Hacker News Reader

Comment by supernova87a | Hacker News Reader

I remember wondering aloud at MIT/Harvard putting their courses online for free for anyone to listen to, and saying, well what is the value add then of paying to physically go to college?

I guess we're going to find out whether the networking, friends, social aspect is worth it. There's also something about the competitive aspect of seeing other people in person do better than you or have better ideas than you. Which is hard virtually.

That being said, the college experience has gone through events like wars, plague, disruption, etc, and has come back before. (Although not at a time when there was such a ready alternative technology at hand.) I'm not yet so worried as some say (wholesale change to the higher education model), unless this goes on for 2-3 years.

View on HN

> well what is the value add then of paying to physically go to college?

You can put on you resume that you were accepted in a selective program, and that you were able to complete it.

Besides, the actual lectures are only the tip of the iceberg. If you attend in person, you have to do tons of assignments, labs, exams. You also have tutors and professors that supervise, pressure, grade you. You work in collaboration with other students. You have access to internship opportunities or to your professor networks.

I once worked through a distributed systems project from MIT. I worked almost full time on it for a couple of weeks. Most of the material was available, but not everything. You certainly don't have the same resources students have. And you don't have the same pressure students have. Also, this project had zero value on my resume. It would be different if MIT could grade it for me, and certify I did well.

>If you attend in person, you have to do tons of assignments, labs, exams

Exactly. Even as a self-taught developer, one of my biggest regrets is how little I appreciated the incredible resources I had available at university and didn't really take advantage of or appreciate.

For me, it rather was "I had so much to do for courses that I had hardly any time to make use of or appreciate the incredible amount of ressources."

Also, a lot of classes are essentially impossible online. Intense lab courses, term-long project courses, and time-sensitive experimentation courses are all screwed. The way classes are coping is interesting:

1. For classes requiring user testing or behavioral experiments, assignments are being processed through mechanical turk.

2. The experimental physics lab is changing from a 'choose-your-own-adventure' model of choosing a final project to a satellite/telemetry/astronomy project.

3. Some classes are providing synthetic datasets or data from previous semesters for lab classes.

4. Some classes are basically finished. It is actually impossible to continue teaching them, so they're purely instructional or seminars, at this point.

A lot of these lab classes are required to graduate, too. They also require so many resources that they can only be taught once a year. It's seriously demotivating, especially for passionate hands-on learners. And paying sticker price doesn't help the sting for many students.

Medium to large lecture classes can probably work pretty well. As can mostly small seminars I would think. Other things are tough or impossible.

For example, I had an undergraduate thesis in my major, which required being physically in a lab and having access to a machine shop. In any case, there are lab/project classes throughout a lot of STEM majors that students aren't going to be able to reasonably finish this semester. I don't know how this is being typically handled with everyone having to drop things ~ 1 month into the semester.

About halfway through my college experience, I switched gears from extreme attendance-and-attentive-note-taking to extreme self-teaching, out of necessity (I was taking too many classes and working simultaneously, so I didn't have time to attend each lecture each week). Of course, I still did the homework assignments, studied for exams, etc.

Except for the occasional "gotchas" where professors say something random in class that shows up on the exam, it didn't adversely impact my GPA, and it saved a ton of time. I think I learned more efficiently on my own, which led to saving a ton of time shuffling from physical location to physical location. It freed up more time for working, exercising, and sleeping.

Given this experience, in retrospect, I think the most valuable parts of my education were not necessarily the classroom, but rather:

50%: The (well-written) syllabi. They point out which resources (books, blogs, websites, etc) are worth your time, if you have an interest in the subject. The best ones are filled with opinions, and "optional resources if you want to dive deeper into X". Even the lazily-constructed ones are a great foundation if you have no idea about how to begin learning X.

25%: The classmates and TAs. Oral/written feedback, delivered efficiently and with empathy is not essential, but a great catalyst to development. This one's a bit personality-dependent, for other's it might be higher or lower.

25%: The assignments, projects, and deadline pressure. To learn things, you really need time-spaced practice. Reading is not sufficient. The deadline and grade pressures erode as you get better, but they're invaluable early on.

I think that you can get 90% of each of these online, just fine. But that's me! It might require personal conversation for others, but in my experience a good collection of reading and tasks, some form of IRC, and rudimentary deadline pressure are more than sufficient to replace the classroom for self-starting individuals. I think Coursera's relative success proven that case.

> 50%: The (well-written) syllabi. They point out which resources (books, blogs, websites, etc) are worth your time, if you have an interest in the subject. The best ones are filled with opinions, and "optional resources if you want to dive deeper into X". Even the lazily-constructed ones are a great foundation if you have no idea about how to begin learning X.

