TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Learning math? Think like a cartoonist

162 点作者 freeCandy大约 3 年前

13 条评论

woolion大约 3 年前
I like the approach, even though in practice I would not necessarily follow the suggested presentation.<p>From my experience, most math courses start with the &quot;dead&quot; book definitions. Which is not the way mathematics are built: people start with intuitive but incorrect ideas, the idea is proven to be useful, and only then made correct. The definitions end up coming from refinements to avoid contradictions arising from edge-cases. This was a common reproach to the Bourbaki group who wanted to formalize mathematics into an almost computer-digestible form; I think it was Grothendieck who said they were &quot;embalming mathematics&quot;. This is absolutely not a good way to learn mathematics, although it is the lowest effort to come up with. I generally think it is much better to go through what do we want to achieve, how it came to be, etc. Which you can generally find with presentation of the Seven Bridges of Königsberg problem, some game mathematics (Rubik&#x27;s cube, etc), or &quot;You could have invented...&quot; type articles (the famous one being for &quot;spectral sequences&quot;).<p>Although there should be different approaches for different learning types. But this is another problem, which is more linked to the (bad) educational structure.
评论 #30906363 未加载
评论 #30907207 未加载
评论 #30909061 未加载
chrisweekly大约 3 年前
Just want to say that <a href="https:&#x2F;&#x2F;BetterExplained.com" rel="nofollow">https:&#x2F;&#x2F;BetterExplained.com</a> has long been one of my favorite sites; IMHO the more math teachers and students know about it the better. The author&#x27;s ability to make potentially daunting subjects comprehensible, even intuitive, is a rare and powerful gift.
Zhyl大约 3 年前
I think the emphasis on this title is &#x27;Learning&#x27; as the principles described don&#x27;t only apply to maths and very much shouldn&#x27;t be the end point for understanding a concept.<p>Reducing a concept to a &#x27;cartoonish&#x27; essence is really the secret to getting an intuition for a new concept, to allow the learner to &#x27;get&#x27; it for the first time. The reason is that when one sees something for the first time, its value or meaning isn&#x27;t apparent. It&#x27;s just shapes on a page. In other subjects it might be words on a page or a list of events without any throughline or narrative.<p>Reducing something to a &#x27;cartoon&#x27; allows you to focus on the orthogonality - the thing that that concept does that can&#x27;t be done elsewhere and the reason why you should pay attention and find somewhere in your mind palace to put this new thing.<p>But I would emphasise that while the ability to understand something well enough to reduce something to its bare essence <i>for a given audience</i> requires a high level of understanding and mastery, the thing the article skims over somewhat is that for the audience this is only the first step and should be re-enforced and expanded upon almost immediately.<p>BetterExplained as an outlet mostly focus on the pop-explanations sector which is underserved in Mathematics (with 3Blue1Brown and Numberphile doing a lot of heavy lifting) but as a <i>general</i> philosophy of learning, this article needs a bit more development.
nagonago大约 3 年前
Mathematics aside, am I the only one who finds this style of cartoon caricature unappealing? I find that these kinds of caricatures often bear little resemblance to the actual person, as well as being simply unpleasant to look at. In this instance particularly:<p>&gt; <i>Technically, his head is an oval, like yours. But somehow, making his jaw wider than the rest of his head is perfect.</i><p>I honestly don&#x27;t understand this point. Looking at the original image, his head and jaw seem perfectly average to me. I understand that caricaturists love to exaggerate features, but why praise them for amplifying a feature that does not really exist?
feoren大约 3 年前
&gt; We agree that multiplication makes things bigger, right?<p>Huh!? I have luckily never heard anyone try to explain multiplication that way before. It&#x27;s a <i>horrible</i> mental model for multiplication!<p>&gt; Imaginary numbers let us rotate numbers.<p><i>Kind of</i> ... it&#x27;s useful to visualize <i>multiplication</i> by imaginary numbers as analogous to rotation, but this makes it sound like imaginary numbers are useful <i>because</i> they let us &quot;rotate numbers&quot;, like that&#x27;s something we always wanted to do. This is neither the essence of nor the impetus behind imaginary numbers.<p>&gt; The number e is a little machine that grows as fast as it can<p>No no no! We happen to use <i>e</i> as an exponential base a lot because it&#x27;s convenient, but it&#x27;s exponentiation, not <i>e</i>, that grows quickly! f(x) = 4^x grows faster than e^x does; e^-x shrinks; e^0 stays where it is ... this is a <i>horrible</i> intuition to have!<p>&gt; The Pythagorean Theorem explains how all shapes behave (not just triangles)<p>It works on other shapes <i>because you draw triangles in those shapes</i> ...<p>These are awful, <i>awful</i> intuitions about math. We should not take advice about how to learn math from someone who so deeply struggles with basic concepts about it!<p>Edit: this article is so bad that I just have to add another example:<p>&gt; Euler’s Formula makes a circular path.<p>But you <i>just said</i> that <i>e</i> is a little machine that grows as fast as it can! How can it just spin in a circle!? His own intuitions are completely inconsistent! Besides, Euler&#x27;s Formula has no free variables ...
评论 #30912431 未加载
t43562大约 3 年前
I think the problem I have with maths is that at the point where I start to realise that the simplifications are wrong, I lose trust in every area of my understanding and want to give up. I can&#x27;t be comfortable feeling that there&#x27;s a missing hole in my mental model.<p>OTOH, it doesn&#x27;t work at all to give exhaustive mathematical proof because it doesn&#x27;t construct a model in my head.<p>So I don&#x27;t know what the solution is really.<p>To take complex numbers for example: the idea that -1 has a square root seemed like utter bullshit as it contradicts other models one has had drummed into one. If you look at it as a means to an end, however, it&#x27;s a useful little bit of machinery that can help to make other problems easier to solve (e.g. converting a differential equation to a quadratic). That makes sense and explains why we&#x27;re doing it whereas talking about SQRT(-1) starts one off with the feeling that everything is nonsense.
