Hacker News Reader

Comment by Animats | Hacker News Reader

This is from the operator of a coding school. They have an interesting payment plan. "There is no upfront cost to join Holberton school. We only charge 17% of your internship earnings and 17% of your salary over 3 years once you find a job." That's more interesting than anything in the article. The school takes the financial risk on students not getting a job. This is the reverse of most for-profit education, which relies on loans and doesn't guarantee a job.

I wonder how this is working out for Holberton School.

View on HN

[I am the cofounder of Holberton] it is working great for us. As you said we do take the financial risk on students not getting jobs. Education is so broken today: students get in debt and don't find jobs because too many universities are in the business of signing in students and not finding them a job :( hopefully we can show the way and we'll make this change. Purdue university is now offering this too, which is a good sign that things are changing. Honestly I think that this is going to become the norm simply because the current system is broken an unsustainable (US students debt is in Trillion USD! and cost of atudies is getting higher and higher -> more devt and more ppl who can't afford Education anymore) 10 years from now ppl will simply look back and laugh at today's system.

There are two other important things that this model solves too: 0. There is no more financial barrier to high quality Education. Anyone can afford Holberton* 1. Colleges need to make sure that you are a fit to their system/culture/field. Today this is not the case so students are wasting time and money in programs they are likely to fail. By having this tuition model they will need to be transparent and make sure everyone succeed

(*) we are in SF so cost of living is high. this is a pb for a lot of candidates. We are activally working on a solution.

> too many universities are in the business of signing in students and not finding them a job

If you are a university in the "business of finding students a job" you are university'ing wrong.

That is not exactly what I said :)

With the name treehau5 I expect there's some affiliation to Treehouse here

> If you are a university in the "business of finding students a job" you are university'ing wrong.

Then everyone is university'ing wrong!

Universities spend an enormous amount of time and money getting students jobs. Maintaining relationships with recruiters, developing alumni networks, and entire departments of full time staff dedicated to "career services".

There's a solid critique of education programs that are overly sensitive to industry fads [1], but IMO universities and educators should definitely include "finding students a job" as an explicit curricular goal -- balanced, of course, with "ensuring students have a long and healthy career beyond that first job".

[1] https://www.joelonsoftware.com/2005/12/29/the-perils-of-java...

Well, the problem is that universities are being co-opted into job training programs, when they shouldn't be. The cost of training people for jobs is being passed on to them through the university, from corporations. It shouldn't be that way. Finding a job should never be the main, or even a large part of attending college. Now, naturally that's not what is currently happening, but we need to work to avoid this as a long-term solution to job training.

You want it to be one way but it's the other way.

Right. We're doing things the wrong way, and I'd like to see that changed.

That seems unrealistic for the most part. No, most universities aren't optimized as trade schools focused on near-term employment prospects. And, you can certainly choose a university and major that, shall we say, sub-optimizes employment prospects.

However, it's certainly the expectation of most people that going to university for four years is going to improve their employment prospects--often in a field directly related to their undergraduate study. And as someone else noted, schools absolutely put effort into relationships and programs that improve their graduates' job prospects.

> And as someone else noted, schools absolutely put effort into relationships and programs that improve their graduates' job prospects.

They also put in efforts in a bunch of other places as well.

I will admit that, coming from a lower middle class background, university education to a 17 year old me was mostly a way to optimize my career prospects. But in hindsight, I would say it is a lot more than that. The two most important skills I learned were probably critical thinking and team work. There were also many side-benefits, that you get merely from being in close quarters with so many different kinds of people... to learn that diversity is important, that even people from different cultures are awesome etc. These soft skills are very important, and I don't see a way of acquiring them otherwise.

However, I think a way to remedy this would be to be very clear with what a university provides instead of being vague about it. e.g. if you pursue CS in Uof Something, you have x % chances of getting a job here right out of college. I agree Universities have been co-opted into being preparatory schools for the job market, but that is too narrow a purpose for them.

I don't really disagree with any of that. There are (mostly good) reasons why most universities look and act differently than they would were they solely constructed for learning some trade whether white collar or otherwise. Part of this is that the training part is mixed in, for many people, with moving out of the house for the first time, being self-directed to a greater degree, etc.

That said, there's also a general expectation that, barring graduate school, university grads will also go on to earning their own living. Some do so with just generalized university education--which is more or less what you have if you majored in medieval German literature. But it's easier with an engineering degree or something else that ties directly to what a company is specifically hiring for.

That's a good point. While they're definitely not the same thing, I wonder if the goal posts are just changing. So at first, it was: complete high school and you will def get a job! And then it was: get a univ education and then you will def get a job!

Maybe the economy is just changing so much that we simply don't have the capacity to absorb so many workers. Or maybe we need to direct more resources to create employment in those areas (i.e. more funding for arts, history, archaeology etc.). I don't know what the solution is.

> Then everyone is university'ing wrong!

Hey hey! I tend to agree!

Finding students a job should be a secondary byproduct of first giving them a world class education where you prepare and enable the student for society through being able to learn. Universities are to teach students how to learn so they can learn and be more aware and active about what is happening around them in society. Universities gives you exposure to a wide array of subjects for a purpose. It isn't random, and if you are attending a university where it does feel random, then find a better university.

"The goal of university education is to help build a fairer, more just society" - Steven Schwartz. [0]

Indeed this aligns with Plato's view on education -- "Plato regards education as a means to achieve justice, both individual justice and social justice." [1]

[0]: https://www.timeshighereducation.com/comment/columnists/the-...

[1]: http://epublications.marquette.edu/dissertations/AAI9517932/

I agree with you, philosophically.

