I'm pretty amazed at all of the negative responses here saying that we should just hire our unemployed engineers. I've worked several big-name tech companies including, Microsoft, Apple, and Google, and we couldn't find enough qualified engineers at any of them—US citizens or H1Bs. You'd think that these top companies would be able to easily hire good engineers, but it's really tough to find good people, even when we were in the depths of the recession.
At Apple we would go months without filling some critical positions because we couldn't find anyone qualified. After interviewing 10 or 15 people on site, none of them made the cut. I'm constantly amazed at how poorly the people I interview do—and they're the top 1–5% that make it past the resume screeners.
Maybe we don't need more engineers, but from what I've seen, we definitely need more good engineers.
--Bruce
PS—I've never seen any evidence that the companies I've worked for preferentially hire H1B employees. It's a lot of red tape for the company, and they get paid the same as US-born employees. The fact is that they're often simply more qualified than Americans.
I don't think it's so much about embarrassment; it's more about providing a stable target for other developers. Google doesn't want to have to deal with people building software on top of a foundation that they're just going to rip out and replace with a better one in the next version. They don't want to support APIs that aren't solid yet.
--Bruce (who is not speaking for himself, not for Google)
How about measuring that in actual computer usage? X MHz on Y cores per Z nodes over A hours? Or at least say it would have taken one X MHz processor 35 years to compute it.
Um, they did. From the article:
it would take a good desktop PC (Intel Nehalem, four-core, 2.8GHz) 1.1 billion seconds, or about 35 CPU years, to perform this calculation
Heck with the way things are now, the Auto Pilot can nearly land a plane by itself.
Actually, autopilot can land a plane without any human help, and in some cases it's even required to. I was talking to a pilot for United (friend's uncle) a couple years ago, and he said that in high winds or poor visibility, airline regulations prevent the pilot from landing the plan manually. The pilot is required to allow the autopilot to land the plane. Pretty crazy stuff.
An interesting, relatively unknown fact that I picked up while working on telephony systems a while back: carriers get paid (by other carriers) for incoming calls.
Not only do you pay more to your carrier to listen to the inane voicemail prompt (since you might use more minutes), but your carrier also pays more to your friend's carrier. For example, if I'm an AT&T customer and I call a Verizon customer to leave a voicemail, AT&T has to pay Verizon for every second that I'm on the phone. This (perverse) incentive makes more sense than charging people for more minutes, since often the company charging for minutes (AT&T in this case) is not the company that controls the recorded message (Verizon).
This sounds a lot like the 40-year-old Carterfone decision, where AT&T argued that allowing people to connect third-party devices to their network could disrupt or degrade service. I'm pretty sure that modems and Panasonic phones didn't ruin the telephone system, and I have a feeling that jailbroken iPhones wouldn't be the end of the world, either.
It's not a "problem" that can be "worked on". It's the character of the author. As any decent psychologist will tell you that character is inborn and cannot be changed or "worked on".
That's a pretty dismal view of human nature. I, on the other hand, believe people can change.
It sounds to me like the EFF is overreacting again. When the new iPod shuffle came out, the they sounded the alarm about headphones requiring "DRM authentication". They later retracted their statement. Turns out that the "DRM" was just an electrical signaling protocol that allowed the headphones to send button press events to the iPod. You knowâ"the kind that any EE would design. The headphones and the iPod have to communicate somehow.
It sounds to me like this is more of the same. From the article:
car manufacturers now severely limit the number of repair shops that are allowed to have the tools, diagnostic codes and updated repair information essential to being able to repair late-model cars
A lack of information, or the need for sophisticated test equipment does not automatically imply DRM.
I'm all for making information on car diagnostic protocols and diagnostic code semantics more freely available, but let's call it what it really is.
Why does every manager not screen all applicants? Takes 5 minutes.
Because in many places it's illegal to make hiring decisions based on anything but whether the person is qualified for the job. If they go snooping around on the internet, they might discover your marital status, national origin, sexual orientation, or any of the other categories that are legally protected in many jurisdictions. They may or may not make hiring decisions based on that information, but you can imagine how that would look in court.