Do you have any pointers to online examples of syllabi you found useful? My syllabi are always to the point and focussed on the structure of the class; it's never occurred to me that someone might want to use them as a roadmap for independent learning, and I'd be interested in facilitating that.

I did not understand this in the first 2 years of uni; I was in a very easy going high school before uni (this had something to do with my grades as well) as in, I did not have to attend classes, so I was used to just going to the exams. Which was great because I could make money writing software while others were sleeping in class. When I went to uni I just continued that pattern; it alienated me from my peers and as a result I did not like. Until (around 2 years in) I got bored and picked a graduate course (kolmogorov complexity) which could not be completed without attending; it was excellent; very small group (6) and the prof. That changed my ideas a bit ; I still wouldn't go to the auditorium stuff (I have/had (ex death metal band gitarist...) tinnitus; I cannot hear anything in full rooms with people whispering and sound echo-ing anyway, so what's the point). Well recommended; going to those gave me businesspartners + colleagues + funding for my first 'real' company as well.

Small classes are the best. I had an intro to the Standard Model class which had 4 students and the teacher... possibly the most enjoyable class I've had. There's no other occasion where a teacher will dedicate so much time to you.

I had a very intense 6 person class learning Russian 12 hours a day for three months. One of the most incredible experiences of my life.

That sounds like one of those experiences where, at least for the first few weeks, you return home and promptly fall asleep, because your brain is too full for anything else.

And you get to meet a lot of people who will tomorrow be influential in your industry.

> You can put on you resume that you were accepted in a selective program, and that you were able to complete it.

BTW via Coursera (perhaps only through white label?) you can enroll at some universities and get an ordinary degree (no qualifying language like “online”, join the alumni association etc). So you can get that benefit from some online programs, though perhaps not the top echelon.

Despite that, I agree with you that the realtime meatspace stuff is at least 50% of the value for the higher end schools (say “top” 100, however you define “top”).

> Also, this project had zero value on my resume. It would be different if MIT could grade it for me, and certify I did well

This is hugely important not just on the CV. The feedback you get from grading helps you understand what mistakes you made, and the prospect of receiving an official pass or failure grade is a big incentive to work hard.

I know this might not be received well by the HN crowd, but I feel like the classes and networking are secondary, the main value of a reputable college is having a reputable college you can put on your résumé.

I think that is an opinion very popular on HN, so don't worry about the reception.

Yeah in the US that’s exactly it.

Administrators for the sake of administrators everywhere.

I've heard that a lot but I did get a degree from a reputable college in my home country (one of the best in France for my subject matter), yet I've only worked outside of France for my entire professional life and pretty much no French colleges are well ranked on international rankings (mostly because the rankings are biased towards much bigger institutions).

And, well, the fact that it is isn't ranked well on any of those international rankings has not been a problem for my carrier at all. I have rarely been asked where I went to university and I don't think that it's affected my carrier at all.

However the fact that I have no student debt because tuition was pretty much free has been a boon for my professional life.

I have the feeling like OPs reputation theory is self fulfilling in the US and a narrow effect that came up with those paid education thing they have over there. It's like that artificial value you get for your extra money. Like wearing the brand of cloth you bought on the outside of it for everybody to see.

It's different if you worked in France though, right? From what I understand, the grandes écoles are very similar. It's not so much that you'll get an education that's ten times better, but you'll have a degree from a school that says you've been selected for leadership.

Yes, probably, I mean I come indeed from a Grande école but since I've never actually worked in France, it's never really been asked for. And the few times when ranking was important (visas for Japan and Hong Kong), the fact that it's not well ranked internationally means that I got no exta points.

In the end, I do think that degrees matter less than what people think and that it's completely possible to have a good career without the school name or university name having much of an influence (obviously it's anecdotical and probably depends on the field). I do strongly believe that being saddled with lifelong student loans is probably more detrimental than the positive aspects of a degree in most cases (maybe with the exception of lawyers but the only those who can cope with the lifestyle)

I absolutely agree that you can have a good career without having a degree (or a degree from the right school), but you will have a very tough time getting beyond a certain threshold. I understand it's very obvious in France for upper management and public/political higher positions that it's not merit-based. The same is true for most countries, with varying degrees of impact and obviousness.

But if you're not going for that threshold, you'll likely not run into issues. And if you're in niche-careers, it's probably true as well.

The fact remains that most educated people are aware of the French approach to higher education and the exacting filter applied for entrance and matriculation.

