An insight into our tax dollars and tuition at work.
"During the heyday of Napster, the University of Wisconsin - Madison had a difficult decision. As it watched the traffic for Napster consume over 70% of total inbound bandwidth at its peak, we asked ourselves: do we start blocking Napster? After all, it's mostly used for stealing music. Right?"
What the administrators *should* have done is an analysis on that traffic, and examined what exactly it was being used for. If 70% of your bandwidth in one direction is being smoked because of non-essential traffic, in any business, you stop that traffic...period. It doesn't matter whether the traffic is legal or illegal. If it could cause others to not be able to get their jobs (or in this case their bonafide homework) done, then you put an end to it.
I'd say they didn't think too hard about this one...nor do I think this applies to the original post at all. Napster was used for downloading music...that's it...that was always illegal.
Bittorrent is used for downloading all kinds of content...basically anything you want to create a torrent for...whether that be CD images of legit software (Linux), software updates like someone alluded to for gaming, or anything else. It can get used for legit as well as non-legit purposes. However, Napster was pretty much illegal, whether anyone wanted to admit it or not. I'm not passing judgment on anyone who used it...hell, I used it. But I KNEW it was illegal, and I still did it. It was used for nothing but *sharing* music...not distributing software updates. Bittorrent can't even be compared to Napster...it's completely different, and it's uses are endless.
My overall gist? Napster has nothing to do with academics, so your analogy is laughable at best...it's about increasing your music library for free. Bittorrent IS about freedom. It gives you a way of getting and distributing any type of content (legal or otherwise) by using the rest of the world's bandwidth if they so desire.
Don't know if it covers it or not, because I tend to find everything I need to know in Google for free, but if you've ever tried to clone a few machines the same way simultaneously, you use NFS...easiest to set up, and the fastest, and is the only method that allows up to use the full GUI (Anaconda) in Redhat and Fedora through the whole install process other than from CD-ROM. If you use HTTP or FTP as your install method, you can only use the text menus.
And if you want to tighten up your sendmail config without regenerating.cf files from.mc files, yeah, you just modify the.cf file and restart the daemon. Any good system admin should know those things, so yeah, it belongs in the book.
You also seem to think that I want the manufacturers to pretty-please help us out here, whereas most of the people who share my views are calling for law requiring a solution like the one I described. If that law has to replace some current law that specifies the machines as they are, fine.
Cool. Get it done then. Because there are already machines on the market that support the process that you want to take place...every vendor has them. Fact is, 99% of the jurisdictions out there don't want them for whatever reason. Until those laws pass and those county clerks pull their heads out, these problems will continue.
They don't have to show the public anything. The laws in place prevent them from having to do that. The only reason they as an industry would, is out of good will. Election software is what they make their living on. Why would a company open source the code when that's how they make money?
If there are laws in place that TRULY protect both the voters and companies in question because a truly independant source is looking at the code (not some jackass on 60 minutes trying to get his 15 minutes of fame), then that's one thing.
But the general public, and Avi Rubin for that matter, don't have a clue about the technology or about the laws in each jurisdiction that determine why those voting machines are made the way they are.
Hate to spook you out, but the majority of voting machines are why they are because the county clerks WANT them that way, not because the companies designed them that way. And if anyone here has dealt with a county clerk in Podunk, Wisconsin, you know what I'm talking about.
Election companies don't determine what the public uses...it's the other way around. The laws in each jurisdiction determine what the company's equipment does. Maybe everyone bitching should go look at the laws on their books in each state and examine why those laws don't call for a voter-verifiable-paper-trail.
And this:
1. Provides the voter assurance that the machine read the voter's input correctly.
That doesn't do anything. As a programmer, I can tell that machine whatever I want to print on the paper, and still tell the machine internally to record something else. There's a fundamental problem with that. When that voter leaves, and they start tallying up all the WRONG votes later on, what are they going to do about that? Answer: Nothing. The voter has already left, the polls are closed. Basically, when they find those faulty votes in machines, the machine doesn't count. So X number of voters just lot their right to vote.
As a software engineer, you still haven't solved the problem. You've just solved a symptom while creating another. What you should be looking at are convuluted laws on your books that make these machines the way they are. Stop trying to fix the end-result of your bad laws...it doesn't work.
