Just to give a sense of scale to this: my wife was backing out from a covered parking lot. While going backwards at maybe 2 mph and turning, she hit the right front of the car on the steel roof support post (a 4" diameter steel pipe). The pipe didn't even get the paint scratched, but the Volvo S80 suffered $3k worth of damage. Over the 8 years of our car ownerships, we've paid about twice that amount in insurance premiums.
In Ohio, where insurance is mandatory, you don't have to pay a dime to insurance companies. In lieu of standard insurance, you can have:
1. a surety bond of $30,000 issued by an authorized surety company; 2. a BMV bond secured by real estate having equity of at least $60,000; 3. a BMV Certificate for money or government bonds in the amount of $30,000 on deposit with the Ohio Treasurer.
Option #3 actually earns you interest!
If you have $60k of equity in real estate (typically your home), option #2 will cost you the least amount if you gamble on not having an accident where you'd injure someone/damage someone's property.
So, no, you don't have to pay an insurance company. OTOH, maybe you should get insurance from a mutual insurance company where any profit is shared by the policyholders.
Are you sure "finger on this spot, phone doesn't work"? I could certainly make it not work when signal would be so weak on 3G that it'd be dropping out. But I couldn't find a single situation where 4G, with finger on the spot, would work perceptibly worse than 3G. Maybe my finger wasn't sweaty enough, I don't know. But I sure played with both for half a day, in different locations -- outdoors, indoors, in residential buildings, in commercial buildings, etc. YMMV of course...
Disclaimer: I'm an Apple iMac and Macbook Pro user, but that's out of practicality -- my previous desktop OS was Fedora Core. I even own AAPL stock.
I don't have any iPhones, heck, I only have a prepaid Tracfone for emergency use. But I did play with iPhone 3G and iPhone 4 in the same location, and surely enough I could find many places where iPhone 3 had marginal reception or it'd drop calls or stall data. But iPhone 4 did work just fine in the same spot -- and surely enough, touching the magical spot on the metal band did cause it to drop out. By my very rough guesstimate, iPhone 4 receiver system drops back to iPhone 3G's sensitivity when you touch it like they wish you weren't.
How would that be considered a very bad thing -- I just don't know. The worst case is you're no worse off, reception-wise, than with iPhone 3G. That's from my limited experience, from tests done specifically to see WTF is going is on, since I just couldn't wrap my head around the fact that Apple's engineering would not test the darn thing. Well, my guess is that they did test it, and decided that this behavior is OK. It may be bad politically, and bad for the stockholders (yes, I'm one), but it's a passable engineering tradeoff methinks. Antenna engineering on mobile devices is pretty damn hard, or so I'm told.
That's a fairly ridiculous price, but given that there are no other products that can stay aloft that long, it'd probably be acceptable. I wonder if there are any jets out there that could take enough fuel in extra internal tanks to stay aloft for 4 days... Wikipedia doesn't have enough details to tell, say, how far a 787-8 would fly if all wet weight were to be fuel.
Decently designed XP applications store data in user's profile. This can be very easily backed up and restored. So I don't know how would you "lose quite an amount of data" simply by doing an XP reinstall.
Of course as an administrative user you can store your stuff all over the place, but if you do -- so what, presumably you remember where you put stuff. Simple suggestions: shell $50 for a new HD, plug the existing one as secondary or via a USB enclosure, and reinstall. Then move your data. If anything fails, you swap the HD and are back to where you left.
This seems like system admin 101, and I really can't quite see how SP3 would "bring your machine to a near halt". Halt HOW? You must have something messed up pretty badly -- malware?
As far as copyright and patent laws go, in the U.S., they are there only and specifically to implement the U.S. Constitution, Art. I, Sec. 8, Cls. 8
To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.
This is the sole raison d'être for copyright and patent laws on the U.S. books. If those laws poorly support their mission as outlined in the Constitution, they should be repealed and replaced by laws that do the job right. It is my view, that the copyright and patent laws on the books do little to nothing to "promote the Progress of Science and useful Arts". Copyright protection lasts too long, and patent laws do little to promote progress as written.
The blatant opportunist makes a case for U.S. patents to make economic sense if sales are to exceed 12M USD. I agree with him, and he further shows that for most small businesses patents make no sense whatsoever. This is reason enough, for me, to argue that patent laws as written don't do the job they are supposed to do. Either we claim small businesses do nothing to further the "Progress of Science and useful Arts", and thus they need no protection by limited-duration exclusivity, or the laws are missing their Constitutional Target.
