And when that link fails, when you can get a certificate in 3 days as opposed to 3 years, when you can study interactive online courses with top-notch content at your own rhythm as opposed to a 4-year outdated programs, then it will happen: certification will be independent and traditional universities will have serious disruption in their tuition prices and budgets.

What will not happen is the demise of high-end education and student loans. The senate & president won't let it fall; US needs to keep itself competitive, especially versus EU where most states heavily subsidize education from taxes. The society and the police can't afford uneducated children.

A good STEM program teaches you how to think, and that never gets outdated.

Colleges offer a smorgasbord of courses to take. If you select courses for the easiest route to a degree, you won't be that employable. If you select courses that will give you the best foundation for your chosen career, that'll do much better.

For example, although I got a degree in engineering, I made sure I also took a class in business accounting. (That class has paid off well for me.)

Honest question: does one not learn to think before college? Where are the parents? What happened in high school?

All these skills are developed in the process of STEM training. I know I got an awful lot better at it after 4 years of that, and I can see the deficits in those abilities in people who have not undergone such.

For example, given someone debugging their program - I see many programmers who will just try guessing things randomly. If they get lucky with this approach, they often have little idea of why their change made it work. (Often, nor do they care.) But they do spend an awful lot of time with this method, compared with someone who sets about finding the problem in an organized fashion.

Maturity plays a huge role. A lot can happen in 4 years of independence.

So since you're going to learn to think after 18, a good college education provides the shortest cut to learning to think WELL, based on the accumulated experience of thousands of years of human history.

I've never seen evidence that specific courses matter much for employment post college. Do you have data on this?

And yes, I've seen EEs flummoxed when faced with a bit of noise in their digital circuit, when a guy who knew some analog had it fixed pronto. Which engineer was more employable?

I slightly disagree with this position, difficult doesn't mean better (depending on your definition of better). For example, I would have been better served to take a class on databases rather than a class on computational geometry.

So many smart people apply to Google that they could (at least historically--maybe it's different now) get away with a hiring process optimized for false negatives over false positives. This is only fair, but also probably makes them a poor example as far as hiring practices go.

So yes: they can figure out if you're smart in 5-6 hours. But they can't do the converse reliably.

That means a lot of smart people never get the opportunity to try out for the interview. In my opinion, you have to market yourself better and differently if you want to increase the odds of getting to the interview, where you get a chance to prove yourself beyond what is on paper.

On a general note, is there any company that tries to extract patterns out of their employees performance track records; like a feedback loop into their interview process?

I know about people with no college education being interviewed at google

You can blame google for having a broken interview process but it appears 99% of their assessment comes from the interview and it really doesn't matter if your major is in pet grooming or math

Edit 'cause I can't answer: Not "friend of a friend" account, but directly from the concerned person

'"99% of their assessment comes from the interview." Indeed, it's the very opposite since the interview is not where you show that you can |excel in multiple areas."'

I don't believe you are familiar with the Google selection process. The importance of academic achievement (for engineering positions especially) is underrated by google.

You may shine in academia, if you don't get almost a perfect score in the interview you're out.

You said they were interviewed. You didn't say if they were hired. If they weren't hired then was it partially due to a bias against people without a college degree?

You also said "know of", which means this is a friend-of-a-friend account, with bearing on the usefulness of that information for this question.

Better would be a pointer to http://management.fortune.cnn.com/2011/04/07/so-you-want-to-... where someone who was on a Google hiring committee "acknowledges that the 3.7-or-higher-GPA myth is widespread, but she discounts it." and says "Academia is merely one way to distinguish yourself, and there are plenty of others. So if your GPA, or your school, doesn't stand out, look for additional avenues. Besides, you'll need to excel in multiple areas to get your resume selected."

I read this as saying that good grades from a good college help, because it's one of the ways to show that you excel in an area. This also implies that excelling in college is one possible indicator on success in working at Google.

That's very different than saying that "99% of their assessment comes from the interview." Indeed, it's the very opposite since the interview is not where you show that you can |excel in multiple areas."

They were in the early days, but it doesn't seem to matter anymore. Some regular posters to HN are Google employees void of any college degrees. The change indicates to me that their attempts to filter by education was failing them, which is interesting in itself.

