Because fabrication is too general a word. The proper name is "additive manufacturing".
But fabrication can include many processes, including subtractive manufacturing (usually CNC machine), simply bolting two parts together to form a third, nailing things together, injection molding, etc. And it can involve multiple processes - additive followed by subtractive, etc.
In fact, combining the existing tools of being able to injection mold some stuff, add and subtract bits can easily manufacture parts that once had to be manually assembled or took many parts to accomplish.
Like Slashdot Beta, this is probably being driven by Ãoeweb designersà and marketers. It's not good enough that something have reached a state of maturity that works well with users, and they like. Throw away the furniture and toss out the Persian rugs, white carpet and a do-over by Ikea is what we need, right?
The problem is a very vocal minority call for redesigns because designs get "stale".
The most obvious is to take a look at UIs that haven't changed too much - OS X and iOS. They were functional and they worked. But more and more, people see how Android changes its UI practically every version, and seeing that OS X and iOS stay the same and change little, call it "stale", "dated" and "not evolving".
So you get people who are always looking for the new shiny convincing everyone that to look and act different is good, it shows you're "evolving" and "changing" and "innovative".
And then there's the rest of us who are trying to get shit done, and having the whole base of our work ripped from under us continually.
There's no love when you keep things the same - the users get their work done the same. But change it up, and those minority say "cool" and "innovative" and stuff, even though to everyone else, it just means they have to spend more time doing what they were doing because the new UI is less efficient.
More shiny, that's it. Keeping something the same for more than two versions is stale, old, passe.
I just wish these people would realize that if you need to sell someone on the new UI, you did a bad job. Like iOS7, the new Firefox, Windows 8, Gmail, etc. People want to get shit done, not figure out where the (*&@#% you put the commands now.
Perhaps a lot of server administrators are simply tired of dealing with the unending farce that constitutes modern internet security, and have simply decided to give in. What's the use in spending time and effort on security measures which frequently fail, sometimes spectacularly so in the case of heart-bleed. In particular, what's the point of protecting customer data if organizations like the NSA can simply walk in and take it, or if you're already selling it en-masse to marketers.
I suspect a lot of them are like that. Or rather, the devs are too lazy. One of the reasons is simple - the 30% that Apple, Google, Microsoft, Valve etc., take for product distribution. Why do that when you can just set up your own Linux server, write a bit of code to get personal information and then use Paypal to handle the money?
Of course, such systems remain vulnerable because it's too expensive to maintain (you want to write product software, not maintain the damn website) - forgetting part of that 30% is so you don't have to hire someone to keep up to date on security patches and other flaws.
Meanwhile your customer's data remains vulnerable, and most likely since it was hacked together, lacks any detection to know if it was broken into and data stolen.
Yes, a lot of people feel the likes of Google/Apple/Microsoft/Valve/Shopify/Amazon/etc take too much money as their "cut" and want it all for themselves, without realizing it's actually a lot of ongoing full-time work, and the services take that cut so you can concentrate on more important things.
You can say "Google is evil" all you want, but are they anywhere near as evil as, say, Comcast?
Of course not, the telecom companies are pretty much the most hated companies in the world, and it's fairly universal.
Google is evil, but on a relativity scale, Google, Apple, Microsoft, etc., appear to be saints compared to telecoms.
It's just like airlines - we keep hearing how Air Canada is "the best airline" in North America, yet most Canadians specifically AVOID flying Air Canada because well, they're awful. It's just US airlines are even worse. (There are other carriers like WestJet and the like for within-Canada flights, and using the Asian and European air carriers when going transcontinental.).
However, the value that they bring has to be WORTH IT. That's how capitalism works: if someone is willing to do the job cheaper than you and they do it "well enough", they will get business. My cab driver can also do surgery and quote Tennyson? Great - I'm not paying for it. It's not the purchaser's responsibility to offer a 'living' or 'fair' payment for what you're bringing. They are going to buy the cheapest service possible that does what they want.
And that's the problem the cabbies have with Uber - the cabbies are forced to take these extensive tests (the thing is, they MUST know the entire map of London by heart).
So they're going to cause disruption until either Uber forces its drivers to the same standards, or the livery commission drops the requirements.
So it's really also a political move - because either one or the other has to give. Either the law has to change to permit lower standards to compete, or there will have to be enforcement. And in the absence of either, well, they'll take matters into their own hands.
After all, the cabbies didn't just practically spend a year studying to become a cabbie for nothing (yes, it can take that long to get a license).
I suppose we'd see similar actions if instead of hiring certified welders, construction sites started just hiring random people who know how to weld. The uncertified welders are just as cheap because well, anyone can buy a welder and start sticking metal together at home.
I used aircraft parts as an example, and the likelyhood that the correct parts are made on the same assembly lines in China as the 30 cent Walmart versions is vanishingly small.
Actually the chances of them being made on the same assembly line is pretty high. The difference of course is that the line that has to "have" the certification, they'll use a higher grade material and take random samples for stress testing to ensure that it's right. They may even go as far as x-raying the materials before it goes through processing, and after to look for material defects.
I used to work in heavy industry back oh 15 years ago now. The stuff we sold went to the US military, and was used for scraping your ICBM's(particularly the minutemans). Everything had to be checked like that before it went out, but the differences were trivial in terms of what we sold to the general public and what went to the military.
The other thing is that the parts are traceable. That bolt from Wal-Mart probably can't be identified who was the original manufacturer, when it was manufactured, who manufactured it, who was working at the factory that day, where the metal came from, etc.
The item itself is cheap. The documentation is not. An aircraft bolt will have a serial number engraved on it that traces it back to the factory it was produced from, and when it was produced, who was working and signed off on it, tests on random samples during that shift, and even where the wire the bolt was made from originated and their composition and lab tests.
Because in case the bolt actually fails, they can see if it's a one-off failure, or if every bolt of that batch needs to be recalled and changed because it's defective.
The problem happened in the 80s - there were tons of parts brokers who were shady and often re-tagged expired parts as new or factory overhauled. It was so bad that the faulty counterfeit parts made it into Marine One (It's not Air Force One until POTUS is onboard) which sent the FAA into an enforcement tizzy that put a whole bunch of brokers forging tags into prison.
Like I said, the documentation on these parts is insane. It's also why a 30 cent bolt from Wal-Mart costs at least $5, $10-30 being more typical.
We all know that things like screen grabs can "save" a snap chat on the other side. There is always an analog hole. To get a fine over it is stupid. Snap chat's promise is that THEY don't save it - either on their server or in their app. The summary doesn't seem to indicate they did otherwise either.
