working for what I want.
Want to see the power of citizen lobbying? We're not drilling in ANWR, despite 6 years during which the Republican party controlled the White House and most of Congress. Whether you think that's good or not, it certainly demonstrates that corporate interests are not the only players in Washington.
If members of Congress received 10 calls every day about copyright issues, they would listen. A call is worth 10 letters because it interrupts what they're doing--they can't put if off or ignore it.
Yeah, obviously the U.S. government is incapable of running any large complex program successfully. That's why we in the U.S. live under the umbrella of global deterrence provided by Canada's large and complex military, and not vice versa.
The purpose of a government is to make and enforce laws, thus any government will have the power to impact the future earnings of a business. It's impossible to imagine a government so small that no business would ever feel threatened by its decisions.
Congrats! Now if you ever actually played basketball competitively you'd know that almost any good team employs the full court press, but does so selectively. That's because it is not hard to beat with a bit of good coaching, particularly if the pressing team is relying more on effort than skill.
Gladwell did what he always does, which is blow an interesting story way out of proportion.
Slashdot Mirror
I see a fundamental issue with the central storyline, which is that drug companies are seeing a stronger placebo response in drug trials. But drug trials are not designed to measure the placebo response; they are designed to measure the drug against the placebo. It would be like comparing 100 different scales for accuracy, and then going back into the data set to try to discover any differences in the standard weights that were used. A placebo can either be a control or an effect; you can't run one experiment and then treat it both ways. Based on your article it sounds like this is what the drug companies are trying to do though.
The specific test of a double-blind drug trial is to measure the effectiveness of the drug under study. The point of using a placebo for control is that it mirrors the drug--same size and color for the pill, same presentation and language from the doctor, etc. In fact the doctor does not themselves know whether they are administering the drug or placebo. This controls for all those extraneous factors that might affect "normality". Since all the details are the same, any "pleasing effect" (no matter how strong) will confer upon the drug as much as on the placebo, creating a common baseline.
The only controlled--and thus meaningful--result of a drug test is how the drug performs vs. the placebo. If the drug-takers do not show significant improvement compared to the placebo-takers, all that tells you is that the active ingredient in the drug does not have much effect--thus answering the trial.
If both the drug-takers and placebo-takers show significant improvements, that is an interesting result but it is of questionable scientific value if it is not properly controlled itself. To study WHY they all improved would take another, different test. The Wired article gets into the aspects of really testing the placebo effect scientifically.
Basically because the controls are so different, you can run a good study of the effectiveness of a drug, or a good study of the strength of the placebo effect, but not both at the same time. You can't use a placebo as both the independent and dependent variable simultaneously and expect to get a good result. It's got to be one or the other.
You will lose some good people who do not want to take the test. But without the test, you might hire an incompetent employee. Which is worse?
Personally I'd take the former failure mode. In that case your job search just takes longer. I'd rather that, than hire a bad employee.
I work for myself. I use my boss and my employer to provide money and health care to myself, giving them some of my talent, time*, and effort in return. I'm an equal partner in my working conditions; if I don't like them I have great freedom to go get another job or start my own company.
I have 2 computers at work and 3 at home. I have a Blackberry. Plus I travel and sometimes do not bring any of these with me.
I can pick my e-mail from any of these computers, with perfect sync between all of them, because I use Gmail. Which is a cloud service.
I remember the days of having a dedicated e-mail application on my home computer that reached out to my ISP's mail server to download my e-mail. But, I don't remember it well because I switched to Hotmail just about as fast as I could when it came out. It's incredibly inconvenient to have my communications tied to one particular device sitting on a desk somewhere.
People like cloud services for the same reason most people like cell phones better than landlines: convenience and portability.
Every month with them its some new "thing" with its own name and brand. "The Long Tail." "Free." "Good Enough." It's not insightful! It's an exercise in meme-making and marketing.
"Good enough" has always been, um, good enough. For a given need, people always take something over nothing and then competitive markets walk the quality up and the cost down. A CD player has far, far better fidelity and portability than a phonograph did. But a phonograph was better than no music, so it sold well when it came out. MP3s players are already walking the quality curve--iTunes is now the #1 music retailer selling AAC (better than MP3), recently bumped to 256 kbps with "iTunes Plus."
Look at TV, which became ubiquitous using high-contrast, grainy, black and white screens. Now you can buy a 32" 1080p resolution HDTV for like $400.
Off the top of my head, these are three fields where small smart teams can still do research and produce game-changing ideas.
Game-changers are sometimes only visible after they've changed the game. I bet we'll look back at this time in 30 years and the early signs of the next "big thing" will be obvious. The hard part is spotting them now.
"A foolish consistency is the hobgoblin of little minds." :-)
http://en.wikipedia.org/wiki/The_Dying_Night
CD-RW is more archival than CD-R because it uses re-meltable metal instead of dye to hold the data. It's a good middle ground in terms of archival quality, and should last 16 years in dark cool storage pretty easily.
I think I have that bumper sticker on my...hang on, just let me check this e-mail...and get this call...
If someone defaces it or uses it as a means to alter someone's reputation (for good or ill), it will lose credibility.
Can we please let this argument die? It has no factual support. There have in fact been a number of very high-profile instances of defacing or altering entries and they have had no measurable effect on the growth of the content or the traffic. It's a meme that continues to live on solely by repetition.
