Please create an account to participate in the Slashdot moderation system


Forgot your password?

Comment Evil Things RDRAND Could Do (Score 3, Informative) 566

Yes, RDRAND could do evil things. It could go play Towers of Hanoi when you execute it. It could Halt and Catch Fire. It could email your MAC address to the KGB. So could any other instruction, if Intel wanted to be malicious, just when you thought it was safe to go back in the register pool.

If the NSA has convinced Intel to do evil things with RDRAND, the most likely one would be to hand out low-quality entropy when claiming that it's high-quality. It's still useful, and like any entropy source, it shouldn't be the only entropy source you use, and you shouldn't use it without hashing it together with a bunch of other hopefully-not-broken entropy. But it's still useful, and as somebody said, the NSA isn't your only enemy.

Especially when you're starting up a machine (physical or virtual), you really need good entropy and you don't have a lot of sources available yet. If you don't trust RDRAND, or even if you do, hash it together with some secret password and the clock and whatever else you've got.

Comment Mmmh. What is happening to Torvalds? (Score 0) 566

I have not met him, and since I am not a hacker I don't know if this is standard behaviour or not.

The only thing I would say, being an admirer and supporter of Linux and of him personally, is a Mexican saying: "lo cortes no quita lo valiente" (that loosely translated is "be brave, but be courteous").

Comment You missed the Golden Age Pulps, I guess (Score 1) 322

Somebody at Worldcon said that the circulation of paper magazine SF is down to something like 20,000 copies/month, from millions back in the day. Short stories work differently from books, and the choices are pretty much to do them in magazines, or in anthologies, or now the internet, but it's a tough market. (And short story writers are getting paid far less per word today, adjusted for inflation, than back in the 50s. The going rate is under 10c/word, while the 3c/word the old guys used to get would be maybe 30c-$1 in today's money.)

Comment Books also have better pictures and audio (Score 1) 322

There are exceptions - "2001: A Space Odyssey" was a better movie than book (even without drugs :-) and "Star Wars" (the original non-re-edited one) was too. But for the most part, the visual effects and sound are a lot better when you're reading a book and playing them inside your head than when some movie producer with a finite budget tries to interpret them.

Comment How can you not read a Wordstar 1.0 floppy? (Score 1) 440

All these mechanisms have their limitations, and if you've ever tried to do real-world data collection from a wide-ranging group of people who have data in random formats, it's a mess. People used to send me tapes in VMS Backup format, or with a duct-tape label indicating which tape it was and an Nth-generation photocopy of what some of the fields on the tape were, or 8" floppies in RSX-11 format. I've got useful data on Sun cartridge tapes, ZIP drives, and several generations of floppies, not that I've got readers for all of them (or ways to plug the readers into my current computers.) My department at $DAYJOB had the last 800-bpi 9-track tape drive in my building 20+ years ago; these days I don't know anybody with a 1600- or 6250-bpi tape drive, though I suspect there are some here in Silicon Valley besides the Computer History Museum and Digibarn.

Data formats rot. Hardware formats rot. The only way to keep the stuff is to keep copying onto newer media, and keep extensive documentation.

Comment X-Don't-Wiretap-Me-,-Bro!: (Score 1) 202

Yeah, that'll work.

Protecting your messages with crypto is a start, and using traffic mixers like Tor and Mixmaster to resist traffic analysis, but it's a hard job when the Bad Guys have Moore's Law on their side and unlimited unaccountable budgets and politicians who want to keep it that way.

Comment Big Government is a Right-Winger thing (Score 1) 202

Look, you right-wing trolls like to talk about how liberals and progressives want big government, but we're dealing with Bush's Homeland Security Mafia here, and the right-wing Drug War, and the right-wing Big Military-Industrial-Complex which goes conquering other countries on behalf of Big Oil and Hating Foreigners. And you guys talk about "Intellectual Property" like it's as sacred a thing as owning real dirt property that we stole from the Indians, so the Copyright Police are as much your fault as they are the liberals' fault. And if Obama were actually a liberal, we'd have some Hopey Changey Stuff and the warrantless wiretappers and Gitmo torturers would be in jail, instead of him telling his Justice Department to defend the Bush Administration policies.