Except when they did.
First, the app is SUPPOSED to let the other side know if you took a screenshot. On Android, if you do it, it works very reliably. On iOS, not so much (because Apple didn't provide an API for that). Instead, what people did was put their finger on the screen then did the Power+Home screenshot trick. The app detected this and used it as a "screenshot taken" notification back to the sender. (The goal is social - someone who did this a lot would probably get less photos from friends who really wanted them to disappear). Of course, iOS 7 broke all this so there's no way for the app to know if a screenshot was taken.
Second, the app itself was completely vulnerable. The photos were easily accessed (unencrypted, even). And even when viewed and supposedly "gone", they were still retrievable from the phone very easily (i.e., they weren't actually deleted).
Then there's the whole "well, we'll let you review any image from the day again, but just once" which implies that you can see any image anyone sent you again. Of course, the sender gets notification when this happens, too.
So it's a case of well, photos that supposedly disappear, but didn't. And the fact you can re-view any image sent again (but just one of them, once a day).
So the claim that the photos simply disappear after viewing turned out to be false advertising when there were so many easy ways to retrieve those photos, ignoring even the analog hole. Even then the claim they disappear is meaningless because of the analog hole, so it's an intentional fraud since users believe that once it's gone, it's gone. No more compromising images hanging around!
I fully expect our civilization to collapse when we run out of oil. Any "migration off fossil fuels" scenario at present technological level means drastically reduced ability to support population. This will all but inevitably lead to nukes flying, making problem that much worse. Save black-swan technological breakthrough Western Civilization has "best before" date of 2070-2120.
Hence the need to well, do the migration as soon as possible. I mean, instead of waiting for the inevitable oil crash, why don't we start migrating and researching alternatives?
We can still use fossil fuels, and there are plenty that are plentiful and "renewable" like natural gas. Then there's a whole pile of geothermal and other energy sources that are basically unlimited (i.e., the sun will basically engulf the Earth).
The thing is, we need to wean ourselves off the "oil is mandatory" mentality and realize there's plenty of alternative fuels out there that can keep us going for a while yet, sustainably.
Some industries, like air transport and ship transport, will probably require a lot of research, so they will need oil the longest. Others, like basic cars and inner-city transportation, have plentiful alternatives, including electricity, making the switchover not only less painful, but cheaper as well.
The real problem is inertia and stubbornness. You want to know why one of the reasons Tesla did a sports car? Because they wanted to show that electric cars can perform just as well as gas ones when you want to floor it and "have fun" (because well, back in the 60s and 70s, the electric cars were decidedly non-sporty, slow vehicles that couldn't keep up).
Apple is somehow (allegedly) pulling these things from your device remotely (heaven knows why the security model even allows this to happen) at the behest of law enforcement.
Not remotely, unless your definition has changed to "using some other computer hooked up to it".
Apple needs to be in physical possession of the suspect device, AND said device needs to be delivered with warrant simultaneously.
Likely this means the phone needs to be hooked up to a special test rig to actually work.
Older iPhones and other phones often have special download rigs that police can have and they use over USB to extract data, but later ones only Apple can retrieve data from. With physical possession. And that hasn't been remotely wiped. And with a warrant.
It's actually kind of refreshing that Apple details what it can and cannot retrieve and the conditions for it publicly
Yet Japan, the country mentioned in the article, has a much higher suicide rate than the United States despite their strict gun control policies
And there are many reasons for that, culturally.
First, educational systems in Asia tend to be brutal. Basically, at the end of high school, you take an exam. That exam determines your future. If you score well, you can go to university (overseas! scholarships! fully paid!) and study what you want and get a job doing what you want.
Score lower, well, you may be able to get into a trade school and do some blue-collar work.
Fail, and well, your life is over. End it.
And there's incredible pressure to get into the university and professional track. So much so that if you don't make it into a local college, it might as well be over - your family will practically disown you.
It's not a surprise that many students crack.
Then there's the whole work thing. Especially in Japan where there's a whole "job for life" thing. Getting fired is a great shame (even just getting laid off) because it implies failure. So the only out is well, suicide.
Hell, the Japanese have a word for killing oneself honorably.
Of course, I suppose if we gave the Japanese guns, they'd off themselves so much that there will be a visible population drop the next year
Yes, but they are not doing another offering. If you buy Tesla stock today, you are not buying it from the company. You are buying it from a third party who holds it today. Tesla does not see a dime of the exchange of their previously issued shares between you and someone else. Additionally, secondary offers of new shares are very uncommon.
Long story short: when you buy a stock, you are not giving any money to the company you invested in. Your purchase can help them indirectly - pushing up their stock price improves their ability to borrow, attract new employees with stock incentives, etc. - but if you want to help them directly, buy a car. Or go to one of their concerts.
It also affects what the company does. As a stockholder, you have a say in the way the company is run.
Now, if you buy stock because you believe in Tesla, then you're likely to vote in ways that are good for the company's future. If you bought the stock from an investment company, it means the voice of "profits profits profits dammit" goes down, and the voice of "environmentally friendly cars" goes up.
For example, a few months ago, an Apple stockholder demanded Apple stop trying to do so many profit-cutting "green" initiatives, in the name of more profits and money. He was soundly voted off so well that even Tim Cook could say "If you're not aligned with Apple's principles, including environmental practices, do not own Apple stock".
Remember when complaining about CEOs who only care about this quarter and not the future of the company? That's an edict usually issued by the Board of Directors, who gets their orders from the shareholders. If all the shareholders cared about are short term profits, well, the company will focus more on short term profits. If the shareholders cared about long term prospects, environmental responsibility, and other factors, well, the board and CEO will be forced to care about them too.
And this is a debugger. So what are you on about? Do you normally ship the debugger with your game?
A lot of places are very wary of the GPL. Especially GPLv3.
One place I work now audits all the open-source/free-software code that enters the codebase. If it has a hint of GPL, it's almost always auto-rejected.
Oh, and by codebase, that includes the tools as well. Because some tools emit code, or otherwise touch code, and it's better to be safe than sorry.
Some GPL programs were allowed - e.g., gcc, but only the versions approved.
Other licenses, like BSD, Apache, MIT, no problem - those tend to sail through the legal license audit easily. They even went out to say no new GPL code would be accepted - look for alternatives.
Me personally, I release code under the 3-clause BSD or GPLv2. I don't agree with the GPLv3, so it's v2-only or 3-clause BSD (incompatible with GPL).
And I guess Crytek feels the same - they don't mind anyone using it, just not locking it up under the GPL.
ever before have documentation and good tools been so available to even indie developers. Never before has it been so easy to actually earn money with indie game development.
