Are you sure? I was under the impression that it was unsigned 32 bits. 4294967296 / 365 / 86400 = 136
2147483648 / 365 / 86400 = 68
(yeah, leap years conveniently ignored)
Naively, though, there seems to be limited reason to store integers that are on the order of the square of the number of electrons in the observable universe. This means that we should stop at 256, if not earlier. There are pure physical reasons for why will never use up that address space for any kind of real memory. With limited use for the numbers, and limited use for the addresses, increasing the width seems quite strange. The bus width is a different thing, of course, and you can have a maximum word for SIMD operations, but by that logic all Pentium Ms are 128-bit chips already.
There might still be enough file formats and other stuff where the word width is preserved, or embedded systems that people simply forget about (or where an existing design continues to be used simply because there is nothing wrong with it, except the epoch that happended to be used for the dates). It might seem like just another reason to bash MS, but a few years ago (hm, maybe it was back in 2001 or something) the RTC in my machine went awry. I didn't notice for a few hours/days (not sure), as only the date was affected. The first thing I noticed consciously was that Age of Empires II wouldn't play. When I tried to fix it, I realized that some other seemingly arbitrary things on the Windows machine didn't work. Finally, I noticed that the year was well into the 2040s. By trial and error, I concluded that most of the issues I experienced were related to the 2038 bug. Note that none of these things actualy showed as a date handling error, just sudden termination and crashes.
Actually, it would probably be in 1901, since the overflow only takes place for signed ints. We'll have another bug for unsigned 32-bit ints in 2106 or so.
Yeah, spatial or temporal dithering to simulate 24-bit color really eliminates the point of subpixel text rendering, which in turn makes text look like crap (compared to what it looks like on a proper 24-bit screen).
Well, the lack of an Ethernet port can be justified with the whole idea of a portable machine. The lack of an optical drive can be justified by the size, and the fact that the spindle would suck power anyway. Higher battery time isn't unusual for ultra-portables, for the very reason that they are supposed to be mobile. 5 h is still far from impressive enough to solve the battery needs for many users, and it will certainly be even less so after a year of use (i.e. when you would need to switch the battery even for some continental flights, while it would be hard to justify yanking out the now maybe 60 % good original battery).
I really don't see the conflict between serious science fiction and some degree of artistic license. The point is that the scenario should seem conceivable, "this is one possible future development". It doesn't have to be the one that the author her/himself deems the very most likely. After all, if every story written was part of the author's personal ML estimate optimum of the future at the point of writing, one would either run into a complete inability to write any coherent works (the prediction would continuously change), or lock it at one point in time and then go on writing about the same stuff.
To say that Niven predicted that synthetical organs wouldn't be possible for hundreds of years is like saying that Clarke predicted that a 1:4:9 monolith should have been found on the moon about ten years ago, and that the creats of that monolith should have seeded human intelligence. Despite those aspects, both authors try to give a somewhat "realistic" view of a possible future, but that doesn't change the fact that some aspects are chosen more for the benefit of the story or to explore an interesting issue, rather than for the purpose of prediction.
Or, maybe, like in many kinds of SF, the specifics of technology available is just as well chosen to make the story interesting, even in hard SF. It's supposed to tickle your imagination, not as a technology roadmap. Hence, to paint the picture of a society where this becomes common practice over the course of generations, of course you need to stipulate that the problem is hard, just like some people chose to assume that somewhat-strong AI or FTL drive is far more feasible than it was maybe reasonable to assume.
You realize that the problem with modchips are related to their semilegal status? BTW, you perfectly illustrated why publishers love consoles over PC, the bar for piracy is quite a bit higher.
Now, I wouldn't point to Civ IV to illustrate that using Python won't hurt performance. The reasons for the performance issues (especially before the patches) were complex, but Python didn't exactly help.
I would think that fuzz testing would be worthwhile. Take a valid packet or file and just randomize limited sets. Then, get crash dumps for those cases that fail.
The problem of TSP is not the n^2 aspect, but the 2^n aspect. While searching, you might have done 50 errands of the total 100. You need to know what those 50 are, as the number in itself doesn't tell what parts are left. Consider errands placed in a downtown area and some suburbs. A greedy method may tick off all errands in the center and then start zigzagging to the outside malls, rather than going in some radial/spiral manner that would be more efficient in total.
In real-life problems, you can make some assumptions: "ok, this solution where 50 errands used 2 days of driving is probably not worth extending" and come up with more efficient algorithms, but a brute force 100 city or 133 city problem is simply not possible with no assumptions on graph structure is simply not possible. Note that even assumptions like "all distances are positive" is useful in this context, but it's not a part of the actual distance definition. "A to B to C >= A to C" is another one that is quite useful. I guess that what you did 20 years ago might have been an iterative depth-first search, or something, or you are missing some crucial part of the story. How would you store the data about what paths have already been tried? How many paths did you end up testing?
