So you do accept: - That it is possible for a corrupt programmer to manipulate the results. - That Outcome testing may not be enough to test for this - That we should just run with the system
Is this correct? On Tue, Oct 13, 2009 at 10:14 AM, Oscar Plameras <[email protected]>wrote: > Because when I put in 1000 ballots and the resulting count is 1000 ballots > and > correctly categorized as valid, invalid, and uncategorized, is all that > matters > in the election. > > On Tue, Oct 13, 2009 at 1:02 PM, Michael Janapin > <[email protected]> wrote: > > Sir, > > > > How can it be well and good if your check do not catch the absurdity in > any > > coding? It's bad. Because that code Mr. Locke just produced as an example > is > > a mischievous one. It will only take just one freaking trigger hidden > inside > > the source code. > > > > You said your 'check' will not look for such things. Again, that's the > > weakness of your proposal. On the other hand, Source Code review will > allow > > us to look for such things to assure integrity. > > > > You also said that what matters are the outcome. Don't you realize that > the > > outcome is determined by the code? > > > > And finally, I think sir, that you are working on a "Default Accept" > policy. > > This may not be so assuring when it comes to the code that will determine > > our country's next leaders. A saner policy should be "Default Deny." > > > > > > On Tue, Oct 13, 2009 at 9:09 AM, Oscar Plameras <[email protected] > > > > wrote: > >> > >> The check I propose is by Outcome. > >> > >> If my check will not catch the absurdity in any coding, that's well and > >> good. > >> > >> My check is not going to look for such things. What matters are the > >> outcome > >> or results. > >> > >> On Tue, Oct 13, 2009 at 1:40 AM, Robert Locke <[email protected]> wrote: > >> > Oscar, > >> > > >> > If I had a closed system that did the following: > >> > > >> > if (current_date < '2010-05-10) { > >> > do_a_normal_tally(); > >> > } else { > >> > do_something_slightly_different_but_not_too_obvious(); > >> > } > >> > > >> > How would your proposed "Testing the System by Outcomes" catch this? > >> > > >> > Maybe you set the system date to be 2010-05-10, and the ruse is > >> > revealed. > >> > > >> > So the programmer does this: > >> > > >> > if (current_date < '2010-05-10 && !obscure_hot_key_pressed) { > >> > do_a_normal_tally(); > >> > } else { > >> > do_something_slightly_different_but_not_too_obvious(); > >> > } > >> > > >> > How do you catch it then? Obviously this is an oversimplified > >> > example, but I'm curious to hear how you would expose it. Or are we > >> > supposed to blindly "trust" that this won't happen? If that's your > >> > position, then I would say it's a bit naive. > >> > > >> > "There is one safeguard known generally to the wise, which is an > >> > advantage and security to all, but especially to democracies as > >> > against despots. What is it? Distrust." - Demosthenes > >> > > >> > Rob > >> > > >> > > >> > > >> > > >> > > > > > -- > > Michael R. Janapin > > Systems Administrator > > PBTS Baguio City, Philippines > > http://mulingsilang.wordpress.com > > > > > > _________________________________________________ > > Philippine Linux Users' Group (PLUG) Mailing List > > http://lists.linux.org.ph/mailman/listinfo/plug > > Searchable Archives: http://archives.free.net.ph > > > _________________________________________________ > Philippine Linux Users' Group (PLUG) Mailing List > http://lists.linux.org.ph/mailman/listinfo/plug > Searchable Archives: http://archives.free.net.ph > -- Regards, Danny Ching
_________________________________________________ Philippine Linux Users' Group (PLUG) Mailing List http://lists.linux.org.ph/mailman/listinfo/plug Searchable Archives: http://archives.free.net.ph