And things might be getting even better.
I agree - now is probably amongst the most fertile time for indie developers. There are tons of platforms to choose from, even ones that were previously hostile to indie development are fairly open now.
It used to be you could only do it on the PC. But Apple opened it up on mobile (previously it was REALLY HARD to develop on mobile - you couldn't get SDKs unless you had carrier agreements in place, and carriers were extremely picky.) - for $99 you could develop on a reasonably (back then) powerful device and within limitations that's all you needed. No more dealing with Verizon, Sprint, AT&T and having to deal with incompatible SDKs, etc. etc. etc. As much as I'd like to say Android, well, it came later and offered more of an alternative at the time. These days, both are equally compelling platforms to write for. Though Apple is more "console like" and gets you prepared for a world dominated by approvals like Sony, Nintendo and Microsoft, and even Valve.
Plus, the consoles opened up as well - Microsoft did it best on the Xbox360 by having practically an open community-regulated way to put your games on it. And there have been notable mentions from there.
These days really are the golden age - both current gen consoles are establishing Indie development guidelines, you have mobile and tablet gaming on iOS and Android, and there's always the old standby of well, the PC.
Hell, even Steam's opened up to indie games to help solve distribution and payment issues. (It was a lot harder to get on Steam in the past unless you were a big publisher. The domination of the App Store started forcing Valve's hand to open things up).
But 'right now' is possibly the hardest that it's ever been to *make money* from indie development - simply because there's so many people making games (due to much-improved tools), it's incredibly hard to get noticed, and the bulk of the media attention goes to the already-successful 'super-indies'.
And even with all the digital distribution options out there, there are new all-powerful middlemen controlling what has a chance of real success - Steam, Humble, Apple (featured content), etc
Personally, I loved the 90s, when the technology was really exciting and evolving fast. The indie boom of the late 2000s was cool too, but now we seem to be facing oversaturation and race-to-the-bottom pricing (even beyond mobile).
You think the middlemen are bad because they control and take a share? I say for indies, they actually provide a very valuable service - payment and distribution.
Sure it's easy to create a website and stick a Paypal button on it, but you won't believe how many issues there are. First, you need to figure out a way to ensure paid users can get at their content any time of the day or night. Usually that means an account system, but now you've just opened yourself up to data breaches and having to ensure your account security is always up to snuff. Then there's dealing with payment problems - the whole "I paid and I got nothing" aspect. Software upgrades, additional content purchases, etc just add to the complexity. And website coding - just because you can write a spiffy iOS/Android/PC/Console game doesn't mean you're a uber web coder. Plus refunds, taxes, international sales. A developer's day can easily be tied up in administrivia so they can't really spend time doing what they want - developing the game!
So paying Valve, Apple, Google, etc., to handle that aspect (re-downloads, receipts, taxes, payments, upgrades etc) relieves the developer to concentrate on supporting their game and development. And in-app purchases can be used to provide additional content on all platforms (extra levels, pinball tables, etc).
Slam the brakes on and don't swerve either way. It's by no means optimal, but as far as lawsuits are concerned, it's much easier to defend "the car simply tried to stop as soon as possible" than "the car chose to hit you because it didn't want to hit someone else".
Actually, I think that IS the solution. Because an autonomous car should be designed to drive safely and keep a distance. Which means not following the car in front too close that there's no way it can brake safely should it stop suddenly (and by "suddenly" it doesn't stop on a dime, but it undergoes maximum braking so it still travels a good 10-20m, if not more - no car can stop on a dime). And even then an autonomous car should be able to stop quicker by threshold braking. (Yes, ABS does lengthen stopping distance compared to perfect threshold braking, but if the driver just slams on the brakes and locks up the wheels, the loss of control and traction means they stop further. Plus, ABS lets you steer around the problem).
And if the traffic flow is such that there are cars on both sides, the speed is limited anyways.
This limits the reasons to avoid hitting something in front to either someone cutting you off and suddenly stopping, or a pedestrian crossing. In the first case, it was unavoidable anyways, and to be honest, if there's traffic on both sides, crashing in front is the safest - front/rear crumple zones tend to be most generous and safest. Side impacts are among the most dangerous.
If it's a pedestrian, to be honest the car should be deliberately driving slower if it has blind corners where people can step out of and where "see in front and around" sensors are obscured. But if there's cars around you, it either means you're in the middle lane of traffic, in which case a pedestrian would be struck by someone on the lane beside you first (i.e., not an issue since if you see that the car should already by slowing down), or it's a row of parked cars, in which case swerving into that is the safest option since most parked cars don't have an occupant.
You have to remember that autonomous cars would be among the most polite on the road because the computer can easily see far ahead and anticipate traffic, and be one of the best defensive drivers out there. And with the mountain of sensor data, the idiot that cut you off and ensured you ran into him would have to face a pile of evidence showing that no, there was no way to avoid an accident in the circumstances.
Seems like we could probably stop at about 187 digits, really. The radius of the observable universe in planck lengths (call it X) is about 2.7*10^61, which makes the observable volume (4*pi*X^3) about (8*10^184)*pi cubic planck units. The value of the 186th digit of pi (after the decimal) should only affect the final volume by about 0.7 units; going much beyond that seems unnecessary:)
To be even more dramatic, while the planck length determines the ultimate resolution, you only need 39 digits of pi to calculate the circumference of the observable universe to within the width of a hydrogen atom.
Any two-bit ISP (and in this context, that includes even the likes of Comcast) that thinks they can twist L3's arm has one hell of a nasty surprise waiting for them when their current contracts expire. This doesn't work quite the same as not getting to see this week's episode of Glee because of a pissing contest between cable companies and content providers - A week where Comcast customers can't get to Por... er... Google, means a week where Comcast loses half its customer base.
No, they just drop Level3. That's it.
Unless you're a tiny rural ISP, you're multi-homed. You peer with Tier 2 ISPs like Level3 and others to provide you service. Level3 peers with other Tier 2 ISPs as well as Tier 1 ISPs.
Typically, an ISP will pay for service from a higher tier ISP, and get paid for service from a lower one - i.e., Level3 gets paid from a last mile ISP, but Level3 has to pay a Tier 1 like Sprint. Other Tier 2's usually have an equal-sharing arrangement where it's free.
Naturally, everyone wants to keep traffic on the "free" peering (i.e., same level), which works when traffic is roughly equal. The problem occurs when it isn't, in which case one side or the other has to pay up for the differential traffic. (If the traffic was truly equal, then both would upgrade the ports together because both sides are dropping packets).