As an example, in a previous job, I noted that the new French software engineer had come from INRIA and was a graduate of a Grand ecole. The fact that this resource went on to demonstrate well above average work and domain abilities was not a surprise to me.

So note that some of us do pay attention even if it is not mentioned during recruitment.

Interestingly, though, it seems that in France, people place a lot of weight on which university you went to long after it becomes irrelevant... especially for the people who went to grandes écoles. I still don't understand how it matters that somebody, for example, want to les Mines 20 years after they've graduated.

But I agree with you; I've worked in a lot of places, and in all of those except France, once you have real work experience, your university degree doesn't mean a whole lot for normal engineering jobs.

This is also the main problem, because it suppresses any real competition.

All the while these organisations are grading their own students we will continue to get a somewhat mediocre system.

> having a reputable college you can put on your résumé.

At what point will firms rank new grad candidates inversely proportional to the prestige of their university?

Prestigious universities equate with more expensive candidates. Many positions probably don't demand skills beyond the ability to write mostly CRUD apps. The pedigree doesn't really matter in that case.

All companies have these roles. Somebody needs to do the mundane work.

> At what point will firms rank new grad candidates inversely proportional to the prestige of their university?

Never.

> Prestigious universities equate with more expensive candidates.

But that doesn't mean you rank those people lower. It means you offer them the job at the budgeted salary. If they don't accept the job offer at the salary range you've budgeted for the position, then you move on to the next candidate.

This might make less sense if there's a strong opportunity cost (i.e., missing a candidate because they already accepted a job), but that's not typically the case for new grads.

Bryan Kaplan argues that the majority of the value of a reputable college is signaling.

> Suppose your law firm wants a summer associate. A law student with a doctorate in philosophy from Stanford applies. What do you infer? The applicant is probably brilliant, diligent, and willing to tolerate serious boredom. If you’re looking for that kind of worker—and what employer isn’t?—you’ll make an offer, knowing full well that nothing the philosopher learned at Stanford will be relevant to this job.

If the value of education was the information taught in the classes or the network, you should see a somewhat linear increase of value as you progress. For instance, after completing your second year out of four, you should receive ~50% of the college premium.

> Most of the salary payoff for college comes from crossing the graduation finish line. Suppose you drop out after a year. You’ll receive a salary bump compared with someone who’s attended no college, but it won’t be anywhere near 25 percent of the salary premium you’d get for a four-year degree. Similarly, the premium for sophomore year is nowhere near 50 percent of the return on a bachelor’s degree, and the premium for junior year is nowhere near 75 percent of that return. Indeed, in the average study, senior year of college brings more than twice the pay increase of freshman, sophomore, and junior years combined. Unless colleges delay job training until the very end, signaling is practically the only explanation

[0] https://www.theatlantic.com/magazine/archive/2018/01/whats-c...

This is very true! I believe (don't have a good reference) that companies in the past had more incentive and means to cultivate or train people on the job, and thus, a formal signal of education wasn't as important. (and until 1950s(?), wasn't as available to many)

I seem to observe that now many companies generally want to pick the already-developed crop of people who fit the bill to a T, have had education or previous employment do the job of training / filtering who to consider, and not engage in the training part of it much any more.

Maybe it's due to the increasing cost of hiring people that you must only take a chance on a sure bet, and that there is less and less appetite (financial return-wise) to spend money on training people who are just as likely to pick up and leave for another company.

> Maybe it's due to the increasing cost of hiring people

... perhaps put another way, the 'professionalization' of HR?

I also meant, when employing someone for any kind of position of responsibility costs $100k minimum, you don't make big bets on unproven people.

Especially in "tech," job tenures are generally shorter than they used to be. It wasn't at all unusual at the big computer companies of the 80s and 90s to have a fairly large group of employees who had worked at a company for a decade or two (or more).

For various reasons, on both the employer and employee side, this has changed a lot especially when you're talking startups--but also companies like Amazon from what I understand. In any case, if the expected tenure of an employee is only a couple of years, it's probably unreasonable in most cases to have a 6-12 month training period during which the employee isn't especially productive.

I guess there's a case that someone who got an online degree, even from a reputable university, doesn't allow you to draw the same inferences quite as strongly?

That's what I gather. Colleges in the US have always been a bit of a gatekeeper. If its too easily accessible to students (e.g. lower costs, distance learning, ability to work while attending, etc), then the signal is weaker since more people finish college. It's kind of like an expensive bar charging a lot for drinks. The high price ensures a certain clientele.

I think that ties in to the proliferation of higher degrees, where college is not a strong enough signal so students go on to receive higher degrees that signal more time, money and investment has been spent

Which unfortunately means the most productive and creative years of their lives are spent piddling about in libraries. Woohoo.