What everyone doesn't understand/get:
1. The paper receipt is there as a justification tool against what's on the memory cards or electronic storage media. It doesn't guarantee though that the vote hasn't been tampered with. It could very weel be tampered with while the person is pushing the "vote" button.
2. The purpose of the DRE (touchscreen), is to prevent over and undervotes. Overvotes *confuse* optical scan machines. Remember the standardized tests back when you were in grade school? This is why they told you to darken ONE oval...the machines are intelligent enough to determine what's what...so if someone darkens two ovals for the same candidate, it doesn't count either...it records it as an error--in this case an overvote...so that vote doesn't count. DRE's prevent this from happening. You can only choose Kerry OR Bush...you can't choose both.
3. You can't just take the memory cards out and change the ballot or the results. It doesn't work that way. Different companies use different ways of encryption and verification. Basically, if that key on the memory card doesn't match one on the aggregating machine that also programmed those memory cards, as well as every file validity check --depending on the company, this could be CRC, PGP, MD5, and the list goes on--but the files just aren't there waiting to be modified/deleted/replaced. The machine/process ceases to work if one file is changed/deleted/modified in any way...period. That's how at least two company's technology works.
One thing I find funny, is that since all this proverbial shit has hit the fan starting a couple years ago, Avi Rubin in one year has all of a sudden become it seems the world's expert on voting machines. There are very talented programmers who work on this stuff every day...and have worked on this every day for the past 20 years. And before you can understand the issues that may plague an election system, you have to understand the laws in whatever jurisdiction those election systems will be deployed in. And that's one HUGE issue that no one wants to address or take the time to learn. I'm pretty confident Avi Rubin doesn't know why some Florida laws prevent touchscreens from being used in say, Texas...and vice versa. Any jackass can get on 60 minutes and say "This sucks, that sucks, it all sucks, and my vote isn't secure." But it takes a person of a little bit more intelligence to understand why it is that way.
Example: I hear arguments all the time (from Computer Science people like Avi Rubin) that say that relational databases and other technology like that should be used to validate votes vs voters coming into the polling place. Wrong. The whole democratic system in the USA is based upon the fact every voter should be able to remain anonymous in the polling place regarding what/who they voted for. Introduce a database to keep track of voter and their ballot results and you've just violated the very law/premise that our democracy stands on.
My message to everyone including Avi Rubin and anyone else in Academia who thinks they are an election system expert after one year: Learn every state law...then try to build an election system that conforms to every single state law with the same piece of software.
If anyone can do that within 5 years, I'll be very impressed...
If you want a system that can't be electronically compromised, do it like the jurisdictions in the UK. They scan all the paper ballots electronically, then recount them by hand until the numbers match. That's the only way to ensure they aren't electronically altered, and that no over/under votes are incorrectly counted.
Or is the editing in that Q&A just horrible? Spelling and grammatical errors abound...
Someone should invest some time in Hooked on Phonics before writing an article about something they know nothing about...
Results are what matter. If someone can get their required work done with 10% of their time, no one should really care about the use of the other 90%.
Sorry...gov't or not. That's what happens in companies who are concerned about the bottom line. You get your shit done, you have a job. You don't, you're fired.
If higher ups thought he was getting work done, Solitaire is not an offense. If he's slacking, Solitaire becomes an offense. That simple.
Dunno why there's 200+ posts debating this.
Re:What would I do with this much bandwidth?
on
Ethernet at 10 Gbps
·
· Score: 1
It's not bandwidth that's the problem...it's network saturation of many people connecting to many machines at high speed and saturating your switching backplane. Your port speed is simply your maximum potential...it doesn't imply you'll actually ever get that speed. If you use any kind of data encryption at all, that's obvious.
I can see practical applications in some scenarios...but not any scenarios that involve any less than 50 users on a switch at one time.
And you might learn something you didn't know between ass-kissing sessions with professors.
It's about giving people what they want, getting their approval, and moving on. It's a giant, real-world, real-time video game really. There are rules. If you don't follow those rules, you fail.
That's college in two paragraphs.
What I suggest you do is find what you like, find out how to make some money at it, retire early, and do what you want for the rest of your life...whether that includes going to school or not, doesn't matter. If you are as smart as you think you are, you should have no problem making it happen with a good work ethic.
Clear as mud? Thought so.
Life/college/whatever else is what you make of it. You gotta figure it out for yourself.
