Ok, I've been thinking about this each time a/. article comes up about electronic voting machines. Here is what is needed at a minimum:
Have a printed copy that is stored in the voting maching. Make it be like a calculator/cash register roll of paper, but have it not physically accessable to the voter, just visible.
You would vote on the screen, and on the right it would print your vote so you could verify that the machine was recording accurately how you voted. Perhaps in an abbreviated format like:
-------------------- |C1 C2 C3 M1 M2 M3 |<- Electronic line |AA BB CC YY YY NN | <- Electronic line |------------------| |AA BB CC YY YY NN | <- Paper roll printed line --------------------
"AA," "BB," and "CC" are the initialize of the canditates this example voted for (use 3 initials or 2 initials an a number or whatever for uniqueness), and Yes to Measure 1 and 2, No to Measure 3. It would display this summary off to the right of the main graphical display, and perhaps at each graphical display you had to say "CONFIRM" before it would print and "finalize" the vote.
It would only move forward, so you couldn't see the previous voters votes.
From here, someone who is intent on rigging an election still has the same power to do so as someone with a classic ballot: the could diddle the data and print out matching paper tape rolls to match.
An extra nice feature, but not required:
Once you were thru all the voting screens, it should print you a copy to keep that has a history of your voting. An md5 hash that used your name + voting choices should be included.
First of all, please use a BitTorrent client to mirror and leave the client running for some time for others to download from you as well.
Once you have the client, tell it to download the torrent: http://torrent.dulug.duke.edu/yarrow-src -i386-iso. torrent (mind the extra space in the URL thanks to Slashcode)
We mirrored the 1.8GB worth of 3 ISOs with BitTorrent in just under 2 hours with two T1s (Sprint & UUNET). During that time and the next hour we'd sent out 1.8GB as well. We'll continue to run BitTorrent until 7am PST when users would complain about the speed.
I suggest downloading each file completely, and then viewing locally (vs viewing with a plug-in, downloading one page at a time in your browser). Trying to pull down a page at a time as you scroll isn't efficient with PDFs unless you're local (10mbit+):
Actually, there would be nothing wrong with this. If it's good for GPL'd software, it's good for all software to protect IP.
The real trick is what if some non-GPL code is watermarked, but in fact watermarked by the real author (in other words they "borrowed" the code and then watermarked it)?
My riding skills are (nearly) flawless, which accounts for why I can lane split and drive past parked traffic the way I do.
I would say others need to work on their danger perception skills. Driving 10-20 over the limit on a clear open freeway isn't dangerous. Driving 10-20 over (or even the limit) and zipping between and cutting off other cars in a hulking vehicle that can't stop in time is dangerous.
I can stop in 20-30 feet on my bike, can your 4-wheel road-hogging beastie do the same?
Just some thoughts from a two-wheel riding pro-motorcycle guy.
Shortage, perhaps not, however why do I have to pay more for more than 5 static IPs with SBC? Why does another local SDSL provider (Arrival.net) charge $5/month/ip? If there was no shortage, and some ISPs were just giving away addresses like it was nothing, then I would say it wasn't a problem. Right now, the problem is that there is a finite amount of addresses, so ISPs will only give out as many as you can justify, as they have to justify them to ARIN/RIPE/APNIC.
The same is somewhat true of IPv6 allocations, except that the allocations are HUGE in comparison. From Sprintv6 we received a/48. That's/16 worth of/64s (/64s being what you commonly assign to a LAN to use automatically addressing with EUI-64). In other words, I've got the equivalent of a Class B worth of networks to play with, or 65536/64 netblocks, just for my company.
The advantage with IPv6 deployments is that every subscriber of an ISP could easily have a/64 and then have a nearly unlimited amount of nodes online without the ISP having to micromanage IP allocations.
My point is that you don't think about if you can justify allocating a/64 for a single end user or a/48 for a company, you just do it. Yes, a company with a/48 should allocate it internally as wisely as possible to minimize routing tables, etc., but that's a given already with IPv4 RFC1918 addressing.
While the 6bone 3ffe:: address space is not meant to run "production" services, it is just as viable as the 2001:: production address space. Both networks are connected at various points. Just do an ns dig on my artoo.net domain and you'll see a mixture of both addresses, both equally well connected.
