If the government can use a backdoor, than so can anyone else

Feds Pushing For New Legally Required Wiretap Backdoor To All Internet Communications
[Via Techdirt]

The unfortunate, if not surprising, news story making the rounds today is that the feds in the US are looking to pass new laws to legally require a wiretap backdoor in every kind of internet communication offering. Yes, you read that right. If there’s any way to communicate online, the US government is demanding the right to be able to wiretap it. Any company that doesn’t comply will face fines. This despite the long history of the US government massively abusing its wiretapping privileges repeatedly throughout history.

And, yes, this would supposedly apply to non-US communications services as well:

Foreign-based providers that do business inside the United States must install a domestic office capable of performing intercepts.

Yeah, that’ll go over well. It’s difficult to see how this is any different than foreign governments demanding access to others’ communications as well. It’s pretty ridiculous for President Obama to talk about open internet principles to the UN, while cooking this up at the same time. Pushing for this also means that the US will have no excuse when the governments of Iran, China and elsewhere also demand backdoors into all US-based communications.

And, really, that’s the biggest problem with this law. Beyond the inevitable privacy violations by the feds, putting backdoors into communications technologies guarantees that those backdoors will be used by others (outside of the federal government) to snoop on communications. The FBI and the NSA (who are pushing for this) are being totally and completely naive if they think that they’re the only ones who will use this. We’ve pointed out in the past how large scale surveillance systems mean large scale security risks, and this is no different. We showed how a similar surveillance system in Greece was hacked into to spy on government officials. US officials should be aware that they’re opening themselves up to these same potential risks.

And, the simple fact is: this won’t help and it won’t matter. The people who really want to communicate secretly will still use tools to communicate secretly. The feds are (once again) being naive to think that such tools won’t exist and won’t be widely known and widely utilized. Instead, all this will do is open up everyone else to abuse of the system by other governments, organized crime, people with malicious intent and (of course) the US government.

[More]

Besides the obvious abuses by the government, it also means that bad guys will be able to get access to any corporation’s online communications. Do they really want to hamper these sorts of communications?

And, since anyone can also use encryption to prevent eavesdropping, how will this really do much against the bad guys? It’ll get used against regular people who happen to join a political group the government does not like, even if that group has done nothing wrong, in violation of our Constitution.

Franklin is again correct. How much essential liberty will we give up? especially since it does not really make us safer just more exposed to abuses by the government.

Hope it goes to committee to die

U.S. Senate Poised to Vote on Internet Blacklist
[Via Daring Fireball]

Aaron Swartz has started a campaign to stop COICA, proposed legislation to create an Internet blacklist in the U.S. From the fact sheet:

What exactly does it do?

The bill creates two blacklists of Internet domain names. The first can be added to by a court, the second by the Attorney General. Internet service providers (everyone from Comcast to PayPal to Google AdSense) would be required to block any domains on the first list. They would also receive immunity (and presumably the government’s gratitude) for blocking domains on the second list.

Don’t make me quote Ben Franklin again.

[More]

Giving the government the ability to blacklist people and blocking websites because of the material present on those sites seems unconstitutional to me. It is a pure sop to big media and otther corporations, with no benefits for real people.

We do need to make sure this never makes it out of committee.

‘We need more than 3 mice for this to work’

Paul Krugman: Downhill With the G.O.P.
[Via Economist's View]

Republicans have thoroughly embraced the Irving Kristol strategy for political effectiveness: “say whatever it takes to gain power”:

Downhill With the G.O.P., by Paul Krugman, Commentary, NY Times: Once upon a time, a Latin American political party promised to help motorists save money on gasoline. How? By building highways that ran only downhill.

I’ve always liked that story, but the truth is that the party received hardly any votes. And that means that the joke is really on us. For these days one of America’s two great political parties routinely makes equally nonsensical promises. … And this party has a better than even chance of retaking at least one house of Congress this November.

Banana republic, here we come.

On Thursday, House Republicans released their “Pledge to America,” supposedly outlining their policy agenda. In essence, what they say is, “Deficits are a terrible thing. Let’s make them much bigger.” The document repeatedly condemns federal debt — 16 times, by my count. But the main substantive policy proposal is to make the Bush tax cuts permanent, which … would add about $3.7 trillion to the debt over the next decade — about $700 billion more than the Obama administration’s tax proposals.

True, the document talks about the need to cut spending. But as far as I can see, there’s only one specific cut proposed — canceling the rest of the Troubled Asset Relief Program, which Republicans claim (implausibly) would save $16 billion. That’s less than half of 1 percent of the budget cost of those tax cuts. As for the rest, everything must be cut, in ways not specified — “except for common-sense exceptions for seniors, veterans, and our troops.” In other words, Social Security, Medicare and the defense budget are off-limits.