Slashdot Mirror
Works fine. YMCA does it.
An insight into our tax dollars and tuition at work.
"During the heyday of Napster, the University of Wisconsin - Madison had a difficult decision. As it watched the traffic for Napster consume over 70% of total inbound bandwidth at its peak, we asked ourselves: do we start blocking Napster? After all, it's mostly used for stealing music. Right?"
What the administrators *should* have done is an analysis on that traffic, and examined what exactly it was being used for. If 70% of your bandwidth in one direction is being smoked because of non-essential traffic, in any business, you stop that traffic...period. It doesn't matter whether the traffic is legal or illegal. If it could cause others to not be able to get their jobs (or in this case their bonafide homework) done, then you put an end to it.
I'd say they didn't think too hard about this one...nor do I think this applies to the original post at all. Napster was used for downloading music...that's it...that was always illegal.
Bittorrent is used for downloading all kinds of content...basically anything you want to create a torrent for...whether that be CD images of legit software (Linux), software updates like someone alluded to for gaming, or anything else. It can get used for legit as well as non-legit purposes. However, Napster was pretty much illegal, whether anyone wanted to admit it or not. I'm not passing judgment on anyone who used it...hell, I used it. But I KNEW it was illegal, and I still did it. It was used for nothing but *sharing* music...not distributing software updates. Bittorrent can't even be compared to Napster...it's completely different, and it's uses are endless.
My overall gist? Napster has nothing to do with academics, so your analogy is laughable at best...it's about increasing your music library for free. Bittorrent IS about freedom. It gives you a way of getting and distributing any type of content (legal or otherwise) by using the rest of the world's bandwidth if they so desire.
Don't know if it covers it or not, because I tend to find everything I need to know in Google for free, but if you've ever tried to clone a few machines the same way simultaneously, you use NFS...easiest to set up, and the fastest, and is the only method that allows up to use the full GUI (Anaconda) in Redhat and Fedora through the whole install process other than from CD-ROM. If you use HTTP or FTP as your install method, you can only use the text menus. And if you want to tighten up your sendmail config without regenerating .cf files from .mc files, yeah, you just modify the .cf file and restart the daemon. Any good system admin should know those things, so yeah, it belongs in the book.
You also seem to think that I want the manufacturers to pretty-please help us out here, whereas most of the people who share my views are calling for law requiring a solution like the one I described. If that law has to replace some current law that specifies the machines as they are, fine.
Cool. Get it done then. Because there are already machines on the market that support the process that you want to take place...every vendor has them. Fact is, 99% of the jurisdictions out there don't want them for whatever reason. Until those laws pass and those county clerks pull their heads out, these problems will continue.
Which brings me to my next point.
They don't have to show the public anything. The laws in place prevent them from having to do that. The only reason they as an industry would, is out of good will. Election software is what they make their living on. Why would a company open source the code when that's how they make money?
If there are laws in place that TRULY protect both the voters and companies in question because a truly independant source is looking at the code (not some jackass on 60 minutes trying to get his 15 minutes of fame), then that's one thing.
But the general public, and Avi Rubin for that matter, don't have a clue about the technology or about the laws in each jurisdiction that determine why those voting machines are made the way they are.
Hate to spook you out, but the majority of voting machines are why they are because the county clerks WANT them that way, not because the companies designed them that way. And if anyone here has dealt with a county clerk in Podunk, Wisconsin, you know what I'm talking about.
Election companies don't determine what the public uses...it's the other way around. The laws in each jurisdiction determine what the company's equipment does. Maybe everyone bitching should go look at the laws on their books in each state and examine why those laws don't call for a voter-verifiable-paper-trail.
And this:
1. Provides the voter assurance that the machine read the voter's input correctly.
That doesn't do anything. As a programmer, I can tell that machine whatever I want to print on the paper, and still tell the machine internally to record something else. There's a fundamental problem with that. When that voter leaves, and they start tallying up all the WRONG votes later on, what are they going to do about that? Answer: Nothing. The voter has already left, the polls are closed. Basically, when they find those faulty votes in machines, the machine doesn't count. So X number of voters just lot their right to vote.
As a software engineer, you still haven't solved the problem. You've just solved a symptom while creating another. What you should be looking at are convuluted laws on your books that make these machines the way they are. Stop trying to fix the end-result of your bad laws...it doesn't work.
