The compiler can take the good friendly Open Source, compile like normal (for the most part,) but then inject some nastiness wherever it was programmed to.
No one is claiming the open source solves everything. It's a necessarly feature of a transparent voting system. There are many other rules, procedures, etc. necessary to make it all work correctly.
What I recommend: Humans performing pencil & paper counting under scrutiny of video camera and representatives of competing parties. Distribute the video tapes of the counting process on the Internet, and maintain archives for at least 12 years.
Sounds like a security nightmare to me.
Are you going to prepare the ballots without computers too? That's where many problems occur.
Sure we need representatives of competing parties. Still, consider that Democrats in the South are really Republicans.
Consider a presidential year general election. 200,000 pollsites and an average of 500 ballots per pollsite. How many millions of counters will you need? How many millions of scrutineers of counters will you need? How many millions of cameras? How many people are needed to coordinate, train, supervise the whole process? How much will all that cost?
Computers are great at counting and following rules.
Your proposal (which, fortunately, has zero chance of adoption here in the U.S.) has one major flaw: you want to use people for exactly what they're really bad at doing: counting and following rules.
Industry shills showed up to oppose the resolution that makes this report mandatory. Here are some letters of opposition:
http://www.openvotingconsortium.org/ad/242-opposit ion.pdf
They got no where. We have to show up and make our case.
I talked with Schneier some about OVC around a year ago. I did not contact him about being on the panel because I doubt we'd be able to get him scheduled. I think he lives in Minnesota or Wisconsin -- somewhere around there. If he was really willing and able to do it, I'd try to have him included. At a bare minimum, we'll ask him to provide testimony.
This works fine where ballots are very simple and populations are not dense. Hand-counted hand-marked paper ballots are used for less than one half of one percent of the U.S. vote.
However, 80 percent of the population in the U.S. lives in urban areas. Los Angeles County has to support 9 different languages. By law, we'll also have to have a system in place in every pollsite to accommodate people with various disabilities.
This ballot has 44 contests and 95 candidates. Care to estimate how long it would take to count manually?
http://www.smartvoter.org/uvote4/uvote4.cgi?addr=1 719++WILSHIRE+BL&date=2002/11/05&zip=90403
Now consider this ballot with 44 contests and 95 contests. Please describe how many people will be needed to hand count this and how long it will take.
They vote for one thing at a time -- for example, Member of Parliament. We have many on the ballot. We investigated one in Santa Monica CA that had 44 contests and 95 candidates. Another in NM had 76 contests.
Slashdot Mirror
No one is claiming the open source solves everything. It's a necessarly feature of a transparent voting system. There are many other rules, procedures, etc. necessary to make it all work correctly.
What I recommend: Humans performing pencil & paper counting under scrutiny of video camera and representatives of competing parties. Distribute the video tapes of the counting process on the Internet, and maintain archives for at least 12 years.
Sounds like a security nightmare to me.
Industry shills showed up to oppose the resolution that makes this report mandatory. Here are some letters of opposition: http://www.openvotingconsortium.org/ad/242-opposit ion.pdf
They got no where. We have to show up and make our case.
I talked with Schneier some about OVC around a year ago. I did not contact him about being on the panel because I doubt we'd be able to get him scheduled. I think he lives in Minnesota or Wisconsin -- somewhere around there. If he was really willing and able to do it, I'd try to have him included. At a bare minimum, we'll ask him to provide testimony.
This works fine where ballots are very simple and populations are not dense. Hand-counted hand-marked paper ballots are used for less than one half of one percent of the U.S. vote. However, 80 percent of the population in the U.S. lives in urban areas. Los Angeles County has to support 9 different languages. By law, we'll also have to have a system in place in every pollsite to accommodate people with various disabilities. This ballot has 44 contests and 95 candidates. Care to estimate how long it would take to count manually? http://www.smartvoter.org/uvote4/uvote4.cgi?addr=1 719++WILSHIRE+BL&date=2002/11/05&zip=90403
How many things were you voting on?
1 719++WILSHIRE+BL&date=2002/11/05&zip=90403
Now consider this ballot with 44 contests and 95 contests. Please describe how many people will be needed to hand count this and how long it will take.
http://www.smartvoter.org/uvote4/uvote4.cgi?addr=
See the difference?
I know, you were in Sweden and all. See you soon. Thanks.