Well, speaking of Amtrak employee accountability, I have a story about that. A few years ago my family took a train ride across the country. When we changed trains in Chicago I noticed that the reading light in my sleeping compartment was stuck on, which of course was bad if I wanted to actually sleep. I found the friendly and helpful attendant and reported it, and her reaction was like watching a balloon deflate.

"What's wrong?" I asked.

"If we report damage they take it out of our wages," she said.

"What! What do you mean take it out of your wages?" I asked.

"If a car is damaged under my watch I have to pay for it," she said.

"Well," I said, taking out my swiss army knife, "I guess there's nothing to see here."

I have to say that I've never encountered such a nice, enthusiastic, friendly group of people with such an abysmally low morale as the crew of a cross-country train. With passengers they're great, but all through the trip I'd see two or three congregated having low muttered conversations. It didn't take me long to figure out they were talking about management. And while the experience was wonderful, the equipment was in horrible shape. It was like traveling in a third world country.

With management that bad, more data doesn't equal more accountability and better performance. It means scapegoating.

From what I've read the lack of respect for negative results ties into both the leadership for study funding and to the less informed people from outside the scientific community who often approve the funding.

The person in charge of a larger scientific entity may have even more invested in the "right" conclusion in terms of their leadership potential and may not want to fund or advance studies which could threaten their larger position on the issue.

And people from outside the scientific community may see negative outcomes wrongly as "failed" science -- why look, you couldn't even prove your theory. As you point out, this is wrong, but I think these people look at it kind of like a failed business venture. If Joe Scientist's theory is disproven, he must be an incompetent idiot and we should disown him because clearly he's going down the wrong path.

Yes, we're programmers, so all those so-called experts in user-interface design should just shut up. We don't need usability studies! We wrote it ourselves, and pressing Alt-X and then typing "frobnitz" works and is memorable for us. Why should we adapt to using these stupid mice the way lusers would expect?

Sure, intelligent programmers might realize that there are domains involved with computing and software projects that don't directly involve coding skills (like user interface design, or real-world use cases, or, god forbid, the realities of social interactions with others), but those are all quivering gamma rabbits who cower before us mighty Social Injustice Warriors who lurk in our mothers basements, and never have any interaction with women whose names don't end in .jpg!

Successful ones invariably are, yes. One common mode is the "benevolent dictator" model used by the Linux kernel. Another is the constitutional-democracy-with-technical-oversight model used by Debian. And there are a variety of others. But for an open source project to scale well, it needs some sort of conflict resolution method (at the least), since programmers are about as easy to get pointed in the same direction as cats.

Yes, the first thing Ithought when I saw the article summary was "this is going to bring out the SIWs (Social Injustice Warriors) in force!" :)

Do SAN vendors intentionally mix production runs of drives when they ship them?

I would kind of expect them to, and it might explain why I've never seen a group of drives bought at the same time (installed in a server or SAN) fail as a group.

Although I would kind of expect some logistical challenges if I was a SAN vendor trying to keep inventories of multiple production runs in stock for populating new SANs, especially when some single unit devices can ship with as many as 24 drives. Keeping a half-dozen unique batches on site for populating systems, sure, but 24? I would think some SANs would have to go out with drives from the same production run and the logistics just get more complicated with mismatched supply/demand/production curves.

Which is beside the point. Consensus isn't about truth, it's about burden of proof.

Suppose Alice and Bob both try to make a perpetual motion machine. Alice claims she has failed, but Bob claims he has succeeded. The scientific community treats Alice's claims of failure without skepticism but it automatically assumes that Bob has made a mistake somewhere.

Does that seem unfair to Bob? Well, imagine you're a rich guy and Alice and Bob are both applying to you for a job. Bob says you should give the job to him because he's your long-lost fraternal twin your parents never told you about and which the hospital hushed up for some reason. When you mention this to Alice she freely admits she is not related to you. You automatically believe Alice, so is it fair to Bob to be skeptical of his claims?

It's a case of "extraordinary claims require extraordinary evidence. In either case Bob can prove his claim, it's more complicated and time consuming because he has to explain what went wrong with all the prior knowledge. Alice's claims in either case are consistent with what you reasonably believe to be true so you can reasonably assume she's correct.

Er, no. That's like positing science going off the rails because it replaced instrumentation with data.

I think there's two other interrelated things that contribute to this.

"Big Science" these days, especially in healthcare, often involves long-term, expensive studies which take years to perform. People who commit to this mode of science make both a commitment to the field, but often to the hypothesis being tested.

To get the study funded requires basically betting your career on the validity or at least the likelihood of the validity of the hypothesis.

So, if I've bought into the hypothesis that dietary cholesterol influences serum cholesterol and it takes 10 years to design, fund and implement the study involved in it when the results turn out negative, what of my career? I've invested a good chunk of it basically being wrong.

And I think a fair amount of the people involved in these big theories aren't just scientifically interested in them, they are invested in them in terms of scientific reputation since they kind of have to be to get them funded. They often become advocates for the theory before it's proven, and if it isn't sustained by the study there's the risk of looking foolish because you were wrong.

So between personal reputations and career commitment and the size of the science involved, people have a lot of personal stake in seeing their hypothesis validated.

I think it's more the Peter Principle -- people get promoted for success in their current position and stop getting promoted once they become ineffective.

I think the last "kick upstairs" is done for employees who are too ineffective but too loyal/valuable to have working elsewhere.

I see this at two clients with POS systems. They don't handle any cash or credit card transactions, everything is billed to internal accounts, but they still want to use some of the terminals for productivity software because the POS systems are underutilized as POS systems, they lack the space for additional productivity PCs and don't want to spend money on them anyway.

I opposed it on principle in terms of providing advice, but as a matter of practicality since they're not handling real money or credit card information the risk is a lot less.

A nominal promotion in terms of title, but actually reduced responsibilities in terms of work.

As ShanghaiBill says, Bats aren't rodents. I'll just add that bats and rodents are about as taxonomically unrelated as two mammals can possibly be.

Bats are more closely related to horses, bears, rhinos, even whales -- like most mammals they're members of the huge and diverse superorder Laurasiatheria. Rodents are in the much smaller superorder Euarchontoglires, the only non-extinct members of which are: rodents, rabbits, hares, pikas, tree shrews, flying lemurs, and the various primates.

In the late 1970s in junior high we would ride the bus and get off at random stops and write down pay phone numbers. Then when we got home we would call the numbers and do all sorts of gags.

The one that inexplicably worked well was telling people that had won money from a radio station. Why they believed that an 8th grader sounded like a disk jockey is still beyond me.

It's almost kind of sad that kids of today can't get that experience. There's very few pay phones left and I bet none of them accept incoming calls. It was also pretty safe from a get in trouble perspective. Call logging and tracing would have been a huge endeavor and we never called any one pay phone more than a few times or suggested anything violent or even all that ribald.

They don't pay it off now by printing money because other people keep buying the debt.

The dollar represents 2/3rds of the world's reserve currency. The rest of the reserve currencies combined aren't enough to replace it.

Hyperinflation probably isn't always a guarnateed outcome, I would wager political pressure not to manipulate monetary and fiscal policy that much is a bigger reason.