So what’s left? Howard Gleckman of the nonpartisan Tax Policy Center has done the math. As he points out, the only way to balance the budget by 2020, while simultaneously (a) making the Bush tax cuts permanent and (b) protecting all the programs Republicans say they won’t cut, is to completely abolish the rest of the federal government: “No more national parks, no more Small Business Administration loans, no more export subsidies, no more N.I.H. No more Medicaid… No more child health or child nutrition programs. No more highway construction. No more homeland security. Oh, and no more Congress.”

The “pledge,” then, is nonsense. … So how did we get to the point where one of our two major political parties isn’t even trying to make sense?

[More]

Looks like the GOP is trying to do the experiment with less than 3 mice. Perhaps they should have asked a statistician.

I wrote about the futility of this earlier in the year. Getting rid of any ‘waste’ still leaves us with a huge deficit. We had a $1.4 trillion deficit in 2009. The total amount of discretionary spending that is not defense related is $437 billion. We could completely do away with ALL of that and still be left with $1 billion still to remove.

And things do not get better as we go forward. Deficits begin rising fairly fast 5 years from now unless revenues are substantially increased.

Cutting revenue without specific cuts in spending is simply pandering.

We can not make this work with only 3 mice. There has to be some effort to make some hard decisions and to make the numbers match some sort of reality. The Pledge does no such thing. It would have been better for them to point people to some of the simulations online and let them see the hard choices that have to be made.

Because there are not very many good choices from here on out. Anyone who leads you to believe that there are not any is trying to do the experiment with 3 mice. They will not succeed.

How the ethics of human experimentation have changed

by courtneyBolton

The further adventures of Thomas Edison, asshat
[Via Boing Boing]

Say you were curious about what happened when a human brain came into prolonged contact with a strong magnetic field. To find an answer, you might learn more about the nature of magnetism and the biology of the brain and make some hypotheses. Later, you’d run some experiments using dead tissue, or perhaps live animal models. This is because you are a good person.

If you were Thomas Edison, on the other hand, you’d find the nearest street urchin and keep him inside a giant electromagnet. It’s OK, though, science writers would later say, because there turned out not to be much of an effect, and, anyway, the boy liked it.

Many years ago, Mr. Thomas Edison made the following interesting experiment in his laboratory. Wishing to see what, if any, influence is produced by the passage of strong magnetic flux through the brain, he kept a boy for a long time inside a huge electro-magnet with his head placed between the poles so that the flux passed directly through his brain. If now a magnet is capable of producing any effect whatever on the body it should certainly have done it in this case. But as far as could be seen no effect whatever was felt by the boy. When asked privately how he felt he replied confidentially, “The experiment is bully. I am all right in the magnet. I like to be here for I do not have to work while the experiment is going on and I can take a nap occasionally. But don’t tell Mr. Edison. I hope he will keep me here for a long time.”

[More]

I’d love to hear his conversation with the biostatistician. ‘Can I do this experiment with one human? My grant is due tomorrow.’ How does one experiment with one human prove anything? It really does not.

I wonder what he would have done if the poor kid’s head exploded? Sometimes we just forget why we have well vetted clinical trials backed by good animal data, with rigorous statistics.

Seat of the pants experiments such as Edison’s really reveal little at all.

When science parody becomes reality does it remain parody?

by Husky

The science media make my head hurt
[Via Pharyngula]

First, read this parody of science journalism. It’s the template for just about every science story you’ll find in a newspaper, and it’s so depressing.

Second, imagine something even worse. Hint: it’s the media’s coverage of every scientific “controversy” you might think of. It takes a few of the tropes mentioned in the parody, like “shift responsibility for establishing the likely truth or accuracy of the research findings on to absolutely anybody else but me, the journalist” and “quotes from some fringe special interest group of people who, though having no apparent understanding of the subject, help to give the impression that genuine public ‘controversy’ exists.” and “Special interest group linked to for balance” and expand those to fill the allotted space. There is no possibility that a journalist will actually examine the evidence and show which side is clearly bonkers.

For an example of this phenomenon in action, examine this article about a teacher in Modesto, Mark Ferrante, declaring that he will teach intelligent design in biology classes. It’s a moist sopping wallow in the so-called middle ground, getting quotes from teachers on both sides of the issue, and making special care to include a theist teacher mumbling platitudes about “Let science tell us what and how. Let religion tell us who and why.”

And of course, they go to the Discovery Institute for their story about ID, and set them against the NCSE, as if these two groups have an equal investment in the scientific truth. They do not. Intelligent Design has no credibility, no empirical support, and no reasonable proposals for scientific investigation. When will the media wake up and realize that their constant pushing of a false equivalency is a major factor in feeding this pseudo-controversy?

[More]

The parody is excellent, exactly demonstrating some of the things I recently discussed.

But then to be juxtaposed with a real article displaying almost all of the same items from the parody? Nan Austin of the Modesto Bee displayed little journalism when she just checked off many of the points from the template provided in the parody.

