It is not uncommon to hear somebody, perhaps yourself, comment that a certain product on Amazon got really bad reviews and therefore you will avoid it, or reviewing the comments about a doctor or RMT and seeing 10 negative posts and 2 positive posts, then concluding the person should be avoided.

The problem with taking online reviews about anything seriously is that without additional, hidden data the statistics just do not work the way you may think it does.

Firstly, there is obviously the problem of bots and hired guns that try and skew the reviews by getting paid to post positive or negative reviews. Ignoring this for a moment, the second issue I am addressing here is the fact that you normally do not know the total size of the population that interacted with the product or service.

Make/Model: BMW M235i Gran Coupé 2020
Colour: Boring Alpine White
New?: New (Demo)
Engine: 2.0l I-4 Turbo
Power/Torque: 224kW (301 BHP) / 449Nm (331 lb·ft)
Average Fuel Consumption: 7.6 - 10.3 l/100 km (Highway - City)
Performance: 0-100km/h in 4.8s, 250 km/h top speed
Transmission: 8 speed automatic
Owned: September 2020 - October 2023

If a small company with limited budget and resources releases a product or enhancements to a product, I am fine with the odd bug here and there. Not because I like bugs, but I thoroughly understand the challenges that go in to developing, testing and deploying code. The fewer resources and budget you have, the harder it is to eradicate bugs.

However, if your company is worth USD2,000,000,000,000 then you have NO excuse to release iOS 14 / tvOS 14 with these major bugs:

1. Airplay from an iPhone 11 Pro on iOS 14 to the latest Apple TV with tvOS 14 no longer works reliably - I get disconnected every 5 seconds, with the tvOS crashing and rebooting every time. This worked perfectly for months on iOS/tvOS 13 with no issues.
2. With the new sound detect accessibility feature turned on, it fails to recognise my cat miaowing. This system is supposed to aid hard of hearing people, but it discriminates hard against my Bengal cat. Did Apple not test this on different breeds of cats? So way-cist.
3. When receiving a new sound notification alert, if you fail to unlock the screen properly the notice disappears and there is no history of it anywhere - so you do not know what sound caused it.

I am not sure who is Director of UI and other visual experiences over at Apple but their web page for the upcoming Big Sur release of macOS has to be the most awful design in Apple's history (fonts too big, too many obnoxious colours, non-harmonious colour schemes, erratic zoom scrolling effect, way too many bubbles, too many images, disorganized layout, headache, pain, suffering...) :

Big Sur, almost illegible

I actually miss the old UI design:

A person-woman-man-camera-TV just announced his new creation - DiceKeys. The point is that you roll a set of 25 special dice, each with 6 sides I presume, that will land in a truly random pattern that can be scanned using a custom application, that will then interpret the orientation and faces of these dice and generate a master key with lots of entropy, about 2¹⁹⁶. That is the same entropy of an alphanumeric password of 33 characters [a-z, A-Z, 0-9] which is 62³³.

But why overcomplicate something already so confusing to most people as good security? He claims you can place a lid on the box and preserve the layout of the dice as a secure way to store your dice keys.

However there is absolutely no added security compared to writing a secure 33 character password on a piece of paper and storing it where you would have stored the DiceKeys. Ink will last a long, long time. As far as getting a good secure password? Easy. If you really want to use a mechanically random method, write the letters a to z, A to Z and 0 - 9 on small pieces of paper. Toss them in a basket. Cover the basket and shuffle them by shaking the basket. With the cover on, pick a piece of paper. Write it down, return it and repeat 32 more times. There is your person-woman-man-camera-TV über secure password.