At my local grocery store, they sell three different types of bananas: "standard" (49c/lb), organic (79c/lb) and fair trade (89c/lb). Without some form of labelling, the cashier cannot distinguish between the types.
Of course, the self-checkout lanes make this entire argument almost moot...
You know, I used to think it was solely an OpenOffice.org issue, and while the application certainly could be snappier and less bloated in some areas, I don't think it's nearly as bad as people make it out to be.
As the only Computer Science teacher at my high school, I often inherit lots of older hardware. The other day our local technician gave me three old Celeron 600MHz laptops that he was just going to throw in the trash. I decided to turn them into workstations for my students who did not have easy access to computers at home. As they were slow machines, I set them up with a basic Ubuntu install running IceWM -- it runs quite snappy, and is a bit more user-friendly than say DSL.
I considered installing AbiWord for a word processor, but chose OO.o mainly because we have it installed on the school servers anyway. The result? It takes about 6-7 seconds to load up, and responds quite well. As a point of comparison, the 2GHz Pentium 4 machines running XP and Word in our department office launch Word in approximately 5-6 seconds, so it's pretty much on par.
I know that OO.o is a beast sometimes, but my experiences have been fairly positive.
I know you were going for funny, but in reality Jeopardy never uses one(or two)-word answers.
They often do, but they're typically framed within a narrow, specific category. For example, the category might me "National Drinks" or some such thing. Typical Q/As might be Japan (Sake), Russia (Vodka), and so forth. Jeapoardy! also has a few other categories (anagrams, for example) that frequently use one or two words.
I frequently hear this argument, but it presupposes a truly "random" sick day. I suspect that the dates are indeed biased, with more people being "sick" on a Monday or Friday. I know that would certainly be my choice.
Somewhat related, it's like flipping a coin 999 times and coming up heads each time. What is the probability that it will come up heads again? If it was truly random, the probability would be 50%. More likely, however, the coin is weighted on one side and biased toward heads. Bias is definitely a factor.
Well I thought to myself I better keep others from being ripped off so I looked high and low on how to 'add' to my comment, restract or change my comment, now that I had found out this individual had stolen hundreds of dollars from me. Guess what? THERE ISN'T ANY Ebay purposely makes it so there is no way *I* can tell the world that this user I interacted with stole from me!!!
That was probably the case back when you sold your goods, but IIRC, there's a follow-up feature now where you can comment on your comments.
I like many of the add-on features on more powerful systems, but I pine for a browser that I can run quickly and easily on low-end machines.
Well, you've got two main options:
Use Firefox as it is, sans memory-hogging extensions. This is the default installation.
Seek out a lightweight browser like Dillo and use it for surfing on your low-end machines.
Re:Not necessarily bad in all cases...
on
Open Source Math
·
· Score: 1
But with the pseudocode, you can write your own program in whatever language you like to verify the results.
Of course. There's no real reason to restrict a proof to a particular language. I would expect, however, that if a computer was essential in producing the proof, that the source code that was used would also be released. In the case of the FCT, the program ran for several hundred hours. In the spirit of mathematical disclosure, it would be nice to see someone provide their own particular program that they used, so that peers could inspect the code, rather than have them recreate it themselves. This way they're not reinventing the proverbial wheel, although they could certainly do so if they wished.
I think people are forgetting about printed computer magazines - e.g. Linux Journal, APC, etc. They have a restricted column magazine format, and they often use TinyURLs when publishing links.
Which has always made me wonder why they don't simply provide a link to their own site, from which you may be redirected. For example, SpiffyPC Magazine might be doing a review on the new XYZ 123 motherboard, and configure spiffypc.com/XYZ123 as a referral link.
Actually, given most magazines' enthusiasm for advertising, don't most magazines include links back to their own sites, where they can include as many other links and feastures as they wish?
Re:Not necessarily bad in all cases...
on
Open Source Math
·
· Score: 2, Insightful
In other cases, like the proof of the four color theorem, it seems like the source code is important to see, but not essential. Pseudocode should suffice. Providing pseudocode is akin to saying things like "Simplifying expression (1) yields..."; we don't have to provide EVERY step, but with pseudocode you have enough to determine whether the algorithm itself will work. Checking the source code beyond that is akin to checking someone's algebra.