The truth is that code was reused (if not copied, exactly, in the same way you don't submit a copied essay which you've taken from a classmate) from a UNIX derivative, which is now (somewhat disputably) owned by SCO.
Um, citation needed? Nothing in TFA suggests to me any code reuse.
Don't mistake writing for the same API as "code reuse". You seem to have no clue that APIs are not a subject of copyright protection -- either you're badly underinformed, or you're a troll.
If you write for given API, where variable/parameter/function/field names are either part of the API or are a de-facto standard, there is no helping creating structures that look the same, creating same function declarations that look the same, etc.
Again, let there be no mistake: in my opinion, there is no code "reuse" or "plagiarism" in any of the samples TFA refers to. So most of your rant is a waste of steam. The fact that Linux-is-a-UNIX makes certain things look similar, but that's simply because to be a UNIX, you must implement a bunch of APIs, and by necessity functions and structs will have same signatures, and there will be a bunch of C macros that are same as well, perhaps to the letter. When it comes to C macros, sometimes there is only one way to write them correctly, save for whitespace, and function dictates form, to the letter. This does not imply any reuse of code, merely implementing same thing.
I don't really see how the non-buildable source can be generally useful. Certainly, some things can be examined on a printout. Perhaps most of interesting things.
But there are still some pieces of code where it's hard to reason about their execution paths without seeing them in action. Thus you really need to build them, hack them, run under debugger and see how they behave in action, and how they react to your changes.
I was not referring to golden rice at all -- I merely pointed out that many people are vit. D deficient in first-world countries. As for skin exposure to sunlight -- that's fine and dandy under two conditions:
1. You should be young, not old -- folks over 50 should not depend solely on their own vit D production.
2. You should not be deficient to begin with. Skin's production is naturally limited at the source, in spite of systemic deficiencies if any to prevent possible toxicity. So if you are deficient the sunlight won't do you much good.
Gibson (a 50+ guy IIRC) had himself tested for vit. D levels while sunbathing a whole lot, and the UV exposure did nothing to recover his vit. D deficiency.
Rickets is the worst you can get from vit. D deficiency, the problem is that even smaller deficiency is bound to have widespread effects -- just look in how many places vit. D is involved! Our bodies have a way of compensating it seems, but if the problem is so widespread you're bound to run out of range of compensation somewhere, or even in a few somewheres...
Going for a fixed distance is easy, other things are not. For the fixed distance, all you need is a powerful enough infrared laser (can be also visible or UV), and a vibrating focusing system that sweeps the focus along a linear path of some length. Along that path, you get a column of ionized air. That due to dielectric breakdown induced by high enough electromagnetic field strength from the focused laser.
The biggest issue is really cooling the laser and possibly optics, one could fit a rechargeable power source, the laser and the focuser into a largish lightsaber-style case using today's tech. But you will need coolant and gas hoses going from that to your backpack, and the backpack will have a CO2 bottle, a motorcycle-sized radiator and a fan. Not very sexy I'd say.
Human synthesis of Vitamin D precursors is highly insufficient unless you want to sunbathe enough to get a serious risk of skin cancer. You can read all about it at GRC's website.
For a typical office dweller, you should shelve photosynthesis of Vit. D precursors as an impractical-to-exploit curiosity. Make sure you get Vit. D in your diet, or take supplements. And get tested for serum levels of 25-hydroxyvitamin D.
Here's the fine citation from Holick's NEJM article
[...] it has been estimated that 1 billion people worldwide have vitamin D deficiency or insufficiency. According to several studies, 40 to 100% of U.S. and European elderly men and women still living in the community (not in nursing homes) are deficient in vitamin D. More than 50% of postmenopausal women taking medication for osteoporosis had suboptimal levels of 25-hydroxyvitamin D — below 30 ng per milliliter (75 nmol per liter).
Children and young adults are also potentially at high risk for vitamin D deficiency. For example, 52% of Hispanic and black adolescents in a study in Boston and 48% of white preadolescent girls in a study in Maine had 25-hydroxyvitamin D levels below 20 ng per milliliter. In other studies, at the end of the winter, 42% of 15- to 49-year-old black girls and women throughout the United States had 25-hydroxyvitamin D levels below 20 ng per milliliter, and 32% of healthy students, physicians, and residents at a Boston hospital were found to be vitamin D–deficient, despite drinking a glass of milk and taking a multivitamin daily and eating salmon at least once a week.