Agamus大约 3 年前
How many other &#x27;mathy&#x27; people with aphantasia have trouble &#x27;thinking like a cartoonist&#x27;? (And how many people with aphantasia are naturally &#x27;mathy&#x27;?)<p>Read Blake Ross&#x27; post for a better sense of why this method simply does not work for many people who are more comfortable with abstraction.
评论 #30906371 未加载
评论 #30907914 未加载
评论 #30906637 未加载
xg15大约 3 年前
I think the bit about the definition of multiplication could be applied to code as well: A source file necessarily contains <i>all</i> possible logic flows, no matter if a flow represents the main operation, an obscure special case or parts that are completely unused or practically unreachable. This can easily hide the &quot;gist&quot; of a function among heaps of clutter.<p>I think some tool would be interesting that took a code traces or coverage information and generated a view of a code file where each line has a font size dependant on how often that line was executed. Ideally, &quot;important&quot; lines would end up with a large font, while special cases, error checks etc would become &quot;fine print&quot;.<p>I&#x27;d be curious if this would help making codebases easier to understand.
评论 #30909103 未加载
goldenkey大约 3 年前
Multiplication only makes numbers smaller if we restrict ourselves to starting with a positive number.<p>If we instead start with a negative number, multiplication will make the number bigger.<p>If we allow for any integer, multiplication on average, will neither be biased one way or another.<p>It&#x27;s a bit of a carny trick to hide the bias in the selection of only positive numbers, and then act like multiplication itself has the issue..<p>This leads me to believe the author doesn&#x27;t know math very well and their learning techniques are not efficient or worth giving credence to.<p>If you give an analogy to explain a concept, it shouldn&#x27;t be fast one, it should be honest!
评论 #30907955 未加载
ar_imani大约 3 年前
I&#x27;m tired of these kinds of titles. They are like news headlines and nothing but a way that maybe it worked for someone. as the adage states: &quot;there is no royal way for learning.&quot;
评论 #30908182 未加载
photochemsyn大约 3 年前
When we talk about &#x27;learning mathematics&#x27; it&#x27;s important to recognize a huge difference between learning how to use mathematical discoveries&#x2F;inventions, and learning how to develop mathematics from scratch, via the whole theorem -&gt; proof -&gt; new theorem route that defines &#x27;pure mathematics&#x27;.<p>Most people are simply not going to get a lot out of learning the latter method and will indeed be turned off by it, much to the disappointment of the professional mathematicians (i.e. most college professors in maths). I&#x27;d guess &gt; 95% of people taking higher maths courses are not going to ever develop new proofs - but they will use what they&#x27;ve learned in other areas, such as physics, biostatistics, finance, etc. Essentially we just take it on faith that the mathematicians got their proofs right, and we gratefully use the fruits of their labor. (They&#x27;re all quite mad, those mathematicians, if you ask me)<p>Now, when you first learn how to apply maths to things like physical problems, this is where tyhe cartoons, or &#x27;simple approximations neglecting complex factors&#x27; becomes really important to learning. You don&#x27;t want to try to include friction when first examining falling weights and springs and pendulums through a physical viewpoint, for example. Later on, when you get that job with SpaceX, understanding friction in depth will be critically important, but if you don&#x27;t start with the simple cartoon approximations, it&#x27;ll be way too much to comprehend.<p>However, this probably wouldn&#x27;t work for the real mathematicians. They&#x27;ve got their axioms, then from the axioms they develop proofs, then from those proofs they develop more proofs - there&#x27;s no approximation or simplification involved in all that, is there?
评论 #30914481 未加载
eternityforest大约 3 年前
I don&#x27;t see what&#x27;s wrong with &quot;Multiplication copies things&quot;.<p>The fact that &quot;copying&quot; can be a continual scale from 10 copies to 1 copy down to half a copy and zero copies captures most of the behavior that a non-math person working as a programmer(One of the main uses for math, right next to using it to learn other math and becoming a math-heavy coder) is interested in.<p>The main actual uses of multiplication before you get to advanced stuff are things like volume controls, total price of N units, area of a square, etc, that are pretty much copying.<p>Except Ohm&#x27;s law. That doesn&#x27;t quite fit intuitive models of math exactly, and when you actually go to use it you usually wind up wanting to know power dissipation which is nonlinear in a very bizzare multi-variable way with some IRL use cases, since you&#x27;re usually looking at systems as a whole.
necovek大约 3 年前
In a nutshell, whatever this article suggests, just don&#x27;t.<p>This oversimplifies to a point where all learners learn the same way. They don&#x27;t.<p>I don&#x27;t even see the wider jaw in that cartoonist depiction, and I&#x27;d never recognize the man from it: a good cartoon will amplify features of someone, but this is a completely non-existant feature of the face being drawn. If anything, the guy has overly oval face compared to an average face.<p>So basically, based on the wrong premise, it happily leads you to a wrong conclusion.<p>A better take-away would be to attempt to recognize multiple ways to learn something, and make an effort to see what works best for any single individual. If you can&#x27;t afford that (too time consuming, thus too expensive, to cater to each individual student), choose what you optimize for: having <i>most</i> kids learn to a particular (likely lower) standard, or having &quot;most-compatible&quot; (eg. in maths, those who kinda already have the mathematical, algorithmic, abstract mind) get the best of their talent. But you will be compromising either way.
评论 #30905638 未加载