Live data from Hacker News

MIT dean to start new university: “No majors, no lectures, no classrooms”

tech.mit.edu

101–110 of 281 posts

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#101
post #31
post #4

The only point of a modern day university is that they are (a) exclusive, and (b) "guarantee" that students "know" what they are "taught". That's at least what people who hire think. I remember someone saying "It wouldn't matter if MIT only taught basket weaving as long as they still had such a thorough selection process". This holds true here. People only care about the pedigree of your degree, it's substance comes…

"guarantee" that students "know" what they are "taught" That used to be the case, but now corporations require you to pass their own exam before they hire you. Let that sink in a moment. No longer are the universities taken at their word about their product. For some of you it has never been any different, but this change came about more or less when I was teaching at university/college. Some of us saw this as a quit…

On higher Ed:

There is far more to a degree than the classes and the piece of paper with a GPA. It is also an environment where young people grow as themselves for the first time in their lives in many ways. There is also more that can be done to ensure strong job placement, like my alma mater, RIT, in requiring a full year of paid co-ops before they grant a degree.

On hiring:

There are infinite metrics that assign value to a candidate regardless of the school they attended; culture fit, for instance.

However, hiring is an expensive and labor-intensive process, and every moment an engineer is looking over resumes or conducting interviews is time that could be spent fixing bugs or creating valuable features. Pedigree, for better or worse, simplifies the decision. When you have a stack of resumes to deal with, half from small colleges you've never heard of, and the other half from, say, Cornell, whom do you prioritize?

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#102
post #37

Earlier quoted context omitted.

> I find that crazy. Both teach basically the same thing. If they are both accredited, then it is almost indistinguishable. Speaking just on computer science here, but the depth of CS courses varies radically across universities. Sure they all might follow the ACM model curriculum, but how much of the book you're actually reading (and which book you're using in the first place), the number and difficulty of practice…

It's not just the depth, but the breadth as well. Better universities do an amazing job of teaching their students how to learn. The curriculum is thoughtfully designed to produce well rounded students who are capable of achieving expertise in their chosen field (they don't leave college as experts by any stretch.. and THAT is the disconnect surrounding education these days). The big difference between someone coming…

For someone who can't see nuance, this is a remarkably perceptive comment.

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#104
post #31
post #4

The only point of a modern day university is that they are (a) exclusive, and (b) "guarantee" that students "know" what they are "taught". That's at least what people who hire think. I remember someone saying "It wouldn't matter if MIT only taught basket weaving as long as they still had such a thorough selection process". This holds true here. People only care about the pedigree of your degree, it's substance comes…

"guarantee" that students "know" what they are "taught" That used to be the case, but now corporations require you to pass their own exam before they hire you. Let that sink in a moment. No longer are the universities taken at their word about their product. For some of you it has never been any different, but this change came about more or less when I was teaching at university/college. Some of us saw this as a quit…

> corporations require you to pass their own exam before they hire you

If you're talking about software interviews, know that you need to have pedigree even to get an interview. See Twitter or Palantir recruiting.

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#105
post #37

Earlier quoted context omitted.

> I find that crazy. Both teach basically the same thing. If they are both accredited, then it is almost indistinguishable. Speaking just on computer science here, but the depth of CS courses varies radically across universities. Sure they all might follow the ACM model curriculum, but how much of the book you're actually reading (and which book you're using in the first place), the number and difficulty of practice…

It's not just the depth, but the breadth as well. Better universities do an amazing job of teaching their students how to learn. The curriculum is thoughtfully designed to produce well rounded students who are capable of achieving expertise in their chosen field (they don't leave college as experts by any stretch.. and THAT is the disconnect surrounding education these days). The big difference between someone coming…

> it sure would have been easier to have done it in my 20's

Not necessarily. It might have required less determination to go along with what others around you were doing, but a lot of people who try learning subjects like calculus as adults report that it seems easier. They have better focus, and most important, they understand why they are doing it.

In short: it's never too late! Study on!

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#106

Earlier quoted context omitted.

It's surprising that it's taking so long. The wealthiest have always found ways to preserve their wealth and Ivy League institutions have played a role in that. Take some smart people, a lot of people that are good at fitting the mold that looks best, mix your rich kids in there and call the paint one color. Remember, George W. Bush (all politics aside, just looking at him in terms of intellectual prowess) graduated…

George W Bush plays much dumber than he is. It's hard to get an unbiased opinion on the matter since it can be used to score political points, but he definitely has >100 IQ.

I'll just leave this here: "George W. Bush, Republican, 1946–present, 2001–2009, IQ:138.5" --> https://en.wikipedia.org/wiki/U.S._Presidential_IQ_hoax#IQ_e...

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#107

Earlier quoted context omitted.

I wouldn't say that the substance of e.g. a M.D. degree comes "second". Maybe you're focusing too much on IT?

