Only one country in the world considers that the "proper" reading of the treaty and that's Israel.

The Geneva Conventions were originally about wars between countries. The additions apply also to military actions against non-state actors.

Additionally, the humanitarian requirements apply anytime military force is used to invade territory not belonging to the invader. Israel was defined by UN Mandate by its pre-1967 borders. Israel may claim that they can annex areas outside pre-1967 borders but that doesn't make it legitimate. Your basic argument as I see it is because Israel claims territory outside pre-1967 border the Geneva Conventions no longer applies. That's ludicrous though because then any invader can take over land, claim it as their own and be exempt from the Geneva Conventions.

Claiming occupied territory as part of your country is not considered acceptable now and has not been considered acceptable for much of the last 50 years. Whether it may have been in the past isn't relevant.

Treaties are international law.

So you can get around Article 49 simply by claiming occupied territory as part of your country? What is the tortured logic you have that makes that acceptable?

Article 49 of the Fourth Geneva Convention of 1949. I guess it didn't legally apply until 1949 so while territory occupied prior to that may not be relevant to the debate, the territory occupied from the 1967 war is very much relevant.

Since when do we have a precedent "that lands seized in a war are given to the inhabitants of the country that ends up with those lands?" Did international law change on this aspect prior to 1948 and then change back right after?

While you don't have separate stacks, all new processors have the NX bit that allows the OS to mark data as executable or not. This theoretically should prevent buffer overflows from succeeding. That said, it can break some old programs so I know in Windows, its only turned on for specific programs.

My only guess for why there is one stack is that its just easier to manage. Granted another stack shouldn't be that much extra work though so maybe I'm totally off base.

Managed code takes some control away from the developer but is the developer having that control for the best?

For example, think of the type of errors leading to security bugs. A lot of them have to do with buffer overflows primarily in the area of string manipulation. These are easy mistakes to make in C or C++. Hell Microsoft and others have tried to modify the C runtime library to have "safer" versions of string manipulation functions because these errors continue to happen. Now consider a managed language like Java or C#. It's not possible to overflow buffers provided the buffer management code is sound. Instead of each mostly average programmer going through the process to manage buffers and reinventing the wheel, we have a few particularly talented people develop the code who specialize in that type of work.

Also think about memory leaks. Firefox is a prime example of a great open source program developed in an unmanaged language. Probably thousands of people have looked at the code and still some of the most common complaints are memory leaks. My feeling as for why that would be is that C++ makes it so easy to forget to deallocate memory that with a million line program, its just too difficult to find every possible leak. A managed language never has that problem. We have a group of particularly talented people develop a garbage collector and memory management system and the average developer will never have to worry about it.

I know its nice to get up on your high horse but let me explain to you our situation. In Wisconsin our polls require at MINIMUM 4 poll workers, a chief voting inspector, an assistant chief voting inspector, and two people to sign in voters. The chief and assistant chief are usually of different political parties and rotate responsibilities for each election. The sign in workers use two different lists and the lists much match at all times.

We receive a set number of empty ballots. The number of ballots cast must match the number of ballots counted and any discrepancies must be accounted for (e.g. a voter made a mistake and needed a different ballot). Additionally the number of ballots cast must match the number of people signed in. I know this may sound silly but this is very important and we have to prevent discrepancies.

We use oddly enough a Sequoia optical scan system. This system is publicly tested a week before the election and any person may view this. Sample ballots are entered and the counts are verified to match. The pcmcia card with the voting software is put into a sealed plastic bag with a numbered pull tie. You can't open the envelope without wrecking it or breaking the pull tie. We verify the pull tie matches the previously recorded one. Once the pcmcia card is inserted a new numbered pull tie is used to lock the pcmcia slot. Also all the bins holding the completed ballots have numbered pull ties. All these numbers are verified when the pull tie is to be broken.

Our ballots require voters to draw a line next to the name of the candidate. The machine will reject bad ballots and give it back to the voter or if there's a write in they will be kept in a separate bin from the normal ballots. At all times we have a ballot where a voter made a mistake and the voter isn't there to fix it we try to figure out the voter's intent. If we can, we make a new ballot with the same votes and insert it into the machine, we must keep the original though.

When counting begins, we get the results from the machine's printout. We don't verify those results but we do verify the number of ballots and make sure that matches the number the machine says were cast. In the event of a close race or a candidate demands a recount, the ballots will be manually inspected. Anytime ballots are moved, any member of the public is allowed to witness it.

I actually think it has more to do with the idiots in public health who think what they say doesn't matter to the public. You can't call something a pandemic without scaring the shit out of people because for most people it implies something often lethal. What the government should have done is considered the effect of using the word pandemic and decided on better terminology that more clearly implies what they're saying. Pandemics are basically fast spreading diseases that infect a large number of people. They could have said that and I came up with that in a minute. I trust they can find something better if they spent a few days on it.

Well my understanding of upgrading to Vista and 7 is that the setup does a clean install and then adds the relevant registry entries from the previous system. It basically is a clean install.

XP Mode doesn't exist for Vista so I'm not sure how that can be.

I'm an Ubuntu user too and I've never had an upgrade work correctly. Not once.

Did you get paid for that post?

I don't know if I'd call it fiddling as much as time. You need to install a clean copy and not enter a product key. After the clean install finishes, run an upgrade on your clean install and enter the product key that time. Microsoft even recommends that if you need to do a clean install.

Uh how could XP Mode NOT run half of their crap? It IS Windows XP.

*nods* That makes sense. Do you know a good reference that talks about that? In my cursory look, I didn't find any details on the NT arch's memory layout, although Win 9x used the upper 2GB for kernel memory. Looking back at your original comment, I guess I'm not totally convinced that there wouldn't be driver issues with PAE enabled, especially considering the thousands of drivers floating around out there.