Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?
Compare cell phone plans using Wirefly's innovative plan comparison tool ×

Comment Re:FIFY (Score 1) 59

"Amazon and NVIDIA, the venture arms of the CIA"

Unsurpringly, Amazon is *the* cloud service supplier for the the CIA...

On the other hand, In-Q-Tel (the *actual* venture arm of the CIA) has partnered with companies like...

* Booz Allen (the actual employer of Edward Snowden during his short stint as an NSA contractor)
* Intelliseek (eventually purchased by Nielsen Online, the folks help decide what programs get air time)
* Keyhole (eventually purchased by Google to become google earth was also partially funded by NVIDIA)
* Safeweb (eventually purchased by Symantec, the same sellers of the popular Norton Anti-virus software)

As always, the truth is out there, and some of it is a little bit unnerving...

Comment Re: Inevitable (Score 1) 59

Are you planning to ... um, VACATE the presidency in 2017? Because there's no US presidential election next year. At least not yet.

Perhaps. If DT this Nov 2016, I'm sure he'll strongly consider how he can VACATE the presidency shortly after the inauguration in 2017 (still no indication he actually wants to be president for 4 years).

Or maybe DT will just keep the office and hold an apprentice contest for the "most powerful apprentice in history". Odds are that Pence probably won't stand a chance against an AI in such an apprentice contest...

Comment Re:Of course. . . (Score 1) 445

& torrential rains soon to hit the Left Coast again soon: see Wikipedia for "Pineapple Express."

The climated do-gooders are going to have a field day with the next mega-rain on the West Coast, but they happen about every 160 years.

Well, on the west coast we are still waiting for the repeat of the 1862 atmospheric river (aka pineapple express) they promised us last year to get us out of the drought. Of course as will all things, there can always be too much of a good thing.


Comment Re:Internet or hyper-linked documents (a.k.a. Web) (Score 1) 70

The write-up and TFA conflate the Internet and (what became known as web). Maybe, the slines don't know any better, but Slashdot users ought to... The hyperlinked documents weren't the first "killer application" — e-mail was. The first systems weren't even using the Internet, but, according to Wikipedia:

And Sir Lee's was not even the first system for linking documents/files across the networks — Gopher was. And Gopher was not merely proposed in 1991, that's when an actual system became available (though protocol was codified in an RFC only in 1993).

If you want to get "technical" the web (aka http/html) was first (1990 vs 1991 for gopher), but the graphical browser mosaic didn't appear until '93 and not to many folks were using the non-graphical web servers that were in existence at the time.

If email was the killer app, inter-domain mail (via unix mail via rmail/UUCP) was probably the real killer app, not ARPANET email as ARPANET was mostly restricted to non-commercial use. Gopher like the "web" didn't really pop up until '91 when the NSFNET (the modern "internet") was winding down and the commercial internet was ramping up (the various NAPs like MAE and CIX, etc were taking off). Prior to inter-domain unix mail, commercial email was generally *unconnected* (needed to be on the same proprietary system like compuserve to send/receive mail).

Comment Re:How durable? (Score 1) 160

What kind of Shingles do you use in the US that you have to replace them so often?

Replacing roofs in Germany is a rather rare activity. I doubt it happens more often than every 50 years.

Cheap crap in some cases, this being America.

In others, I will grant that windstorms in the US may be very common, and there is a greater degree of sun exposure from the lower latitudes.

Some parts of the USA are prone to get hail which causes recurring damage on roofs. Fortunately I no longer live in one of those areas. Insurance companies really don't like paying for hail damage on roofs.

Allegedly, modern solar panels are more resistant to hail damage than typical composite asphalt roofing material (and probably better than wood shingle/shake or spanish tile roofing too) so if over a few year period, it helps save a roof repair (or two), that might be a thing on the plus column for these new fangled solar roofs...

I'm interested in what this solar roof would be like, but if it is like a metal roof, I don't know if that would be any more problem free. Metal roofs can have quite a few problem in residential installations (e.g., condensation, noise, ice dams), so it isn't a panacea. Also the current generation of metal roofs are generally painted with IR reflective coatings, but current generation solar panels are generally transparent to IR. I don't know about solar roofs and there may be some attic heat issues that might be specific issues with a solar roof if they aren't much different than current generation solar panels.

Comment Re:Makes sense for Intel (Score 1) 81

I'd like intel to use their high-resolution fabrication facilities to make systems with new programmable opportunities, like part of the die set aside for FPGA...

Well, intel did buy altera last year and promised such a thing, but we haven't see anything real come of that yet...

Intel did announced that they would ship a multi-chip module with a Xeon server chip and an Arria fpga connected by QPI in the same package, but I haven't seen anything on their roadmap yet that actually has FPGA blocks on the same die as their CPU.

Comment Re:shortsightedness and quarterly investors (Score 1) 26

mystified by the reluctance of some portfolio companies to avoid a stock market flotation. "They would benefit from the rigor and discipline that the public market requires," he said.

