Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lessons of an MIT Education

math.tamu.edu

Post reply on HN
Print view View original

131–140 of 263 posts

Re: Lessons of an MIT Education

#131

Post by wolfgke »

mcguire wrote: post #95
wolfgke wrote: post #62

Earlier quoted context omitted.

> It's also weird to suggest that no historical claims are falsifiable. That seems to put real history on a par with nutty conspiracy theories. In my humble opinion the scarceness of falsifiability in history is the reason why it is so easy to create nutty conspiracy theories about history vs, say, mathematics or physics.

There are plenty of nutty mathematical or physics theories around. History is roughly as falsifiable as astronomy or biology.

> There are plenty of nutty mathematical or physics theories around.

At least for mathematics I have hardly seen any "nutty mathematical theories" (I am a PhD student in mathematics, but not a physicist, so I cannot say anything about physics here). The reason that I see is that for a math text to be considered as solid, it has to contain good, understandable proofs - which are hard to write by "crackpots". To be more precise: On sceptic's blogs I have of course seen links to papers containing "mathematical crackpottery", but I have hardly ever seen those "in the open countryside".

Re: Lessons of an MIT Education

#132

Post by _delirium »

wallace_f wrote: post #124
_delirium wrote: post #118

Earlier quoted context omitted.

> and which makes their lifestyles possible Arguably computing as a discipline is an important part of what makes contemporary lifestyles possible, and most of it, especially the practical parts, resembles the liberal arts more than it resembles mathematics. Especially areas like programming languages overall operate more through argumentation and analogy than through proofs. Sure, there are subsidiary proofs involve…

>Especially areas like programming languages, in their applied parts, operate more through argumentation and analogy than through proofs. I think the virtue is not in the value of the arguments, but whether the argumenrs are falsifiable. Programming has the virtue that it has to work for it to have value. Social sciences and humanities have to convince someone that it would work.

PLs arguments are very rarely falsifiable, if that's your criterion. I mean, yes, a programming language has to work in some sense, but that's a pretty low bar; C and Lisp work and have for decades, so we could just say we're done, disband the research field. The rest of PLs exists because they claim that they're building something that is in some way "better" than C or Lisp. But that's a pretty fuzzy argument, difficult to falsify. There is a small area of empirical software engineering that does try to measure things like whether certain constructs can measurably reduce bugs in real-world usage, which would be a testable hypothesis. But they've been able to establish very, very few solid things about PLs, and the vast majority of PLs doesn't look at all like that. It's more design argumentation.

To pick a concrete example: Rich Hickey introduced transducers into Clojure a while ago, using an argument, illustrated by a number of examples, for why they're useful. Is this argument falsifiable? In principle some version of it might be, if you made "useful" more precise (useful to whom? in which contexts? how would you know?). But the kind of empirical work it would take to measure it in a non-toy setting is quite difficult, so afaik nobody's tested it, or even really formulated the question precisely enough to test it. In practice, you accept or reject the construct based on what you think of the argument, or you try to find a counterargument that makes them look like an inelegant/awkward construct, but in either case you probably aren't attempting to rigorously validate or falsify a scientific hypothesis relating to them. Basically all of PL design and evolution looks more like that than like Popperian science...

Re: Lessons of an MIT Education

#133

Post by foldr »

wolfgke wrote: post #123
foldr wrote: post #98

Earlier quoted context omitted.

So, you believe that the moon landings weren't faked solely becuase of the credentials of the people who say that it was faked? You believe that the Holocaust happened solely because Holocaust deniers say crazy things about other topics? That seems really, really, off base.

> So, you believe that the moon landings weren't faked solely becuase of the credentials of the people who say that it was faked? This is exactly one example for my "where falsification is much better possible" - in this case because of physics.

It would not have been physically impossible to fake the moon landings. Physics tells us that the moon landings could have happened, but it doesn't tell us that they couldn't have been faked.

