Umm... not revealing how a voter voted matters. Your vague hope that no one will go to the effort to correlate the data to get that information isn't good enough security for me.
You're better off without the time at all. Imagine that 3 people voted on a given voting machine from 2pm to 3pm, and they all voted for the same candidate...
I live in Rockport, Massachussetts. I voted by putting an X in a box next to the name of the cadidate of my choice.
This time there were only four or five things on the ballot: President, Senator, House Rep, and County Sherrif I think, but in the past we've had more ballot items. It seems to work here.
This actually doesn't matter as much as you might think it does.
It'd be good to have publicly available source for the voting software.
But it could equally be the Diebold software, and be just as secure, as long as the whole system is designed to audit properly. There need to be paper recipts that act the same way that paper ballots would. Some percentage of precincts should be hand-recounted as an audit mechanism every time anyway.
And here's an imporant part - each town should order an extra voting machine. Before voting, one voting machine should be set aside for auditing. After voting, a second voting machine should be set aside. Those two voting machines should be gone over by computer forensics experts looking for anything sketchy. Propritary software would just make the audit more expensive - decompiling sucks.
The actual text on the voter recipt should be OCRed for any recount. This prevents the "Print one guy's name but the other guy's barcode" exploit.
OCR isn't hard, especially if you design the printout assuming you're going to be OCRing later.
Optimally, any automated recount should be designed so that it's identical to a hand recount. i.e. A human recounter scans the ballot and sees the appropriate number increment on the screen.
The base procedure for a US presidental election would be the following:
For each precinct:
For each voter:
- Voter is looked up on the list
of registered voters.
- If they are on the list, they are
given a paper ballot.
- The voter fills out the ballot,
and it is put in a ballot box.
After the voting:
- The ballots in the box are
counted by hand using some procedure
to minimise the possibility of fraud.
- The totals are called into the
town/county/state tabulation centers
and totalled there. The intermediate
numbers are kept for later analysis. Once all the totals have been reported to the central state tabulation center, it's a simple matter to compare the results.
If there is any challenge to the numbers, the paper ballots are re-counted. The intermediate results from the initial count are available to help track down any fraud.
That procedure works fine, but since it's possible to optimise out some of the slow parts in the general case using computers, people want to do that.
Here's a procedure using electronic voting machines that has a better average-case time to results:
For each precinct:
For each voter:
- Voter is looked up on the list
of registered voters.
- If they are on the list, they are
given a paper ballot.
- The voter inserts the ballot into
a slot in the front of an available
voting machine.
- If the ballot looks real to the voting
machine, it lets the voter vote.
- The voting machine stores the vote
in memory, and prints it on the ballot.
- The voter verifies the vote printed
on the ballot, and puts it in a ballot
box. The ballot box keeps a count of
the votes.
After the voting:
- The number of ballots handed out, votes
stored on voting machines, and ballots
put in the ballot box are compared.
- If these numbers differ, votes from
the precinct will be counted manually.
- If the numbers agree, the votes totals
from the voting machines will be used,
at least initially.
- The totals are called into the
town/county/state tabulation centers
and totalled there. The intermediate
numbers are kept for later analysis. Once all the totals have been reported to the central state tabulation center, it's a simple matter to compare the results.
If there is any challenge to the numbers, the paper ballots are re-counted. The intermediate results from the initial count are available to help track down any fraud.
The key points in both protocals: - Each voter saw a paper ballot with his vote go into a ballot box. - The first intermediate numbers are at the precinct level - this allows for an easier audit later, and isolates any issues to as small a pool of votes as possible.
That's me sarcastically trying to figure out your thought process in coming up with the suggestion of a solar panel window.
At a certain point, you can't both let the light through and generate power from it, so obviously you'd be cutting IR/UV to generate power from and be passing through (most?) visible light.
A secondary question is "Why wouldn't you just use space that *isn't* windows instead of trying to do something complex?" - I mean most buildings have non-window surfaces, like a roof, and walls.
And I just wasted like 2 minutes explaining perfectly obvious sarcasam =P.
There's a couple of big bittorent sites. The ones that are legally easy to hit have already died.
The ones that are left are obviously hard to hit for various reasons. I don't quite see how the MPAA/RIAA/BSA haven't shut down Suprnova yet, but Demonoid.com is a significantly more difficult target - it's based in Hong Kong.
Spam is not a "Relatively Meaningless Incidental" unless you're also willing to classify Junk Faxing, DDoS attacks using zombies, and collect call telemarketing (Where they quickly whisper a relitives name to get you to accept) in that category.
