Shush... the fact that governments - and not only dog-eat-dog-capitalism - achieve useful and economical high-tech goals is such a foreign concept to these people that a sudden exposure to contrary evidence might bring their ideological world crashing down on them. So be careful.;)
Is there an acceptable statistical error in the number of I's compared to the number of D's in that line? Or can you say there are 8 D's and 8 I's?
Write down millions of I's and D's, spread them over several states, make some of them digital and make some of them cardboard ballot cards with hanging chads. Ask people to count them all in 24 hours with the weight of the national media on your back and you're bound to get different results each time. Of course it is acceptable and expected.
I don't quite understand why so people on Slashdot expect that "if it's binary, you can count it error-free". That's just naive - even if it's computerized.
But that's absolutely no excuse not to try and eliminate errors
You're quite correct. However, before such efforts are made one should consider if the errors are statistically significant to warrant the expense.
My point in this thread was simply that even if vote counting is "counting", there is still an acceptable statistical error in the results. I quite don't understand why people got so upset about what I said.
I have 2 pens on my desk. I could count them repeatedly for years
In elections like yours you have millions of votes and 24 hours or less to count them in a distributed manner. Physical ballots get squashed, torn, burnt or eaten. Voters are stupid and they vote for wrong candidates and then want to vote again but leave both ballots behind. Digital ballots get swallowed into/dev/null or multiplied (by a signed constant) by random bugs. Don't tell me you really, really believe that every vote is for real?
Sure, you could spend years really refining the final election results, but it's really not worth it. Most of the fraud or mistakes will soon be statistically insignificant.
I don't think the gambling industry would put up with errors like this, why should we?
You really think the gambling industry relies on spot-on-accurate counting? They just hedge the bets enough to cover all losses due to random mistakes and fraud that's not caught, for instance.
It's the election not some radio receiving test, there should be no anomalies.
Ok. As a physicist, I'd like you to show me how to perform an error-free measurement. You can claim the patent rights. I'd just like to see how you do it.
one of the most entertaining byproducts of Bush's smashing victory has been watching the worldwide Left come unhinged. Anyone remember the exploding heads from Scanners? That is where the fruit cakes are heading.
Hehe. Although I am rather a left-leaning European who's severely disappointed by the results (I want back the USA I liked!), I found your post most entertaining. Maybe it's because I do have colleagues, who did come unhinged at work the next day... Anyway, thanks for the chuckle.
typical government reports ares, even tech, are powerpoint with no relevant info
Huh? Presentations are never supposed to yield in-depth, relevant info. A presentation that does that is torture for the audience. It's just supposed to give an overview on the topic. If you want relevant, in-depth info, you're supposed to read the report, but as you say, a common ground would be nice.
The best way to achieve common ground is that when you give a presentation, you hand out a 1-6 page summary with plenty of figures alongside with the report itself.
As long as MS Office does not run on Linux, Linux desktop is going to struggle.
At work I just switched over to Mac OSX and installed MS Office for Mac. That's all I need: Unix-like CLI, X11, and the essential tools like MS Office.
The next best thing would be a PC with a virtual machine Linux and Windows.
Well, it could have an abstract and a few pretty pictures for us who'd just like to know what it's all about. If us natural science geeks can do it on a regular basis, what's holding you CS people back?
Yes. As I said above, they'd have to buy a full license since the educational license won't do. And sure it would cost a lot of money, but that could be recouped by selling a Windows compatible Xen. That'd be a financial risk, yes, but they should also think of the potential financial rewards.
A port of Windows XP was developed for an earlier version of Xen, but is not available for release due to licensce restrictions
Sigh... how hard would it be to get a license and distribute it as a binary-only module to people like me who'd be willing to pay for it? I'm sure it'd still be less expensive than the existing alternatives.
Otherwise this looks very nice. In fact, I didn't know that there was such a mature free virtual machine available.
I am aware that Mplayer can decode several proprietary codecs by using the code written for Windows. I use it myself. However, to use these same unlicensed codecs in a commercial product would be seriously illegal. That's why I assumed commercial intent in my original post.
I assume Linux is a lot better than MCE for Media Centers as it lets you run any format you want not just propietry codecs+formats. (WMA, WMV, etc.)
Uh. I don't quite get your point. If it's running Linux, sure you can add support for OGG and the likes, but then again you can't play proprietary codecs such as WMA, WMV, QuickTime and others - unless you pay for the license. All this assuming that such a device would be aimed at the mainstream markets, of course.
