This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
There’s nothing I can add other than to reiterate @Corvos and @orthoxerox’s replies. Because classical systems still involve effective, consistent multigenerational meritocracy (ie social mobility), they effectively offer almost the same competence with a huge reduction in unnecessary make-work. Kids studying for 5 hours a day after school is worse for society than them spending that time digging ditches and filling them in; the latter at least involves exercise. Yes, you are impatient, but that is more about you than it is about society.
As for me, I have one correction to make. I’m not old money! I’ve said this before, but I was born upper middle class; my parents became truly rich only in my teens. My mother’s family were a mix of middle class for many decades. My father’s family were a mix of shtetl dwellers and Italian Jews who mostly arrived between the 1820s (early for Ashkenazim) and 1890s (with a few later outliers), and who went on to make and lose several fortunes, but who were at the time of his birth as middle class as you can get (think a small town accountant or government worker).
Hi, can you expand on this point? My position is that there's obviously marginal utility in increasing/decreasing schooling duration/homework, but then that's a very individual thing where some need more and some need less. This individuality expresses itself in extra tutoring (some people need more time than the average to get something) or special classes (some people need less time than the average to get something and is ready for the next thing). Since society most often rule in averages*, then yes an arbitrary fiat of extra 5 hours system-wide is probably going to be bad, probably just as bad if kids spend 5 hours digging. I suppose what I am trying to get at is do you have a different more clarifying example of your position?
* It's been said before but the dream for education is obviously personalized individual study plan suitable for the person. AI and tech seems to be 1 or 2 years from being able to offer this.
First homework is stupid. Even worse it is pointless. It takes the most valuable time of your life for some absolutely marginal advancement in academic performance. If someone can't grok quadratic equations - solving 200 more at home will help? Making them suffer trough war and peace or crime and punishment will make them appreciate literature? That somehow staring at the physics handbook will make them understand relativity?
We treat knowledge as goal not as a tool. Tell the little ones that at the end of the course they will know how to make total synthesis of cocaine and you will have the most attentive chemistry class in the history of the world. And they will do their homework without even being assigned.
I think that math class is probably the worst offender for pointless busywork.
Case in point: Polynomial long division. 100% "we make you learn an algorithm as a proxy for intelligence", 0% "something you will need to know as a prerequisite for understanding something else." The correct place for it would be "Having just discussed the properties of polynomial rings in general, here is as a curiosity a technique of dividing polynomials. You know that it will not be relevant for you exam because it involves just playing an algorithm (possibly even with concrete quantities)."
Instead you get tasks like "one of the factors of x^3+5x^2+7x-3 is (x+3). Factorize x".
People whose skill is to pass 'math' class in high school do not need to worry about being replaced by LLMs, because they were presumably replaced by WolframAlpha in 2011.
This is equivalent of doing multiplication and eventually memorizing the multiplication table during elementary school. With enough "busywork" you will be able to factorize from your head, which will enable you to solve some key problems much, much more quickly "look and see" style ,sifting through unproductive approaches before even trying them. This is cookie cutter thing in many engineering areas, basically anything that uses differential equations which is a really a lot of stuff.
Busywork is really important especially in early stages. You have to go through some shit like memorizing vocabulary when you are learning a new language, before you will be able to do some creative things like translating poetry or doing some Tolkien-like stuff with that language.
No, that is my whole point. Knowing the multiplication table up to 9*9 is a prerequisite for the following:
Anyone who is arguing that kids do not need to learn what 7*8 is denying the need for learning arithmetic at all. (There are probably people arguing that you do not need to know arithmetic because computers exist. I disagree. Knowing how to arrive at a result without a computer in principle, even if a million times slower and error prone, is useful in itself. Besides, with LLMs, this argument generalizes against anything you learn in high school.)
I am fine with busywork which is an instrumental goal for something worthier. I learned my multiplication table because I knew that it would enable me to do long multiplication in base 10. Then in fifth grade, they tried to make me learn 2..9 * 11..19, and I was not having it. If you are in base 10, the utility of knowing non-trivial products of integers takes a sharp dive after 99. I felt that if I gave in and learned what 137 was, they would come around in sixth grade and demand I learned by rot what 1315 or 237 was. (For factorization, the useful thing would be to quickly identify a prime factor in any integer in a given range, but even this obviously does not scale.)
A lot of the stuff kids learn in math class, painful as it may be, is actually useful in some fields. I don't think that youths should be forced to learn what a logarithm is, but they should be clearly told that any subject up to the softest social sciences will expect them to know, so that they can make an informed decision.
Other stuff has value because it teaches core concepts of math, such as proof by induction or the definition of real numbers. Sadly, people only learn what real numbers are when they go to university, and proof by induction in high school is turned into a mockery by turning it to just another pointless algorithm which can be used to prove increasingly pointless sum formulas by rote.
Wait - you value this, but not polynomial division? They're both things you can just ask the computer to do for you instead, but at least polynomial division requires you to hunt down a computer algebra system; long division capability come pre-installed on every phone.
Polynomial division is IMHO a curiosity.
Long division crops up every time you need want to split the bill.
Polynomial division might crop up in the wild if you want partial fraction decomposition, which I guess you might want if you are dealing with rational functions and want to numerically evaluate them or calculate their anti-derivatives. While I am sure that rational functions have their uses, my gut feeling is that they are both too narrow to pop up in physics a lot (where you will frequently have square roots placed so that your functions can not easily be transformed into rational functions) and too inconvenient to be preferred for empirical models.
Factorization of an integer is a hard but finite problem. Factorization of a polynomial is in general just not possible exactly. You can test if 7, 13, 17, 19, 23, ... etc happen to divide 3071 to factorize it. You can not test if x-1, x-1.1, x-sqrt(42+sqrt(42)) etc are factors of 5x5+4x4+7x**2-2x-2, because there are countably many algebraic numbers which could be a root.
I think that we learned both polynomial division and solving quadratic equations around grade eight. Solving quadratics in something which I would call bloody useful. Quadratic functions are the first non-trivial functions students can tackle, and quadratic equations pop up all the time in high school physics.
I strongly disagree with the sentiment that math skills which are less readily automated are more valuable. To grok (I'm reclaiming that word) how multiplication and division work doing long multiplication and division is definitely more useful than just using a calculator. Nobody needs the numeracy to be actually excellent at these operations any more. Anyone whose job actually requires them to multiply five-digit numbers will hopefully have the good sense not to try that by hand.
My more general point might be that I do not want students to be excellent at applying any algorithm. They will always suck very hard compared to the simplest of computers. Still, it is useful to demonstrate that you can apply an algorithm, even if it is just at toy-sized problems.
Also, applying a pre-learned algorithm is not math. Some algorithms (e.g. solving an equation for a variable) are genuinely useful in proofs, and thus are valid technical skills to learn to be able to engage in math, same as being able to write symbols with a pencil. And of course, 'can you apply an algorithm halfway reliably?' is also a good way to check if someone has a basic understanding of the algorithm in question (even if it does not probe if they understand why it works), which is why rotating trees by pencil is a staple in CS exams.
Still, for school math, I feel that 'can apply pre-learned algorithms' should earn a passing grade, not an actually good grade, which should require thinking.
This was an excellent explanation, thank you.
One side note:
Piecewise-rational functions are very popular for two big categories of empirical model: anything where the true behavior can have asymptotically-polynomial singularities, and CAD models. Being able to do sharp corners and spectral approximation refinement and exact conic sections all with the same backend is a very useful trick.
This is irrelevant to your points, though; even when working with NURBS, polynomial long division doesn't really come up.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link