If you are applying for an engineering position and you have MIT on your resume, that is certainly going to imply that you very likely have what it takes. But it still doesn't prove anything. On the other hand, if you have no formal education on your resume but you've built a fusion reactor in your basement, I'm pretty sure your resume wouldn't get thrown away.

"Built a fusion reactor in my basement" is neither of those things.

Similarly, "check out my GitHub" has many of the same problems. I'm sorry, but most employers don't have time to look you up on GitHub to crawl through all your commits. That will happen if you get past the first cut, but it won't do anything beforehand.

Sure, you can put "Main developer for <some project>" on your resume. Hopefully the person who is in charge of culling applications has heard of <some project>. If they haven't, they will hopefully take the time to go look it up. Hopefully.

I'm not saying that this is the way it should be, but you have to understand that the average person in charge of hiring is constantly inundated with applications. They simply don't have the time to do research on each one. As with anything, you have to find some way to stand out, but in a way that doesn't require them to visit your basement.

In that case they shouldn't look too surprised if they end up hiring wrong people. Or they end up hiring a person with a lot of paper degrees but can't move a coin on the table.

Because you ultimately get what your truly ask for. You asked for degrees, you got it. Why is this so surprising?

Ask for innovation, creativity, hard work and productivity and you will get it. If you want to hire such people, you may to have to visit their basement or look at their github logs. If you can't do that you shouldn't complain you can't hire such people.

People often ask why do great hackers, nerds and other geeks don't gel so well large heavy bureaucratic corporates. Its because the rewards vs performance incentives are all badly messed up. More often than not somebody who has a degree from X college, with Y GPA gets the money.

We recently hired a guy straight out of college, extremely good marks on paper. He seems to know every other data structure and algorithm in the book. But doesn't last more than couple of hours. On the other hand, I've known people from ordinary background with medium exposure, able to last through tough deadlines and crisis situations with enthusiasm. The latter people make far more better hires than the previous ones.

Your degrees are of no use if you can't perform. That fact doesn't change.

You have to put yourself in the shoes of the person in charge of hiring. If you advertise a well-paid tech position, you will probably get thousands of applications. You can't read them all in elaborate detail. You can't visit everyone's Github. You just don't have time. So how do you cull? Just randomly select 10% and go with that? What if you happen to pick all the bad ones? That's not going to work. Industry experience? That's easy to fake and also doesn't really mean anything. Contact their previous employers? Those interviews are notoriously inaccurate and also way too time-consuming.

If they graduate from a good college, that's possibly a good sign (but possibly not! easy to get a CS degree and have no work ethic). If their resume lists a bunch of personal projects and GitHub links, that's possibly a good sign (but possibly not! easy to push random crap to GitHub that no one will use).

I'm afraid it all comes down to time. Either you hire an army of extremely well-trained HR people who are all capable of evaluating code quality and project performance (in which case, why aren't they off programming themselves?) or you come up with some heuristic for culling applications that doesn't take an inordinate amount of time. It's harder than it sounds, and that's why employers fall back on do-you-have-a-degree - because it sits at the sweet spot of sort-of-predictive and easy-to-cull. Or they end up asking their current employees who to hire, which means they hire their friends (which is not necessarily a bad thing).

In the world of social networks and artificial intelligence, it is interesting that nothing has come up to provide better recommendations. Perhaps because it's not really needed in the software industry at this point (at least if you believe there is a shortage of talent), so it is not noticed as a problem by the programmers who are capable of making such a system a reality.

I wouldn't be surprised to find this sector increasimgly revisited in years to come. I've recently come to experience that programmers without much experience, no matter how well-intentioned, sometimes just stomp around doing far more damage than good on projects.

The screen would be a quiz, puzzle, or other technical challenge which can root out the obviously unqualified (and possibly net you some viral buzz), while being easy for an HR drone to administer. They don't need to verify the code, just that it produces the correct result. Mind that "correct result" should be sufficiently qualified (e.g.: parsing /etc/passwd for valid user accounts will vary somewhat by 'Nix system and Linux distro).

College degree does make for a useful filter. It's not the only one. For a company which is deluged in qualified candidates, the problem isn't making the exact correct hire, but in eliminating obvious misfits quickly. Now, I'd also hope that they'd make exceptions for personal referrals who lack a college degree.