Wikipedia is never going to be Encyclopedia Britannica. It's never going to be thought of that way. That's ok because in many ways it is much better than Britannica, despite the greater potential for abuse. Where was Britannica's entry on the London Underground bombs one hour after they happened, for instance. (But this new system could easily take that value away!)
So can we please let Wikipedia be Wikipedia and stop comparing it to printed encyclopedias? Give people some credit--they know it's not Britannica.
This requirement first appeared in politically controversial articles. Then it spread to most articles on serious subjects. Now it's applied even to fancruft. ("What do you mean I can't write about 'Zords in Power Rangers: Jungle Fury' because they weren't mentioned in a Journal of Popular Culture article?") The detailed fancruft is gradually moving to Wikia, which has lower standards.
I believe this is a conscious commercial strategy designed to drive more and more content to Wikia, which is a for-profit company founded by Jimmy Wales--who also happens to be the leader of the "inner circle" at Wikipedia.
I've written about it before
Command systems only do good if the commanding authority is good. If the command authority is compromised, the entire system is compromised.
A better, more flexible system is the wisdom of the crowds and the marketplace of ideas, which naturally tempers extremist viewpoints. See: Federalist #10.
I cannot believe I am having to make this point in a thread about Wikipedia.
Wikipedia grew to where it is by allowing anyone to edit it. That is what makes (made) Wikipedia different from other Web sites and sources of information. Yes there has been a lot of criticism, but it is criticism of the core concept...a concept that has proven itself in spectacular fashion! So why would anyone listen to the criticism? Yet over time a small group of people has altered and altered Wikipedia in response to the criticism. This is yet one more step down that losing path.
How should one measure the success of Wikipedia? In "criticism" or pageviews? I would submit that pageviews is the right answer. If it is useful it will be used, criticism be damned.
Wikipedia is not in any less need of updating now than the day it started. In fact because it is so much bigger, it is in exponentially MORE need of updating than the day it started. It would be a huge, huge mistake to think that it is at all "finished" and ready to be put into maintenance mode. If anything the Foundation ought to be finding ways to make it easier and easier to update and correct pages.
Steve Jobs on multiple occasions has said that Apple is at heart a software company, then quotes Alan Kay saying that "people who are really serious about software do their own hardware." The implication (right or wrong, you decide) is that they can't do the best software unless they control the hardware too.
Sorry, I once worked on a site, where we got 16 million visits *a day*! And that's only for the top country.
Yeah, we've all "worked on" big sites. It's quite another thing to be financially accountable for a big site. You'll find it's a little less easy to cast away 10% of your users overnight when your profit margin is only 1% to 3%.
The proper way to build a big site is to build to standards and then add exception handling for any significant user bases. Over time some of these will shrink below the "who cares" limit and you can get rid of that exception. Obviously those limits will vary per site and audience.
Your rant about leadership is fine, but you do not lead your customers, you serve them. You lead your employees, and then, yes, you sometimes make decisions that are necessary but not popular.
Big difference. The "rubber sheet" is a simplified model of Einstein's approach to gravity, which is descriptive. In other words general relativity doesn't tell us "why" gravity works, it just provides the maths for modeling the "shape" of a gravitational field and predicting what will happen to objects moving within it. So the GP is exactly right--general relativity doesn't tell us why things roll around on the rubber sheet. But it does tell us with great accuracy where they will go. This practical approach goes all the way back to Newton.
If you think i'm going to sign that, you're crazy.
Duly noted, jimmydevice (699057) ;-)
Just let me know if they spot a ring that looks like it's been braided.
The F-35 is being built by Lockheed as well, and much of what they learned on the F-22 project has gone into the F-35 project. The planes even look fairly similar.
Remember that these "greedy schmucks" are the ones lobbying and influencing the law. You, I, the Slashdot community, we do not. But we are tax paying constituents. The only time we influence this is when we vote--and let's face it, it's not a voting issue.
Speak for yourself. I call or write my elected representatives all the time about issues. I donate money to nonprofits who have professional lobbyists just like the RIAA does--but working for what I want.
Want to see the power of citizen lobbying? We're not drilling in ANWR, despite 6 years during which the Republican party controlled the White House and most of Congress. Whether you think that's good or not, it certainly demonstrates that corporate interests are not the only players in Washington.
If members of Congress received 10 calls every day about copyright issues, they would listen. A call is worth 10 letters because it interrupts what they're doing--they can't put if off or ignore it.
Are you one of those guys who looks at the Google homepage and say "pffff, I could that easy"?
Yeah, obviously the U.S. government is incapable of running any large complex program successfully. That's why we in the U.S. live under the umbrella of global deterrence provided by Canada's large and complex military, and not vice versa.
Thanks, Canada! ;-)
The purpose of a government is to make and enforce laws, thus any government will have the power to impact the future earnings of a business. It's impossible to imagine a government so small that no business would ever feel threatened by its decisions.
Congrats! Now if you ever actually played basketball competitively you'd know that almost any good team employs the full court press, but does so selectively. That's because it is not hard to beat with a bit of good coaching, particularly if the pressing team is relying more on effort than skill.
Gladwell did what he always does, which is blow an interesting story way out of proportion.