I would be really amazed if 1 B people ACTUALLY lost $142 in a year, it's like 30 M average Americans losing $1500 each. There is ransomware and credit card theft, but former is still rare (1 % of net users is likely high ballpark) and latter covered by credit card companies.

If corporate cybercrime is counted, sure, but the $172 B figure sounds like those sums that people calculate are lost standing in red lights. Kinda money... but actually really not.

Of course, didn't yet RTFA, checking that out next!

There should be a +6: Insightful for posts like these. Thank you for aptly summarizing the absurd tone of many Torvalds-related Slashdot discussions on the topic of his "aggressive attitude". It is a small wonder that even after 20 years of posting online, the people complaining about this usually have to dig up the same few instances of Linus "behaving badly" (and even in those cases usually justifiably so). I'd say it's almost a miracle he does not get pissed of more often and lash out, given the scale of responsibility (and possibilities to get drunk on power).

Furthermore, as a fellow Finn, I think most people judging Linus' posts as aggressive have a poor radar of sarcasm and witty hyperbole Linus uses in his hotter emails. For example, I think the "why I don't use C++ for Git" email is one of the most hilarious pieces of tech writing, and you should not take everything he says 100 % seriously all the time.

Sure there will be those who'll comment that "in his position it's enough to be 99.9 % benevolent and 99 % friendly, he should be 101 % benevolent and 99.999 % friendly", but those are just people really detached from reality, who'd rather have an imaginary superhero to worship in their shrine of ideal world. :)

Rest in peace, Sir Lee.

I heard that when he died, a grey mist gathered, and rising slowly to a great height like smoke from a fire, as a pale shrouded figure it loomed over the Hill. For a moment it wavered, looking to the West; but out of the West came a cold wind, and it bent away, and with a sigh dissolved into nothing.

There have been nice suggestions for buckling spring (IBM Model M and Unicomp) and Cherry MX switch (Das Keyboard and dozens of others) mechanism keyboards, but personally I think Topre keyswitches are superior to both of them. Granted, they are among the most expensive, but the price is minimal when compared to rest of the rig, and they last a lifetime.

I have been using Happy Hacking Keyboard 2 Professional (I did a review of it a while ago: http://codeandlife.com/2013/10/12/hhkb-professional-2-keyboard-review/ ) for two years, but recently switched to Topre Realforce 88UB (US version is 87U). It's probably the best keyboard money can buy, the feel is superb and I even like the black-on-black color a lot. Details: http://codeandlife.com/2013/10/12/hhkb-professional-2-keyboard-review/

I think the switch type and layout are the primary factors, so I'd advise you to read a bit of geekhack forum and decide what you'd like to try. A word of warning: Mechanical keyboard can be addictive!

Alternate explanation would be that usually economic downturn is linked with low inflation, whereas when economy is thriving, inflation is often a concern.

Could it just be that people tend to laugh more when things are going well in general (omitting inflation), and are in turn more somber at hard times?

I have to say I'm not certain why all these "Rob Malda gets a job" and "Rob Malda discusses Reddit" articles that have appeared since he resigned from Slashdot are relevant news for nerds. I've been following the site since the nineties, and with the exception of few posts (like the proposal mentioned earlier), the editors were never a focus in the posts, but the news instead.

I would assume the people who wanted to follow CmdrTaco more carefully could, like, follow his personal website or something. At least to me, the Slashdot culture has never centered around any figureheads, but the community and commenters (and of course Rob Malda was an integral part of both for a long time).

At least this three-part interview series could've been timed to occur more closely with the 15 year anniversary, now it just feels like free PR push.

I wonder what the unexplained "sine wave" is doing on the background of that image. When you replace it mentally with a linear best-fit line, you get an increasing trend. Using an arbitary non-linear function, especially the sine wave that implies "we're going down next" feels like deliberately pointing the viewer to a certain conclusion.

I haven't read Roy Spencer's arguments behind this graph, but I'm quite sure I could apply the sine wave to many statistical datasets that do actually increase over time to create the illusion that it's "just oscillating, not rising".