In all likelihood, you are vitamin D deficient, your health may be suffering for it. This includes mental health!
Methinks that stupid/useless apps are not an issue. There's a lot of crappy books in every bookstore, and I have no problem with that. But the issue is that people's iTunes credentials got stolen, and I don't think it was Apple's fault unless the exploits were running on OS X...
Train carriages differ between countries. Pretty standard CSX (north american) rail boxcars can be bigger than studio-sized apartments. Think 60'-9" long, 9'-6" wide, 13' tall -- that's 577 square feet, or 54 m^2. Presumably you would install insulation on the outside -- at least in Europe there are plenty of exterior-grade acrylic-finished styrofoam systems, that are easy to install.
Architects and contractors are licensed because they are taking on liability for the specifications and buildings they produce.
Architects are licensed for that reason, yes. Contractors -- not so. Contractors are licensed not because they are "taking on" liability. Quite to the contrary -- the problem is that some would not feel liable even if they were. They are licensed so that they can be held financially liable for money owed to subcontractors, suppliers, and customers. In many US localities, all it takes to become a licensed general contractor is to post a bond and fill a form. The bond is a financial instrument that will have the bond issuer pay your obligations, up to a certain limit. The issuer can then try to get the money back from you via usual processes for that.
I'm out of mod points, but the above is very insightful. I can relate seeing, on many occasions, where a dead simple UI, no more complex than say that of an ATM, becomes utterly confounding just because it's being presented with a PC in sight. In one case: as long as the PC was hidden, and the UI was accessed via a touchscreen --- everything was fine. As soon as mouse, keyboard and the PC case became visible, people would say that "something broke" and that I should bring it back to the "way it was before". This was a big eye opener when it comes to usability: users are not rational. Not at all.
Re 2: those printers, by the virtue of being "professional", usually have documentation available that lets you format data for their consumption. Those are pretty much "gimme a bitmap" devices -- there is no real driver development per se. Once you get your image bitmap rendered, you format it for the printer and ship it out. Usually all it takes is some colorspace conversion and dithering, prepending a header, and off you go.
Re 3: You don't need/want any off the shelf "tools" to administer such devices. Everything should be built-in within the administrator's console view. There is a very limited set of things you will do on those machines anyway, and perhaps except for time-limited storage of copies of images, and a maintenance log, the main image should be read-only.
There is no "Windows UI" developers anymore. There are people who use MFC, Forms, WPF, or some 3rd party development frameworks. Many won't care about any winapi calls. So, since the development really uses some Windows-decoupled framework, one may as well use something that is cross-platform.
When it comes to frakking photo kiosks -- yes, Linux is the answer, and the "lifecycle" management can be as simple as a public-key-authenticated pull over ssh from the update server. The development can be done using Qt Embedded -- then you can run it on your dev platform just as well as on the target. I don't know how Windows would ever be somehow a better solution for a photo kiosk system.
I'm just wondering how do the card readers interface to said ATM/photo kiosks? Maybe they are just plain old USB HID devices, mimicking as keyboards? Surely those readers have some backdoors/debug "bits" left over that allow you to have a magstripe that has some special header followed by raw keycodes? Ctrl-Alt-Del won't be a problem, then.
That sorta points fingers to potentially killer transients on the 3.3V supply -- RAM and CPUs don't see 3.3V methinks. The regulators that feed the CPU/RAM core voltages have very good transient response, with spare capacity.
I would like to know, though, if there are any tantalum capacitors on the motherboards. Maybe the issue is as simple as a shorted tantalum...
I meant to say that out of all the logic supply voltages on a modern motherboard, only 3.3V comes directly from the ATX supply. I don't think that anything uses 5V for digital signaling these days -- maybe I'm wrong. But memory, northbridge and CPU will run at voltages lower than 3.3V AFAIK. Maybe external buses (pin drivers) on those run on 3.3V? That I don't know.
