ign

Послесловие о выступлении Dr. Shiva

Dec 05, 2020 20:14



Несколько дополнений по поводу двух последних записей о выступлении Dr. Shiva в Аризоне,  раздва.

Во-первых, я совсем пропустил один совершенно замечательный момент. Вот Dr. Shiva сравнивает свою «модель» с реальными результатами, здесь:



И никто из тех уважаемых людей, кто это смотрел, даже не обратил внимание, что в этой «модели» допущена очевидная ошибка: в матрице сверху явно переставлены местами республиканцы и демократы! Иными словами, кривая, построенная на основе этой модели, вообще не может чему-либо соответствовать, ни плохо, ни хорошо, никак. Ну или они использовали для расчетов одни данные, а в презентации рисовали совсем другие, что тоже довольно мило.

И ещё, очередно раз, повторю: на основании вот этой вот брехни нам предлагают признать выборы (как минимум в Аризоне) фальсифицированными.

Во-вторых, несмотря на все мои попытки хоть как-то попытаться понять, что Dr. Shiva пытался нам сообщить, я так и не смог найти смысл в его поисках корреляции между явкой и регистрациями. Это никак не помогает приложению его моделей, где мы используем конкретные известные нам числа зарегистрированных избирателей. Более того, сама идея по отдельность искать наилучшую линейную зависимости между процентом зарегистрированных республиканцев, демократов, и независимых (назовём их R, D и I) явкой T, математически абсурдна, потому что R, D и I не являются вполне независимыми, R + D + I ⪅ 1.

В третьих, после примерно 10 минутной презентации следует 20 минут ответов на вопросы, которые по уровню абсурда превосходят всё предыдущее. Вот небольшой отрывок:

Dr. Shiva: We know that starting in 2002 that these voting vendors it was brought out in one of default manuals have a feature called “weighted race feature”, it’s documented feature, the feature in the software which the secretaries of states have certified knowingly or maybe unconsciously, that must have carried this voting system software has a future called a “weighted race feature”, which allows voter’s vote, a candidate’s vote to be multiplied so what that means is that central principle that Justice Earl Warren ruled on in 1964, one person one vote… can be violated, it could be one person not one vote, and that is one explanation, but people in the public need to understand it, these electronic voting software systems, have a documented feature called “weighted race feature”, so the implementation of this is very possible, in fact it’s a documented feature starting as early as 2002.

Giuliani: Do you know if Dominion machines have that “weighted vote” feature?

Dr. Shiva: As I understand, they do have this feature, and remember, the more important thing to also understand is, are votes are stored not as while integer numbers, but as decimal fractions, which means if a candidate gets 100 votes, it’s not stored as 100, it’s 100.0215. Which is also quite alarming, which means, the fact that votes are stored as decimal fractions, means you can have external programs intersect, you know, communicate with those programs and alter them using external software programs which can also do weighted race features.

Giuliani: Would I be overly simplistic to conclude that the only reason you would want a machine with weighted vote feature is in order to cheat?

Dr. Shiva: It’s enabling cheating…

[…]

Rep. Cook: I just have a question for the doctor, early we heard from a state senator, I believe, in his presentation, when it had the vote totals, it has a decimal point, and numbers after, so this perfectly makes sense of what you are saying, one vote is one vote, two votes is two votes, but if ether is some way there is something added to it electronically to have a weighted vote as 1.25, so every vote for this person is 1.25, would that be the way these decimal points got there?

Dr. Shiva: Yeah, the issue is why our votes are being stored not as whole integers versus what it called floating point variables in computer science or in math, so when developer writes a program it should be very important, and I have been programming for many, means years, and the first thing you learn is when you are writing a software program you have to be extremely conscious on how you are declaring variables, this variable is an integer which means it can only be, you know, 0 1 2 3 4 5 6 7 8, this variable is a decimal which is also called a float or a double, which means it can be 1.1 or 1.3 and so on. So the fact that the vote count is stored as a decimal, when I first leaned about this, in my own race here in Massachusetts, was quite disconcerting, this means that consciously people are putting into the software a capability to have a fraction vote, and that’s what this leads to, and why is there such a feature or a capability? And I think this is probably the central question of our time, going back to 3/5 compromise of 1787, do we have one person one vote, or do we have one person not one vote, and in the era of electronic software systems, I think the courts need to step in and re-clarify one person one vote going back to 1964 the Equal Protection Clause of the Constitution.

Senator Borrelli: I am not surprised, it’s pretty much modus operandi of the Democrat history anyway, to have a poll tax and actually tax the republican to that vote goes to a democrat, so that’s what they did, they took that 30%, tax the republican vote and and sent this over to a democrat, so I am not surprised, that the democrats would want to tax the vote.

Я думаю, люди с техническим образованием оценят полное непонимание докладчиком разницы между fixed point и floating point, и особенно его «экспертное мнение» о том, что дробные числа отличаются от целых тем, что их может поменять “external program”. Забавно и то, что Dr. Shiva повторят придуманную им ранее концепцию “weighted race”, причём это учёное название ни один из присутствующих не в состоянии повторить. Джулиани говорит “weighted vote”, и.т.п.

Но самое бесценное в этом отрывке, это реакция сенатора Sonny Borrelli (между прочим, Majority Whip, не абы что). У него в голове «демократы забирают голоса» наложилось на poll tax (практика, используемая когда-то в бывших рабовладельческих штатах для ограничения доступа к голосованию), это в свою очередь наложилось на «демократы за налоги», и всё, его уже убеждать ни в чём не надо, мозаика сложилась.

В четвертых, мне всё это немного напомнил одну давнюю историю студенческих лет с лекции в МГУ. Молодой лектор, в ходе каких-то крайне запутанных вычислений, сначала взял производную какой-то функции, а потом начал полученный результат … интегрировать.

Я слегка офигел и посмотрел вокруг себя. Человек 20 в аудитории сидело и тщательно записывало всю эту галиматью, нимало не смущаясь. Набравшись храбрости, я обратился к докладчику:

- Простите, но вы интегрируете только что полученную вами производную. Результат у вас уже есть, вот он здесь чуть выше прямо перед вами.

- А да, вы правы.

И дальше пошёл. Никто даже не удивился.

Я смотрю на этих людей, которые внимательно слушают Dr. Shiva (и даже аплодируют), и вспоминаю ту лекцию.
See copy at dreamwidth for more comments; currently it has
.

go home, dr shiva, fake news

Previous post Next post
Up