IP and MAC address can be spoofed, so that the ISP will not know the true address and location of the originating client endpoint. Also if WiFi is used, then it becomes easier to hide the true endpoint from the ISP and mail server.
In Canada you only vote for a single canidate in an election, you need to understand is that in The USA we are normally voteing in 10 to 30 races and questions, so that if we used a paper ballot an an "X" the ballots would have to be physicaly counted 30 to 90 times or more to validate all elections. It would probaly require weeks to certify an election. All races would have to be completely counted before any race or question could be certified under the election rules as the certifiaction is for the entire election.
I am an election official and we haven't used paper ballots for the past 40 years. When we switched from OCR to Touchscreens, the time it took to vote was cut in half and the time go get the results went from days to hours. The OCR form was 17 by 40, both sides in the last election I worked that the OCR was used.
I remember a race in Califonia using OCRs where someone changes the markers and the OCR would not corectly read the ballots that were marked using the substituted pens. They had to hand recount all ballots for all elections.
NO. You gave the source to everyone you gave a binary. You only have to make the offer of the source if you don't give (distribute) the source and binary together to your customers. The only way a third party has a right to the source is if you make the offer instead of providing the binary and source at the same time. Now everone that redistributes the binaries has to redistribute the source as well, either bundled with the binaries or are responsible to make the offer to provide the source later themselves. If you distribute the source and binary together, you are off the hook for providing the source to anyone to whom you haven't provided a binary.
The letter of the GPL allows one to the use of a non GPL compiler/translator. If a vendor wrote their on compiler/translator, then the source is whatever the vendor uses to generate the binary. The module could load the firmware with a statement such as:
"load_firm(board_id,port,parm1,...)".
The compiler/translator could generate an inline routine to load the firmware and buffer with the firmware data. The module is compliant with the GPL as the source is written in The language of the compiler/translator. The Compiler/translator is not under the GPL, but if all Macros to generate the output are static loaded into the compiler/translator then the source of the firmware does not have to be delivered, nor the I/O sequences, as it is embeded in the tool. If all scripts, Makefiles, source for the module, and libraries used by the module are delivered, the the module is under the GPL. This satisifies the letter of the GPL, but not the spirit. But as the actual source of the tool used to generate a GPL module is not required, the source that is delivered does not have to give a view of whats under the hood, only a Macro view of the program flow. Therefore the source delivered is not as useful for modification, but the output could be linked into the kernel and still have a GPL kernel. This loophole could allow a vendor to provide a GPL driver for linux and still keep the actual interface closed, if they provide the Macro source for their compiler/translator and the support files to build a binary/module for linking into the kernel.
What happens to the piece of paper if it is incorrect. Printing it out before it is recorded and saving it for recounts allows fraud by printing out more than one slip of paper by the voter and having more votes in the recount than in the election. If it is printed out after the vote is recorded, how does the electronic recorded vote be corrected by the voter and another paper receipt generated to be put in the recount ballot box. Also leading to the possibility of a voter putting multiple paper recipts in the ballot box for recounts. Both leading to how to handle the problem if the number op paper receipts do not agree with the electronic ballots. Not to mention the problem if the printer ever jammed. This would require the ability to reprint receipts and learn how a voter voted or create extra receipts. Thus a paper receipt could create as many probles as it solves. It also does not help stop fraud by multiple use of the smart cards.
Slashdot Mirror
IP and MAC address can be spoofed, so that the ISP will not know the true address and location of the originating client endpoint. Also if WiFi is used, then it becomes easier to hide the true endpoint from the ISP and mail server.
In Canada you only vote for a single canidate in an election, you need to understand is that in The USA we are normally voteing in 10 to 30 races and questions, so that if we used a paper ballot an an "X" the ballots would have to be physicaly counted 30 to 90 times or more to validate all elections. It would probaly require weeks to certify an election. All races would have to be completely counted before any race or question could be certified under the election rules as the certifiaction is for the entire election. I am an election official and we haven't used paper ballots for the past 40 years. When we switched from OCR to Touchscreens, the time it took to vote was cut in half and the time go get the results went from days to hours. The OCR form was 17 by 40, both sides in the last election I worked that the OCR was used. I remember a race in Califonia using OCRs where someone changes the markers and the OCR would not corectly read the ballots that were marked using the substituted pens. They had to hand recount all ballots for all elections.
NO. You gave the source to everyone you gave a binary. You only have to make the offer of the source if you don't give (distribute) the source and binary together to your customers. The only way a third party has a right to the source is if you make the offer instead of providing the binary and source at the same time. Now everone that redistributes the binaries has to redistribute the source as well, either bundled with the binaries or are responsible to make the offer to provide the source later themselves. If you distribute the source and binary together, you are off the hook for providing the source to anyone to whom you haven't provided a binary.
The letter of the GPL allows one to the use of a non GPL compiler/translator. If a vendor wrote their on compiler/translator, then the source is whatever the vendor uses to generate the binary. The module could load the firmware with a statement such as: "load_firm(board_id,port,parm1,...)". The compiler/translator could generate an inline routine to load the firmware and buffer with the firmware data. The module is compliant with the GPL as the source is written in The language of the compiler/translator. The Compiler/translator is not under the GPL, but if all Macros to generate the output are static loaded into the compiler/translator then the source of the firmware does not have to be delivered, nor the I/O sequences, as it is embeded in the tool. If all scripts, Makefiles, source for the module, and libraries used by the module are delivered, the the module is under the GPL. This satisifies the letter of the GPL, but not the spirit. But as the actual source of the tool used to generate a GPL module is not required, the source that is delivered does not have to give a view of whats under the hood, only a Macro view of the program flow. Therefore the source delivered is not as useful for modification, but the output could be linked into the kernel and still have a GPL kernel. This loophole could allow a vendor to provide a GPL driver for linux and still keep the actual interface closed, if they provide the Macro source for their compiler/translator and the support files to build a binary/module for linking into the kernel.
What happens to the piece of paper if it is incorrect. Printing it out before it is recorded and saving it for recounts allows fraud by printing out more than one slip of paper by the voter and having more votes in the recount than in the election. If it is printed out after the vote is recorded, how does the electronic recorded vote be corrected by the voter and another paper receipt generated to be put in the recount ballot box. Also leading to the possibility of a voter putting multiple paper recipts in the ballot box for recounts. Both leading to how to handle the problem if the number op paper receipts do not agree with the electronic ballots. Not to mention the problem if the printer ever jammed. This would require the ability to reprint receipts and learn how a voter voted or create extra receipts. Thus a paper receipt could create as many probles as it solves. It also does not help stop fraud by multiple use of the smart cards.