But by picking over the details of a particular example you're missing the point. Surely you don't want to say that you can reject crazy historical claims solely because the proponents of those claims make outragerous claims in other areas too. If you think the moon landings are an exception, for some reason, then let's look at other examples. E.g. Holocaust denial.

Re: Lessons of an MIT Education

#134

Post by ooronning »

pkrumins wrote: post #125

Lesson zero: You don't need an education.

found the self-taught programmer :)

No offense intended; I'm self-taught as well and can appreciate the sentiment immensely. But damn do I wish I made a few different decisions during high school. I'm ~1 year into my web development career and I'm concerned about what happens when I get bored of building APIs and web apps -- I seem to have a talent for picking up new skills pretty quickly, but I'm afraid my lack of formal math and algorithms training is going to limit the career trajectory of self-taught programmers like me. That's without considering the ramifications of an irrelevant college degree.

Re: Lessons of an MIT Education

#135

Post by tome »

wenc wrote: post #22
icc97 wrote: post #3

> The world and your career are unpredictable, so you are better off learning subjects of permanent value. I've been looking for a way to express this idea for years. Doing a core subject at University like Mathematics or Physics can be used in a million different careers. In a similar way I'd recommend learning say Functional Programming over React or Scala. Plus I guess learning Category Theory over functional prog…

> learning Category Theory There is a level of abstraction which for most people is the wrong end of the stick to start from. Category theory is one of them. (and I'm a big fan of category theory and Haskell). Category theory may occasionally lead to elegant, orthogonal and consistent solutions, especially if you are building new abstract tooling (LINQ in C# is a good example). It gives you tools to reason about stru…

Yes! Absolutely. Unless you already have quite some proficiency with mathematics and/or generic programming then learning category theory is pretty pointless. In fact I don't even know what it would mean to "learn" it if you don't have a suitable background to build it on.

Re: Lessons of an MIT Education

#136

Post by nothis »

dougmwne wrote: post #6

The author was quite harsh on the soft sciencies and humanities and I'm going to attempt a defense. First, he characterises these subjects as merely what, facts to be learned, instead of how, skills to be mastered. Let's take history, which seems very what, full of dates and names. A great course in history teaches you how to think differently and with deeper context about world events and your country's politics. I…

I'd argue that "learning how" to write good texts or understanding a person psychologically, for example, is a rather valid example of a soft science having a real impact on the world that can easily compete with engineering problems. It's a bit like arguing that construction work is more valuable than planning architecture, which might be true in a hierarchical sense but ultimately it's just a matter of perspective and scale.

Re: Lessons of an MIT Education

#137

Post by ThenAsNow »

bjourne wrote: post #41

> Lesson Six: You must measure up to a very high level of performance. I can imagine a propective student or parent asking, "Why should I (or my child) take calculus at MIT rather than at Oshkosh College? Isn't the material practically identical, no matter where it is taught, while the cost varies a great deal?" Is there any truth to this? Because it seem like classical university jingoism to me. "My institution is b…

I enjoyed most of Professor Rota's comments, but couldn't disagree more with his Lesson 6 and think it is pretty destructive.

In my view, the ultimate proof as to whether you deeply understand something is when you can engage in rigorous synthesis using that thing, or correctly explain it to others without using terminology as a crutch.

Getting to that level of understanding is not merely a function of the recognized accomplishment level of your peers or your instructor. What imbues people with a deep understanding varies from person to person, and many really benefit from a deliberative, instruction and discussion/conversation-focused instructor. My casual observation of MIT is that the level of instruction can be far from the best I've seen, because the students are so bright or so fast that they don't need much from the instructor in order to immediately see the conceptual space.