Also: college is itself a place to seed referral networks. Again, not the only one, but there.

Just enough to show basic competency.

A shell one-liner, a brief function in a programming language, some basic domain knowledge.

Those people you request send a code sample or their Github profile.

You've already knocked off about 75-90% of the applicant pool.

The reality is that they get so many resumes, like thousands, they are forced to do an initial culling. That cull takes about 2 to 3 seconds per resume.

If you don't have a degree, you won't make the cut. We all know that means some great candidates will be missed. But any means of whittling down those thousand candidates runs such a risk.

They shouldn't, but they probably will anyway. I think you have to assume that lots of companies will be stupid and work with them or work around them.

Most people miss the point on why this is important. MIT doesn't teach its undergrads anything they couldn't learn at the local state school for a fraction of the price. But you have to be really smart to get in to MIT. That's what employers care about - not what you did when you were there, but that you got in.

The root of the problem is Griggs v Duke Power. This problem isn't going to go away until companies can give IQ tests again without the legal peril, because otherwise employers don't have much to go on besides where you went to school.

I think it's more subtle than that. If you start with people who are very good at learning (which is all you can know about them at college entry time) then you can teach them more in 4 years than you can the "average" person. So the MIT guys probably really do learn more. The question is of course the extent to which that translates to on-the-job performance, as good learner doesn't imply good doer.

As a matter of fact, yes, UMass Amherst is quite competitive with MIT in STEM departments. Thanks for noticing.

I work with students for development projects. We don't get to pick them, they pick the projects we propose. Here's a question that tells us in seconds whether we're going to have a tough project with a lot of pussyfooting or one that'll go mad with ideas:

Do you have tech-related hobbies?

Most barely have a few hobbies but some are pretty much involved in every electronics/software niche they come across.

You want to find the people who care enough about technology (of whatever type) to invest their own time and resources in it. I've met too many so-called professionals who see development as their job, and see no reason whatsoever to do it for recreation.

The only "no" answer that I ever respected in this regard was "Hobbies? I am working two jobs to get through school. I have no time to do anything but work and study."

I take questions like this to mean that the employer expects the employee to work outside of work hours, to not have a family, and and to not do other things. Perhaps that isn't the intention, but it seems to be the implication.

Your average lawyer/doctor/accountant ends up in a very predictable career, just like your average engineer.

People who end up in politics tend to have a degree in law or finance and debating as a hobby. For jobs in politics, that's the kind of folk they look for, not just the "I have the right degree so hire me" types. Look at doctors who work in non-profits abroad: you'll find people with a very active engagement in social groups in their youth (I know a few in that profession) .

But that did happen. Yale may not "go out of print", but I don't find it hard to imagine new online schools competing and winning in the battle to redefine education.

You need to be calling your contacts at Mark Cuban and saying, "Yeah, I'm in good shape here, but I'm willing to look at other things."

After a few years of work experience, all that really matters is the last 3-5 years.

You could pretty much omit everything before that, because just the last 3-5 years is usually enough to figure out the caliber of work a candidate may have been doing before that.

If all résumés only included the last 3-5 years of an employees life whether or not that includes their education, would it really be worth that much less as a factor using in the decision to interview or hire?

Universities have requirements you need to cover in order to graduate. So when a job lists a BS CS degree as a requirement, they're saying they expect you to be familiar with the core subjects of the field. Certainly not an expert in any one subject, but at least familiar. (If they want subject area experts they'll usually be advertising for MS/PhD with a focus in [blah]) Obviously this still gives the Universities a great deal of trust, with the implication that in order to receive the degree, you had to demonstrate your capability. That's a sizable question mark, even though it's what University accreditation is supposed to be for...

Sorry, went off on a bit of a tangent there. Anyways, four years of self-directed learning won't necessarily cover the same subject matter they may expect you to be familiar with.

You know how much I paid for college? $0.

You know why I paid that? Because in high school I did a lot less than what you outlined above, but it was enough to get a scholarship to go to college. Six great schools looked at my application and offered me somewhere between a discounted rate and a full ride + stipend on top of that. And I was only 20th in my high school class by GPA rank.