Maybe an idea which needs a year to properly develop an idea is justifiably afraid of quarterly meddling at the beginning.
That's discipline to avoid that short term trap

I assume he's not talking about small companies, but GV portfolio companies like...

Uber, Slack, Cloudera, Blue Bottle Coffee, etc...

Which seem to be happy to continue to take new rounds of VC money leaving GV and early investors with no exit. W/o an IPO it is more challenging for GV to get their money back to invest in other companies which means their returns might look good on paper, but these companies are essentially holding GV's investment hostage and using the new VC money to invest in new projects w/o paying back the original investors for the risk they took on previous projects. Essentially these companies are effectively forcing GV to invest in their new projects (instead of taking that money and investing in another company).

If you want a car analogy, it's like Slack telling GV about a plan to drive from SF to LA to sell a car but Slack doesn't have any money for gas. GV gives Slack the gas money with the promise that Slack will pay GV back the gas money and some part of the money from the sale of the car as a bonus when they get to LA. Somewhere near Bakersfield, Slack decides he wants to go to Las Vegas instead because Slack could get even more for the car in Vegas, much more than the cost of gas to get there.

However, GV is starting to have second thoughts about Slack's Vegas plan and wants out of the car and Slack to pay back his gas money and suggests Slack crowdsource his trip to Vegas and use the money raised to payback GV (maybe use the money to fund someone else's trip to LA to sell a car). Instead, Slack decides to pick up a VC hitchhiker that willing to pay for the gas to go from Bakersfiled to Vegas, but not enough to pay back GV's gas money.

GV thinks Slack won't crowdsource because they would price in the risk of Slack's plan to go to Vegas instead of LA to sell the car and reduce Slack's expected return of going to Vegas vs LA and it was simply easier to find some VC hitchhiker to get some Vegas gas money and hold GV's gas money hostage instead. GV thinks Slack could use the discipline of the crowdsource wisdom to see the error in it's way.

Comment Re:Darkening the skies as well! (Score 1) 323

All that solar is making the sun dimmer! Oh and all that wind power is slowing the rotation of the earth! Senator Joe Bartan (R) said it was true!

On the other hand, the three-gorges dam in china apparently slowed the earth's rotation by about 0.06 microseconds, and shifted the pole position by about two centimeters.

Of course if we extrapolate this data and we continue building dams at the current rate, we will probably cause catastrophic damage to the earth in the next million years (unless we go extinct by then). Maybe we should stop building dams now before it is too late.

FWIW, what Mr. Barton said was actually this...

I am going to read a paragraph which is, if true, very ironic. And this is from Dr. Apt’s paper, and I quote: ‘Wind energy is a finite resource. At large scale, slowing down the wind by using its energy to turn turbines has environmental consequences. A group of researchers at Princeton University found that wind farms may change the mixing of air near the surface, drying the soil near the site. At planetary scales, David Keith, who was then at Carnegie Mellon, and coworkers found that if wind supplied 10 percent of expected global electricity demand in 2100 the resulting change in the earth’s atmospheric energy might cause some regions of the world to experience temperature change of approximately 1 degree Centigrade.’

This Dr. Apt's paper was poorly paraphrasing this 2013 Harvard report which quotes research that was partially funded by the Natural Sciences and Engineering Research Council of Canada (a country with major petroleum reserves and skin in the game)...

It's easy to blame politicians for ignoring science (as if many of them were scientists qualified to analyse data), but it is often the Universities that tend to confuse issues. Hey we've got researchers from Princeton, Haarvaard, and CMU saying something ironic *if-it-were-true*. Then confirmation bias sets in (happens on both sides of the aisle). The spin cycle kicks in to make your political enemies look like idiots to please your audience. Rinse, repeat (god forbid don't lather, think of the environment).

Comment Re:SLS? (Score 1) 81

NASA surely could have come up with a better acronym than SLS (OK, the SLA could have).

The way over-cost and doomed Space Shuttle Program bore the acronym STS.

Branding-wise, that is a little bit too close.

If you are attempting to "sell" a program to a funding strapped congress, the block-1 SLS being derived from the STS might imply you aren't reinventing the wheel and probably the same pork is involved. Funding a similar program carries the implication that nobody getting money/contracts the last time is likely to going to go home hungry on your new program. Assuming you got enough votes last time to get funded, that's an important selling point.

If you name your program something totally different, your lobbyists will have to waste part of their elevator pitch time on explaining the what it is and why they should vote for it to some uninformed congress-dweller's staff.

Comment Re:Holmes (Score 4, Informative) 40

My question is: Did the FDA properly audit Theranos, and if not, why? From what I understand, this wasn't found out until some doctor saw his patient's results and questioned it, and then it got blown wide open from there. It shouldn't have even made it to that point really.

As I understand it, the FDA *did* audit Theranos on a single test (HSV-1 IgG aka herpes test). That only tests for the presence (or absence) of antibodies, not a test that requires determining a reasonably accurate concentration of certain chemicals in blood so in a sense it was a somewhat easy test.