Perhaps I'm being too pessimistic, but shouldn't the source code have to be provided alongside the pseudocode? If the pseudocode is 100% spot-on, then there would really be no need for the computer-assisted proof in the first place --- you will have provided a proof in the form of verifiable instructions. But the FCT was proved by some amount of brute-force, IIRC. Who is to say that the coder who translated from pseudocode to source code didn't mess something up? I mean, if my pseudocode reads
INCREMENT current value by ONE ... OUTPUT result of long computations
and my source code is entered as
value += value++; ... printf("%d",result);
then even if the pseudocode is verified, the program may still be producing an erroneous result. In other words, you're assuming that IF the pseudocode is correct THEN the program itself is also correct, which may not be the case.
Vista should be testing its incoming audio to detect whether it matches any outgoing audio that Vista is playing.
I imagine it's not quite so straightforward. You'd need to take into account room acoustics, hardware effects, generic ambient noises, or even other interfering sounds in the same room that could all interfere with a comparison of outgoing sound to incoming sound. It's very rare that you'd ever have a time where your outgoing sound file exactly matches one that is sensed coming from the speakers.
I found someone charging $60 to ship a USB Thumb drive via the slowest cheapest USPS method possible.
In a case like this, you have to look at what the item's price is. A number of sellers jack up the shipping, but drop down the item price, so that they don't have to fork over as much "royalty" fee to eBay when it sells. For instance, your $60 thumb drive might end up selling for $2 + $60 S&H = $62. If they set shipping at $20, the item might sell for a higher price, say $42. In this case, you'd pay the same amount, but the seller would be charged less by eBay, because shipping does not count in the royalty calculations.
I always get a quote for shipping before bidding, as I am sure you do as well. I factor this into my decision. If I'm willing to spend $50 on a item, and the shipping is $35, then I can max out my bid at $15. To me, it's all the same. Either way, I'll only spend up to $50.
Your other points? I've been lucky enough to avoid them. One day I am sure I'll be bitten.
It's open source. We'll just change what's inside the gifts from Greeks so it won't be dangerous.
Of course, that's only if you *notice* that it looks dangerous. Even Open Source can contain some subtle things that appear innocuous. See the classic "Reflections on Trusting Trust" -- great little read.
Try representing 1/3 in any finite number of digits...
...
The rounding problem he describes is simply us getting confused and wondering why a fraction with an exact representation in base 10 doesn't have an exact representation in base 2. The obvious solution is the one he alludes to at the end: don't use base 2. Computers have had base-10 arithmetic in them for decades...
Maybe I'm missing something, but I don't see how this is obvious at all. Something does not have an exact representation in base 2, so we should use base 10. This still does not solve problems like 1/3 above, which has no "finite" representation in base 10 (unless you move into using a ratio class)? Short of having a "has an exact representation in base N" lookup table, I don't see how choosing a specific base can help, except for some cleverly contrived examples.
As others have pointed out above, a better solution is to use decimial arithmetic. You'll pay a performance penalty, but you won't run into these same errors.
Slashdot Mirror
So... you're saying I should reevaluate my KDE Ultimate Network Tool then?
It's even easier than that these days:
Done in two. I have no idea why OP is having such a difficult time.
At my local grocery store, they sell three different types of bananas: "standard" (49c/lb), organic (79c/lb) and fair trade (89c/lb). Without some form of labelling, the cashier cannot distinguish between the types.
Of course, the self-checkout lanes make this entire argument almost moot...
You know, I used to think it was solely an OpenOffice.org issue, and while the application certainly could be snappier and less bloated in some areas, I don't think it's nearly as bad as people make it out to be.
As the only Computer Science teacher at my high school, I often inherit lots of older hardware. The other day our local technician gave me three old Celeron 600MHz laptops that he was just going to throw in the trash. I decided to turn them into workstations for my students who did not have easy access to computers at home. As they were slow machines, I set them up with a basic Ubuntu install running IceWM -- it runs quite snappy, and is a bit more user-friendly than say DSL.
I considered installing AbiWord for a word processor, but chose OO.o mainly because we have it installed on the school servers anyway. The result? It takes about 6-7 seconds to load up, and responds quite well. As a point of comparison, the 2GHz Pentium 4 machines running XP and Word in our department office launch Word in approximately 5-6 seconds, so it's pretty much on par.
I know that OO.o is a beast sometimes, but my experiences have been fairly positive.
They often do, but they're typically framed within a narrow, specific category. For example, the category might me "National Drinks" or some such thing. Typical Q/As might be Japan (Sake), Russia (Vodka), and so forth. Jeapoardy! also has a few other categories (anagrams, for example) that frequently use one or two words.
