"Power management isn't magic, though - there is no great secret about shutting down hardware that isn't being used."
Powering down is the easy part, sometimes the power company does that for me. Bringing it back up is the hard part. Having an excellent software architecture for power savings will help, but not as much as you would hope.
SoCs are incredibly black-box when you power devices and/or the CPU down. You have nothing to help you -- especially logs -- until you bring it back up. Often, even JTAG cannot attach depending on how deep it went. There are fewer signals leaving the SoC, so there's less to probe. You'll find yourself desperately grasping for any sign of life. If you guess wrong too often, you might even fry portions of the chip.
If they provide no documentation, it's even harder. Bringing it back up is all-or-nothing and you're at the mercy of the configuration and register sequences you did. If you do it wrong, you might introduce the internal bus to APIC #2: Destroyer of Worlds.
Sorry to focus so much on a bummer. If the Atom really is %25 efficient as ARM while running, these power cycles are important to compete with Windows over battery life.
Evolution isn't irrefutable. I can't stand hearing ID advocates saying ID is science, but "irrefutable" is something scientists can't claim. It's sort of like claiming your program has no bugs.
Scientists only create models based on what they can observe, with a black-box mentality, reverse-engineering nature. To claim "irrefutable" requires you to know everything about how it works.
Without asking some supernatural being, we never can...
I have a younger sister in high school, and one of the classes she's taking is called "BCIS". I don't know what that stands for, but I do know what they teach: Microsoft Applications.
That isn't an exaggeration. They teach students (at a public school, no less) how to use Word, Excel, Powerpoint, etc. And no, they don't teach stuff that can be generalized to other office suites, such as OpenOffice. The course material describes how to navigate the menu structure of the MS Office suite, and specifically how to use the features in MS Office.
My typing class I took there also taught students how to do format and layout with MS Word in particular. Plus, that typing class was a pre-requisite for almost all other technology classes.
This, coupled with the fact that unfamiliar applications intimidate many people, is probably contributing to the perception that linux is "out of touch" with users. I don't think it's a simplicity issue per se; it's closer to "does this look enough like Windows et alii".
Why stop at violating layering? We should reach across the subsystems and modules! We could have every part of the system involved in performing every system task! Once no one understands how our system works anymore, we could mark it as proprietary and have our own monopoly on GNU+linux maintenance!
Better yet, keep GoogleEarth up, but change some things up..."secret bunker" is actually a booby trap..."headquarters" is just a particularly large dune...
Wrong information is worse than no information. Once they learn they can't rely on GoogleEarth, they'll stop using it. They feed intelligence agencies wrong information all the time, so it's time to return the favor.
First, I doubt it is possible to be purely unbiased (in an theoretical sense) in web publishing. I mean, if you are going to pay for net-hosting, etc., you have some stake in the fact that the information is viewed by others. It could take the form of desiring an audience to be aware of the issue at all, the desire for them to have the correct information, or perhaps the desire to have your information. (forgive me, bloggers!)
That being said, a major factor in the inception of blogging was that it was fairly cheap and easy to do. This lowered the "stakes", and helped against bias. It also gave viewers a much-needed reliable source of what a product is actually like. Since then, many probably still view blogs as unbiased and independent. That's where the wannabe-viral PR comes in...
Paid testimonials from celebrities who've probably never actually used the product is one thing; thinking that an unbiased person loves their product so much they make a blog about how great it is is another.
Given how cheap this sort of marketting probably is, it'll probably never go away (like reality TV). However, the cultural perception of blogging may adversely change as the astro-turf outnumbers the real grass. It'll lose its effectiveness, and they'll probably find something else. The blogging community, however, may lose their perceived credibility. ergo ipso facto, we need AstroTurf-icide!
I don't know why anyone would blindly download something without researching it, but I wouldn't put it past him. I always search for its name to look up its reputation, and to dig up the dirt [if any] on it. More often, I go off of my friends' tips for good apps.
Plus, I think that paid software actually has the potential to be more dangerous. You could either download an app and find out it's malware...or you could *pay* for an app and find out it's malware! At least the free versions let you see if it's good before you give them your credit card info. Other than that, they are capable of the same things.
A price isn't the defining factor. You hardly need any money to collect payment. There's probably a ton of web-apps that do it available. Paypal would work too, but it leaves an incriminating paper trail. Name-recognition and/or a good reputation are what you really need.
If anything kills the security apps, it'd be a real-live secure Windows Vista. Until then, I cross my fingers for it, and brace for the imminent blunt impact of reality.
Since you're running multiple nodes, you should try to decrease the inter-node dependencies as much as possible. You don't want an error in one cascading to the others.
You will also want to error check as much as possible. Don't let functions/methods operate on corrupt arguments! If a function gets some bad values, don't accept it and try to get it to error gracefully. The more assumptions you make about the input to a function, the more potential problems you can have. For instance, always check the bounds of any index to arrays or STL vectors.
You will also want to be able to detect errors in other nodes. You might be able to make the nodes "back up" their progress in case one fails, so it doesn't have to start from the beginning.