I worked for one employer (a big computer company you've heard of) that had a policy that explicitly forbid people making hiring decisions from doing an internet search on applicants, for this very reason.
No local decision are made. It's just another algorithmic approach to combinatorial optimization with centralized management, which till I see a Big O notation, and some papers, I withhold comment on the computer model.
It's true that swarm theory isn't being applied directly by the drivers--after all, they're getting instructions from corporate, as you said. However, I think it's unfair to say that swarm theory isn't being applied at all; it's just being used to compute a solution beforehand rather than on the fly. From the article's description, it sounds like the software creates billions of autonomous agents that all follow simple rules and arrive at a solution without any central planning or coordination. That sounds like swarm theory to me.
Maybe I'm just getting old or spoiled by Macs, but is there an end in sight to the mantra of fetch driver and reboot and accept things as they are?
I don't know about you, but I have to reboot my Mac every few of weeks for a security update, or even worse, a QuickTime(!) update. I reboot it about as much as I do my Win2k3 machine at work.
AJAX (at least as it's usually implemented) relies on the XMLHttpRequest object, which was created by MS.
Now it's true that noone really used it for a long time, partly because it was only implemented by IE. It's also true that you can simulate asynchronous requests using hidden frames (something my company did back in 99), but that also never really took off (and probably won't now).
Actually, hidden frame requests are still used (e.g. in Gmail), and for good reason: they don't break the back button. XMLHttpRequest communication has no notion of page transitions, and so the back button doesn't go back to the previous state; instead, it usually exits an AJAX app completely.
In contrast, IFRAME navigation does implement page transitions, which can make web apps more user friendly since the back button does what people expect.
As a bit of a side note, you can write your resume/CV in XML using the XML Resume Project and then easily generate PDF, HTML, or plain text from a single source document. You can even tag elements with keywords and then automatically generate targeted resumes for different audiences.
Slashdot Mirror
It does have a humidity sensor: http://support.nest.com/customer/portal/articles/181007-what-kind-of-data-does-the-nest-learning-thermostat-track-or-measure-
I'm pretty amazed at all of the negative responses here saying that we should just hire our unemployed engineers. I've worked several big-name tech companies including, Microsoft, Apple, and Google, and we couldn't find enough qualified engineers at any of them—US citizens or H1Bs. You'd think that these top companies would be able to easily hire good engineers, but it's really tough to find good people, even when we were in the depths of the recession.
At Apple we would go months without filling some critical positions because we couldn't find anyone qualified. After interviewing 10 or 15 people on site, none of them made the cut. I'm constantly amazed at how poorly the people I interview do—and they're the top 1–5% that make it past the resume screeners.
Maybe we don't need more engineers, but from what I've seen, we definitely need more good engineers.
--Bruce
PS—I've never seen any evidence that the companies I've worked for preferentially hire H1B employees. It's a lot of red tape for the company, and they get paid the same as US-born employees. The fact is that they're often simply more qualified than Americans.
I don't think it's so much about embarrassment; it's more about providing a stable target for other developers. Google doesn't want to have to deal with people building software on top of a foundation that they're just going to rip out and replace with a better one in the next version. They don't want to support APIs that aren't solid yet.
--Bruce (who is not speaking for himself, not for Google)
How about measuring that in actual computer usage? X MHz on Y cores per Z nodes over A hours? Or at least say it would have taken one X MHz processor 35 years to compute it.
Um, they did. From the article:
Sounds like bad news for these guys. :)
Yes, today's placebo is almost twice as powerful as those used as little as 5 years ago.
You joke, but there's actually quite a bit of truth to that statement.
Heck with the way things are now, the Auto Pilot can nearly land a plane by itself.
Actually, autopilot can land a plane without any human help, and in some cases it's even required to. I was talking to a pilot for United (friend's uncle) a couple years ago, and he said that in high winds or poor visibility, airline regulations prevent the pilot from landing the plan manually. The pilot is required to allow the autopilot to land the plane. Pretty crazy stuff.
--Bruce
Not really.
--Bruce
An interesting, relatively unknown fact that I picked up while working on telephony systems a while back: carriers get paid (by other carriers) for incoming calls.