Slashdot Mirror
Shush... the fact that governments - and not only dog-eat-dog-capitalism - achieve useful and economical high-tech goals is such a foreign concept to these people that a sudden exposure to contrary evidence might bring their ideological world crashing down on them. So be careful. ;)
Ok. I hereby give up. You win. Go ahead and believe in your infallible vote counting system. I'm going to sleep.
Write down millions of I's and D's, spread them over several states, make some of them digital and make some of them cardboard ballot cards with hanging chads. Ask people to count them all in 24 hours with the weight of the national media on your back and you're bound to get different results each time. Of course it is acceptable and expected.
I don't quite understand why so people on Slashdot expect that "if it's binary, you can count it error-free". That's just naive - even if it's computerized.
No. I'm just cynical by nature and in a foul mood.
How about bugs casting your votes to /dev/null or multiplying them by a signed, random constant?
Or are you expecting software to be 100% bug free?
You're quite correct. However, before such efforts are made one should consider if the errors are statistically significant to warrant the expense.
My point in this thread was simply that even if vote counting is "counting", there is still an acceptable statistical error in the results. I quite don't understand why people got so upset about what I said.
In elections like yours you have millions of votes and 24 hours or less to count them in a distributed manner. Physical ballots get squashed, torn, burnt or eaten. Voters are stupid and they vote for wrong candidates and then want to vote again but leave both ballots behind. Digital ballots get swallowed into /dev/null or multiplied (by a signed constant) by random bugs. Don't tell me you really, really believe that every vote is for real?
Sure, you could spend years really refining the final election results, but it's really not worth it. Most of the fraud or mistakes will soon be statistically insignificant.
You really think the gambling industry relies on spot-on-accurate counting? They just hedge the bets enough to cover all losses due to random mistakes and fraud that's not caught, for instance.
Life's all about statistics.
Things like that just don't happen.
Ok. As a physicist, I'd like you to show me how to perform an error-free measurement. You can claim the patent rights. I'd just like to see how you do it.
Yes, but are any of these anomalies statistically significant? If not, it's just random noise regardless of the source.
Uh... no.
Hehe. Although I am rather a left-leaning European who's severely disappointed by the results (I want back the USA I liked!), I found your post most entertaining. Maybe it's because I do have colleagues, who did come unhinged at work the next day... Anyway, thanks for the chuckle.
Heh. I guess that's like during the good old Cold War. If you just got some sort of an idea of how to beat the enemy, you've got a blank check.
Huh? Presentations are never supposed to yield in-depth, relevant info. A presentation that does that is torture for the audience. It's just supposed to give an overview on the topic. If you want relevant, in-depth info, you're supposed to read the report, but as you say, a common ground would be nice.
The best way to achieve common ground is that when you give a presentation, you hand out a 1-6 page summary with plenty of figures alongside with the report itself.
That is an interesting approach. Make it run native on Windows and run Linux/BSD/whatever under it.
Somehow I don't think you get low-level enough in Windows (without the proper licenses) to make it happen.
As long as MS Office does not run on Linux, Linux desktop is going to struggle.
At work I just switched over to Mac OSX and installed MS Office for Mac. That's all I need: Unix-like CLI, X11, and the essential tools like MS Office.
The next best thing would be a PC with a virtual machine Linux and Windows.
Well, it could have an abstract and a few pretty pictures for us who'd just like to know what it's all about. If us natural science geeks can do it on a regular basis, what's holding you CS people back?
Heh. Such a typical government tech report. No pics, just text and tables.
Yes. As I said above, they'd have to buy a full license since the educational license won't do. And sure it would cost a lot of money, but that could be recouped by selling a Windows compatible Xen. That'd be a financial risk, yes, but they should also think of the potential financial rewards.
As I said above, I don't think that Xen + licensed Windows-compatibility module would cost as much as VMware and other alternatives.
What other issues could there be preventing the purchase of a full license? VMware developers must have bought a license, so what's the problem here?
Sigh... how hard would it be to get a license and distribute it as a binary-only module to people like me who'd be willing to pay for it? I'm sure it'd still be less expensive than the existing alternatives.
Otherwise this looks very nice. In fact, I didn't know that there was such a mature free virtual machine available.
I am aware that Mplayer can decode several proprietary codecs by using the code written for Windows. I use it myself. However, to use these same unlicensed codecs in a commercial product would be seriously illegal. That's why I assumed commercial intent in my original post.
Uh. I don't quite get your point. If it's running Linux, sure you can add support for OGG and the likes, but then again you can't play proprietary codecs such as WMA, WMV, QuickTime and others - unless you pay for the license. All this assuming that such a device would be aimed at the mainstream markets, of course.