I frequently hear this argument, but it presupposes a truly "random" sick day. I suspect that the dates are indeed biased, with more people being "sick" on a Monday or Friday. I know that would certainly be my choice.
Somewhat related, it's like flipping a coin 999 times and coming up heads each time. What is the probability that it will come up heads again? If it was truly random, the probability would be 50%. More likely, however, the coin is weighted on one side and biased toward heads. Bias is definitely a factor.
Don't be too harsh, now. 5 out of 4 people can't work with effectively with fractions.
I use a dvorak keyboard, you insensitive clod!
That was probably the case back when you sold your goods, but IIRC, there's a follow-up feature now where you can comment on your comments.
I think this is the link.
...or posting to Slashdot under that user name, Jack. ;)
Well, you've got two main options:
Of course. There's no real reason to restrict a proof to a particular language. I would expect, however, that if a computer was essential in producing the proof, that the source code that was used would also be released. In the case of the FCT, the program ran for several hundred hours. In the spirit of mathematical disclosure, it would be nice to see someone provide their own particular program that they used, so that peers could inspect the code, rather than have them recreate it themselves. This way they're not reinventing the proverbial wheel, although they could certainly do so if they wished.
Cheers.
Which has always made me wonder why they don't simply provide a link to their own site, from which you may be redirected. For example, SpiffyPC Magazine might be doing a review on the new XYZ 123 motherboard, and configure spiffypc.com/XYZ123 as a referral link.
Actually, given most magazines' enthusiasm for advertising, don't most magazines include links back to their own sites, where they can include as many other links and feastures as they wish?
Perhaps I'm being too pessimistic, but shouldn't the source code have to be provided alongside the pseudocode? If the pseudocode is 100% spot-on, then there would really be no need for the computer-assisted proof in the first place --- you will have provided a proof in the form of verifiable instructions. But the FCT was proved by some amount of brute-force, IIRC. Who is to say that the coder who translated from pseudocode to source code didn't mess something up? I mean, if my pseudocode reads
INCREMENT current value by ONE...
OUTPUT result of long computations
and my source code is entered as
value += value++;...
printf("%d",result);
then even if the pseudocode is verified, the program may still be producing an erroneous result. In other words, you're assuming that IF the pseudocode is correct THEN the program itself is also correct, which may not be the case.
Pretty soon, the "mute" button will be considered a theft-enabling device.
Reminds me of "Join the military. Travel the world. Meet many interesting people. Kill them."
Only on /. would parent be modded "insightful" instead of "funny".
I imagine it's not quite so straightforward. You'd need to take into account room acoustics, hardware effects, generic ambient noises, or even other interfering sounds in the same room that could all interfere with a comparison of outgoing sound to incoming sound. It's very rare that you'd ever have a time where your outgoing sound file exactly matches one that is sensed coming from the speakers.
In a case like this, you have to look at what the item's price is. A number of sellers jack up the shipping, but drop down the item price, so that they don't have to fork over as much "royalty" fee to eBay when it sells. For instance, your $60 thumb drive might end up selling for $2 + $60 S&H = $62. If they set shipping at $20, the item might sell for a higher price, say $42. In this case, you'd pay the same amount, but the seller would be charged less by eBay, because shipping does not count in the royalty calculations.
I always get a quote for shipping before bidding, as I am sure you do as well. I factor this into my decision. If I'm willing to spend $50 on a item, and the shipping is $35, then I can max out my bid at $15. To me, it's all the same. Either way, I'll only spend up to $50.
Your other points? I've been lucky enough to avoid them. One day I am sure I'll be bitten.
Of course, that's only if you *notice* that it looks dangerous. Even Open Source can contain some subtle things that appear innocuous. See the classic "Reflections on Trusting Trust" -- great little read.
Never tried C/C++ programming, have you?
Yes, I realize you're being facetious.
...
Maybe I'm missing something, but I don't see how this is obvious at all. Something does not have an exact representation in base 2, so we should use base 10. This still does not solve problems like 1/3 above, which has no "finite" representation in base 10 (unless you move into using a ratio class)? Short of having a "has an exact representation in base N" lookup table, I don't see how choosing a specific base can help, except for some cleverly contrived examples.
As others have pointed out above, a better solution is to use decimial arithmetic. You'll pay a performance penalty, but you won't run into these same errors.
Yes, something tells me you'll be getting a lot of use out of that palm after trying that approach...
It doesn't matter which way you turn -- a sphere's a sphere.
e=2.71828183, so this would have happened long ago, say 2/7/2 or 2/7/18.