Buy a shredder. Don't throw out VISA/Mastercard receipts (many contain your full CC number, exp. date and full name). This one is a big of common sense, if you ask me.
Would you throw away a credit card? Would you just cut it in half and toss it? Perhaps you don't realize how many folks have access to your trash... I'm a bit paranoid, but I cut mine into about 20 pieces and toss them randomly into 5 different trashcans in my house.
In fact some spammers are mining "spamproof" addresses.
3 months ago I did a test after I was getting more and more spam to my "clean" account. I changed the email address on my webpage from original-aliasNO@SPAMmydomain.com to test-aliasNO@SPAMmydomain.com. Within a month, I was getting spam to test-alias@mydomain.com.
My solution now is is to use a PHP-based script so that my email address is never seen by the public. When I reply to people I don't know, I use a limited lifetime account that turns into an auto-reply in a month directing them to my webpage's PHP contact form.
No more spam... zip, zero, ziltch.
I don't even bounce email to 'spam-collected' accounts... it just wastes my mailservers CPU cycles and instead I send it to a single account I'm using to track all wildcard account spam.
Right now, in just the past week, I've collected 94MB of spam to wildcard accounts (whatever@mydomain.com, whatever2@mydomain.com).
At least for AT&T and Nextel, here in the US, you only pay for outbound email or SMS, not inbound.
My brother went camping last weekend in a 'remote' area with no cell coverage. When he got back to civilization, he had 44 spam text emails on his cell phone.
AT&T says that he can turn off the feature, but then he can't get any.
What we want is a way to say, "only accept email from my forwarding service email address of: mycellphone@mydomain.com" or whatever. But I agree with another poster: a customized email account would be great... I'd set it to the maximum allowed characters and have it be pseudo-random so noone would be emailing it (except my own procmail script).
I know you were just trying to be funny, but Freeciv is a multiplayer strategy game, released under the GNU General Public License. It is generally comparable with Civilization II(R), published by Microprose(R).
Heh, then it occured to me, no need to strap a GPSr on your car if you have a cell phone and keep it on... I wonder how well you can track where someone is going or has been by looking at what cell towers they've been connected to as one travels? Yeah, nowhere near as precise, but still a very good vague idea when you piece other bits of info together.
The real question to me is, how long until they do away with the TIA blocker: cash? I use cash to by stuff I don't want tracked (2600 at B&N, lunch with a friend I don't want someone else to know about, etc.).
I'm not really worried until they decide to abolish cash. Then you'll have no choice but to have everything you purchase or anything you go being tracked (think toll booths, subways, buses, taxies).
Of course, all they really have to do is just strap a GPSr on the bottom of your personal transportation device. Not warrant required (ask Scott Pederson).
Hmm, and don't even get started with what is tracked with cell phone calls.
We've set a/32 route to one of our webservers and have a *.net and *.com alias for http://wildcard.artoo.net. At least this way customers know WTF is going on and can complain to Congress/ICANN, or just go try Google, but mainly VeriSign gets no traffic.
DNS admin needs to add Microsoft-Antitrust.gov. How many folks don't type 'www' these days? Rather, how many non-tech folks won't think to try http://www.Microsoft-Antitrust.gov when http://Microsoft-Antitrust.gov fails?
Slashdot Mirror
Those from the #fedora IRC channel on irc.freenode.net have started an unofficial FAQ.
I highly suggest browsing through the various issues others have had, before you decide to upgrade from RH or try a fresh install.
fedora.artoo.net.
Ok, I've been thinking about this each time a /. article comes up about electronic voting machines. Here is what is needed at a minimum:
Have a printed copy that is stored in the voting maching. Make it be like a calculator/cash register roll of paper, but have it not physically accessable to the voter, just visible.
You would vote on the screen, and on the right it would print your vote so you could verify that the machine was recording accurately how you voted. Perhaps in an abbreviated format like:
"AA," "BB," and "CC" are the initialize of the canditates this example voted for (use 3 initials or 2 initials an a number or whatever for uniqueness), and Yes to Measure 1 and 2, No to Measure 3. It would display this summary off to the right of the main graphical display, and perhaps at each graphical display you had to say "CONFIRM" before it would print and "finalize" the vote.