1. Pun in the sentence. Check.
2. ‘Some trustees and other science teachers say.’ Check.
3. Soundbite quotations. Check.
4. Subheadings that only attempt to add context. Check.
5. Padding out with useless, irrelevant facts. Check.
6. Last paragraph demonstrating how nothing as really changed. Check.

A true test of whether a piece of writing is hackwork or not is whether it can be reduced to a check list. Looks like a lot of science journalism fits. But then, as Sturgeon’s Law supports, a lot of normal journalism is crap also.

So, the parody may not be really funny anymore but perhaps it can now actually be used as a guide to help identify the 10% of science journalists who are not hacks.

If it can not be reduced to a checklist, then it is most likely not hackwork. I expect that it would likely no be possible to make a parody of good writing.

Lab humor – it is funny because it is true

How not to consult your biostatistician before doing an experiment
[Via Science-Based Medicine]

A friend of mine at work sent this video to me in great amusement.

I just hope he wasn’t making a comment on my behavior when it comes to dealing with our biostatisticians. I have, of course, seen investigators approach biostatistians this late in the game. Not that I’ve ever flirted with this sort of behavior, of course. At least the researcher in the video above actually consulted the biostatistician before doing the experiment, rather than after doing an experiment with inadequate statistical power to answer the question asked. On the other hand, I guess it doesn’t matter if the researcher doesn’t listen, does it?

[More]

Many people go into biology because the math requirements are lower. So, their understanding of statistics and probability may be rudimentary. Of course, this causes a problem since so much of biology is based upon random or stochastic events.

Thus, placing two identical mice into the same cage and treating them with the same drugs can result in very different results. So we try and define the range of possible results that occur so that 95% of the time, the actual results fall in that range.

So, in a simple case, I guess we could examine a hundred mice and make the range fit 95% of them. But for ethical reasons we want to experiment on the fewest number of animals as possible. It is cheaper, faster and less destructive to the animals to use few animals. We need statistical rigor to be able to get similar results by using much fewer mice.

Of course, the drive to reduce the number of mice results in the plea to use only 3 mice. But think of those 100 mice. If you chose 3 of them, what are the chances that those 3 would encompass the entire range of ‘real’ results? Not very likely unless the range was very, very small – something seldom seen in biology.

Now three mice might be useful in a pilot study in order to just get an idea of the variability. But in a comparison study between two regimens, three will most likely not provide any real information. Thus all the mice will have been wasted.

But, and this is so true, when writing a grant, you spend a lot of effort making sure the work can actually get done in the budgeted amount. Making sure you have covered all the details is what can take so long. Months.

And then, a friend reviewing the draft of the grant asks the dreaded question “Are three mice enough to get good data?” and you realize that if it is not, you will have to rework the budget, the budget that you spent weeks mashing around to get to fit.

And you know that will be one of the questions a real reviewer will have abd the answer can determine whether you get a grant or not.

So you shop around for a biostatistician who will sign off on the 3 mice study because you really have no time to rework the budget.

And that is when a biostatistician comes in and provides a real object lesson. A humorous one also.

How science works when people cheat

Gene therapy researcher Savio Woo retracts two more papers
[Via Retraction Watch]

Mount Sinai School of Medicine researcher Savio Woo, whom Retraction Watch reported last week has already retracted four papers from major journals as two postdocs have been fired from his lab, has retracted two more from Molecular Therapy: The Journal of the American Society of Gene Therapy.

The two papers, both from 2007, were “Metabolic Basis of Sexual Dimorphism in PKU Mice After Genome-targeted PAH Gene Therapy” and “Correction in Female PKU Mice by Repeated Administration of mPAH cDNA Using phiBT1 Integration System.” As Nature noted in its coverage of the other retractions, the papers apparently followed from a now-retracted paper in the Proceedings of the National Academy of Sciences that, as we noted in a previous post, “claimed to have discovered a possible cure for phenylketonuria, or PKU, in mice.”

The work was of course already under scrutiny. We’ve asked Mount Sinai for comment, and will update if we hear anything back.

Update, 4:15 p.m., 9/23/10: Mount Sinai confirmed “that these two are part of the six total that were retracted because of the two dismissed postdocs.”

[More]

With over 22 million scientists just in the United States,and over 500,000 papers published a year, it is not surprising that some people take shortcuts, either through mistakes or outright fraud.

But, most cheaters want to get recognition for what they did. Why take a shortcut on something that no one cares about? And since it is something others care about, people check it out and discover the cheats.

You can not fool reality for long. If you say you may have found a cure for PKU then people will check it out.

I guess someone could cheat in an area no one cared about. Fraud through obscurity might be successful but if it is something no one cares about then it really does not matter. It will have no effect one way or the other.

Besides, you do not get ahead by publishing obscure papers. Fraudulent papers are created because someone wants to get ahead unfairly.

It has been my experience that someone cares enough about almost everything to eventually expose the fraud in science. So what we see is the normal process of science correcting itself – sometimes because of error and sometimes because of fraud.