If you think that you can match a college degree based solely on self-guided learning, then you should be smart enough to get a scholarship. It's the same damn process. You need the self discipline to stick with a program of learning and working, and you need the intelligence to learn without support. It's just a question of whether you realize that in high school or later in life.

I had the same experience as you (college paid for via scholarships), but the scholarships available differ from school to school and community to community. So, not everyone has that as an option.

0 < correl(school chops, business chops) < 1

No HR department can interview people without a degree when there are so many people with degrees. A degree doesn't get you in the door, but is a minimum standard for just about any job.

It is too bad school is too expensive, but it doesn't matter.

Nonsense. I have never been to college and I am currently responsible for the IT Security of a mid-sized health network. Not only was I hired without a degree but I've been promoted several times.

I also have no technical certifications, at various points in the past I had some but have let them lapse. Having a piece of paper may make things easier, but in the end it comes down to whether or not you can sell yourself to the organization.

It's a great company, but my experience doesn't matter to them. Without a degree, I cannot be hired. Schools in my state don't even offer Security degrees, so it's not a matter of what the degree says, it's a matter of having the degree. I literally cannot do anything more to prove my value to the company. They don't care. It's about the paper.

In what sense? In a social-justice sense it seems like it does matter quite a bit. In a national-productivity sense "the college question" also matters quite a bit. In what sense doesn't $trillions of spending matter?

It does matter. The schools that are worth anything are too expensive (exceptions are there of course) while the cheap ones (read < 10K a year) are probably good for nothing. You are better off trying to get a job instead of attending those cheap schools.

Cheap schools are not good for nothing, and price does not equal quality.

You are saying this: Ivy/Private School > No college > State School

That statement is absolutely ridiculous.

I am talking about accredited universities, big state schools that are very affordable and provide fairly good educations.

I would rather attend one of those than no school at all.

The domain learning useful for my career came after.

You want to be a janitor? Congratulations, you can now get a master in "facility management". Teaching special needs children? You better get started on that degree! You want to be a librarian? Well, what about getting a degree in "information management"?

You got a degree in communications and have freelanced for a couple of years as a copy writer? What you gonna do? Ah right, you start to teach "corporate PR" at a university. Demand is that high that I could probably start teaching shit like Java, which I don't even know, in a heart beat. Without having any pedagogic qualities whatsoever.

We live in a day and age where everybody wants a slice of that big fat cake. Everyone is entitled to 10k a month. Everybody wants a designer apartment, a BMW. Every year, a new smartphone, vacations in the Maldives. Living the dream. On top of that, costs are rising across the board, mandatory health insurance is killing everyone. Prices are inflating. So what you gonna do? Ah, right, you get a degree, it doesn't even matter which one. Shit, three years ago my mom got a degree (she was a Kindergarten teacher). She is 53 now.

I really don't know how this could last. Maybe it can last for a small country like Switzerland, but I have a very bad feeling. A lot of the people will one day wake up and realize that just because the studied something, they're not going to earn 200k a year, even not 10 years down the road, yet they already live a lifestyle that assumes this to be true.

edit: My girl friend proof reads master theses for people. You wouldn't believe the shit she has to read and correct. Without outside help, many people would fail miserably. If you pay enough you will get your degree in one form or another. And that is precisely what is going on.

I used this example because it relates to my mother, who recently got a master in this field. First of all, I'm proud of how my mother pulled that off at that age, while still working at her old job!

That said, the courses were largely theoretical mixed with a minuscule amount of practical studies. How many of the graduates will be able to pursue an academic career in this field after this degree? Maybe 1% (Note, this number is pure guesswork). The rest will actually end up assisting kindergarten teachers with their "disorderly" children.

In many cases, they should function as a buffer, protecting the child from overzealous kindergarten teachers who want to get rid of the troublemakers asap. This is important because a lot of developmental "deficiencies" are not there forever or can be mitigated. I digress. From my point of view, the actual theoretical work was frankly substandard. I don't think it has to do with me applying the standards of "hard science" to a "soft field". No, simply put, those professors didn't know that much to begin with and neither did many of the students.

As an aside if you were to (honestly) think about almost all fields of inquiry (maths may get a pass, possibly) then I think that you would have to agree with the statement "those professors didn't know that much to begin with and neither did many of the students". We are pretty ignorant about more or less everything when you look closely.

