Benchmarking Reality

How to bring a Science that has wandered off into the realms of fantasy, back down to Earth.

Defining Benchmark Testing

Benchmarks - also known as metrics, outputs, or reference points - are a means by which businesses measure the effectiveness of their software or service. Benchmark testing sets expectations for the quality of a product, and then attempts to measure how well those perceptions match reality.

A benchmark must be quantifiable, measurable, and repeatable. If it does not meet these three specifications, then it does not qualify as a metric. 
‍

This type of testing is a mainstay in industry and business where the actual performance of your product is paramount. A company or busniess cannot afford to dwell totally in the realms of fantasy with their product. If a product that they made didn’t do what they claimed, the market would very quickly find out and that company would go bust very quickly.

Let’s find an anology: A company designs a product that they call a Magic Bean, they spend a Million Dollars in RnD that transfects this Magic Bean with the genetics of a Giraffe. Their Computer models show that when someone eats this Bean they will trasnfect themselves with Giraffe genetics and grow 4 inches in a year.

This company accept their computer models as reality and they stand to make $10m profit in the first year. They release their product and 100 people buy and eat the Bean. 98 people don’t grow 4 inches that year and demand their money back. The Company goes bust.

It seems absurd to even have to make an analogy of it in business terms. Not a single business would release a product that had never been tested in the real world! Well… Welcome to the World of Modern Science!

Circular Reification

This is my own concoction of two logical fallacies , Circular Reasoning and Reification fallacy. This is how I would describe the Benchmark Testing done on large swathes of Modern Science.

I am going to focus on this post and in the Control Studies Project on Gene Sequencing but this fallacious way of “testing” can be seen right the way across science, especially all science used in the political sphere.

This is a very quick *Primer* to how Genes are sequenced.

What is happening in its simple form. 1. A Dye is INTENTIONALLY added to a sample. 2. This Dye is attached to a primer which ONLY Flouresces when bound to target sequence. 3. An Led is shined at the sample. 4. A Camera measures how much of the Dye has fluoresced.

To simplify this even further: They are INTENTIONALLY adding something to a sample and when they SEE it they CLAIM that is because of some incredibly complex and accurate UNSEEN process.

This raises two main questions:

1. Does the Fluorescent Light show actually represent what they claim it does?

2. Is the specific sequence found accurate?

When dealing with the first question, it can again be further split into two questions.

1. Do Nucleotide exist and act in the way they are claimed?

2. Are the chemicals and Dyes added to the sample interacting with the assumed nucleotide in the way they claim?

When dealing with the second question this can be split into two further questions

1. Can the sequences be replicated?

2. Can you find a known real and quantifiable outcome from the found sequences?

The first Substack I did describes the power of Controls. Where you A/B test ingredients to make sure the assumed independent variable is causing the observed effect. This can be very effectively employed to answer all of question 1. but when answering question 2. it may not be so clear.

The Alignment process of “finding” a complete genome is done totally In Silico ( in a computer). It takes the Fluorescent light display output in small snippets of nucleotide and tries through programming to force them into one continuous sequence. This is done by digitally altering the the small sequences by removing parts or slightly adding parts or taking rough alignments. This is entirely subjective and human manipulated. They have never just found one complete genome in one go and in the case of Sars Cov 2 have never been able to assemble the exact same genome twice even using the same sample and machinery!!

Now where Controls are great for showing observed effects may not be caused by the Independent Variable in question, we have a slightly different problem when it comes to Sequencing and alignment:

All of the parts of the scientific method have not been shown to be real (It’s all Unicorns LOL), there is no BENCHMARK. The observed effect or outcome is a long string of the same thing that is the independent variable. I call this Circular Reification.

So let’s make another analogy: Take a calculator, it is objectively extremely accurate. It would only not be accurate with Hardware/battery issues which are rare. It is extremely repeatable, every time you input the same data, you get the same answer. On the surface of it, no sane person would question its reliability.

However you would never use a calculator to prospect. For instance you would never use a calculator to find the perfect spouse or to find a murder suspect.You would have to allocate real variables to numbers and give outcomes reified meaning:

If the suspect has Brown hair let that be the number 9, If they are right handed let that be the number 6, If they are tall let that be the number 25. If you add them all together if the number is more than 100 you have found your murderer. Absurd right?