> Most of the salary payoff for college comes from crossing the graduation finish line. [...] Unless colleges delay job training until the very end, signaling is practically the only explanation

Well, or employers think people drop out due to poor grades.

Three years of 4.0 GPA then dropping out due to pregnancy and three years of 1.0 GPA then dropping out before your expulsion hearing for cheating both put "3 years of college" on your resume.

Fundamentally attending college has an implied understanding of completing college. That almost completely explains the non linear gains from graduating.

It’s similar to closing a house sale. The money earned by brokers who get the buyer to sign the contract is exponentially more than the money earned by the brokers who only get 90% of the way there.

That doesn’t mean that the value of the initial 90% of the work is 0.

Of course, this is an extreme example for illustration purposes, because real estate is a step function where the returns go from 0 to 100 at a single point.

The exponentially higher earnings of grads can similarly be explained as a step function where earnings go from 50 to 100 after a certain event. And it’s not even that surprising, because the event in question is completing the thing that you actually stepped out to do.

I really enjoyed Bryan's book. I'd always thought uni was ridiculously overpriced, but hearing it straight from the mouth of a professor really hammered the point home for me. I believe we are on the verge of the education bubble bursting, just like the housing bubble a decade earlier

There is no doubt signaling at play, but one important signal conferred by graduating is “can slog all the way to completion of a multi-year, somewhat difficult, often tedious project, following somewhat arbitrary rules along the way.”

That’s close to actual job requirements in a lot of places.

That is indeed the big signaling aspect Kaplan talks about. It's still signaling, that is, what the student learns during that is considered less important.

A large amount of the value results from signaling but don't underestimate the value in taking a smart, ambitious 18 year old and putting them in a dense concentration of similar, like-minded individuals for 4 very formative years.

Before i had a degree, my father said it's really important etc because of all the things you'll learn. Now after he says it's just about making sure doors are open that would have otherwise been closed.

I guess maybe i needed to do my studies under one set of assumptions to have the motivation. Whilst only finding out on the otherside that those years were really only spent to get me my first job. And now my degree doesn't matter.

I would agree the door-opening effect is true for many walks of life but it doesn't apply to software engineering and similar disciplines. I'm a drop out; I studied economics and have been doing very technical work (low latency application development) for over 15 years now. Worked for UBS, Goldman Sachs, Morgan Stanley, HSBC, etc. No one ever asked me about my degree. They were all (rightfully) interested in what I can do for them.

True passion for the job will carry you much further than a piece of paper from a uni.

Did you have a hard time getting your foot in the door?

After 15 years, most people do not care anymore about your education; but when getting that first job, unless you have an impressive portfolio (and even then...), that piece of paper is huge.

With a recognised piece of paper, you can get much more easily get into a recognised company, which has a trickle down effect for the rest of your career.

I was looking for my first job during the dotcom crisis and it took me about 18 months. I took a semi-programming job at slavery-level salary just to get my foot in. What really helped was the fact that I had spent 3 months as an unpaid intern at a reputable software house 2 years earlier. I remember my friends mocking me back than for being a looser and an idiot for "working for free". This sentiment (entitlement?) is even stronger among young people today. They refuse to acknowledge that when you come in with zero experience and nothing to offer and get free training - you are the real beneficiary.

I think it's almost exactly the opposite. No one cares what college I went to, but the network I built during my time at University is still paying dividends.

Top tier universities offer both. Proof that you can get in, and once you’re in, a network of others who also can get in.

Well, it definitely serves a strong filtering mechanism to find people who have a certain level of probably-going-to-behave-in-ways-you-expect when you're trying to hire. Whether that leads to the most optimal outcome is not always clear, but it seems reliable enough (though boring) that many use it as a filter.

We don't have great alternative mechanisms to find employable people otherwise, and given the current unpopularity of standardized tests / uniform measures of performance or skill, what can an employer do?

This is a pretty common opinion here on HN, and colleges regularly get criticized for being nothing but a brand name.

This is the popular view on HN.

I take the opposite stance - being in person makes a huge difference, possibly one not appreciated by the demographics here.

People continually reduce the experience to “sitting in a box and ingesting information.

But a classroom is much more than that - it’s the ability to interact with other students, to deal with real world issues and practice even basic skills like asking questions.

It’s chances to interact in larger groups.

It’s resources which can’t be shared online at all (libraries, information terminals, printers.)

There are also people who depend on interacting with people to process ideas.

All of this modulated by the degree.

If you are doing some courses, you very much depend on interacting with people - doing a negotiations module online is very different from an in person class to pick an example.