It would only move forward, so you couldn't see the previous voters votes.
From here, someone who is intent on rigging an election still has the same power to do so as someone with a classic ballot: the could diddle the data and print out matching paper tape rolls to match.
An extra nice feature, but not required: Once you were thru all the voting screens, it should print you a copy to keep that has a history of your voting. An md5 hash that used your name + voting choices should be included.
Not to drop any names, but Amazon.com has it for $25.99 and free shipping (for orders over $25). Anyone know of anywhere else that has it cheaper?
Oops, that would be the torrent if you want the source. The binary torrent is:
i so.torrent
http://torrent.dulug.duke.edu/yarrow-binary-i386-
Once you have the client, tell it to download the torrent:
http://torrent.dulug.duke.edu/yarrow-sr
(mind the extra space in the URL thanks to Slashcode)
We mirrored the 1.8GB worth of 3 ISOs with BitTorrent in just under 2 hours with two T1s (Sprint & UUNET). During that time and the next hour we'd sent out 1.8GB as well. We'll continue to run BitTorrent until 7am PST when users would complain about the speed.
We're running an FTP mirror for IPv6 as well:
ftp://r2.ipv6.artoo.net/pub/fedora/linux/core/1/i
(mind the extra space in the URL)
However you get the ISOs, get the MD5SUM direct from RedHat and checksum those ISOs:(mind the extra space in the URL)MD5SUMs match (other than extra spaces thanks to slashcode), so the ISOs haven't been tampered with.
To store it locally and view later or without jitter:Mind the slashdot spaces.
Doh, ZIP really is your friend with these PDFs (60% compression, ~2mb vs ~6mb):
IPv4:
diversityanalysis.zip
doj-attorney-diversity-unredacted.zip
IPv6:
diversityanalysis.zip
doj-attorney-diversity-unredacted.zip
I suggest downloading each file completely, and then viewing locally (vs viewing with a plug-in, downloading one page at a time in your browser). Trying to pull down a page at a time as you scroll isn't efficient with PDFs unless you're local (10mbit+):
IPv4:
diversityanalysis.pdf
doj-attorney-diversity-unredacted.pdf
IPv6:
diversityanalysis.pdf
doj-attorney-diversity-unredacted.pdf
non-free or open source (ie commercial) entities.
Actually, there would be nothing wrong with this. If it's good for GPL'd software, it's good for all software to protect IP.
The real trick is what if some non-GPL code is watermarked, but in fact watermarked by the real author (in other words they "borrowed" the code and then watermarked it)?
My riding skills are (nearly) flawless, which accounts for why I can lane split and drive past parked traffic the way I do.
I would say others need to work on their danger perception skills. Driving 10-20 over the limit on a clear open freeway isn't dangerous. Driving 10-20 over (or even the limit) and zipping between and cutting off other cars in a hulking vehicle that can't stop in time is dangerous.
I can stop in 20-30 feet on my bike, can your 4-wheel road-hogging beastie do the same?
Just some thoughts from a two-wheel riding pro-motorcycle guy.
Shortage, perhaps not, however why do I have to pay more for more than 5 static IPs with SBC? Why does another local SDSL provider (Arrival.net) charge $5/month/ip? If there was no shortage, and some ISPs were just giving away addresses like it was nothing, then I would say it wasn't a problem. Right now, the problem is that there is a finite amount of addresses, so ISPs will only give out as many as you can justify, as they have to justify them to ARIN/RIPE/APNIC.
/48. That's /16 worth of /64s (/64s being what you commonly assign to a LAN to use automatically addressing with EUI-64). In other words, I've got the equivalent of a Class B worth of networks to play with, or 65536 /64 netblocks, just for my company.
/64 and then have a nearly unlimited amount of nodes online without the ISP having to micromanage IP allocations.
/64 for a single end user or a /48 for a company, you just do it. Yes, a company with a /48 should allocate it internally as wisely as possible to minimize routing tables, etc., but that's a given already with IPv4 RFC1918 addressing.