Dijkstra dosen't approximate the shortest path, it finds it, fair and square, although I understand that you might choose to limit the search depth in practice (but even with a huge set of users, determining only the single distance from A to B should be a matter of seconds at most, even in a weird SQL implementation).
TSP is the linking of several shortest paths, which would get really strange in the social graph, something like: "I want a path of friends where A, B, C, D, E, F are included, but also allowing other nodes, and I don't care where I start or where I end".
You don't need to traverse every road connecting your errands, since you're not selling door to door along the way. TSP is not a filling problem. You only need to traverse every road when doing a brute-force TSP search for an arbitrary graph, since you cannot assume any specific distance constraints (like the triangle inequality or a close-to-planar structure). The problem itself is just about getting a(n optimal) permutation of the nodes. I can't imagine why you were modded insightful.
He's talking about a surge traveling over the Cat5, not the power wire. UPS won't help, since the current enters the motherboard via the network port.
And there are Ethernet surge protectors, but they are very expensive. As in around $250 at the low end from what I've seen. The interesting issue is whether the lightning actually hit the CAT-5 cabling, or if it simply got into one machine and then got into the network from there. I would find the second option more likely, so surge protecting every device could indeed help.
Huh? What do you suppose the mouse experiments did show? They induced a partial injury and could show that rerouting did happen within the short-fiber tissue. We are already able to grow neural tissue, it's far more problematic to imagine recreating all the long nerves.
Considering what Flash tends to do in VNC/RDP/similar settings, and with no info on what kind of VM you used, a severe lack of accelerated video might explain something.
Re:The best tools stay out of the way...
on
Goodbye Cruel Word
·
· Score: 1
It defaults to the one used by the document when it was opened, just like Office 2003. There is no new format with a default extension of.doc in Word 2007. Although you can force a docx to be doc, I think it will cause "security" warnings in some cases, as it could be seen as an attempt to deceive the user about the actual type. (Like.jpg.vbs.)
Re:The best tools stay out of the way...
on
Goodbye Cruel Word
·
· Score: 1
What do you mean, specifically, about Excel? There's nothing (AFAIK) you could do that you can't do now. Quite a few things are simpler. However, I've been astonished by the slow performance when the amount of data grows, and if that's what you're talking about, I can only agree.
Re:The best tools stay out of the way...
on
Goodbye Cruel Word
·
· Score: 4, Informative
Then press Alt in Office 2007 and you are shown what the shortcuts are. For commands that you do often enough that keyboard menomnics are worthwhile, the interface still works fine.
The airship traffic back in the 30s was totally negligible compared to the airlines of today. The single Hindenburg (and that was not the only accident) is enoguh to bring the flights over the two decades or so to a very bad point compared to what we have today... kind of like the Concorde.
Dell has stated "dynamic" before the contrast number in the specs, so I think they reference the temporal contrast between the brightest white and the darkest black ever to be shown by the same monitor. Two pixels shown at the same time will always be in a narrower range.
I think that even rececnt 6-bit panels really tend to negate the pros of subpixel rendering (ClearType). I've mostly seen it on cheapo Dell laptops and the MacBook, but also on some desktop displays. I know that there are different opinions of ClearType out there, but I really, really like it -- on a good display. It's one of the first things to suffer if you use VGA or a self-dithering monitor, as that will disturb the very fine balance between fake resolution and blur...
Slashdot Mirror
Do you include birth in that number, because if not, I think you need to back it up further.
Naively, though, there seems to be limited reason to store integers that are on the order of the square of the number of electrons in the observable universe. This means that we should stop at 256, if not earlier. There are pure physical reasons for why will never use up that address space for any kind of real memory. With limited use for the numbers, and limited use for the addresses, increasing the width seems quite strange. The bus width is a different thing, of course, and you can have a maximum word for SIMD operations, but by that logic all Pentium Ms are 128-bit chips already.
There might still be enough file formats and other stuff where the word width is preserved, or embedded systems that people simply forget about (or where an existing design continues to be used simply because there is nothing wrong with it, except the epoch that happended to be used for the dates). It might seem like just another reason to bash MS, but a few years ago (hm, maybe it was back in 2001 or something) the RTC in my machine went awry. I didn't notice for a few hours/days (not sure), as only the date was affected. The first thing I noticed consciously was that Age of Empires II wouldn't play. When I tried to fix it, I realized that some other seemingly arbitrary things on the Windows machine didn't work. Finally, I noticed that the year was well into the 2040s. By trial and error, I concluded that most of the issues I experienced were related to the 2038 bug. Note that none of these things actualy showed as a date handling error, just sudden termination and crashes.
Actually, it would probably be in 1901, since the overflow only takes place for signed ints. We'll have another bug for unsigned 32-bit ints in 2106 or so.