And fundamentally- human brains on average are designed to work with people.

Mine isn’t, and being at home and studying isn’t a bad deal- but i would cease being objective if I didn’t recognize that there are many ancillary and direct benefits to studying in college, and for some people in person studying is the only way they can achieve it.

Finally I present to the people here Blooms 2 Sigma problem.

https://en.m.wikipedia.org/wiki/Bloom's_2_sigma_problem

People sorely sorely underestimate what “education” is actually about.

I don’t disagree with anything you said. I think it strongly argues for university being highly inefficient compared to optimal education strategies. University cohorts should probably be smaller, with smaller class sizes and more stringent entry requirements to work toward the ideal 1:1 student:teacher ratios. Sure it’s more expensive, but if you get 2 std deviations out of it, that’s huge.

For certain gen ed classes, maybe have the whole cohort in a single class, not for lectures or exams, but for discussion and collaboration. Make those courses fully project based and assign them uniformly across degree boundaries.

Right now, graduate level courses and prestigious institutions are the only ones even trying to offer a higher education. The majority of universities at this point are paper mills, rushing hundreds or thousands of students through programs without most of the benefits you described. It’s a system optimized for making money or university survival, rather than student achievement or societal benefit.

This can’t happen (in the United States) until there are alternate ways of making a decent living for normal people who can’t self-teach math and programming skills. College is a mandatory job “certification” for a lot of office and other types of jobs so college has to be approachable enough that almost anyone can be accepted.

You are absolutely correct, which is why I keep harping on the Bloom 2 sigma problem. Its an easy way to highlight to readers how significant human 1 on 1 mentoring is to academic outcomes.

That immediately makes it clear to people that we dont have the money or people to afford that outcome, and after a few traversals on that ladder, the issue of education being a huge method to minimize the cost to educate people without overloading on number of teachers and cost of HR becomes apparent.

Mentorship is my new educational holy grail today frankly. Gone is the hope of the MOOC, now we need the Young Lady's Illustrated Primer, a mentor that customizes information that matches a student.

Fundamentally though people learn through play, thats how humans are designed.

Kids pick up 5000 combinations from playing league of legends, and do stupid number crunching on the fly.

However play is to hard to create, because we don't associate fun with a lot of work.

If we ever solve education, it will be solved by using play, even Mentors must understand fun.

> ...now we need the Young Lady's Illustrated Primer, a mentor that customizes information that matches a student.

In case you didn't get the reference: https://proto-knowledge.blogspot.com/2011/11/building-young-...

And you aren't even mentioning the case of degree programs that are in some essential sense about learning to use equipment costing in the tens to hundreds of thousands of dollars, and in some cases whose availability is strictly controlled by the US government.

I got my undergrad in Nuclear Engineering, and I learned to use Gamma Ray Spectrometers costing easily 30,000 dollars and did Plasma Physics experiments in a Vacuum Chamber setup of similar cost, all the while supervised by people who knew how to safely perform experiments with radioactive material and comply with relevant government regulations. And I was mainly about Computational Neutron Transport - I got what was considered a bare-minimum grounding in Experimental Nuclear Science. This is to say nothing of programs with research reactors students can use for experiments (whose ranks my alma mater sadly left just as I arrived).

This was basically true of all the non-CS engineering disciplines. Perhaps one could cobble together an equivalent to a Mechanical Engineering or Civil Engineering degree from my alma mater with CnC Routers at Hackerspaces and the like. But I really can't imagine that working for Nuclear, Aerospace, Biomedical, the Semiconductor side of EE...

Or rather, I can, but it would require an enormous investment into high-fidelity simulation of the relevant equipment. And that is a dangerous road. At the moment, we verify Engineers' ability to use equipment in a very robust way - having them actually use the equipment to perform engineering tasks and measurements. If you rely instead on training simulators you need to be extremely vigilant about their fidelity (and you will almost certainly fall back on minimum duration of instructor-monitored usage of real equipment, like the nuclear and aerospace industries require).

Also, chemistry/biochemistry. How would the labs work? I mean, get real. For anything software, sure, that can be taught almost totally remotely. But most actual engineering disciplines just aren't going to have effective remote coursework with hands-on labs any time soon.

There is almost no demand for 'actual' engineering a ton of this is completely offshored into oblivion, my girlfriend has a degree in chemistry and now works in a bank. There is extremely little demand for everything outside of software.

I don't buy that in the slightest. IT is, hands-down, the most commonly outsourced. The building engineers who inspected my foundation made me wait for 3 months because of demand.