Of course, the Internet is supposed to route around congestion - the overloaded port means that instead of Level3 continuing to send data that way, they need to send data to another port, usually through another provider which can mean $$$ gets paid.
The fact that's not happening means either or both sides are playing shenanigans - it's a trick either side can use to force the other to comply. E.g., if Level3 has multiple ports with an ISP, they can easily force all traffic through one port, ignoring the other ports and force the ISP to upgrade that port. Then they send traffic to another port, to force that one to be upgraded as well until they're all upgraded.
It's a nasty trick, but it usually works. (A more relatable example would be if you paid for network access - you only use say, 250Mbps, so you buy three Fast Ethernet ports. But your provider wants you to buy Gigabit ports, so they basically strangle traffic through two ports, forcing it all on one port until you agree to upgrade that port to GigE. Then they send all the traffic through another port so you upgrade that, until you're paying for 3 GigE ports when 3 FE ones was sufficient).
the real reason bluray ''won'' the hd format war wasn't the slightly higher quality audio or video possible due to the higher data capacity.... it was **THAT** -- region locking.. bluray has it, hd dvd did not, and hollywood chose the format that had it.
Blu-Ray also had two other features that Hollywood liked.
1) Mandatory AACS encryption. HD-DVD made AACS optional, which meant "amateur" videographers could publish their own HD-DVDs without paying for an expensive AACS key. This also meant that self-publishing via HD-DVD recordables was possible.
2) Profile Locking. A Blu-Ray movie uses the BDMV profile which gives you full access to interactivity features of Blu-Ray. This was only possible through pressed media enforced by ROM-Mark. People who burned their own Blu-Rays were forced to use BDAV instead, which meant you basically got a collection of videos. Again, it's not an attempt at the home videographer, but more for the independent filmmaker - because they couldn't make Blu-Rays as slick as what Hollywood could.
The whole point of it all was less about home videos, and more about locking out the indies - if you weren't part of the MPAA, you couldn't make your own Blu-Rays, effectively. Of course, many third party publishers eventually bought their own AACS keys and mastering hardware and have contracts with (highly-regulated, again, supposedly to limit piracy) Blu-Ray disc pressers these days, so it's no longer a limitation.
But back then, Hollywood used it as a way to block indie films from high-def. Heck, you couldn't put a burned BDMV disc into a commercial player other than a PS3 (because the players needed to read the key from ROM-Mark, and you couldn't burn that).
These days, most players don't bother anymore - all that excess protections aren't used or needed because the original goals have been defeated - we can rip Blu-Rays even with BD+, indie filmmakers have lots of publisher choices to make their own professional Blu-Rays, etc.
(And Blu-Ray took a couple of years for Profile 2.0 to come out - something HD-DVD had at launch).
The extra space Blu-Ray had was because the tools were immature (HD-DVD forced their hand) so the only format available for compress was... MPEG2, while HD-DVD used either AVC (h.264) or VC1 (WMV9), and HD-DVD also had the space advantage - 30GB on a dual layer disc while Blu-Ray only had 25GB because pressing dual-layer BDs wasn't available and yields were too low.
Is AMD just around so Intel doesn't get bogged down by anti-monopoly or antitrust penalties?
Somehow these days, I think it's yes. And I think Intel's lobbing customers AMD's way to ensure that AMD survives. E.g., the current generation of consoles now sport AMD processors. I'm sure Intel would be more than happy to have the business, but not only do they not need it, they see it as a way to give AMD much needed cash for the next few years.
Hell, I'm sure part of the whole Intel letting others use their fabs thing is to figure out a way to get AMD to use some of their spare capacity. Of course, it has to be done in such a way that it doesn't run afoul of any anti-trust and all that.
Right now, AMD is in a good spot for Intel - big enough to count as competition, small enough to not really matter..
You can bet many other companies pay lots of money for a competitor to stay in it - I can think of Google and iAds, for example. Google got AdMob because Apple introduced iAds, yet iAds is completely worthless to any advertiser - it's too expensive, too limited, and all around a bad dead, whereas Google is cheap and easy. And yet, Apple keeps iAds around , despite practically no one supporting it. Apple's killed other stuff for less. Only reason I can see is Google pays Apple for that to keep anti-trust at bay.
Isn't this what used to be called natural selection, evolution, if you will, and isn't this how living things have adapted throughout the planet's history of continuous "climate change"? I call complete and utter (pardon the pun) bovine excrement.
Evolution is a slow process - it takes millions of years for it to work. But in 200 odd years, we've basically changed the atmosphere enough that historical records show it points to a natural ELE (extinction-level event) that has occurred a few times in Earth's past.
Yes, the dinosaurs were wiped out by a meteor. But prior to that, there were other events of climatic disruption that killed almost all life on Earth.
And 400ppm of CO2 is one of them.
Evolution works, but only over time. It's why these mass-extinctions happened - because the climate changed faster than anyone could evolve, so the ones that survived did so more by luck than anything else.
Ununseptium's name has been up for debate for years.
Hell, I'm surprised this thread hasn't even suggested one based on Halo - Master Chief's designation number was, after all, 117. (He is "officially" known as John-117, or Spartan-117, depending on whether you want to personify him or not).
The 'true' version can be found in Sled Driver which is phenomenally hard to get hold of in dead tree form:
Given it's only in deadtree form, it's actually not that hard to get. The special edition is out of print, but there's an unlimited one for sale. Online shopping, though not at Amazon, but it's not hard to get. Though I suppose for a year or two between the special editions selling out and the unlimited printing it might have been hard.
The big problem is the price tag - $250! Yes, Two Hundred Fifty Dollars. (The special edition was $450, Four Hundred!).
I don't know why it's so expensive, nor why there aren't cheaper editions out there. You'd think for that price they'd include a ride or something.
And no, for whatever reason no one's posted it for download.
I suppose it's "hard to get" purely because it's so expensive. There are more expensive books out there, but those tend to be speciality and collector type books, or overinflated textbooks. But something that appears to be a coffee table book?
The ironic thing is, by doing all these countermeasures, she appeared MORE unique to Big Data.
Everyone loves to try to hide, but the thing is, hiding makes you more obvious. Making large purchases from Amazon, but no obvious Amazon account usage or credit card/debit billing? Well, you must be an Interesting Person to follow. And not only that, there are so few Interesting People that you stick out like a sore thumb.
