Another similar “shortcut” I’ve heard about was that a system that analyzed job performance determined that the two key factors were being named “Jared” and playing lacrosse in high school.
And, these are the easy-to-figure-out ones we know about.
If the bias is more complicated, it might never be spotted.
Yes, “Bias Automation” is always an issue with the training data, and it’s always harder to resolve than anyone thinks.
The annoying thing is that “I’ll go to work tomorrow” and “I’m going to go to work tomorrow” have subtly different meanings to English speakers, but good luck trying to come up with a rule to explain the difference to someone learning the language.
The spoken language has more vowel sounds than the alphabet has vowels, even when Danish has added three extra (æ, ø and å).
That’s one thing I think English at least did right. Other languages added extra letters and/or diacritics to try to capture all the variations on vowel sounds. But, in most languages there are far more vowel sounds than there are vowels. So, don’t make things unnecessarily complicated by adding extra letters. The one language that seems to do it basically right is Spanish, where there are only about 5 vowel sounds and they use the accent character not to show that a letter is pronounced differently, but just to cue you in on which syllable is accented when it might not be the one you expect. (With a few minor annoying variations, like el and él).
English’s future tense
There are various future tenses.
Future Simple / Simple Future: Will + [base form] – I’ll eat that later; or Going + [infinitive] – I’m going to eat that later.
Future Continuous: Will be + [present participle] – I’ll be eating that later.
Future Perfect: Will have + [past participle] – I’ll have eaten that later.
Future Perfect Continuous: Will have been + [present participle] – I’ll have been eating that later.
There’s also using the present continuous to talk about the future – I’m eating that tomorrow.
Also, the simple present – I eat that tomorrow.
English is flexible, but it’s also weird. There are a lot of distinctions that matter to native English speakers but that are really hard to put into rules. Like “will” vs. “going to”. They have slightly different meanings, but good luck coming up with an easy to understand rule about when to use each version.
What are some of the issues with Danish?
Do you know if that’s unique to Estonian, or also true of Finnish? AFAIK, Finnish (and Estonian) are a weird language branch in that most of Europe is Indo-European. Even distinct languages like Italian and German are more related to each-other than Finnish.
What does willpower have to do with it?
Phonetically, put: /pʊt/ vs. putt: /pʌt/
ʊ is the sound from words like “book”, “hook”, “pull” or "should.
ʌ is the sound from words like “gut”, “double”, “butter”, “luck”, etc.
I think English allows you more different ways of doing things than most other languages. The future tense being “going to X” and one of the past tenses being “used to X” means that new English learners don’t need to spend as much time studying yet another verb tense.
OTOH, the spelling and pronunciation is such a massive hurdle compared to a simple language like Spanish.
That’s part of it, but I think a bigger part is all the countries that colonized the British Isles. English has elements of Germanic languages like German, Dutch, Old Norse, etc. It has elements of Latin languages from Latin itself to French. The British Empire definitely resulted in words being brought back from the various colonies, but the English they spoke then was fairly similar to what we know today. It was already this weird, bastardized Germanic / French language.
All languages have quirks, but English is awful.
I only realized that the more I studied other languages, making me reflect on English.
Like, English doesn’t have a future tense. It seems like a pretty basic thing, but in English you say “I’m going to X”. Why do you use the verb ‘to go’ there? Why is that the way English creates a future tense? If you’re headed to the store now: I’m going to the store. If it’s happening later: I’m going to go to the store. WTF is this bullshit? “going to go”? Just stop and think for a second about “going” and “go” in that phrase.
And the verb “to do”, why is that part of questions in English? Statement: You have a dog. Question: Do you have a dog? What does “to do” have to do with any of that? Why is “doing” the verb that somehow is used to turn a statement into a question?
And then there’s “to use”. Using is to take, hold, deploy, consume… so why is it sometimes part of the past tense. Sure, you can say “I walked to school”, but if you want to talk about habits or routines: “I used to walk to school”. Why is “to use” even involved there at all?
That’s not even accounting for spelling and pronunciation which is just ridiculous in English.
We have a letter ‘k’ that reliably makes a certain sound, and a letter ‘s’ that reliably makes another sound. But, a huge variety of words use “c” which can make a ‘k’ sound like cat, or an ‘s’ sound like city. The letter ‘c’ has no sound of its own, it’s just a randomizer machine for one of the other useful sounds. The letter ‘g’ has one sound that no other letter makes, in words like “grip” and “great”. There’s another letter “j” that makes a different sound, like in “jet” and the name “Jim”. But, for some reason, sometimes the “g” makes a “j” sound, so “Jim” and “gym” have the same sound but completely different spellings, leading to bullshit like the confusion over how to pronounce “gif”.
English has roughly 20 vowel sounds, depending on the accent, but the vowel letters are ‘a’, ‘e’, ‘i’, ‘o’, ‘u’, and sometimes ‘y’. So, you’d think that at least those 5/6 are sorted and the other 20 come from combinations, right? Nope. In British English, for some reason “can” and “can’t” get a different vowel sound for the ‘a’, despite “can’t” being a contraction for “can not”, which literally contains “can”. The letter “u” can sound different between “put” and “putt”, even though you’re just tacking a ‘t’ to the end of that combination of letters. If you tack an ‘s’ on the end it doesn’t change, but if you tack an ‘e’ on the end… whoa, an entirely new sound different from both “put” and “putt”.
I’m glad the world is slowly converging on one language that allows everyone to communicate with everyone else, but it sucks that the language that came out on top is English.
See, at the end? What you’re describing is timezones with a different name, and more fine grained so we have more of them. This makes it harder.