Slashdot Mirror
Just to give a sense of scale to this: my wife was backing out from a covered parking lot. While going backwards at maybe 2 mph and turning, she hit the right front of the car on the steel roof support post (a 4" diameter steel pipe). The pipe didn't even get the paint scratched, but the Volvo S80 suffered $3k worth of damage. Over the 8 years of our car ownerships, we've paid about twice that amount in insurance premiums.
In Ohio, where insurance is mandatory, you don't have to pay a dime to insurance companies. In lieu of standard insurance, you can have:
1. a surety bond of $30,000 issued by an authorized surety company;
2. a BMV bond secured by real estate having equity of at least $60,000;
3. a BMV Certificate for money or government bonds in the amount of $30,000 on deposit with the Ohio Treasurer.
Option #3 actually earns you interest!
If you have $60k of equity in real estate (typically your home), option #2 will cost you the least amount if you gamble on not having an accident where you'd injure someone/damage someone's property.
So, no, you don't have to pay an insurance company. OTOH, maybe you should get insurance from a mutual insurance company where any profit is shared by the policyholders.
Are you sure "finger on this spot, phone doesn't work"? I could certainly make it not work when signal would be so weak on 3G that it'd be dropping out. But I couldn't find a single situation where 4G, with finger on the spot, would work perceptibly worse than 3G. Maybe my finger wasn't sweaty enough, I don't know. But I sure played with both for half a day, in different locations -- outdoors, indoors, in residential buildings, in commercial buildings, etc. YMMV of course...
Disclaimer: I'm an Apple iMac and Macbook Pro user, but that's out of practicality -- my previous desktop OS was Fedora Core. I even own AAPL stock.
I don't have any iPhones, heck, I only have a prepaid Tracfone for emergency use. But I did play with iPhone 3G and iPhone 4 in the same location, and surely enough I could find many places where iPhone 3 had marginal reception or it'd drop calls or stall data. But iPhone 4 did work just fine in the same spot -- and surely enough, touching the magical spot on the metal band did cause it to drop out. By my very rough guesstimate, iPhone 4 receiver system drops back to iPhone 3G's sensitivity when you touch it like they wish you weren't.
How would that be considered a very bad thing -- I just don't know. The worst case is you're no worse off, reception-wise, than with iPhone 3G. That's from my limited experience, from tests done specifically to see WTF is going is on, since I just couldn't wrap my head around the fact that Apple's engineering would not test the darn thing. Well, my guess is that they did test it, and decided that this behavior is OK. It may be bad politically, and bad for the stockholders (yes, I'm one), but it's a passable engineering tradeoff methinks. Antenna engineering on mobile devices is pretty damn hard, or so I'm told.
That's a fairly ridiculous price, but given that there are no other products that can stay aloft that long, it'd probably be acceptable. I wonder if there are any jets out there that could take enough fuel in extra internal tanks to stay aloft for 4 days... Wikipedia doesn't have enough details to tell, say, how far a 787-8 would fly if all wet weight were to be fuel.
Decently designed XP applications store data in user's profile. This can be very easily backed up and restored. So I don't know how would you "lose quite an amount of data" simply by doing an XP reinstall.
Of course as an administrative user you can store your stuff all over the place, but if you do -- so what, presumably you remember where you put stuff. Simple suggestions: shell $50 for a new HD, plug the existing one as secondary or via a USB enclosure, and reinstall. Then move your data. If anything fails, you swap the HD and are back to where you left.
This seems like system admin 101, and I really can't quite see how SP3 would "bring your machine to a near halt". Halt HOW? You must have something messed up pretty badly -- malware?
As far as copyright and patent laws go, in the U.S., they are there only and specifically to implement the U.S. Constitution, Art. I, Sec. 8, Cls. 8
To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.
This is the sole raison d'être for copyright and patent laws on the U.S. books. If those laws poorly support their mission as outlined in the Constitution, they should be repealed and replaced by laws that do the job right. It is my view, that the copyright and patent laws on the books do little to nothing to "promote the Progress of Science and useful Arts". Copyright protection lasts too long, and patent laws do little to promote progress as written.
The blatant opportunist makes a case for U.S. patents to make economic sense if sales are to exceed 12M USD. I agree with him, and he further shows that for most small businesses patents make no sense whatsoever. This is reason enough, for me, to argue that patent laws as written don't do the job they are supposed to do. Either we claim small businesses do nothing to further the "Progress of Science and useful Arts", and thus they need no protection by limited-duration exclusivity, or the laws are missing their Constitutional Target.