In the age of Google knowing everything, the name of the game is to NOT stick out. Be like your neighbours, and your traffic won't be immediately "more interesting" to investigate. I mean, in a hypothetical case where some terrorist was a newborn mother, are you going to investigate everyone on Amazon who bought diapers and a stroller recently? Or the 10 people who did it using Amazon gift cards bought from a local gas station (with cash)?
The caveats are that map updates tend to be released quarterly (a problem if you're looking for that super-new restaurant the next town over)
The big map companies (Navteq and Tele-Atlas) only do quarterly map updates. The raw map data is provided to the companies that subscribe who then transform that data into the native format required for their apps.
You can't update faster than the map provider gives you.
You'd think Google would have the advantage here, but their map data can be years out of date...
The big scientific screwup on the show is exactly how Walt was managing to get nearly 100% purity from a process that in real life would result in 50% purity at best. Unlike starting from pseudoephedrine, the P2P process results in a racemic mixture of 2 different stereoisomers of methamphetamine, only one of which has any recreational value.
Off-topic, but hydrogenation of fats results in a similar thing (it's a process used to saturate fats with hydrogen).
Hydrogenation can create two forms - "cis" and "trans" - or chirals.
The "trans" form is what is really bad for your heart - hence the bans on trans-fats. But it isn't trans-fats that makes food "taste good" - since you can fry food in oils that are naturally low in trans-fat, and avoid using hydrogenated oils.
Can someone explain this "no-poaching pact" thing to me? I'm a software developer in the UK, and it's not uncommon on employment contracts to have something where you have to agree not to work for a competitor (or even with any of your company's clients) for a year after you leave the company. Is this not usually allowed in the US?
Basically, when a bunch of companies agree to a no-poaching agreement, it just means they will not call up employees at other companies in that agreement to offer them jobs. In this case, Apple will not call (either directly, or through recruiters) Google employees and extend them job offers. Or vice-versa.
Now, that distorts the job market a little bit because companies won't hire from other companies.
HOWEVER, this does not prevent employees from seeking employment at companies making the agreement. That is, if an Apple employee wants to work for Google, or sees a public Google job opening they want, they are free to apply for it.
That is, the first-contact MUST be initiated by the employee, not the company. And the usual way to do this is, well, through a job application - Apple employee submits a job application with their resume, and waits for Google HR to come back and ask for an interview, etc.
And more importantly, if they want you, there are ways around it. Just because Google and Apple have an agreement, if someone at Apple was REALLY wanted by Google, there are ways to do it. One such way is to post a job opening publicly and have the employee made aware of it and apply (through mutual friends - networking. If you don't know anyone outside of the company, why not? Do you not keep in touch with people who leave?). As long as the employee applies for the job, all is clear.
How do I know this? I was with a company with a no-poach agreement with another. That company wanted me, and did exactly that - I was made aware unofficially of the job, and was told to put in my resume and application form. Skipped the interview and everything, etc.
No-poaches are informal gentleman's agreements, which is why there is extensive documentation on who contacted whom, etc. And because they're informal, it also means that companies aren't prohibited from hiring from no-poach companies, just that they're not to make first contact.
Slashdot Mirror
Because fabrication is too general a word. The proper name is "additive manufacturing".
But fabrication can include many processes, including subtractive manufacturing (usually CNC machine), simply bolting two parts together to form a third, nailing things together, injection molding, etc. And it can involve multiple processes - additive followed by subtractive, etc.
In fact, combining the existing tools of being able to injection mold some stuff, add and subtract bits can easily manufacture parts that once had to be manually assembled or took many parts to accomplish.
The problem is a very vocal minority call for redesigns because designs get "stale".
The most obvious is to take a look at UIs that haven't changed too much - OS X and iOS. They were functional and they worked. But more and more, people see how Android changes its UI practically every version, and seeing that OS X and iOS stay the same and change little, call it "stale", "dated" and "not evolving".
So you get people who are always looking for the new shiny convincing everyone that to look and act different is good, it shows you're "evolving" and "changing" and "innovative".
And then there's the rest of us who are trying to get shit done, and having the whole base of our work ripped from under us continually.
There's no love when you keep things the same - the users get their work done the same. But change it up, and those minority say "cool" and "innovative" and stuff, even though to everyone else, it just means they have to spend more time doing what they were doing because the new UI is less efficient.
More shiny, that's it. Keeping something the same for more than two versions is stale, old, passe.
I just wish these people would realize that if you need to sell someone on the new UI, you did a bad job. Like iOS7, the new Firefox, Windows 8, Gmail, etc. People want to get shit done, not figure out where the (*&@#% you put the commands now.
I suspect a lot of them are like that. Or rather, the devs are too lazy. One of the reasons is simple - the 30% that Apple, Google, Microsoft, Valve etc., take for product distribution. Why do that when you can just set up your own Linux server, write a bit of code to get personal information and then use Paypal to handle the money?
Of course, such systems remain vulnerable because it's too expensive to maintain (you want to write product software, not maintain the damn website) - forgetting part of that 30% is so you don't have to hire someone to keep up to date on security patches and other flaws.
Meanwhile your customer's data remains vulnerable, and most likely since it was hacked together, lacks any detection to know if it was broken into and data stolen.
Yes, a lot of people feel the likes of Google/Apple/Microsoft/Valve/Shopify/Amazon/etc take too much money as their "cut" and want it all for themselves, without realizing it's actually a lot of ongoing full-time work, and the services take that cut so you can concentrate on more important things.
Of course not, the telecom companies are pretty much the most hated companies in the world, and it's fairly universal.
Google is evil, but on a relativity scale, Google, Apple, Microsoft, etc., appear to be saints compared to telecoms.
It's just like airlines - we keep hearing how Air Canada is "the best airline" in North America, yet most Canadians specifically AVOID flying Air Canada because well, they're awful. It's just US airlines are even worse. (There are other carriers like WestJet and the like for within-Canada flights, and using the Asian and European air carriers when going transcontinental.).
And that's the problem the cabbies have with Uber - the cabbies are forced to take these extensive tests (the thing is, they MUST know the entire map of London by heart).
So they're going to cause disruption until either Uber forces its drivers to the same standards, or the livery commission drops the requirements.
So it's really also a political move - because either one or the other has to give. Either the law has to change to permit lower standards to compete, or there will have to be enforcement. And in the absence of either, well, they'll take matters into their own hands.
After all, the cabbies didn't just practically spend a year studying to become a cabbie for nothing (yes, it can take that long to get a license).
I suppose we'd see similar actions if instead of hiring certified welders, construction sites started just hiring random people who know how to weld. The uncertified welders are just as cheap because well, anyone can buy a welder and start sticking metal together at home.