No, timezones are intended for people who live in them to be in a time that’s roughly coordinated with other people living in the same area. I’m saying that’s unnecessary. There’s no reason that 12:00PM should be close to the time that the sun is at its peak. That already isn’t true for people in the west of China. For them it’s normal to think that 3PM is when the sun is at its peak. What I’m suggesting is that that be applied worldwide.
If, for some reason, you want to know where the sun is relative to someone else on the planet, there are plenty of ways of doing that. I suggested some. That doesn’t mean that you need time zones.
Business hours are correlated to where the sun is
There’s a correlation, sure. But that isn’t enough information to know if a business is open, especially if it’s a business in another country which has different cultural ideas about when things should be open. Business hours are no reason to stick with clunky time zones.
People communicate with people in parts of the planet where everyone would say it’s a different time because the sun is in a different part of the sky.
No, they say that because it’s what they’re used to. If they were used to using UTC they’d say it’s the same time. They already do that for some things, because time is understood to be related to causality. As in, “Did that happen before or after the bridge collapsed?” People in different time zones will agree that in that sense, time is the same for everyone, even if they’re using a different time zone for historical reasons.
Again, we already have UTC. People use it where it makes sense.
And don’t use it where it would also make sense for historic reasons. People also use US customary units not because “they make sense”, but because of historic reasons.
Like I said, people who use Celsius know when to wear a coat.
So do people who use Fahrenheit.
But if we’re maintaining that 0 and 100 are special numbers, then Fahrenheit maps hazardous conditions more neatly to those numbers.
I completely disagree. 0 Fahrenheit is very cold, but there’s nothing special about that temperature. You need to start dressing for cold conditions long before it gets that cold, and if you dress for cold conditions you can easily handle temperatures well below 0F. 100F is also nothing useful. Yes, it’s very hot, but you start needing to take precautions for heat long before it hits 100F.
Basically the Fahrenheit scale has nothing particularly useful at 0 or 100F. The Celsius scale has useful things at 0C and 100C. 100C is not useful for weather, but 0C is very useful for weather because it tells you whether it’s likely to be icy out.
Crude is approximately the same as water, about 0.8 to 0.9 g/mL. But, even if it were significantly less dense, like gasoline (0.74 g/mL) it’s still good for an order-of-magnitude calculation. Knowing that 1L has a mass of 1kg is especially useful since many of the liquids we commonly encounter are water-based.
I’m imagining a tank wearing a fez.
Below zero is when air temperature starts to get hazardous
Hazardous in what sense? If you’re not wearing proper clothing, lower than 10C can be hazardous. Many hikers who get lost get hypothermia even if it’s above zero because they were dressed for an energetic hike, not sitting around waiting for a rescue.
If you are properly dressed, -10C is no big deal. Many people do outdoor sports for hours when the temperature is well in the negatives.
IMO, if you’re within 10C of ideal room temperature, you may be uncomfortable but you’re probably not in danger. But, if the temperature is above 30C or below 10C you need to take precautions: shade and water in the case of hot weather, warm clothing in the case of cold weather. I don’t think there’s anything special about 0C for humans, except for the fact it’s when water turns to ice, rain turns to snow, etc. If you have the right gear, 5C, 0C, -10C and -20C are all survivable, possibly even comfortable. You just need more and more specialized gear as the temperature gets lower.
I gotta ask, do you use a thermometer to boil water?
No, but I use a thermometer (built into the electric kettle) to prepare tea. Greens want to be brewed at 75-80C. Whites are often about 70C. Oolongs are about 95C.
The numeric value associated with boiling water has no impact on cooking, because the boiling water doesn’t care.
But the human who is doing the cooking might care.
Tineszones exist because we have two uses for time
Not really. Time zones exist for 1 reason: it was too difficult for each town to have its own time, especially when it came to train schedules. So, they were organized into zones so that 6pm in Baltimore and 6pm in Philadelphia were the same. But, people were still used to having 12 pm being the time when the sun was at its peak, so NYC was put in a different zone from Los Angeles.
To communicate across wide stretches of the earth, you need a way to know where the sun is wherever the person you’re talking to is
You normally don’t need to know where the sun is, you need to know if it is normal business hours. Or, if it’s a friend, what their schedule is like and if this is a convenient time for them. You can search for the time in that other place and guess that maybe their business hours are 9 AM to 5 PM, but that isn’t always true across companies and especially across cultures. What you really need to know is something like “what are Dimitri’s business hours” which is easier if everyone uses UTC. If you ask “What are Dimitri’s business hours” and you get the answer 8h - 16h EET, now you need to figure out what “EET” means. But, if you get 6h - 14h UTC and you’re also using UTC, there’s no conversion needed.
is the man in Madrid likely asleep if I’m eating lunch?
If that’s what you need to know, what you really need are the current UTC offsets used to describe time zones. Just store those as “sun offsets” relative to cities and nuke the time zone aspect.
The full scene if anyone’s interested.
Carl Weathers is an interesting contrast to Arnold. Arnold got his muscles by focusing on how he looked, on bodybuilding. Weathers developed his physique training for professional gridiron football. He played college football, then in the NFL (coached by John Madden) and CFL. Arnold used his unique physique as a way to get into movies without ever training as an actor. Weathers was studying theatre arts while playing college football, and finally finished the degree in 1974 just after retiring from pro football. He went on to get a master of theatre arts later.
Both of them pivoted extremely successfully to comedy later in their careers, with Weathers doing Happy Gilmore and Arrested Development, and Arnold doing his whole variety of comedy movies.
Also, good to know that Weathers loved this scene:
https://www.thedailybeast.com/baby-carl-weathers-has-a-stew-going