<quote><p>They can crash without crashing, although a serious crash might cause a crash.</p></quote>

Good point, I was already concerned that crash could cause downtime!

On a more serious note, I wonder how they are going to connect to internet from these servers - maybe a ground station serving as relay link? If so, how is this any better than having the whole server in that ground station?

<quote><p>And the LORD said, My spirit shall not always strive with man, for that he also is flesh: yet his days shall be a hundred and twenty years.</p></quote>

I loved it when in Terminator II Arnold explains to John Connor that he has internal battery which will last for 120 years. I don't think that amount is a coincidence. :)

<quote>For values of "anyone" meaning the first 10000 people to get in before the server implodes.</quote>

I'm pretty sure they uphold their commitment to not taking orders before they have the actual units in stock, so once the first batch runs out, the store closes again and you won't be able to order until the second batch arrives. I can already see the Slashdot headlines: "Truck with 24th batch of R-Pi spotted on premises - expect the store to open once more in 30 minutes. Hurry because they'll run out of them again in 33 minutes!"

Point of sarcasm: I don't exactly understand why R-Pi is refusing to take orders just few days before they get the first batch - how is it different after they get 300 000 orders the first day and run out of units - I'm pretty sure they won't be sending emails to 290 000 people saying "thanks for ordering but sorry, you have to do it again when the second batch arrives, better luck then!".

As I see it, there's not much difference letting people fill the order forms today and wait for three months for supply to catch up with demand and waiting 10 more days, so people only need to wait two months and 20 days. Unless Paypal charge cannot be deferred to later point in time, in which case my horror scenario might just happen.

I find it really hard to understand all the people who are so obsessed about this. "Oh my, the pureness of this project has been tainted by the horrible non-OSS nature of GPU drivers and now I have to respond to every post about Rasp-Pi about this!!!". Come on people, it's a $25 computer, it is running Linux, you can do great stuff with it and unless you have an altar devoted to Richard M. Stallman in your home, you can hardly complain about the influence this small issue will have to your kids learning experience with the device.

I understand that the closed nature of this part of Rasp-Pi may be a personal letdown for some 1 % of geeks who are mainly interested to hack the graphics drivers, but I could bet that nine out of ten people who constantly whine about it when any possibility surfaces would not personally ever tinker with that part of Rasp-Pi even it were completely OSS - they just have the unrealistic world view where no compromises need to be made to provide a cheap learning platform with "sexy" graphics functionality for education.

To end with a car analogy, if there was a non-profit organization making a $100 electric DIY moped that could, thanks to a closed design of the battery circuit (everything else of course would be completely open), achieve a speed of 30 mph instead of 5 mph open designs, the same people would probably be here pointing out how their kids enthusiasm for DIY projects and the usability of the whole moped would be compromised because this one part was not open for everyone.

iPads would probably be a huge hit with alien tourists, much like boomerangs are to us human tourists visiting Australia. For an alien, nothing would remind them of their primitive stone age roots more than a crude device such as the iPad.

Actually, perhaps he really wants to create magnetic patterns to a medium (in this case, a hard drive platter). That would explain the need, and modern harddrive capacity would allow for some really intricate ones.

Other option is that the guy wants to do research on how long strings of ones and zeroes are really stable on modern hard drive platters, or some other such thing - it's hard to simulate this reliably, especially if you're more of a DIY type of guy than a physicist.

Your story, sir, is just not plausible. You really had me going with the monastery stuff (monks need their ones and zeroes), but everyone knows printed zeroes weigh more than ones. See: 0, 1!

OTOH: On a hard drive, it's all digital so ones and zeroes generally have the same weight, the difference comes whether they are just random noise or contain valuable information - for example the entire library of congress stacked on one platter would need to be padded out with proper amount of useless stuff like slashdot comments and lolcats. This is consequently also why writing ones and zeros directly to the platter without having a hard drive controller circuitry handling error correction and lolcats is a really bad idea.

It all makes sense when you really think of it!