The other thing is that the parts are traceable. That bolt from Wal-Mart probably can't be identified who was the original manufacturer, when it was manufactured, who manufactured it, who was working at the factory that day, where the metal came from, etc.
The item itself is cheap. The documentation is not. An aircraft bolt will have a serial number engraved on it that traces it back to the factory it was produced from, and when it was produced, who was working and signed off on it, tests on random samples during that shift, and even where the wire the bolt was made from originated and their composition and lab tests.
Because in case the bolt actually fails, they can see if it's a one-off failure, or if every bolt of that batch needs to be recalled and changed because it's defective.
The problem happened in the 80s - there were tons of parts brokers who were shady and often re-tagged expired parts as new or factory overhauled. It was so bad that the faulty counterfeit parts made it into Marine One (It's not Air Force One until POTUS is onboard) which sent the FAA into an enforcement tizzy that put a whole bunch of brokers forging tags into prison.
Like I said, the documentation on these parts is insane. It's also why a 30 cent bolt from Wal-Mart costs at least $5, $10-30 being more typical.
Except when they did.
First, the app is SUPPOSED to let the other side know if you took a screenshot. On Android, if you do it, it works very reliably. On iOS, not so much (because Apple didn't provide an API for that). Instead, what people did was put their finger on the screen then did the Power+Home screenshot trick. The app detected this and used it as a "screenshot taken" notification back to the sender. (The goal is social - someone who did this a lot would probably get less photos from friends who really wanted them to disappear). Of course, iOS 7 broke all this so there's no way for the app to know if a screenshot was taken.
Second, the app itself was completely vulnerable. The photos were easily accessed (unencrypted, even). And even when viewed and supposedly "gone", they were still retrievable from the phone very easily (i.e., they weren't actually deleted).
Then there's the whole "well, we'll let you review any image from the day again, but just once" which implies that you can see any image anyone sent you again. Of course, the sender gets notification when this happens, too.
So it's a case of well, photos that supposedly disappear, but didn't. And the fact you can re-view any image sent again (but just one of them, once a day).
So the claim that the photos simply disappear after viewing turned out to be false advertising when there were so many easy ways to retrieve those photos, ignoring even the analog hole. Even then the claim they disappear is meaningless because of the analog hole, so it's an intentional fraud since users believe that once it's gone, it's gone. No more compromising images hanging around!
Hence the need to well, do the migration as soon as possible. I mean, instead of waiting for the inevitable oil crash, why don't we start migrating and researching alternatives?
We can still use fossil fuels, and there are plenty that are plentiful and "renewable" like natural gas. Then there's a whole pile of geothermal and other energy sources that are basically unlimited (i.e., the sun will basically engulf the Earth).
The thing is, we need to wean ourselves off the "oil is mandatory" mentality and realize there's plenty of alternative fuels out there that can keep us going for a while yet, sustainably.
Some industries, like air transport and ship transport, will probably require a lot of research, so they will need oil the longest. Others, like basic cars and inner-city transportation, have plentiful alternatives, including electricity, making the switchover not only less painful, but cheaper as well.
The real problem is inertia and stubbornness. You want to know why one of the reasons Tesla did a sports car? Because they wanted to show that electric cars can perform just as well as gas ones when you want to floor it and "have fun" (because well, back in the 60s and 70s, the electric cars were decidedly non-sporty, slow vehicles that couldn't keep up).
Not remotely, unless your definition has changed to "using some other computer hooked up to it".
Apple needs to be in physical possession of the suspect device, AND said device needs to be delivered with warrant simultaneously.
Likely this means the phone needs to be hooked up to a special test rig to actually work.
Older iPhones and other phones often have special download rigs that police can have and they use over USB to extract data, but later ones only Apple can retrieve data from. With physical possession. And that hasn't been remotely wiped. And with a warrant.
It's actually kind of refreshing that Apple details what it can and cannot retrieve and the conditions for it publicly
And there are many reasons for that, culturally.
First, educational systems in Asia tend to be brutal. Basically, at the end of high school, you take an exam. That exam determines your future. If you score well, you can go to university (overseas! scholarships! fully paid!) and study what you want and get a job doing what you want.
Score lower, well, you may be able to get into a trade school and do some blue-collar work.
Fail, and well, your life is over. End it.
And there's incredible pressure to get into the university and professional track. So much so that if you don't make it into a local college, it might as well be over - your family will practically disown you.
It's not a surprise that many students crack.
Then there's the whole work thing. Especially in Japan where there's a whole "job for life" thing. Getting fired is a great shame (even just getting laid off) because it implies failure. So the only out is well, suicide.
Hell, the Japanese have a word for killing oneself honorably.
Of course, I suppose if we gave the Japanese guns, they'd off themselves so much that there will be a visible population drop the next year
It also affects what the company does. As a stockholder, you have a say in the way the company is run.
Now, if you buy stock because you believe in Tesla, then you're likely to vote in ways that are good for the company's future. If you bought the stock from an investment company, it means the voice of "profits profits profits dammit" goes down, and the voice of "environmentally friendly cars" goes up.
For example, a few months ago, an Apple stockholder demanded Apple stop trying to do so many profit-cutting "green" initiatives, in the name of more profits and money. He was soundly voted off so well that even Tim Cook could say "If you're not aligned with Apple's principles, including environmental practices, do not own Apple stock".
Remember when complaining about CEOs who only care about this quarter and not the future of the company? That's an edict usually issued by the Board of Directors, who gets their orders from the shareholders. If all the shareholders cared about are short term profits, well, the company will focus more on short term profits. If the shareholders cared about long term prospects, environmental responsibility, and other factors, well, the board and CEO will be forced to care about them too.
A lot of places are very wary of the GPL. Especially GPLv3.
One place I work now audits all the open-source/free-software code that enters the codebase. If it has a hint of GPL, it's almost always auto-rejected.
Oh, and by codebase, that includes the tools as well. Because some tools emit code, or otherwise touch code, and it's better to be safe than sorry.
Some GPL programs were allowed - e.g., gcc, but only the versions approved.
Other licenses, like BSD, Apache, MIT, no problem - those tend to sail through the legal license audit easily. They even went out to say no new GPL code would be accepted - look for alternatives.
Me personally, I release code under the 3-clause BSD or GPLv2. I don't agree with the GPLv3, so it's v2-only or 3-clause BSD (incompatible with GPL).
And I guess Crytek feels the same - they don't mind anyone using it, just not locking it up under the GPL.
I agree - now is probably amongst the most fertile time for indie developers. There are tons of platforms to choose from, even ones that were previously hostile to indie development are fairly open now.