Slashdot Mirror
"Power management isn't magic, though - there is no great secret about shutting down hardware that isn't being used."
Powering down is the easy part, sometimes the power company does that for me. Bringing it back up is the hard part. Having an excellent software architecture for power savings will help, but not as much as you would hope.
SoCs are incredibly black-box when you power devices and/or the CPU down. You have nothing to help you -- especially logs -- until you bring it back up. Often, even JTAG cannot attach depending on how deep it went. There are fewer signals leaving the SoC, so there's less to probe. You'll find yourself desperately grasping for any sign of life. If you guess wrong too often, you might even fry portions of the chip.
If they provide no documentation, it's even harder. Bringing it back up is all-or-nothing and you're at the mercy of the configuration and register sequences you did. If you do it wrong, you might introduce the internal bus to APIC #2: Destroyer of Worlds.
Sorry to focus so much on a bummer. If the Atom really is %25 efficient as ARM while running, these power cycles are important to compete with Windows over battery life.
Evolution isn't irrefutable. I can't stand hearing ID advocates saying ID is science, but "irrefutable" is something scientists can't claim. It's sort of like claiming your program has no bugs. Scientists only create models based on what they can observe, with a black-box mentality, reverse-engineering nature. To claim "irrefutable" requires you to know everything about how it works. Without asking some supernatural being, we never can...
I have a younger sister in high school, and one of the classes she's taking is called "BCIS". I don't know what that stands for, but I do know what they teach: Microsoft Applications.
That isn't an exaggeration. They teach students (at a public school, no less) how to use Word, Excel, Powerpoint, etc. And no, they don't teach stuff that can be generalized to other office suites, such as OpenOffice. The course material describes how to navigate the menu structure of the MS Office suite, and specifically how to use the features in MS Office.
My typing class I took there also taught students how to do format and layout with MS Word in particular. Plus, that typing class was a pre-requisite for almost all other technology classes.
This, coupled with the fact that unfamiliar applications intimidate many people, is probably contributing to the perception that linux is "out of touch" with users. I don't think it's a simplicity issue per se; it's closer to "does this look enough like Windows et alii".
Why stop at violating layering? We should reach across the subsystems and modules! We could have every part of the system involved in performing every system task! Once no one understands how our system works anymore, we could mark it as proprietary and have our own monopoly on GNU+linux maintenance!
Better yet, keep GoogleEarth up, but change some things up..."secret bunker" is actually a booby trap..."headquarters" is just a particularly large dune... Wrong information is worse than no information. Once they learn they can't rely on GoogleEarth, they'll stop using it. They feed intelligence agencies wrong information all the time, so it's time to return the favor.
First, I doubt it is possible to be purely unbiased (in an theoretical sense) in web publishing. I mean, if you are going to pay for net-hosting, etc., you have some stake in the fact that the information is viewed by others. It could take the form of desiring an audience to be aware of the issue at all, the desire for them to have the correct information, or perhaps the desire to have your information. (forgive me, bloggers!)
That being said, a major factor in the inception of blogging was that it was fairly cheap and easy to do. This lowered the "stakes", and helped against bias. It also gave viewers a much-needed reliable source of what a product is actually like. Since then, many probably still view blogs as unbiased and independent. That's where the wannabe-viral PR comes in...
Paid testimonials from celebrities who've probably never actually used the product is one thing; thinking that an unbiased person loves their product so much they make a blog about how great it is is another.
Given how cheap this sort of marketting probably is, it'll probably never go away (like reality TV). However, the cultural perception of blogging may adversely change as the astro-turf outnumbers the real grass. It'll lose its effectiveness, and they'll probably find something else. The blogging community, however, may lose their perceived credibility. ergo ipso facto, we need AstroTurf-icide!
I don't know why anyone would blindly download something without researching it, but I wouldn't put it past him. I always search for its name to look up its reputation, and to dig up the dirt [if any] on it. More often, I go off of my friends' tips for good apps.
Plus, I think that paid software actually has the potential to be more dangerous. You could either download an app and find out it's malware...or you could *pay* for an app and find out it's malware! At least the free versions let you see if it's good before you give them your credit card info. Other than that, they are capable of the same things.
A price isn't the defining factor. You hardly need any money to collect payment. There's probably a ton of web-apps that do it available. Paypal would work too, but it leaves an incriminating paper trail. Name-recognition and/or a good reputation are what you really need.
If anything kills the security apps, it'd be a real-live secure Windows Vista. Until then, I cross my fingers for it, and brace for the imminent blunt impact of reality.
Since you're running multiple nodes, you should try to decrease the inter-node dependencies as much as possible. You don't want an error in one cascading to the others.
You will also want to error check as much as possible. Don't let functions/methods operate on corrupt arguments! If a function gets some bad values, don't accept it and try to get it to error gracefully. The more assumptions you make about the input to a function, the more potential problems you can have. For instance, always check the bounds of any index to arrays or STL vectors.
You will also want to be able to detect errors in other nodes. You might be able to make the nodes "back up" their progress in case one fails, so it doesn't have to start from the beginning.