Comment You secure it with Crypto, not Guns. (Score 1) 202

You and your friends don't have enough guns to outgun the NSA (who are typically not armed), much less the FBI, Pentagon, and Copyright police. If you want your data not to get wiretapped, you need to use crypto, end-to-end, and use various traffic analysis obfuscation services in the middle, and get enough people doing it to have some actual cover traffic (because being the one person using an anonymity service doesn't do the job.)

Comment Re:Correlation is not causation, FFS. (Score 1) 417

I'm not insinuating anything, I've said it several times!

I don't believe that many climate scientists deliberately falsify data. I'm sure there are some because there are people in every field that falsify data. I do believe that when funding is dependent on grants, and grants are highly available for AGW research topics, that researchers have a completely rational motivation to find AGW in places where there is perhaps ambiguity. I'm surprised this is a controversial statement. If you work in academia, finagling to get grants is just part of life. Five years ago just about every grant request in certain branches of physics and chemistry were modified to include something about "biosensors" or "chemical sensors" because there was a huge amount of funding available for homeland security projects. That's just what goes in academia--it is what it is.

Regardless of all of this, I absolutely stand by my statement that "97% of climatologists agree," as sourced in Cook et al., is totally false.

When I was an undergrad, the Republican group at my school pulled every faculty member's political affiliation. History, Art History, Women's Studies, African American Studies, sociology, anthropology, etc, had between them a large number of faculty (100, give or take) and not one Republican--all Democrats. Engineering, computer science, math, etc were spit more much closer, and in the economics department, there were more Republicans still. Why? Ideology obviously has a great deal to do with what people choose to work on. Correlation or causation? I know what I believe.

I have no doubt that most climate scientists do believe in AGW (hesitation over the 97% claim notwithstanding). I believe there are many reasons for this.

The good news is, we should no for sure within say, 10 years. If the models are right, we should see a statistically significant change by then. If the models are wrong...? Until then (and even then, unfortunately), factions will no doubt keep arguing.

Comment Re:Correlation is not causation, FFS. (Score 1) 417

Yeah. Who cares if cities drown, countries disappear under the ocean, and Africa starves if Tropicana can plant some orange groves in Greenland?

Yours is EXACTLY the kind of statement that makes me so uncomfortable. That is such an incredibly hysterical statement. Not even the most extreme estimates of sea rise produce anything like you say within centuries. Africa starve? 10,000 years ago the Sahara was green and wet. Maybe it will be again. There's so much uncertainty, we really, really don't know.

Republican doves were driven out of the party by 2003. And Democrats have shown their propensity for hackery since they've defended actions from their Dear Leader that would have had them in the streets if it was Bush doing the same things for the same reasons.

The Tea Party and libertarian strands of conservatism have made more traditional (paleocon if you will) noninterference conservatism more resurgent than it's been in years. The present Republican dove (nee neocon) can make some excuse by saying that "we learned from our mistakes in Iraq and Afghanistan."

What the heck excuse do the Democrats have? "We did such a bang up job in Afghanistan and Iraq (plus our limited interference in Libya) that we thought we ought to help out in Syria as well! Never mind the hundred thousand that have already died, a couple hundred deaths from chemicals weapons is the red line!"

Comment Re:Correlation is not causation, FFS. (Score 1) 417

Well no, that's actually not what you accused me of initially, but ok. I think there's a pretty huge difference between the statement that "97% of climatologists agree" and "97.1% of a subset of climate papers that contained a certain set of (in my view potentially biasing) words and in which the authors made some kind of judgment call about AGW posit that humans are responsible for at least some portion of global warming." At this level of disagreement, we're down to a religious wedge issue, so I don't expect us to suddenly come to an agreement, but that's how I see it. I do not believe that mine is an unreasonable statement.

To put it more waggishly (or "partisanly" as you may see fit)--is it a surprise that papers (funded by a government that believes in AGW) discussing anthropogenic global warming believe in anthropogenic global warming?

Slashdot Top Deals

A computer scientist is someone who fixes things that aren't broken.