Are the civil engineers running the local water treatment plant -- which can't seem to get enough qualified people -- going to hire people who don't meet the requirements to be called an "engineer" (unlike the US, these reqs are like doctors and lawyers).

Bloom's two-sigma problem is only tangentially related to your argument. What Bloom found was that one-on-one instruction using a very specific type of teaching (Direct Instruction) coupled with mastery-based learning, results in some pretty spectacular gains (at least according to a few, rather underpowered studies).

If anything, the two-sigma problem argues against in-person teaching. Mastery based learning requires that a student prove that they have mastered the current concept before they are allowed to move on to the next concept. That is much easier to do in an online setting, where each student can be presented with customized material, than it is to do in a class of 30-or-so bored undergraduates.

i dont see why the two are mutually exclusive or something. You can both have a reasonable education in a college AND have it play a role as signaling as well, because nobody is going to check what you studied in details.

They’re not unrelated. Part of going to an elite school is the ability to network with older elite school members; this is literally what fraternities are for.

You dont need to have that to get a great high paying job. Most jobs in the tech industry are talent based not merit based.

There are some areas like doctor or experimental physics where physical presence makes a lot of sense. But most education can be found and consumed much better by thinking of the internet as your entire college and then work much more on self-motivation and curiosity which are the real secrets to educating yourself.

> "...the main value of a reputable college is having a reputable college you can put on your résumé."

That might be true for software developers but it's certainly not true for any other engineering field, medicine, most sciences, etc. that require labs or other equipment. The university I attended had its own research reactor on-site for its nuclear engineering program.

It's not even really true for software developers at the top end of the spectrum. Sure, if you're entering a software career trajectory where your education is unnecessary (which is most of them these days), then, yes, going to an elite school is a waste. If you plan to be building OSes, programming languages, doing AI/ML, etc, then learning that stuff from people who know where the cutting edge is invaluable.

Why do you feel like that, what evidence is there for your hypothesis in your personal life, or in statistics that you’re aware of?

There’s a wide variety of viewpoints on HN, so I think it’s unnecessary to speculate on what other people think, but personally the thing that doesn’t sit that well for me is that you’ve stated an opinion with nothing at all to back it up.

FWIW, having gone to both a not super reputable college (state school), and also a very reputable one, I can’t say that reputation is the “main” value, I might have benefitted from it but I wouldn’t even know. The math classes at the reputable school had much higher standards, they were harder and I learned more. I have some lifelong friends from both schools, but the group from the reputable school is generally much higher profile and further reach. I sold a business to someone I met through the reputable school. I also now work for someone I met through the state school.

I highly disagree with this one. I have one of those brand name diplomas, but the piece of paper is largely irrelevant. Nobody has cared or even asked about it since maybe 3 years out of college.

As a restrospective thought experiment, if you offered me that piece of paper in exchange for a student loan the size of my college education, I would turn it down. But if you offered me the 4-year experience I actually got for that same loan, without the brand name, I would take it in a heartbeat. (You can't actually separate the two since the brand name in part enables the college to implement the product.)

Aside from actual academic knowledge like practice with low level OS and network concepts, here's what I got from those four years that I think made a concrete difference in my life and career:

1. The personal mentorship I got from a professor, who I met during a summer internship, but he was from my comp sci department. It was serendipitous that the two circles collided, but it was crucial, and, to the point, it was enabled by the college network.

2. Meeting people through friends in college who would become co-founders or connections into other jobs.

3. Taking very specific classes on soft skills, such as writing workshops. One in particular, which I took early in college, was like getting two in depth code reviews a week for twelve weeks, except it was for my writing. That course made me fluid and precise at writing. I still suck in comparison to many of my former classmates who write professionally, but it made me a stand-out effective communicator in engineering. In retrospect I would have paid 1/4th of a semester's tuition in a heartbeat for that course (took 4 courses that semester).

4. Being surrounded by people who weren't engineers, who were formidable doers and thinkers, about other fields, such as global health policy and politics. I still remember getting my ass handed to me in a dinnertime debate by someone who was obviously more knowledgeable and more thoughtful about public policy than I was. Getting intellectually smacked down by friends regularly taught me humility and appreciation for the value of expertise in other fields.

5. Spending four years in the company of driven people with high standards. For the people who embrace it, it rubs off on you the same way working on a strong team does. It demystifies high performers and teaches you that you can turn yourself into one.

6. Due to spending time around people who had money/power, you learn to not be (as) intimidated by those people. It's a crash course in walking and talking like an "elite" whether or not you are one.

I think it depends if you actually intend to learn something or not. College is really only what you make of it.