Bandwidth isn't free. Having an abuse department, or wasting your company's primary admin's time with spam complaints isn't free. Buying heavier duty servers to deal with the spam load isn't free. And as we get more bandwidth/etc this won't reduce the problem. There'll just be more spammers, all sending 30 second TV ads as attachments.
Additionally, as electronic communications systems get used more heavily, we're getting to the point where it will be possible to contact most people through stuff like text messaging every waking hour. Marketing people (with legitimate reason) see this as an opportunity - they'll be able to make your cell phone (or wrist computer or whatever) ring and deliver you an ad whenever.
However much Marketting teams think this is great, it's not feasable to allow it as a society. A cellphone would become useless if you were getting telemarketing calls every 30 seconds.
So, at this point - we can set up social conventions (they're gonna have to be laws, or team marketing won't think they apply to them) that make spamming/spimming/spitting/etc unacceptable - OR everyone can set their stuff up so they only recieve messages from people on their white list. I sort of like being able to email anyone. That's kind of nice.
Email/Spam is only setting the precendent for the future issues.
People got used to buying software in 8x6x2 boxes, and now when they go into a store they trust software in a box more than CDs in bare jewell cases - even though that packaging works fine for Music which is conceptually similar in this context.
The FoodSaver devices are designed to take an unnessisary inch off of the reusable plastic bags every time you use them.
They charge a good chunk of money for the plastic bags you use with those things - and you need to use their bags or the air can't get out.
Now, you aren't wasting enough plastic for it to really matter from an environmental perspective - but you are just giving the FoodSaver company free money.
Rape is generally accepted as being a horrible crime - everyone knows it's horrible, and normal people just don't do it. Basically if someone is committing rape, it's a good indication that they're eithor mentally unstable or otherwise are not considering the results of their action. The difference between 15 years in prision and like 50 years for rape would be pretty minimal, and keeping it shorter allows for some chance of the criminal becoming a useful member of society at some point in the future.
Spamming is a non-violent but financially costly crime. Since it's never been a criminal act before, the people doing it don't have an innate feeling that they're doing something wrong - they don't understand that society is going to *put them in prision* if they're caught spamming.
The absolute best thing that could happen here would be for the judge to rule that the spammers get FIFTEEN YEARS IN PRISION (quietly, under breath: with possibility of parole in 6 months with good behaviour). That will give us a headline that will scare some of the other spammers, but will wreck the fewest people's lives.
Bittorrent is no different than, say, HTTP when it comes to this sort of thing.
If you're bittorrenting down a ISO from, say, the Knoppix official tracker - You know it's fine - same as if you downloaded it by HTTP from the same site.
Now if you're randomly downloading stuff from Hack-My-Computer.com, that's a different issue.
Slashdot Mirror
Awk is just an alias for perl -MEmul::Awk -lpe, and sed is the same but with -MEmul::Sed.
How does DSL suck?
I just finished ordering a business 768 up connection for $90/month - that's half a T1 right there, and I get 3 meg down for "free".
Umm... not revealing how a voter voted matters. Your vague hope that no one will go to the effort to correlate the data to get that information isn't good enough security for me.
You're better off without the time at all. Imagine that 3 people voted on a given voting machine from 2pm to 3pm, and they all voted for the same candidate...
I live in Rockport, Massachussetts. I voted by putting an X in a box next to the name of the cadidate of my choice.
This time there were only four or five things on the ballot: President, Senator, House Rep, and County Sherrif I think, but in the past we've had more ballot items. It seems to work here.
Actually, if there was voting fraud that altered the outcome of the United States presidental election, that's a far bigger issue than terrorism.
This actually doesn't matter as much as you might think it does.
It'd be good to have publicly available source for the voting software.
But it could equally be the Diebold software, and be just as secure, as long as the whole system is designed to audit properly. There need to be paper recipts that act the same way that paper ballots would. Some percentage of precincts should be hand-recounted as an audit mechanism every time anyway.
And here's an imporant part - each town should order an extra voting machine. Before voting, one voting machine should be set aside for auditing. After voting, a second voting machine should be set aside. Those two voting machines should be gone over by computer forensics experts looking for anything sketchy. Propritary software would just make the audit more expensive - decompiling sucks.
The actual text on the voter recipt should be OCRed for any recount. This prevents the "Print one guy's name but the other guy's barcode" exploit.
OCR isn't hard, especially if you design the printout assuming you're going to be OCRing later.
Optimally, any automated recount should be designed so that it's identical to a hand recount. i.e. A human recounter scans the ballot and sees the appropriate number increment on the screen.
You can't keep a timestamp with a vote - that would allow an observer to determine the vote of a specific voter.
Why aren't exit polls, random audits, required hand counting of close or should-be-close races, and the possibility of challenges good enough?