What everyone doesn't understand/get: 1. The paper receipt is there as a justification tool against what's on the memory cards or electronic storage media. It doesn't guarantee though that the vote hasn't been tampered with. It could very weel be tampered with while the person is pushing the "vote" button. 2. The purpose of the DRE (touchscreen), is to prevent over and undervotes. Overvotes *confuse* optical scan machines. Remember the standardized tests back when you were in grade school? This is why they told you to darken ONE oval...the machines are intelligent enough to determine what's what...so if someone darkens two ovals for the same candidate, it doesn't count either...it records it as an error--in this case an overvote...so that vote doesn't count. DRE's prevent this from happening. You can only choose Kerry OR Bush...you can't choose both. 3. You can't just take the memory cards out and change the ballot or the results. It doesn't work that way. Different companies use different ways of encryption and verification. Basically, if that key on the memory card doesn't match one on the aggregating machine that also programmed those memory cards, as well as every file validity check --depending on the company, this could be CRC, PGP, MD5, and the list goes on--but the files just aren't there waiting to be modified/deleted/replaced. The machine/process ceases to work if one file is changed/deleted/modified in any way...period. That's how at least two company's technology works. One thing I find funny, is that since all this proverbial shit has hit the fan starting a couple years ago, Avi Rubin in one year has all of a sudden become it seems the world's expert on voting machines. There are very talented programmers who work on this stuff every day...and have worked on this every day for the past 20 years. And before you can understand the issues that may plague an election system, you have to understand the laws in whatever jurisdiction those election systems will be deployed in. And that's one HUGE issue that no one wants to address or take the time to learn. I'm pretty confident Avi Rubin doesn't know why some Florida laws prevent touchscreens from being used in say, Texas...and vice versa. Any jackass can get on 60 minutes and say "This sucks, that sucks, it all sucks, and my vote isn't secure." But it takes a person of a little bit more intelligence to understand why it is that way. Example: I hear arguments all the time (from Computer Science people like Avi Rubin) that say that relational databases and other technology like that should be used to validate votes vs voters coming into the polling place. Wrong. The whole democratic system in the USA is based upon the fact every voter should be able to remain anonymous in the polling place regarding what/who they voted for. Introduce a database to keep track of voter and their ballot results and you've just violated the very law/premise that our democracy stands on. My message to everyone including Avi Rubin and anyone else in Academia who thinks they are an election system expert after one year: Learn every state law...then try to build an election system that conforms to every single state law with the same piece of software. If anyone can do that within 5 years, I'll be very impressed... If you want a system that can't be electronically compromised, do it like the jurisdictions in the UK. They scan all the paper ballots electronically, then recount them by hand until the numbers match. That's the only way to ensure they aren't electronically altered, and that no over/under votes are incorrectly counted.
Or is the editing in that Q&A just horrible? Spelling and grammatical errors abound... Someone should invest some time in Hooked on Phonics before writing an article about something they know nothing about...
Results are what matter. If someone can get their required work done with 10% of their time, no one should really care about the use of the other 90%. Sorry...gov't or not. That's what happens in companies who are concerned about the bottom line. You get your shit done, you have a job. You don't, you're fired. If higher ups thought he was getting work done, Solitaire is not an offense. If he's slacking, Solitaire becomes an offense. That simple. Dunno why there's 200+ posts debating this.
It's not bandwidth that's the problem...it's network saturation of many people connecting to many machines at high speed and saturating your switching backplane. Your port speed is simply your maximum potential...it doesn't imply you'll actually ever get that speed. If you use any kind of data encryption at all, that's obvious. I can see practical applications in some scenarios...but not any scenarios that involve any less than 50 users on a switch at one time.
And you might learn something you didn't know between ass-kissing sessions with professors. It's about giving people what they want, getting their approval, and moving on. It's a giant, real-world, real-time video game really. There are rules. If you don't follow those rules, you fail. That's college in two paragraphs. What I suggest you do is find what you like, find out how to make some money at it, retire early, and do what you want for the rest of your life...whether that includes going to school or not, doesn't matter. If you are as smart as you think you are, you should have no problem making it happen with a good work ethic. Clear as mud? Thought so. Life/college/whatever else is what you make of it. You gotta figure it out for yourself.