The main value is extra curriculars being of a higher caliber at top colleges than at lower tier ones. Getting to engage with other very talented people building organizations and so on in a much more serious way matters.

There's a reason why ivy league college newspapers are far more impactful than at lower tier schools.

For me, university was a bit more about direction-finding and serendipity. Few places can expose you to very different things in both a structured and unstructured way as college.

When I started university I was a returning student with a budding tech career. I promised myself I would get a degree, but I wasn't totally sure what benefit it would be. I thought I already had the tech skills needed to do what I wanted to do.

While taking some required political science classes I discovered the world of development cooperation. I did a bit of traveling. I decided to use my time at a 4-year to explore the world of NGOs and other large-scale initiatives. I did a study-abroad project to do some free tech work for a global nonprofit. I fell in love with a new city. I did a short masters to further my knowledge, and also see if it would be possible to combine my tech background and this new world. It was, so I did.

It's hard to know if something similar would have happened without that experience. It's hard to put a dollar amount on it, since I a) paid for my own education and b) probably earn less now that I've changed directions than I would if I hadn't. But I'm so incredibly glad I did. I also don't know how to compare this story to the many stories of people who graduate with 80k in debt and can't find a job in field in which they studied.

> I also don't know how to compare this story to the many stories of people who graduate with 80k in debt and can't find a job in field in which they studied.

I have a friend going through this scenario minus the 80k debt. The field might not exist in their location. She’s looking at relocating and losing custody of her daughter. Or working another job and staying put for her family. I’ve met a lot of brilliant people who’s ambition has been bypassed or delayed due to external factors.

Sometimes getting the brass ring means sacrificing more than you can give.

Something curiously missing from HN threads about uni is the social aspect. I consider my social experience in uni, from living with other students (instead of mom and dad) to studying abroad for a year, just as if not more valuable than the academics, and a major catalyst for finding what I wanted out of life.

There's a lot more to life than credentializing for wage slavery.

One thing I've definitely noticed in CS is that having a TA sit next to you in the computer lab and help you debug your code is a huge value add, and if it's only to track down a misplaced comma in "Introduction to C".

Doing the same over $ONLINE_SERVICE is not quite the same experience, just like a terminal-emulator-in-a-browser isn't quite the real thing when you have slightly higher and jittery lag. And an async/await model (post code on forum, get back answer anything from minutes to hours later) is not the same thing at all.

Currently a student, and I completely agree. The strain is both ways, too. Since timezones are now a very fluid concept, you're seeing courses accommodating completely new challenges in TA-ship. Some TAs have relocated entirely, which means that the course staff can be much smaller than anticipated.

For some students, they can only receive advice asynchronously and it makes it difficult to meet deadlines based on EDT timezones. Even video debugging/office hours for people in EDT is awful, despite advances in screen sharing/Zoom.

Also, I helped transition a professor to virtual instruction. They had never installed the Internet in their house before. They own no smartphone. They communicate entirely through landline and a dated desktop PC in their office. When absolutely necessary (course admin, journal paper updates) they will use Pine for email. This prof taught a 30-person graduate course in neurobiology having never touched any of the technologies they had to use. Bless that class.

I think there is a lot of value in learning in the same place that others are learning. You have a set schedule, you have people to discuss the material with, you can learn from other students, you gain experience working with peers in a group. That being said, all of this can be done without the actual institution of college as it now exists. If you could organize a small group of people that want to learn the same material then all you need is a place to meet. The only things lacking then would be access to the professor and access to a lab but those have solutions as well, you can look for someone who has already graduated from a similar program at a high level or someone working in the field and ask them if they could work with the group or answer questions and for lab work you can search out a hacker space or other community group that has the required materials or maybe split the cost of buying the equipment with your group. I'll admit there's a hard limit to how far this will go, graduates != experienced educators and some equipment will only exist at universities or research institutions but I'm pretty sure that 80% of what is currently done at college could be done in a coffee shop with good wifi.

From my experience the social aspect of college is definitely worth it. Ofc, you may find a group outside college then there is no need for that. Otherwise college is the best place to get socially better.

When MIT put their course material online (thank you Hal Abelson of SCIP fame) they explicit that weren’t thinking it would substitute for an MIT education but just as a resource for others to use for their own courses. I found it handy when I needed to brush up on my thermo (ironically I hated it when I was in school) to review the current materials (as I’d already taken the course years ago The problem sets were enough)

Yeah, the original concept of OCW wasn't really as materials directly for "end users" at all. I imagine this also helped sell the concept; "people like our current students aren't even the audience for OCW." Over time, OCW has added more things like video lectures as video became cheaper and more ubiquitous but it never had a stated objective of being a course in a box for students to consume directly.