If there were two medical schools, one a state funded community medical school and the other a big name like John Hopkins, what school would you want your doctor to have attended? Even if you were assured that they both teach the same material, you would want the Coca-Cola over the Pepsi.

I know what you mean but the soda analogy doesn't really make sense to me. How is Coca-Cola more "prestigious" or something than Pepsi?

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#108
The "42" (school) seems very promising, it's a private French computer programming school created and funded by Xavier Niel.

To apply, there is no degree or diploma requirement, and it's completely free.

The training focuses on project-based learning via a peer-to-peer pedagogy and allows students to set their own pace for learning.

IMO The 42 school is not research-oriented, but rather emphasizes practical training suited to the needs of startups.

www.42.fr

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#109
post #49
post #7

Earlier quoted context omitted.

Honestly, I wish there were major respected trade schools that taught CS/Software Engineering/IT/IS. The university environment, in my opinion, is adverse to the development of good CS knowledge. Most of us on here, I'd say, have learned through trying and experimentation and that is why we are fairly good at what we proclaim to do. University doesn't foster that kind of understanding. I also hate that currently I ne…

"Good CS knowledge" isn't 'learning git'. It's mathematics, theory of computation, recursion theory and type theory and PL, compilers, complexity theory and algorithms, advanced data structures, concurrency and distributed systems etc. If you know these, you should be able to pick up more applied subjects on the job quickly. You very much need a university to teach you this. Social sciences will make you a more well-…

> theory of computation, recursion theory and type theory and PL, compilers, complexity theory and algorithms, advanced data structures, concurrency and distributed systems etc

All of those are fairly basic topics that most people can pick up in an afternoon or two. These are all things that anyone who has been programming, reading, and interacting in the field of computer science understand. I could understand someone requesting that knowledge be provided to them in a college environment.

What I object to is universities charging for the "well-rounded"-ness that you speak of. Can you please define for me what it means to be well rounded?

From what I have seen is that no, no one can define what it means to be well rounded.

What do you have to do to be well rounded? No, no one can tell me that either.

Some how having gone through a university you are proscribed this trait of being in the "know". You automatically are transformed into a scholar, a person who is now magically "well rounded".

Don't get me wrong, I think that there needs to be a structured curriculum, but only of things you need to understand make it in your respective specialization.

For me, I'd love to get into building compilers, operating systems, digital signal processing, and general software development.

For me, I'd say this is the perfect regiment for me:

  * Full understanding of the computer booting process, understanding of modern processor features.

    - Write a self hosting operating system, bootloader, and learn to write a BIOS for an example computer

 * Software maintenance and revision control

    - Following the current industry best-standard practices

 * High level math with a focus in signal processing

 * Familiarization with X top programming languages at the time of graduation

    - This should include standard formatting for each language, understanding implementation details, and full familiarization with X top used libraries and standard library

 * Understand logic and problem solving
You may notice that I have annotated some of the things I feel I should learn, others I have not. If I have annotated it, it is because I have learned to do this on my own time and I want to expand that knowledge and I know it will help me professionally. The things I have not annotated are things that I have a basic understanding of, but I need someone to teach me although I am not getting that at a university.

I don't know, maybe I'm wrong about this. I think these are fairly rational claims that I am making. Maybe I'm completely wrong and being able to quote shake-spear will help me when I need to learn the documentation for int 0x13 when I need to load a file on a 30 year old IBM PC when fixing a kernel level bug at some company, or when I need to write software to process data coming from sensors at 100GHz.

Edit: Updated formatting.

Re: MIT dean to start new university: “No majors, no lectures, no classrooms”

#110
post #31

Earlier quoted context omitted.

"guarantee" that students "know" what they are "taught" That used to be the case, but now corporations require you to pass their own exam before they hire you. Let that sink in a moment. No longer are the universities taken at their word about their product. For some of you it has never been any different, but this change came about more or less when I was teaching at university/college. Some of us saw this as a quit…

> That used to be the case, but now corporations require you to pass their own exam before they hire you. You mean an ... interview?! Ok, that was purposely obtuse, but I don't see how an extension of a company's interview process (ensuring, through their own testing, that candidates meet the proper qualifications without school bias) is a bad thing. Or how it signals an 'inevitable disruption of modern higher ed'. W…

Your conclusion is right. To me, that helps keep schools honest and continuously adapting their curriculum's (except for the misplaced apostrophe).

In other words: Sure, you have a 4.0 in engineering from Stanford. But now let's see if you can actually compute a Fourier transform. Or even if you know when to use a FT.

To my generation, that's damning. In other words, (quoting you here) ensuring, through their own testing, that candidates meet the proper qualifications without school bias is the new normal, and it's different in a way that says we don't/can't trust the universities to deliver their product any more. The status of the universities has changed in our society, and that, in my opinion, is an early signal for a coming disruption.

Post reply on HN