https://en.wikipedia.org/wiki/Shooting_of_Oscar_Grant
Similar circumstances. The cop got 2 years for involuntary manslaughter.
Spoiler alert: The deplorables are the heroes of that movie and the people hunting them are the villains. The deplorables triumph in the end. The whole movie is about how things are not what they seem.
Yes you are on the right track. Here's the methodology they used:
First they plotted registered voter turnout (number of actual votes / number of registered voters) against age.
Then, they fitted a 6th-order polynomial to this plot. This function lets you predict voter turnout at each age.
Then they multiplied these predictions by the number of registered voters at each age. This gives them the predicted number of votes at each age.
Finally they compared the predicted number of votes at each age against the number of actual votes at each age. This give them a correlation of 0.997.
The issue is they are just making predictions on the data they trained on to begin with, so of course the fit is near perfect.
What is interesting is that they trained on data aggregate across countries but the correlation is very high when making single county predictions. This seems to speak to correlations between counties.
What that proves is that voting turnout is highly correlated between counties, which is not a surprising result.
The problem is that they fitted a curve, then used the curve on their training data, and found a correlation of 0.997. There was no attempt at generalization.
Yes exactly. To be precise and fair, it should be noted that they fitted a curve to the aggregate data across the 9 counties, then compared their curve fit to each county individually. So it does say something about the correlation between the counties.
For the statistically-minded people here, this is the methodology they used to arrive at the 0.997 correlation.
First they plotted registered voter turnout (number of actual votes / number of registered voters) against age.
Then, they fitted a 6th-order polynomial to this plot. This function lets you predict voter turnout at each age.
Then they multiplied these predictions by the number of registered voters at each age. This gives them the predicted number of votes at each age.
Finally they compared the predicted number of votes at each age against the number of actual votes at each age. This give them a correlation of 0.997.
Yeah, it's crazy just how many we're having in such a short time. I don't remember it being this insane in recent years.
El Paso, Pittsburgh synagogue, Parkland school, Sutherland church, Las Vegas concert, Plano...
El Paso, Pittsburgh synagogue, Parkland school, Sutherland church, Las Vegas concert, Plano...
I found the details in this article to be very interesting: https://nypost.com/2021/04/02/text-messages-that-allegedly-sparked-matt-gaetz-probe-emerge/
It seems this incident with Matt Gaetz and Joel Greenberg sparked the investigation into Gaetz. It looks like Gaetz was suspected of being involved in the operation to forge IDs for trafficked minors when he visited Greenberg in this office.
However, later in the article, it says Gaetz actually lost his own ID, and Greenberg was helping him expediting getting a replacement ID for himself.
Could it be that Gaetz was simply visiting Greenberg to get a replacement ID for himself, and was swept up into the investigation into Greenberg because Greenberg was using the same office to forge IDs?
Wait what? Are your saying kids aren't eating lunch?
Ah yes, thanks. I had missed that on my first read through of your post. I found his book on network security and his papers on distributed systems. I agree that this person has a lot of expertise around computer systems that would be central to electronic voting. However, I'm unable to draw any conclusions beyond this. i.e. how does one go from "could do a thing" to "did do a thing"?
I'm not familiar with his whole body of work, I'm just basing my opinion on this one paper, so if you have other papers of his that might be more relevant, please post them, I'd be interested in analyzing them.
But based on this one paper, I'd say he's working on pretty mundane stuff as far as machine learning is concerned. It's basically statistically sampling techniques for training classifiers. It's foundational machine learning stuff that probably thousands of researchers are working on every day. it doesn't tell us a whole lot about his expertise in voting or anything like that.
To be more concrete to others reading this post who may be less technically inclined, the kind of manipulation the paper is talking about is limited to training data preparation for machine learning algorithms. It's no more complicated than writing an Excel macro to change some data in a spreadsheet. It's the kind of mundane data manipulation that anyone with a computer can do, so it doesn't tell us anything about whether they are an expert at manipulating voting systems.
As a machine learning practitioner, I can tell you that your understanding of SMOTE is not correct. I suspect it's because you are not familiar with the jargon used in the field of machine learning. Boosting is a technical term that's common in the field. Here's a primer if you are interested: https://en.wikipedia.org/wiki/Boosting_(machine_learning)
Because it's not a law, it was an executive order.
The question you really should be asking is, why did President Trump have to resort to executive orders to accomplish most of what he wanted? Why were there so few actual legislations?
Corporations help draft legislation all the time. You should look into the American Legislative Exchange Council.
Now apply this idea to voter registration...
The word you're looking for is "reactionary": people who don't want things to stay the same, they want to return to an earlier state. However, most right wing people don't like being labeled reactionaries, due to its negative connotations.
What about the non-whites who are already in?
How will you get 50 state voter ID without federal intervention in state election law? You think California will pass voter ID willingly?
Are there any sources backing up the claim that this guy was known to the FBI?
Is that really how you want EOs to work? To have that much power? Keep in mind who's writing EOs right now when you ponder this question.
Well the El Paso shooting happened in 2019. Then the Coors Brewery shooting in early 2020. After that, things became relatively quiet. Until recently.
If you want to look for patterns, I'd say this coincides with covid lockdowns.
Are you talking about this? https://en.wikipedia.org/wiki/Shooting_of_Oscar_Grant
That was a male officer.