The same is somewhat true of IPv6 allocations, except that the allocations are HUGE in comparison. From Sprintv6 we received a
The advantage with IPv6 deployments is that every subscriber of an ISP could easily have a
My point is that you don't think about if you can justify allocating a
While the 6bone 3ffe:: address space is not meant to run "production" services, it is just as viable as the 2001:: production address space. Both networks are connected at various points. Just do an ns dig on my artoo.net domain and you'll see a mixture of both addresses, both equally well connected.
If you run a dns server, create a bogus gator.com zone. The only records (other than the base SOA and NS records) needed are:
@ IN A 127.0.0.1
* IN A 127.0.0.1
Then anyone "infected" with Gator just tries to talk to themselves.
Buy a shredder. Don't throw out VISA/Mastercard receipts (many contain your full CC number, exp. date and full name). This one is a big of common sense, if you ask me.
Would you throw away a credit card? Would you just cut it in half and toss it? Perhaps you don't realize how many folks have access to your trash... I'm a bit paranoid, but I cut mine into about 20 pieces and toss them randomly into 5 different trashcans in my house.
Like I said, we're a really small ISP, but it appears we caught 281 typo's (excluding anything that was referred from Slashdot).
It's pretty amazing to look at the common sites that folks typo.
In fact some spammers are mining "spamproof" addresses.
3 months ago I did a test after I was getting more and more spam to my "clean" account. I changed the email address on my webpage from original-aliasNO@SPAMmydomain.com to test-aliasNO@SPAMmydomain.com. Within a month, I was getting spam to test-alias@mydomain.com.
My solution now is is to use a PHP-based script so that my email address is never seen by the public. When I reply to people I don't know, I use a limited lifetime account that turns into an auto-reply in a month directing them to my webpage's PHP contact form.
No more spam... zip, zero, ziltch.
I don't even bounce email to 'spam-collected' accounts... it just wastes my mailservers CPU cycles and instead I send it to a single account I'm using to track all wildcard account spam.
Right now, in just the past week, I've collected 94MB of spam to wildcard accounts (whatever@mydomain.com, whatever2@mydomain.com).
Just allow it to go to small claims court, no need for a lawyer.
At least for AT&T and Nextel, here in the US, you only pay for outbound email or SMS, not inbound.
My brother went camping last weekend in a 'remote' area with no cell coverage. When he got back to civilization, he had 44 spam text emails on his cell phone.
AT&T says that he can turn off the feature, but then he can't get any.
What we want is a way to say, "only accept email from my forwarding service email address of: mycellphone@mydomain.com" or whatever. But I agree with another poster: a customized email account would be great... I'd set it to the maximum allowed characters and have it be pseudo-random so noone would be emailing it (except my own procmail script).
I know you were just trying to be funny, but Freeciv is a multiplayer strategy game, released under the GNU General Public License.
It is generally comparable with Civilization II(R), published by Microprose(R).
Heh, then it occured to me, no need to strap a GPSr on your car if you have a cell phone and keep it on... I wonder how well you can track where someone is going or has been by looking at what cell towers they've been connected to as one travels? Yeah, nowhere near as precise, but still a very good vague idea when you piece other bits of info together.
The real question to me is, how long until they do away with the TIA blocker: cash? I use cash to by stuff I don't want tracked (2600 at B&N, lunch with a friend I don't want someone else to know about, etc.).
I'm not really worried until they decide to abolish cash. Then you'll have no choice but to have everything you purchase or anything you go being tracked (think toll booths, subways, buses, taxies).
Of course, all they really have to do is just strap a GPSr on the bottom of your personal transportation device. Not warrant required (ask Scott Pederson).
Hmm, and don't even get started with what is tracked with cell phone calls.
We've set a /32 route to one of our webservers and have a *.net and *.com alias for http://wildcard.artoo.net. At least this way customers know WTF is going on and can complain to Congress/ICANN, or just go try Google, but mainly VeriSign gets no traffic.
DNS admin needs to add Microsoft-Antitrust.gov. How many folks don't type 'www' these days? Rather, how many non-tech folks won't think to try http://www.Microsoft-Antitrust.gov when http://Microsoft-Antitrust.gov fails?
Raymond and Perens respond to McBride's open letter