Imagine a course on compiler design where the professor doesn't know what top-down parsing and lexical analysis is. I'm certain that you would call that class a failure, or not?

I was astonished that someone with a master's degree would need help putting together a resume. Something is terribly wrong.

College is what you make of it.

How many recent grads outside of engineering do you know? Now how many of them are gainfully employed?

Now how much did they pay for school and how much pay and experience did they miss out on by not working for the time they were in school?

I just went through the post graduate job cycle along with a huge batch of my peers and any of us will be the first to tell you that college is broke. The only reason I am employed is because of projects I put together myself, online, outside of the scope of a university education.

If you are looking to meet girls and learn how to drink to excess by all means college is a great place to do it. Otherwise college education in the liberal arts is a complete waste.

Also, most of the useful information I obtained in my degree was which resources are the best.

Would you trust a doctor that only learned on the internet to operate on you?

/nuclear engineer

And also medicine, biology, most of engineering, etc

Sad, yeah.

Not everything is like the software industry. Indeed, most things aren't.

Obviously all those skills can be learned elsewhere, but as a hiring manager, it's a major time investment to try and suss that out whereas having a college degree with good grades means more likely than not that the candidate possesses those skills.

I agree that some things like research can't be done online. But I don't think those will remain at the university either. Graduate/research degrees must represent such a small % of a university's revenue that when the majority goes away they simply won't be able to survive.

Just ask the RIAA if they can survive by selling mp3s when they're used to selling overpriced CDs.

universities exist solely for the purpose of enabling professors to do further scientific research

to do this they need:

* a supply of intelligent grunts (phd candidates)

* an administration capable of turning grant money into a well stocked lab

and

* a supply of very rich young people who will be taught the basics of science such that when the young rich people run the country they might get something right

unfortunately we have forgotten that one genius well supported can provide value and economic growth for a whole planet (who invented steam engines, fire, wheels, transistors) - and  million English graduates wont make up for it.

I would be perfectly supportive of a university system that publically funded scientific (empirical,  provable) courses, had industry funding for vocational work (engineering) and left arts to do what they have always done. 

But saying let's get rid of the lot? Nonsense.

First of all college teaches knowledge that you would need relevant to your major. Then college is a signal that you can learn stuff that is required (but not necessarily that you would like to learn). For example sometimes you might just not know what you need to know. Like math. I am glad I was forced to take calculus because I find it useful in my programming job, years later. Taking that made it easy to take understand and analyze algorithms. I am also glad I was forced to take English and Critical Thinking. I stuck with it for 4 years. That also tells something about me to a potential employer (maybe the wrong thing, maybe it shows I can be a submissive grunt that will willingly follow orders, or maybe it shows I can get stuff done to a future startup partner...)

So, taking costs aside, I think there is some value in college.

Now bringing the costs back into it, I agree with you, that all this should probably somehow be obtained for less than $100k. There is a terrible inefficiency some place if it actually costs that much. There is terrible waste some place. I remember my University was building large stadiums, and huge gyms with lazy rivers and other crap in them, while some colleges couldn't afford paper to print exams on and was using single spaced double-sided, small fonts.

There are efforts underway to provide the education, filtering, verification and signaling services for $0-10k per person. It can be done. We just have to cut out all the fat.

This seems intuitively true - 4 yrs/50-100k looks like an unjustifiable expense - but efficiency is relative. If we don't have another way of systematically vetting knowledge workers, then talk of efficiency is moot. There's room for improvement/disruption, but the parent's response to the gp's "fuck college" sentiment was well measured and in keeping with the sentiment of the linked article.

And from what I can see most of the budget comes from and is devoted to college football.

This couldn't be further from the truth. I'm unaware of any major universities in which sports money comes from or is applied to general budgets. Athletic departments are generally self-contained and there is very little (if any) financial overlap into the academic portions of a university. Payroll, and even scholarships, are paid from athletic income, and any left over is not put into general budget.

Personally, I wish the myth of "the cost to the university/tax payers/etc" would go away.

I believe this mostly because the big conferences have fantastic TV deals and huge research schools.

I've heard this many times. Why is this? I get some research requires access to physical equipment that is out of reach of the common man, but it seems something like algorithmic research in the field of CS could often be done online.