The truth is that code was reused (if not copied, exactly, in the same way you don't submit a copied essay which you've taken from a classmate) from a UNIX derivative, which is now (somewhat disputably) owned by SCO.
Um, citation needed? Nothing in TFA suggests to me any code reuse.
Don't mistake writing for the same API as "code reuse". You seem to have no clue that APIs are not a subject of copyright protection -- either you're badly underinformed, or you're a troll.
If you write for given API, where variable/parameter/function/field names are either part of the API or are a de-facto standard, there is no helping creating structures that look the same, creating same function declarations that look the same, etc.
Again, let there be no mistake: in my opinion, there is no code "reuse" or "plagiarism" in any of the samples TFA refers to. So most of your rant is a waste of steam. The fact that Linux-is-a-UNIX makes certain things look similar, but that's simply because to be a UNIX, you must implement a bunch of APIs, and by necessity functions and structs will have same signatures, and there will be a bunch of C macros that are same as well, perhaps to the letter. When it comes to C macros, sometimes there is only one way to write them correctly, save for whitespace, and function dictates form, to the letter. This does not imply any reuse of code, merely implementing same thing.
What code of yours was plagiarized, again?
I don't really see how the non-buildable source can be generally useful. Certainly, some things can be examined on a printout. Perhaps most of interesting things.
But there are still some pieces of code where it's hard to reason about their execution paths without seeing them in action. Thus you really need to build them, hack them, run under debugger and see how they behave in action, and how they react to your changes.
Just a satisfied user, I guess.
I was not referring to golden rice at all -- I merely pointed out that many people are vit. D deficient in first-world countries. As for skin exposure to sunlight -- that's fine and dandy under two conditions:
1. You should be young, not old -- folks over 50 should not depend solely on their own vit D production.
2. You should not be deficient to begin with. Skin's production is naturally limited at the source, in spite of systemic deficiencies if any to prevent possible toxicity. So if you are deficient the sunlight won't do you much good.
Gibson (a 50+ guy IIRC) had himself tested for vit. D levels while sunbathing a whole lot, and the UV exposure did nothing to recover his vit. D deficiency.
Rickets is the worst you can get from vit. D deficiency, the problem is that even smaller deficiency is bound to have widespread effects -- just look in how many places vit. D is involved! Our bodies have a way of compensating it seems, but if the problem is so widespread you're bound to run out of range of compensation somewhere, or even in a few somewheres...
Going for a fixed distance is easy, other things are not. For the fixed distance, all you need is a powerful enough infrared laser (can be also visible or UV), and a vibrating focusing system that sweeps the focus along a linear path of some length. Along that path, you get a column of ionized air. That due to dielectric breakdown induced by high enough electromagnetic field strength from the focused laser.
The biggest issue is really cooling the laser and possibly optics, one could fit a rechargeable power source, the laser and the focuser into a largish lightsaber-style case using today's tech. But you will need coolant and gas hoses going from that to your backpack, and the backpack will have a CO2 bottle, a motorcycle-sized radiator and a fan. Not very sexy I'd say.
Something has to make money for George Lucas Educational Foundation, right? /ducks and runs
Human synthesis of Vitamin D precursors is highly insufficient unless you want to sunbathe enough to get a serious risk of skin cancer. You can read all about it at GRC's website.
For a typical office dweller, you should shelve photosynthesis of Vit. D precursors as an impractical-to-exploit curiosity. Make sure you get Vit. D in your diet, or take supplements. And get tested for serum levels of 25-hydroxyvitamin D.
Here's the fine citation from Holick's NEJM article
[...] it has been estimated that 1 billion people worldwide have vitamin D deficiency or insufficiency. According to several studies, 40 to 100% of U.S. and European elderly men and women still living in the community (not in nursing homes) are deficient in vitamin D. More than 50% of postmenopausal women taking medication for osteoporosis had suboptimal levels of 25-hydroxyvitamin D — below 30 ng per milliliter (75 nmol per liter).