Unfortunately, Theranos used the aura of this single approval to build their lab business which they apparently used their proprietary "nanotainer" collection tube. Regardless of the ultimate accuracy of their "Edison" machine used to gain their singular test approval (nobody knows outside the company how many samples were tested on Edison vs diluted and tested on standard lab equipment in production which is another FDA complaint that they didn't track this), the collection procedure also has to be FDA approved to assure audit-able quality and repeatably. This is where Theranos failed big time.

Not only was the "nanotainer" not approved to be listed as collection device, but FDA audits revealed all sorts of problems with its quality control, including no supplier auditing, no validation procedures, nobody in the company approved the final design specifications before manufacturing commenced, no clear customer and lab complaint logging and handling procedure, etc, etc...As the FDA audited the complaint handling procedures it discovered complaint that involved the failure or accuracy of the device that were received and not logged as required (to establish a validation history of a procedure) which brings in to question the reliability of the tests results in actual field conditions.

Comment Re:How to make a defense under 17 USC 117 (Score 1) 158

The big advantage of cartridges is that they are more rugged, easier to transport and harder to duplicate.

IANAL, but I'm pretty sure you are incorrect about the IP protection available to "mask-works". The mask-work protection laws protect the mask itself from being copied or being used w/o permission to create computer chips. The theory of mask-work protection is that the actual mask sets used to fabricate integrated circuits weren't sufficiently protected by copyright or patents. As a rule, copyrights do not offer protection against most utilitarian aspect of industrial design objects (e.g., so you can't get around the fact you can't patent a fork by allowing it to be copyrighted). Since a mask is pretty much all utilitarian, and although you can patent the chip that it makes, how do your protect the mask itself from being duplicated or used w/o permission? That's where these smask-works protection laws came from.

It is unclear that mask-work protection laws extend to actually protect the ROM contents represented by the mask (although in their "dumped" form, the software and game artwork is most certainly protect-able by copyright). In fact many statutory overviews I've read about mask-work protection seem to indicate that protection is only given to masks used to make a specific topological pattern of circuits on an integrated circuit, but not protect a circuit that is potentially functionally the same but a different pattern (e.g., a ROM that is topologically different, but functions the same). You could of course attempt to protect the functionality of your IC by a patent, but it would not be protected by mask-work protection laws. It might be hard to assert a new patent on a ROM device, except for potentially a novel copy protection scheme, so I'm not so sure how mask-work protection helps from a legal point of view over simple copyright.

Comment Re:74 at time of crash (Score 1) 623

> designed to ignore large flat signs that cross above the road.

Yeah, maybe ones the car can fit under. But there are no signs 'above' the road that are only 4 foot above it, that's an object you need to avoid.

AFAIK, the radar array in the Tesla S has mostly a horizontal footprint with only a single vertical positioned antenna element for detecting height. Given the typical noise in radar, that probably doesn't give much angular resolution to accurately determine the height of an object w/o lot of tracking and noise rejection. With low angular vertical resolution, it might be hard to tell if it is 4 foot or 8 foot above the road (which isn't always flat**) until the car is too close to do anything about it.

If the software was taught to think these kind of targets are usually signs that generally have enough clearance, the software might just wait it out until it gets a better read on the situation rather than spuriously apply the brake.

That being said, I'm sure they will put in better software in the future to get better read on the situation and/or track/filter more noisy radar returns over time to interpolate better vertical resolution. Perhaps they could simply cheat the tuning to be more defensive (e.g., brake at a different threshold of ambiguity), but I suspect that is a losing battle and it is better to make the software smarter than to attempt to tune a more primitive design.

Comment Re:doesn't matter (Score 1) 704

Doesn't matter why he did it
Doesn't matter who he did it for

What matters is what DNC did, they played a game, they lost, and now they say Russia is winning so vote for us because Russia is bad, but infact Russia is irrelevant, it's just a tactic to make Republicans look bad now, even though Republicans did nothing wrong (in this context).

Maybe we can morph that into a Chewbacca defense.

Ladies and gentlemen, this is Guccifer 2.0. Guccifer 2.0 is a hacker from the country Russia. But Guccifer 2.0 claims to live in Romania. Now think about it; that does not make sense! Why would a hacker, a sophisticated state-sponsored hacker, want to live in Romania, instead of Russia? That does not make sense! But more important, you have to ask yourself: What does this have to do with this case? Nothing. Ladies and gentlemen, it has nothing to do with this case! It does not make sense! Look at me. I'm a pundit defending a gaffe by a major political party, and I'm talkin' about Guccifer 2.0! Does that make sense? Ladies and gentlemen, I am not making any sense! None of this makes sense! And so you have to remember, when you're in that jury room deliberatin' and conjugatin' the Democratic National Coronation Proclamation, does it make sense? No! Ladies and gentlemen of this supposed electorate, it does not make sense! If Guccifer 2.0 is from Russia, you must elect HRC! The defense rests.

Slashdot Top Deals

We can found no scientific discipline, nor a healthy profession on the technical mistakes of the Department of Defense and IBM. -- Edsger Dijkstra