Imagine however that you are no intellectual slouch, but maybe are a more contemplative and deliberate thinker. You may be penalized in an environment like this, which bears no representation on whether or not you can get to the same quality and depth of understanding as your peers. I once had an MIT-educated professor say, "I know how to grade these exams. My 'A' students will finish the exam in the time allotted, my 'B' students will finish most of the exam, the 'C' students less so." This infuriated me then, and infuriates me now. The counterpoint was a (well-known) UMich-educated professor, one of the best I had, that gave challenging exams with no practical timelimit. He inspired me to be very ambitious as the lead on a free-form semester group project, so much so that it was clear we wouldn't finish during the semester. He saw how ambitious it was, and gave us an incomplete until we could turn it in the next semester. We did, and it was a really nice body of work.

I've seen similar patterns in the rest of my education. I've always gotten the most not from the best-credentialed professors, but from those who themselves understand deeply and make the material and themselves open to the interested learner.

This is ultimately what teaching and learning should be about, and can easily take place at any good institution, not just the MITs of the world. In my field of engineering, there is no significant correlation between those who were educated at MIT or Stanford, and the impact of their contributions to what is in the field and has truly advanced the state-of-the-art.

Re: Lessons of an MIT Education

#138

Post by wallace_f »

_delirium wrote: post #132
wallace_f wrote: post #124

Earlier quoted context omitted.

>Especially areas like programming languages, in their applied parts, operate more through argumentation and analogy than through proofs. I think the virtue is not in the value of the arguments, but whether the argumenrs are falsifiable. Programming has the virtue that it has to work for it to have value. Social sciences and humanities have to convince someone that it would work.

PLs arguments are very rarely falsifiable, if that's your criterion. I mean, yes, a programming language has to work in some sense, but that's a pretty low bar; C and Lisp work and have for decades, so we could just say we're done, disband the research field. The rest of PLs exists because they claim that they're building something that is in some way "better" than C or Lisp. But that's a pretty fuzzy argument, diffi…

I basically agree with you, but we're arguing to different ends.

What I am saying is that there is nothing wrong with these arguments or with using them usefully. What is wrong is when nothing is falsifible or able to proven, either. So in other words, arguments are useful, but it does matter if it has to agree with scientific experiment.

Programming does, at the end of the day, agree with fundmantal truths for it to work. Its foundations are on the metal, and everything is reducible experimentability.

Perhaps everything except the human element: most aspects of modern product development involve programming language improvements that have to do with improving human interaction with a computer. But even here, we have a sort of market for ideas in that developers who adopt better ideas will be more successful.

Re: Lessons of an MIT Education

#139

Post by __float »

ploggingdev wrote: post #48

(Off topic) > Scientific biographies often fail to give a realistic description of personality, and thereby create a false idea of scientific work. Any recommendations for biographies which give a realistic look into famous scientists' lives? I am reading the Einstein biography by Walter Isaacson and it's pretty good so far. Completely off topic : how do you configure a LAN to have /~username urls (such as the linked…

https://httpd.apache.org/docs/2.4/mod/mod_userdir.html

Re: Lessons of an MIT Education

#140

Post by sinab »

swiley wrote: post #107
bjourne wrote: post #41

> Lesson Six: You must measure up to a very high level of performance. I can imagine a propective student or parent asking, "Why should I (or my child) take calculus at MIT rather than at Oshkosh College? Isn't the material practically identical, no matter where it is taught, while the cost varies a great deal?" Is there any truth to this? Because it seem like classical university jingoism to me. "My institution is b…

I looked at the OCW stuff when I took calculus at a community college. They cover the same material but a lot of the homework and exams are made of word problems which makes it (arguably unnecessarily) very hard. After that I only used OCW for lectures and book suggestions and never for worksheets/exam/practice problems because if I want to study calculus then I'm going to focus on that. You can take it either way. I…

I think I would disagree. One of the aspects of an MIT education (c/o 2017) was the fact that we were forced to decipher problems from statements in a similar way that one would be forced to do so in real life. Problems should (and almost never do) come packaged in a neat and nice way that is easy to solve, hence why I believe that word problems and sometimes confusing problem statements are beneficial.
Post reply on HN