A lot of people were underwhelmed by OCW early on because they were expecting something a lot more consumable for a self-learner--which led to MOOCs of course.

College has never gone through an event like this at the scale it currently is at. The idea that every American should attend college is a relatively new one; during the last pandemic in America only a tiny percentage of Americans went to college.

As many have said, the bulk of the value of going to college is in the signaling they're able to provide. You're paying (overpaying in my opinion) for the infrastructure they've put in place to adequately assess your abilities in a way that employers can trust.

Say you pay someone to do an inspection of your home that the gov is requiring before you can sell. Has that inspector taught you anything or made any tangible improvements to your home? No. What they've done is signal to the gov and buyers that your home passes inspections.

I think the college experience will survive precisely because the value add is the social aspect and the cross-pollination that comes from being near tons of people in lots of fields working on interesting things.

Colleges even market this aspect as the key differentiator from online classes.

In that case, I do think some refund is due on principle, because students at home in isolation definitely aren't getting that.

Depending on the place and people (i wasn't very well when I was in college), MOOCs can be better in many metrics.

    - less anxiety being in IRC than in class
    - more focused (less distracted by people inviting you, peer pressure)
    - more communication with other students and teachers
    - automatic grading systems being very exciting to keep improving solutions rapidly

You didn't claim they weren't, but I just want to point out how each of those points is very, very subjective. For me:

- I'd much rather be in-person, where I can better gauge what others are communicating and more readily have two-way communication about the material.

- I'm way less focused when on my own computer, on my own time.

- I find I communicate way less purely online than I would in-person.

- Automatic grading sucks, especially for programming, as it can only really test inputs and outputs, and misses a ton of nuance. I want to also know if my code sucks, regardless of it producing the correct output or not.

The main thing MOOCs do well is a solved problem with or without MOOCs--namely lecture videos.

Discussion boards are mostly a tire fire; they're mostly useful for dealing with platform problems. They can probably be better when you restrict access to a qualified class but they don't really scale.

>Automatic grading sucks, especially for programming

I'd probably say automatic grading sucks, "even for programming." At least automatic grading can evaluate outputs of non-trivial programs. Programming is one of the few areas where autograding works for non-trivial, not multiple choice, questions.

I concur with boards being useless in general. It's too slow, not predictable. That's why I liked the moocs IRC channels. It was a live place where we could discuss things and resolves issues quickly.

As a GA Tech student for two years now (GA Tech's OMSCS program is all MOOCS), MOOCS are generally horrible ways to actually teach.

-I have no idea why anyone would have anxiety about being in class, I don't think that's a common problem.

-There's less focus in online classes because now you have to manage all the different ways you have to communicate with everyone, official and unofficial, in the hopes you can get enough interaction to slog your way through the horribly written assignments and figure out what they actually want you to do.

-There's essentially zero communication with teachers -- they're almost totally absent, communication with other students can be helpful, or not, and you don't always know. There are typically two or three TAs for up to 600 students, so any kind of focused interaction with a TA is almost unheard of, and even if you do get that their level of expertise is often in question.

-autograders work well for some very defined sorts of problems, but they don't work at all for most workflows. The TAs, in my experience, spend most of their limited time fixing autograder screwups.

> I have no idea why anyone would have anxiety about being in class, I don't think that's a common problem.

Like 1 in 5 college students[1], I have a diagnosed anxiety disorder. I frequently have anxiety in class, especially before the professor starts talking and is just scanning the room and occasionally making eye contact with me, or if I'm asked to interact with other students.

That said, I still think that MOOCs aren't remotely comparable to real college classes.

[1] pg. 16 - https://www.acha.org/documents/ncha/NCHA-II_Spring_2018_Refe...

> There's also something about the competitive aspect of seeing other people in person do better than you or have better ideas than you. Which is hard virtually.

Maybe even this is not that hard. People already compete virtually on mobile and console games of all kinds of complexity just to see their names at the top of a "leaderboard"

> what is the value add then of paying to physically go to college?

If you can tell from a resume someone attended in-person, it gives their degree more academic integrity. Not that you can't cheat if you attend in-person, but it's a whole lot harder to pay someone to take the exam for you.

How much of Harvard and MIT's catalogs are actually online in their entirety? (incl. textual resources, coursework, answers, etc.) IIRC it was more like a few handfuls of introductory courses.

> well what is the value add then of paying to physically go to college?

Ivy league and other high end colleges are much more a brand than they are a product.

>the networking, friends, social aspect

The what?

Assignments are often locked down.

Reply on news.ycombinator.com