It used to be you could only do it on the PC. But Apple opened it up on mobile (previously it was REALLY HARD to develop on mobile - you couldn't get SDKs unless you had carrier agreements in place, and carriers were extremely picky.) - for $99 you could develop on a reasonably (back then) powerful device and within limitations that's all you needed. No more dealing with Verizon, Sprint, AT&T and having to deal with incompatible SDKs, etc. etc. etc. As much as I'd like to say Android, well, it came later and offered more of an alternative at the time. These days, both are equally compelling platforms to write for. Though Apple is more "console like" and gets you prepared for a world dominated by approvals like Sony, Nintendo and Microsoft, and even Valve.
Plus, the consoles opened up as well - Microsoft did it best on the Xbox360 by having practically an open community-regulated way to put your games on it. And there have been notable mentions from there.
These days really are the golden age - both current gen consoles are establishing Indie development guidelines, you have mobile and tablet gaming on iOS and Android, and there's always the old standby of well, the PC.
Hell, even Steam's opened up to indie games to help solve distribution and payment issues. (It was a lot harder to get on Steam in the past unless you were a big publisher. The domination of the App Store started forcing Valve's hand to open things up).
You think the middlemen are bad because they control and take a share? I say for indies, they actually provide a very valuable service - payment and distribution.
Sure it's easy to create a website and stick a Paypal button on it, but you won't believe how many issues there are. First, you need to figure out a way to ensure paid users can get at their content any time of the day or night. Usually that means an account system, but now you've just opened yourself up to data breaches and having to ensure your account security is always up to snuff. Then there's dealing with payment problems - the whole "I paid and I got nothing" aspect. Software upgrades, additional content purchases, etc just add to the complexity. And website coding - just because you can write a spiffy iOS/Android/PC/Console game doesn't mean you're a uber web coder. Plus refunds, taxes, international sales. A developer's day can easily be tied up in administrivia so they can't really spend time doing what they want - developing the game!
So paying Valve, Apple, Google, etc., to handle that aspect (re-downloads, receipts, taxes, payments, upgrades etc) relieves the developer to concentrate on supporting their game and development. And in-app purchases can be used to provide additional content on all platforms (extra levels, pinball tables, etc).
Actually, I think that IS the solution. Because an autonomous car should be designed to drive safely and keep a distance. Which means not following the car in front too close that there's no way it can brake safely should it stop suddenly (and by "suddenly" it doesn't stop on a dime, but it undergoes maximum braking so it still travels a good 10-20m, if not more - no car can stop on a dime). And even then an autonomous car should be able to stop quicker by threshold braking. (Yes, ABS does lengthen stopping distance compared to perfect threshold braking, but if the driver just slams on the brakes and locks up the wheels, the loss of control and traction means they stop further. Plus, ABS lets you steer around the problem).
And if the traffic flow is such that there are cars on both sides, the speed is limited anyways.
This limits the reasons to avoid hitting something in front to either someone cutting you off and suddenly stopping, or a pedestrian crossing. In the first case, it was unavoidable anyways, and to be honest, if there's traffic on both sides, crashing in front is the safest - front/rear crumple zones tend to be most generous and safest. Side impacts are among the most dangerous.
If it's a pedestrian, to be honest the car should be deliberately driving slower if it has blind corners where people can step out of and where "see in front and around" sensors are obscured. But if there's cars around you, it either means you're in the middle lane of traffic, in which case a pedestrian would be struck by someone on the lane beside you first (i.e., not an issue since if you see that the car should already by slowing down), or it's a row of parked cars, in which case swerving into that is the safest option since most parked cars don't have an occupant.
You have to remember that autonomous cars would be among the most polite on the road because the computer can easily see far ahead and anticipate traffic, and be one of the best defensive drivers out there. And with the mountain of sensor data, the idiot that cut you off and ensured you ran into him would have to face a pile of evidence showing that no, there was no way to avoid an accident in the circumstances.
To be even more dramatic, while the planck length determines the ultimate resolution, you only need 39 digits of pi to calculate the circumference of the observable universe to within the width of a hydrogen atom.
No, they just drop Level3. That's it.
Unless you're a tiny rural ISP, you're multi-homed. You peer with Tier 2 ISPs like Level3 and others to provide you service. Level3 peers with other Tier 2 ISPs as well as Tier 1 ISPs.
Typically, an ISP will pay for service from a higher tier ISP, and get paid for service from a lower one - i.e., Level3 gets paid from a last mile ISP, but Level3 has to pay a Tier 1 like Sprint. Other Tier 2's usually have an equal-sharing arrangement where it's free.
Naturally, everyone wants to keep traffic on the "free" peering (i.e., same level), which works when traffic is roughly equal. The problem occurs when it isn't, in which case one side or the other has to pay up for the differential traffic. (If the traffic was truly equal, then both would upgrade the ports together because both sides are dropping packets).
Of course, the Internet is supposed to route around congestion - the overloaded port means that instead of Level3 continuing to send data that way, they need to send data to another port, usually through another provider which can mean $$$ gets paid.
The fact that's not happening means either or both sides are playing shenanigans - it's a trick either side can use to force the other to comply. E.g., if Level3 has multiple ports with an ISP, they can easily force all traffic through one port, ignoring the other ports and force the ISP to upgrade that port. Then they send traffic to another port, to force that one to be upgraded as well until they're all upgraded.
It's a nasty trick, but it usually works. (A more relatable example would be if you paid for network access - you only use say, 250Mbps, so you buy three Fast Ethernet ports. But your provider wants you to buy Gigabit ports, so they basically strangle traffic through two ports, forcing it all on one port until you agree to upgrade that port to GigE. Then they send all the traffic through another port so you upgrade that, until you're paying for 3 GigE ports when 3 FE ones was sufficient).
Blu-Ray also had two other features that Hollywood liked.
1) Mandatory AACS encryption. HD-DVD made AACS optional, which meant "amateur" videographers could publish their own HD-DVDs without paying for an expensive AACS key. This also meant that self-publishing via HD-DVD recordables was possible.
2) Profile Locking. A Blu-Ray movie uses the BDMV profile which gives you full access to interactivity features of Blu-Ray. This was only possible through pressed media enforced by ROM-Mark. People who burned their own Blu-Rays were forced to use BDAV instead, which meant you basically got a collection of videos. Again, it's not an attempt at the home videographer, but more for the independent filmmaker - because they couldn't make Blu-Rays as slick as what Hollywood could.