Children and young adults are also potentially at high risk for vitamin D deficiency. For example, 52% of Hispanic and black adolescents in a study in Boston and 48% of white preadolescent girls in a study in Maine had 25-hydroxyvitamin D levels below 20 ng per milliliter. In other studies, at the end of the winter, 42% of 15- to 49-year-old black girls and women throughout the United States had 25-hydroxyvitamin D levels below 20 ng per milliliter, and 32% of healthy students, physicians, and residents at a Boston hospital were found to be vitamin D–deficient, despite drinking a glass of milk and taking a multivitamin daily and eating salmon at least once a week.
In all likelihood, you are vitamin D deficient, your health may be suffering for it. This includes mental health!
Methinks that stupid/useless apps are not an issue. There's a lot of crappy books in every bookstore, and I have no problem with that. But the issue is that people's iTunes credentials got stolen, and I don't think it was Apple's fault unless the exploits were running on OS X...
Train carriages differ between countries. Pretty standard CSX (north american) rail boxcars can be bigger than studio-sized apartments. Think 60'-9" long, 9'-6" wide, 13' tall -- that's 577 square feet, or 54 m^2. Presumably you would install insulation on the outside -- at least in Europe there are plenty of exterior-grade acrylic-finished styrofoam systems, that are easy to install.
Architects and contractors are licensed because they are taking on liability for the specifications and buildings they produce.
Architects are licensed for that reason, yes. Contractors -- not so. Contractors are licensed not because they are "taking on" liability. Quite to the contrary -- the problem is that some would not feel liable even if they were. They are licensed so that they can be held financially liable for money owed to subcontractors, suppliers, and customers. In many US localities, all it takes to become a licensed general contractor is to post a bond and fill a form. The bond is a financial instrument that will have the bond issuer pay your obligations, up to a certain limit. The issuer can then try to get the money back from you via usual processes for that.
I'm out of mod points, but the above is very insightful. I can relate seeing, on many occasions, where a dead simple UI, no more complex than say that of an ATM, becomes utterly confounding just because it's being presented with a PC in sight. In one case: as long as the PC was hidden, and the UI was accessed via a touchscreen --- everything was fine. As soon as mouse, keyboard and the PC case became visible, people would say that "something broke" and that I should bring it back to the "way it was before". This was a big eye opener when it comes to usability: users are not rational. Not at all.
Re 2: those printers, by the virtue of being "professional", usually have documentation available that lets you format data for their consumption. Those are pretty much "gimme a bitmap" devices -- there is no real driver development per se. Once you get your image bitmap rendered, you format it for the printer and ship it out. Usually all it takes is some colorspace conversion and dithering, prepending a header, and off you go.
Re 3: You don't need/want any off the shelf "tools" to administer such devices. Everything should be built-in within the administrator's console view. There is a very limited set of things you will do on those machines anyway, and perhaps except for time-limited storage of copies of images, and a maintenance log, the main image should be read-only.
There is no "Windows UI" developers anymore. There are people who use MFC, Forms, WPF, or some 3rd party development frameworks. Many won't care about any winapi calls. So, since the development really uses some Windows-decoupled framework, one may as well use something that is cross-platform.
When it comes to frakking photo kiosks -- yes, Linux is the answer, and the "lifecycle" management can be as simple as a public-key-authenticated pull over ssh from the update server. The development can be done using Qt Embedded -- then you can run it on your dev platform just as well as on the target. I don't know how Windows would ever be somehow a better solution for a photo kiosk system.
I'm just wondering how do the card readers interface to said ATM/photo kiosks? Maybe they are just plain old USB HID devices, mimicking as keyboards? Surely those readers have some backdoors/debug "bits" left over that allow you to have a magstripe that has some special header followed by raw keycodes? Ctrl-Alt-Del won't be a problem, then.
Hey, hey, all those porn sharing sites where people upload gobs of personal stuff have to get their content from somewhere, right?
That sorta points fingers to potentially killer transients on the 3.3V supply -- RAM and CPUs don't see 3.3V methinks. The regulators that feed the CPU/RAM core voltages have very good transient response, with spare capacity.
I would like to know, though, if there are any tantalum capacitors on the motherboards. Maybe the issue is as simple as a shorted tantalum...
I meant to say that out of all the logic supply voltages on a modern motherboard, only 3.3V comes directly from the ATX supply. I don't think that anything uses 5V for digital signaling these days -- maybe I'm wrong. But memory, northbridge and CPU will run at voltages lower than 3.3V AFAIK. Maybe external buses (pin drivers) on those run on 3.3V? That I don't know.