Come on now!!!! The link in the post is broken. That's just sloppy.

Did anyone even look at the post before putting it up?

I would note this Freefall comic.

Just the robot equivalent of organlegging

A three minute check shows they claim a power savings over having no AV installed at all. They claim 4.43 KwH/day for no AV, and 4.39 KwH/day for Abatis. Unless there is some "new math" reason that 4.43 is smaller than 4.39, it would appear to me that they are claiming to have lower power than no AV.

Yes, surprise surprise, I actually RTFA... and not just skimmed, but truly read it.

Chris Howden and John Plumb are the author and approver (respectively) from Lockheed..... Chris and John are lousy scientists.

The kindest way I can figure it is that the driver simply disables disk IO... hence there may be a small power savings from the lack of writes. Less kindly, they happened to measure lower power, and are reporting experimental noise as a solid result (see www-plan.cs.colorado.edu/diwan/asplos09.pdf for instance). We have no error bars (or even a # of runs), so it really isn't possible to say, but disabling disk writes could conceivably reduce power draw. The methodology section is sketchy enough to make solid conclusions impossible; the reporting of experimental details is worse.

Of course, this doesn't (and they admit it) stop me from hacking them in RAM... nor does it stop persistent firmware attacks (e.g. http://www.wired.com/2015/02/n...), nor does it stop me from trapping to ring 0, then trapping to SMM, then just ignoring their F*ING CODE BECAUSE I"'M IN SMM MODE BITCH!!! I GOTZ MY OWNZ ATA CODEZ

Or something.. I'd recommend just cutting the write-enable line on an old IDE drive, or rebooting periodically and running Tripwire from non-writable media (CD?). It's likely cheaper, and probably just as effective.

That is my only question: can I have a flying unicorn? I'm not satisfied with a mere unicorn, or a pegasus. My little girl is turning 2, and it's time she thought about both her data security and her mythical beings. For my baby girl, I won't settle for anything less than the best. Beyond a 100KB 100% effective security module, I want a horse, flying, with one horn, capable of defeating any poison, and only capable of being captured by a virgin.

And she also wants puppy.

I live in the Bay Area.

My boss has told me "you really should be living in a 6-story condo building; if you did then your housing wouldn't so obscenely overpriced". The snarky response would be "well, can I bulldoze your house to build a 6-story condo? Because there aren't any".

I'm not a brogrammer. I'm not a single hipster. I'm married with two young kids. I don't need a trendy city scene; I don't want hip bars; I'm not looking for a luxury condo with all the fancy amenities. I just want a place to live. People who complain about denser development often seem to ask the question "why can't they be happy with what we had back in the day? why do they want so much space?". Well, how much space? In the 1970's, the median household size was 3, and the median house size was 1500 sq. feet. That's 500 sq feet per person.

My family has 350 sq. feet per person, and we pay through the nose for it.

The occasional octogenarian who isn't fit enough to move is a human interest story puff piece; the real political problem is the 55-year-old who simply doesn't want anything to change, and has a "I got mine, what's your problem?" attitude. "Doesn't match the aesthetics of the neighborhood" is code for "we don't want young people here", just as much as "digital native" is code for "people under 40 need not apply". Both attitudes are poisonous.

Actually, with regards to this whole DICSS thing, I was looking for some help with a web-coding project. You see, all of these "high-level" languages are really wasteful, producing lots of garbage (especially if you use a wrapper), and it takes forever for useful output to come out.

Hence, if you're into the cutting edge, I was hoping you might contribute to my project: automatically translating things like DICSS into plain vanilla C. Yes, the C-STRAIGHT project looks to free eveyone of their useless DICSS, and move to a nice, sleek language, with no hairy bits. Even better, if I can get the help of someone who is truly razor sharp on the javascript type system, I think we can do the whole thing by automatic cast-ing nearly painlessly.

Not only that, but basically all gasoline cars have better range than the electric ones, which means a longer time between gas-tank fills vs charge-ups.

So, each chargepoint needs to be used more often, and for far far longer. Calling it "not quite comparable" is quite the understatement.

The developers of the NTFS support for Linux do have a fsck implemented. It does a pretty good job. However, since they've done a black-box re-implementation, they rightfully aren't willing to 100% guarantee they have everything correct. Hence, although it has been almost a decade since I've had trouble with Linux NTFS support causing a problem, the recommendation is still to run the native Windows chkdsk after the Linux one finds a problem, just in case.

So, it isn't that you can't use a pure Linux toolchain to work with NTFS, it's just that the Windows toolchain is a little bit better. All in all, that isn't surprising.

