MrConfusion

What are you talking about? The BBC article references the research paper directly?

https://pmc.ncbi.nlm.nih.gov/articles/PMC4503401/

Yeah, it's not an issue for most people. I think the system works well. And the annual taxes are very simple to fill out. The report comes filled out for you, and you just need to make adjustments or approve it as is. Never had any issues with tracking or reporting my taxes.

I really enjoy living in Norway. I feel the culture and the way of life suits my preferences very well. There are of course always things one hope improves over time and things that could be better. But overall, life is good here I feel.

Employers deduct a part of your salary as taxes on every paycheck (usually monthly). But how much they deduct is based on your tax profile which is created by the tax authorities. It's typically based on expected yearly income. However, you are allowed (and encouraged) to update this number yourself.

So say you swap to a lower paid job, or work less overtime than you planned, you can change your expected annual income to reflect this in taxes paid. If you don't however, then you effectively pay too much taxes as your employer is basing the taxes on too high numbers.

Another reason is that you may have tax deductions that only gets computed into the total when you fill out your full tax report. So if you didn't bake those into your tax profile at the start of the year, you might get returns on your taxes (with interest).

You can also downplay how much taxes to pay through salaries, in which case you will owe back taxes when the full tax report is made.

In the Norwegian tax system, if you pay too much taxes trough the year from your paycheck, you get interest on the amount you paid to much. Likewise, if you pay too little taxes throughout the year, you will have to pay interest on the amount you have yet to pay. So the system is supposed to be balanced in that regard. The interest is on the level of a savings account (3.51% annual atm), so you could make an argument that saving that in a index stock or good bond is a better ROI though, so still recommended to try to not pay too much during the year.

The goal of pasteurization is to kill of harmful pathogens. If you do this early and package and store the milk in the right conditions it can be stable and safe for a long time.

If you don't pasteurize the milk and leave it for a long time, pathogens in the milk, such as bacteria, can potentially produce toxins. Boiling it at that point might not help, no, as it kills the bacteria, but can leave behind the toxins.

So pasteurization is very effective if done early, but you can't do whatever you want to the milk and then pasteurize it right before using it and everything is good.

If all things are equal except for mass. Then the object of higher mass will fall faster.

Hi. Physicist here. You are absolutely wrong. The mass of an object does not affect the magnitude of force of air resistance which acts upon a falling object. But the acceleration that object will have is given by Newton's second law as Force divided by mass. So a heavy and a light ball with the same shape will experience the same air resistance, but the heavy ball will "care less" and thus fall faster.

Well, this is simply incorrect. And confidently incorrect at that.

Vision transformers (ViT) is an important branch of computer vision models that apply transformers to image analysis and detection tasks. They perform very well. The main idea is the same, by tokenizing the input image into smaller chunks you can apply the same attention mechanism as in NLP transformer models.

ViT models were introduced in 2020 by Dosovitsky et. al, in the hallmark paper "An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale" (https://arxiv.org/abs/2010.11929). A work that has received almost 30000 academic citations since its publication.

So claiming transformers only improve natural language and vision output is straight up wrong. It is also widely used in visual analysis including classification and detection.

They are talking about .gov websites. Any website operated by the US government should, at least according to their own standards, develop for and test for users using Firefox. If this is followed in practice the article doesn't really cover.