Not only do you pay more to your carrier to listen to the inane voicemail prompt (since you might use more minutes), but your carrier also pays more to your friend's carrier. For example, if I'm an AT&T customer and I call a Verizon customer to leave a voicemail, AT&T has to pay Verizon for every second that I'm on the phone. This (perverse) incentive makes more sense than charging people for more minutes, since often the company charging for minutes (AT&T in this case) is not the company that controls the recorded message (Verizon).
--Bruce
This sounds a lot like the 40-year-old Carterfone decision, where AT&T argued that allowing people to connect third-party devices to their network could disrupt or degrade service. I'm pretty sure that modems and Panasonic phones didn't ruin the telephone system, and I have a feeling that jailbroken iPhones wouldn't be the end of the world, either.
--Bruce
It's not a "problem" that can be "worked on". It's the character of the author. As any decent psychologist will tell you that character is inborn and cannot be changed or "worked on".
That's a pretty dismal view of human nature. I, on the other hand, believe people can change.
--Bruce
Digital content has no intrinsic cost, so it's not much of a subsidy on their behalf.
I don't the think the publishers are just going to give their content to Amazon for free, even if Amazon gives it away free.
--Bruce
It sounds to me like the EFF is overreacting again. When the new iPod shuffle came out, the they sounded the alarm about headphones requiring "DRM authentication". They later retracted their statement. Turns out that the "DRM" was just an electrical signaling protocol that allowed the headphones to send button press events to the iPod. You knowâ"the kind that any EE would design. The headphones and the iPod have to communicate somehow.
It sounds to me like this is more of the same. From the article:
A lack of information, or the need for sophisticated test equipment does not automatically imply DRM.
I'm all for making information on car diagnostic protocols and diagnostic code semantics more freely available, but let's call it what it really is.
--Bruce
If by "exponential", you mean "large linear", then maybe.
--Bruce
Georgia Tech has a campus in France that teaches (surprise!) technical courses in English.
--Bruce
Am I the only one who read that as "RIAA to Stop Persecuting Individual File Sharers"?
Why does every manager not screen all applicants? Takes 5 minutes.
Because in many places it's illegal to make hiring decisions based on anything but whether the person is qualified for the job. If they go snooping around on the internet, they might discover your marital status, national origin, sexual orientation, or any of the other categories that are legally protected in many jurisdictions. They may or may not make hiring decisions based on that information, but you can imagine how that would look in court.
I worked for one employer (a big computer company you've heard of) that had a policy that explicitly forbid people making hiring decisions from doing an internet search on applicants, for this very reason.
--Bruce
It's true that swarm theory isn't being applied directly by the drivers--after all, they're getting instructions from corporate, as you said. However, I think it's unfair to say that swarm theory isn't being applied at all; it's just being used to compute a solution beforehand rather than on the fly. From the article's description, it sounds like the software creates billions of autonomous agents that all follow simple rules and arrive at a solution without any central planning or coordination. That sounds like swarm theory to me.
The Garmin GPSMAP 60CSx is built to be used in the sun, and its LCD is easily visible in bright light--the brighter, the better, in fact.
--Bruce
I don't know about you, but I have to reboot my Mac every few of weeks for a security update, or even worse, a QuickTime(!) update. I reboot it about as much as I do my Win2k3 machine at work.
He probably learned it from Strong Bad.
Actually, hidden frame requests are still used (e.g. in Gmail), and for good reason: they don't break the back button. XMLHttpRequest communication has no notion of page transitions, and so the back button doesn't go back to the previous state; instead, it usually exits an AJAX app completely.
In contrast, IFRAME navigation does implement page transitions, which can make web apps more user friendly since the back button does what people expect.
Only at Google would a license agreement include the right to create algorithms based on your content. :)
--Bruce
As a bit of a side note, you can write your resume/CV in XML using the XML Resume Project and then easily generate PDF, HTML, or plain text from a single source document. You can even tag elements with keywords and then automatically generate targeted resumes for different audiences.
(Note: I'm a developer on the project.)
--Bruce
RTG = Radioisotopic Thermoelectric Generator
See this site for more info.
--Bruce