Yeah, spatial or temporal dithering to simulate 24-bit color really eliminates the point of subpixel text rendering, which in turn makes text look like crap (compared to what it looks like on a proper 24-bit screen).
Well, the lack of an Ethernet port can be justified with the whole idea of a portable machine. The lack of an optical drive can be justified by the size, and the fact that the spindle would suck power anyway. Higher battery time isn't unusual for ultra-portables, for the very reason that they are supposed to be mobile. 5 h is still far from impressive enough to solve the battery needs for many users, and it will certainly be even less so after a year of use (i.e. when you would need to switch the battery even for some continental flights, while it would be hard to justify yanking out the now maybe 60 % good original battery).
To say that Niven predicted that synthetical organs wouldn't be possible for hundreds of years is like saying that Clarke predicted that a 1:4:9 monolith should have been found on the moon about ten years ago, and that the creats of that monolith should have seeded human intelligence. Despite those aspects, both authors try to give a somewhat "realistic" view of a possible future, but that doesn't change the fact that some aspects are chosen more for the benefit of the story or to explore an interesting issue, rather than for the purpose of prediction.
Or, maybe, like in many kinds of SF, the specifics of technology available is just as well chosen to make the story interesting, even in hard SF. It's supposed to tickle your imagination, not as a technology roadmap. Hence, to paint the picture of a society where this becomes common practice over the course of generations, of course you need to stipulate that the problem is hard, just like some people chose to assume that somewhat-strong AI or FTL drive is far more feasible than it was maybe reasonable to assume.
You realize that the problem with modchips are related to their semilegal status? BTW, you perfectly illustrated why publishers love consoles over PC, the bar for piracy is quite a bit higher.
Now, I wouldn't point to Civ IV to illustrate that using Python won't hurt performance. The reasons for the performance issues (especially before the patches) were complex, but Python didn't exactly help.
I would think that fuzz testing would be worthwhile. Take a valid packet or file and just randomize limited sets. Then, get crash dumps for those cases that fail.
In real-life problems, you can make some assumptions: "ok, this solution where 50 errands used 2 days of driving is probably not worth extending" and come up with more efficient algorithms, but a brute force 100 city or 133 city problem is simply not possible with no assumptions on graph structure is simply not possible. Note that even assumptions like "all distances are positive" is useful in this context, but it's not a part of the actual distance definition. "A to B to C >= A to C" is another one that is quite useful. I guess that what you did 20 years ago might have been an iterative depth-first search, or something, or you are missing some crucial part of the story. How would you store the data about what paths have already been tried? How many paths did you end up testing?
TSP is the linking of several shortest paths, which would get really strange in the social graph, something like: "I want a path of friends where A, B, C, D, E, F are included, but also allowing other nodes, and I don't care where I start or where I end".
Huh? What do you suppose the mouse experiments did show? They induced a partial injury and could show that rerouting did happen within the short-fiber tissue. We are already able to grow neural tissue, it's far more problematic to imagine recreating all the long nerves.
Well, while not as good as proper HD disc, VC-1 or MPEG-4 or anything at 4.5 GB will give far greater quality than DVD MPEG-2.
Considering what Flash tends to do in VNC/RDP/similar settings, and with no info on what kind of VM you used, a severe lack of accelerated video might explain something.
It defaults to the one used by the document when it was opened, just like Office 2003. There is no new format with a default extension of .doc in Word 2007. Although you can force a docx to be doc, I think it will cause "security" warnings in some cases, as it could be seen as an attempt to deceive the user about the actual type. (Like .jpg.vbs.)
What do you mean, specifically, about Excel? There's nothing (AFAIK) you could do that you can't do now. Quite a few things are simpler. However, I've been astonished by the slow performance when the amount of data grows, and if that's what you're talking about, I can only agree.
Then press Alt in Office 2007 and you are shown what the shortcuts are. For commands that you do often enough that keyboard menomnics are worthwhile, the interface still works fine.
The airship traffic back in the 30s was totally negligible compared to the airlines of today. The single Hindenburg (and that was not the only accident) is enoguh to bring the flights over the two decades or so to a very bad point compared to what we have today... kind of like the Concorde.
Dell has stated "dynamic" before the contrast number in the specs, so I think they reference the temporal contrast between the brightest white and the darkest black ever to be shown by the same monitor. Two pixels shown at the same time will always be in a narrower range.
I think that even rececnt 6-bit panels really tend to negate the pros of subpixel rendering (ClearType). I've mostly seen it on cheapo Dell laptops and the MacBook, but also on some desktop displays. I know that there are different opinions of ClearType out there, but I really, really like it -- on a good display. It's one of the first things to suffer if you use VGA or a self-dithering monitor, as that will disturb the very fine balance between fake resolution and blur...