Well this is what happens without a Benchmark tying your proposed outputs to real life.

This is the type of internal Benchmarking that is offered as accuracy rates when it comes to find genomes of viruses. They take the exact same input data,put it into the exact same machines and apply the exact same Alignment manipulations. This is the equivalent to benchmark testing a calculator by continually punching in the same numbers and when the same answer is found they point and say “Extremely Accurate”.

It isn’t even quite a clear cut when it comes to virus Genome Sequencing. They have taken the same sample, put it into the same machne and conducted the same Alignment methods and STILL failed to replicate the genome. They just invent a new term for this benchmark failure “Variants”. There are hundreds of thousands if not millions of variants of “Sars Cov 2”, everyone just an inaccuracy in the procedure.

So how do we conduct a Benchmark?

Benchmarking comes from the idea of construction, literally scoring a mark in a bench to set a level, or in its original design setting in stone.

So we have to take our scientific method and FORCE a Benchmark of an OBJECTIVELY REAL QUANTIFIABLE variable into the observed effect and START there.

Let me describe a Benchmarking Test of the Genetic Sequencing process:

Start with a pool of Known, Quantifiable Outcomes. In this case I choose 100 different animals from across the animal kingdom. Lions, Newts, Eagles, Dolphins etc etc. I take a saliva sample from ONE animal of my choosing and give it blinded to a geneticist. The task is simple; Identify by genetic sequence which animal I chose.

So we have Benchmarked into reality an expected out come to arrive at and IF the process of Genetic Sequencing is accurate *SHOULD* be able to find this animal nearly every time….right?

I have asked this question and proposed this experiment to hundreds of accredited geneticists. Every single one scoffs at the proposal and claims this can be done with ease. It might surprise you that when I follow this up with the question; Could you show me a study like this please? Every single time they draw a blank.

It may surprise you even more (my Jaw hit the floor when I found this information out) that not only had this type of benchmarking never been done for animals, it had never been done for Forensic Science!!!!!!!!!!!!!!

I found this article by the Marshall Project The Imperfect Science of DNA testing. In it it described the search for these very Benchmark tests from the National Institute of Standards and Technology.

They admitted when asked that they had never been done! So the Innocence project set up by Lawyer Barry Scheck forced them to do one. In the article is the link to the 70 page report published by NIST (use Wayback) and it shows the result as above. Out of 108 labs given blinded benchmark tests they returned an accuracy rate of just 6%!!!!!!!

They did even more Benchmarking tests with “single samples” i.e material from one individual. Out of the 17 labs, just One corroborated the initial finding. An accuracy rate again of less than 10%!!!!!

SO we now have identified in our arsenal TWO scientific tests that can falsify practically every part of the virological methodology, not only that but can be used as powerful safety checks for ALL science that is being used as a political vehicle.

1. The Control

2. The Benchmark

So let us just flesh out what a Benchmark test will look like in the Control Studies Project:

Because we have not even identified the independent variable with virology we can’t use the “animal model” as described before. But worry not we can Benchmark our reality with an action that is quantifiable and real: KNOWINGLY adding something or not.

If we take 20 Cultures and a few different “viral titres”. The claim is that these viral titres have very specific measurable genomes. If we chose a “Viral titre” from the ATTC and randomly select a few of the cultures to put them in, this acts as a benchmark whereby an absolute action has objectively occurred.

We get these 20 cultures Gene Sequenced and give the raw data blinded to geneticists.

If they can identify which cultures have the “viral titres” in and what that “virus” is then they have adequately Benchmarked and proved it is accurate in the real world.

If they fail it proves that the entirety of Genetic Sequencing including PCR (because this is predicated on knowing the full genome and is considered ONLY truly verified by WGS) does not constitute as evidence for the existence of viruses and would result in the entire Viral Paradigm being falsified in conjunction with the Cell Culture, TEM and Antibody Controls.

Given the Benchmark failures of the Forensic Evidence how do you feel the Viral Genetics Benchmark tests are going to fair?