Here's a procedure using electronic voting machines that has a better average-case time to results:The key points in both protocals:
- Each voter saw a paper ballot with his vote go into a ballot box.
- The first intermediate numbers are at the precinct level - this allows for an easier audit later, and isolates any issues to as small a pool of votes as possible.
In the case of N required counts where N is greater than, say, 3 - the computer count is about as useful as a count by people.
That's me sarcastically trying to figure out your thought process in coming up with the suggestion of a solar panel window.
At a certain point, you can't both let the light through and generate power from it, so obviously you'd be cutting IR/UV to generate power from and be passing through (most?) visible light.
A secondary question is "Why wouldn't you just use space that *isn't* windows instead of trying to do something complex?" - I mean most buildings have non-window surfaces, like a roof, and walls.
And I just wasted like 2 minutes explaining perfectly obvious sarcasam =P.
Hunt what? Fish?
There's a couple of big bittorent sites. The ones that are legally easy to hit have already died.
The ones that are left are obviously hard to hit for various reasons. I don't quite see how the MPAA/RIAA/BSA haven't shut down Suprnova yet, but Demonoid.com is a significantly more difficult target - it's based in Hong Kong.
Spelling it right works for me
http://azureus.sourceforge.net/
Spam is not a "Relatively Meaningless Incidental" unless you're also willing to classify Junk Faxing, DDoS attacks using zombies, and collect call telemarketing (Where they quickly whisper a relitives name to get you to accept) in that category.
Bandwidth isn't free. Having an abuse department, or wasting your company's primary admin's time with spam complaints isn't free. Buying heavier duty servers to deal with the spam load isn't free. And as we get more bandwidth/etc this won't reduce the problem. There'll just be more spammers, all sending 30 second TV ads as attachments.
Additionally, as electronic communications systems get used more heavily, we're getting to the point where it will be possible to contact most people through stuff like text messaging every waking hour. Marketing people (with legitimate reason) see this as an opportunity - they'll be able to make your cell phone (or wrist computer or whatever) ring and deliver you an ad whenever.
However much Marketting teams think this is great, it's not feasable to allow it as a society. A cellphone would become useless if you were getting telemarketing calls every 30 seconds.
So, at this point - we can set up social conventions (they're gonna have to be laws, or team marketing won't think they apply to them) that make spamming/spimming/spitting/etc unacceptable - OR everyone can set their stuff up so they only recieve messages from people on their white list. I sort of like being able to email anyone. That's kind of nice.
Email/Spam is only setting the precendent for the future issues.
Awesome. That makes the world a better place. Oh, wait, no it doesn't.
Solar Power Window?
"Obviously visible light isn't worth generating power from."
"No way can I use the space on my roof for solar panels, that's valuable sitting space!"
People got used to buying software in 8x6x2 boxes, and now when they go into a store they trust software in a box more than CDs in bare jewell cases - even though that packaging works fine for Music which is conceptually similar in this context.
The FoodSaver devices are designed to take an unnessisary inch off of the reusable plastic bags every time you use them.
They charge a good chunk of money for the plastic bags you use with those things - and you need to use their bags or the air can't get out.
Now, you aren't wasting enough plastic for it to really matter from an environmental perspective - but you are just giving the FoodSaver company free money.
Right - but people do operate stuff like Hotmail.
Rape is generally accepted as being a horrible crime - everyone knows it's horrible, and normal people just don't do it. Basically if someone is committing rape, it's a good indication that they're eithor mentally unstable or otherwise are not considering the results of their action. The difference between 15 years in prision and like 50 years for rape would be pretty minimal, and keeping it shorter allows for some chance of the criminal becoming a useful member of society at some point in the future.
Spamming is a non-violent but financially costly crime. Since it's never been a criminal act before, the people doing it don't have an innate feeling that they're doing something wrong - they don't understand that society is going to *put them in prision* if they're caught spamming.
The absolute best thing that could happen here would be for the judge to rule that the spammers get FIFTEEN YEARS IN PRISION (quietly, under breath: with possibility of parole in 6 months with good behaviour). That will give us a headline that will scare some of the other spammers, but will wreck the fewest people's lives.
Long term, it would be nice to have source to this stuff, but the people we need to bug are at the FCC - not at the hardware manufacturer.
Until it's legal to distribute the source for Wireless NIC firmware, bugging the hardware manufacturer for it is just being obnoxious.
Reverse engineering it on the other hand - that's fair game.
Remember:
Bittorrent is no different than, say, HTTP when it comes to this sort of thing.
If you're bittorrenting down a ISO from, say, the Knoppix official tracker - You know it's fine - same as if you downloaded it by HTTP from the same site.
Now if you're randomly downloading stuff from Hack-My-Computer.com, that's a different issue.