The whole point of it all was less about home videos, and more about locking out the indies - if you weren't part of the MPAA, you couldn't make your own Blu-Rays, effectively. Of course, many third party publishers eventually bought their own AACS keys and mastering hardware and have contracts with (highly-regulated, again, supposedly to limit piracy) Blu-Ray disc pressers these days, so it's no longer a limitation.
But back then, Hollywood used it as a way to block indie films from high-def. Heck, you couldn't put a burned BDMV disc into a commercial player other than a PS3 (because the players needed to read the key from ROM-Mark, and you couldn't burn that).
These days, most players don't bother anymore - all that excess protections aren't used or needed because the original goals have been defeated - we can rip Blu-Rays even with BD+, indie filmmakers have lots of publisher choices to make their own professional Blu-Rays, etc.
(And Blu-Ray took a couple of years for Profile 2.0 to come out - something HD-DVD had at launch).
The extra space Blu-Ray had was because the tools were immature (HD-DVD forced their hand) so the only format available for compress was... MPEG2, while HD-DVD used either AVC (h.264) or VC1 (WMV9), and HD-DVD also had the space advantage - 30GB on a dual layer disc while Blu-Ray only had 25GB because pressing dual-layer BDs wasn't available and yields were too low.
Somehow these days, I think it's yes. And I think Intel's lobbing customers AMD's way to ensure that AMD survives. E.g., the current generation of consoles now sport AMD processors. I'm sure Intel would be more than happy to have the business, but not only do they not need it, they see it as a way to give AMD much needed cash for the next few years.
Hell, I'm sure part of the whole Intel letting others use their fabs thing is to figure out a way to get AMD to use some of their spare capacity. Of course, it has to be done in such a way that it doesn't run afoul of any anti-trust and all that.
Right now, AMD is in a good spot for Intel - big enough to count as competition, small enough to not really matter..
You can bet many other companies pay lots of money for a competitor to stay in it - I can think of Google and iAds, for example. Google got AdMob because Apple introduced iAds, yet iAds is completely worthless to any advertiser - it's too expensive, too limited, and all around a bad dead, whereas Google is cheap and easy. And yet, Apple keeps iAds around , despite practically no one supporting it. Apple's killed other stuff for less. Only reason I can see is Google pays Apple for that to keep anti-trust at bay.
Evolution is a slow process - it takes millions of years for it to work. But in 200 odd years, we've basically changed the atmosphere enough that historical records show it points to a natural ELE (extinction-level event) that has occurred a few times in Earth's past.
Yes, the dinosaurs were wiped out by a meteor. But prior to that, there were other events of climatic disruption that killed almost all life on Earth.
And 400ppm of CO2 is one of them.
Evolution works, but only over time. It's why these mass-extinctions happened - because the climate changed faster than anyone could evolve, so the ones that survived did so more by luck than anything else.
Ununseptium's name has been up for debate for years.
Hell, I'm surprised this thread hasn't even suggested one based on Halo - Master Chief's designation number was, after all, 117. (He is "officially" known as John-117, or Spartan-117, depending on whether you want to personify him or not).
Every other thread on Uus has suggested it.
Given it's only in deadtree form, it's actually not that hard to get. The special edition is out of print, but there's an unlimited one for sale. Online shopping, though not at Amazon, but it's not hard to get. Though I suppose for a year or two between the special editions selling out and the unlimited printing it might have been hard.
The big problem is the price tag - $250! Yes, Two Hundred Fifty Dollars. (The special edition was $450, Four Hundred!).
I don't know why it's so expensive, nor why there aren't cheaper editions out there. You'd think for that price they'd include a ride or something.
And no, for whatever reason no one's posted it for download.
I suppose it's "hard to get" purely because it's so expensive. There are more expensive books out there, but those tend to be speciality and collector type books, or overinflated textbooks. But something that appears to be a coffee table book?
The ironic thing is, by doing all these countermeasures, she appeared MORE unique to Big Data.
Everyone loves to try to hide, but the thing is, hiding makes you more obvious. Making large purchases from Amazon, but no obvious Amazon account usage or credit card/debit billing? Well, you must be an Interesting Person to follow. And not only that, there are so few Interesting People that you stick out like a sore thumb.
In the age of Google knowing everything, the name of the game is to NOT stick out. Be like your neighbours, and your traffic won't be immediately "more interesting" to investigate. I mean, in a hypothetical case where some terrorist was a newborn mother, are you going to investigate everyone on Amazon who bought diapers and a stroller recently? Or the 10 people who did it using Amazon gift cards bought from a local gas station (with cash)?
The big map companies (Navteq and Tele-Atlas) only do quarterly map updates. The raw map data is provided to the companies that subscribe who then transform that data into the native format required for their apps.
You can't update faster than the map provider gives you.
You'd think Google would have the advantage here, but their map data can be years out of date...
Off-topic, but hydrogenation of fats results in a similar thing (it's a process used to saturate fats with hydrogen).
Hydrogenation can create two forms - "cis" and "trans" - or chirals.
The "trans" form is what is really bad for your heart - hence the bans on trans-fats. But it isn't trans-fats that makes food "taste good" - since you can fry food in oils that are naturally low in trans-fat, and avoid using hydrogenated oils.
Basically, when a bunch of companies agree to a no-poaching agreement, it just means they will not call up employees at other companies in that agreement to offer them jobs. In this case, Apple will not call (either directly, or through recruiters) Google employees and extend them job offers. Or vice-versa.
Now, that distorts the job market a little bit because companies won't hire from other companies.
HOWEVER, this does not prevent employees from seeking employment at companies making the agreement. That is, if an Apple employee wants to work for Google, or sees a public Google job opening they want, they are free to apply for it.
That is, the first-contact MUST be initiated by the employee, not the company. And the usual way to do this is, well, through a job application - Apple employee submits a job application with their resume, and waits for Google HR to come back and ask for an interview, etc.
And more importantly, if they want you, there are ways around it. Just because Google and Apple have an agreement, if someone at Apple was REALLY wanted by Google, there are ways to do it. One such way is to post a job opening publicly and have the employee made aware of it and apply (through mutual friends - networking. If you don't know anyone outside of the company, why not? Do you not keep in touch with people who leave?). As long as the employee applies for the job, all is clear.
How do I know this? I was with a company with a no-poach agreement with another. That company wanted me, and did exactly that - I was made aware unofficially of the job, and was told to put in my resume and application form. Skipped the interview and everything, etc.
No-poaches are informal gentleman's agreements, which is why there is extensive documentation on who contacted whom, etc. And because they're informal, it also means that companies aren't prohibited from hiring from no-poach companies, just that they're not to make first contact.