The overall unemployment rate among PhDs in computer science is shockingly low. Per the current Taulbee Survey (see pdf here), unemployment among fresh CS PhD graduates from surveyed institutions (266 North American ones; likely comprising the whole top 100 institutions plus 166 others) is .8%. .8% is well below the frictional unemployment rate; a PhD in CS is almost as good as a civil-service union government job in guaranteeing employment for life.

So ask yourself, what are you totally screwing up. Some previous posters have suggested that perhaps you're shooting way too low (intro programming job) for your talents. This could be the case. It could be that your degree is from a less-than-reputable institution (you didn't say, so we can't comment). You may just be messing up the basics of interviewing --- my PhD prepared me for an academic interview, but not so much for a straight industry job. Asking help from your institution's career services department on interviewing skills could help.

Regardless, there are very well collected statistics that reflect that a CS PhD is a strong benefit to gaining employment; don't blame the PhD.

You seem to have gotten my point backwards

When Joe DiMaggio comes on the radio, and tells you to smoke Winston cigarettes, because they're invigorating and healthful, you should ignore him. When a former playboy bunny goes on television and tells you that vaccines cause autism, you should ignore her. When Rush Limbaugh goes on air and tells you climate change is a conspiracy made up by secret hidden commies working together with Al-Qaeda, you should ignore him.

And when the chair of the divinity department at Harvard tells you climate change is real, you should ignore him too.

If the faculty at a Harvard want to get the administration to not invest in Chevron, that's all well and fine. They want to write a letter, that's all well and fine. But writing a news article about it just because it was Harvard faculty? It deserves a news article just as much as if it was the faculty at the University of Pheonix: not at all. This sort of trash is why the tobacco companies got away with it for so long, why there's an anti-vax movement, and why climate change denial is still around.

Not to be cynical..

OK, that's a lie. Cynical mode is *on*.

How many of these 100 faculty (or is it 93?) are actually qualified to have an opinion about this? How many are involved in hard science (physics, chemistry, engineering)? And how many are in fields that deal in arguments and sophistry above all else?

How many of the signers are in fields that would have been duped by the Sokal Affair and how many have done a good job of curating their facts? How many of those 100 are proprietors of horse-caca? You tell me 100 Harvard faculty want to get out of coal/petroleum... which of them do I care about more than if you told me 100 ballet dancers wanted the same?

FYI, a major value-add the server vendors give is validation, burn-in, and firmware tweaks. So that "same" model of Seagate drive sold by HP might be rock solid, while the IBM version of the WD drive might be absolute junk.

Short version is that everyone ships a crap batch occasionally, and direct from the manufacturer has the worst odds.

IIRC Backblaze's workload is write once read maybe once (I mean, they are a backup company). So it's quite likely that they are massively under the specs for throughput.

The truly interesting thing about this study is that they name names; previous work in the area (lke Bianca Schroeder's FAST 07 paper, http://www.cs.cmu.edu/~bianca/... or Google's FAST 07 paper, http://research.google.com/arc..., or NetApp's FAST 08 paper http://www.usenix.org/event/fa...) doesn't give away vendor names. The Backblaze results broadly agree with the previous results.

but that's a fair accusation, because I don't really.

80-90% of things can be shipped off to software where it's delightfully easy to trace/probe/debug things, and you have a functional unit which is infinitely malleable. What you can't do in software most realistically ends up in an FPGA, where really you're just debugging your VHDL/Verilog, and the simulator is your new best friend for 80-90% of the cases. When the simulator is a lying piece of junk, 80-90% of the time all you need is a good logic analyzer...

But there's still that ~1% of the time where software and/or digital logic just didn't behave right. Something analog is either necessary (e.g. maybe you're doing something actually useful, like driving a motor, rather than just flipping bits), or analog is making your life miserable.

Even professionally, I've found a 2-channel 50 MHz analog scope to be a godsend in some cases; of course, I like my 4-channel 1GHz digital scope more :) If you end up interacting with anything real and physical, or if you you move beyond merely debugging black boxes and into building your own stuff, even a crappy scope can give you information you simply can't get any other way. Who cares if it is uncalibrated and wildly inaccurate if a surplus scope will still show you the shape of what is going on, with all of the noise and ringing and transient under-(and over-)voltages and double bounces and cross-talk and odd harmonics and wtf why was that capacitor in the wrong bin this RC constant is borked and yep that part's dead and oh shit bad solder job and all the other crap that makes me happy I get to spend most of my time in nice clean software?

If you're just putzing around, sure, a DMM will do ya. But if you're actually building something new (even something simple), you need a scope.