But when I apply this philosophy to concrete curriculum design questions in CS, I end up caring about placing students at internships and jobs.

First, internships are a form of education, and I find that students who complete internships come back the next Fall as much more mature programmers. I can then leverage that maturity in programming to dig deeper into interesting theory. Because I'm not helping debug for loops, I can students debug proofs or design more complex algorithms. So I consider internship placement a major goal for the first two years of a CS curriculum, even when my goal is to teach pure theory.

Second, I have a hard time justifying the situation where students are debt slaves to banks. How does that achieve individual or social justice?

Third, your work output is an enormous aspect of your contribution to society. Someone who can build a software platform that helps rural poor in Nigeria get access to micro-loans under a fair terms is making a much greater impact on the world than a philosopher with a perfect understanding of what it means for the world to be just, but without the means to act on that knowledge.

The problem is that it is awfully hard to be a "life long learner" if you are unemployed, desperate, or working a shit job just trying to make ends meet.

Imagine all that extra time college graduates could spend learning if they were easily able to provide for themselves because college prepared them for the real world and they were able to get a good job.

Okay nice goals. But we face a problem today will a shortage of low skilled jobs. The few low skilled jobs that remain often still require university training. The rest offer terrible working conditions and pay.

It's nice to have a society that has people who are more aware and active about what is happening around them in society. But what about the average worker that's been unemployed for six months and just want's to be able to afford to raise a family?

We need to meet our basic needs first. Universities used to be for the upper few, but as they are required more and more for the majority of jobs, either the universities need to change or the job requirements need to change.

Effort doesn't alway equal results and we shouldn't be praising them not attaining their stated goals. (or perhaps they are, and their goals don't matter to their customers -- students)

I'm just a single point of data, but none of my friends ever found the career services people to be helpful in any meaningful way. In my world, their success rate is 0%.

> and we shouldn't be praising them not attaining their stated goals

My post wasn't intended as praise for career services... it was intended as praise for departments and universities that prepare their students for internships and jobs.

> and their goals don't matter to their customers -- students

I don't know this, but I think a lot of the dedicated professionalized career services people are mostly there more to placate clueless parents. It's marketing, and it works.

> none of my friends ever found the career services people to be helpful in any meaningful way

Absolutely. The scare quotes in my original post are there fore a reason...

That said, good career services people are usually work closely with or are even embedded in the department -- not a separate service in the admin building. They ensure that there are regular talks from the alumni network or local community. That companies are invited to come in for an evening and talk about what they're working on / what skills they need. And then ensure those companies are invited back to job fairs etc. in the Fall, with companies who take it as a serious opportunity and who you think are most likely to fast track your students' resumes.

In fact, at smaller universities/departments, the best career services employees are typically professors who keep in contact with graduated students are former colleagues.

University is sold to kids as the thing they'll do after high school and before starting their careers if they want to be successful. There's a grain of truth to that too; I have heard that it's hard to get a tolerable job out of college these days but even harder without a degree.

If that's how universities are positioning themselves, if they're opening their doors to students who are clearly not the best of the best and destined for academia and/or highly technical careers that demand high-level theoretical knowledge, then you bet your ass they should be teaching practical skills and helping students find jobs. Otherwise they're just taking (a lot of) money from kids who clearly don't know any better, and leaving them high and dry after graduation.

Unrelated question, but curious: why do your BART ads that say "I am a software engineer" have a fork bomb in faint text in the bottom-left corner?

But 17% of an entry-level job is HUUUUUGE for the student. If I paid that in my internships and entry-level jobs, I'd be in trouble.

Granted, over the long run, it's a great deal! Just short-term, it's extremely expensive. Personally, I'd opt for a smaller payment that lasts longer, even if it means more money for you.

Not sure if this is it but some guy did an AMA on reddit with a similar payment plan but it was really scummy. They took some percentage of your salary for a given number of years when you eventually landed a job. But there was no stipulation that the job needed to be in tech or anything. So the whole thing might be a total waste of time and you would still have to fork away a cut of your salary when you find a job as a kitchen hand or whatever.

Tell me how this is different from the rest of the student loan industry. My student loan is until I've paid it all back, or I reach 65, or I die. It adds a top up to my income tax, taking 9% of everything over £17kpa earnt.

It doesn't care that I've got a really good job, or that it's relevant to the courses I took. Or that I even finished my degree. It's just the same in that respect.

Holberton takes a higher percentage, but earlier on and only for three years. When you're earning less. You might thing 17% is shitty but coming from low or no income, you're probably not going to notice it. Three years will fly by and by the fourth, you get a massive pay jump.

I prefer Holberton's take to my up-to-48y lien.

> Tell me how this is different from the rest of the student loan industry. My student loan is until I've paid it all back, or I reach 65, or I die. It adds a top up to my income tax, taking 9% of everything over £17kpa earnt.

Student loan industry in the UK. In the US and most other places, income-based repayment is not a thing, the payments are due no matter how much or little you earn.

I don't think that's accurate. When I graduated I got a zillion papers in the mail for applying for various hardship deferments and payment options. And then there's this https://obamawhitehouse.archives.gov/blog/2012/06/07/income-... which is explicitly income-based.

Income-based repayments are definitely 'a thing' in the US.

At least for now the US does have several income based repayment plans. The best one for most circumstances is PAYE (pay as you earn). It requires payments of 10% of disposable income, i.e. dollars earned after 150% of the poverty line, and the debt is discharged after 20 years of payments or 10 years while working for certain government or non-profit entities.

This program is partly the result of regulatory rule making and so could be sharply curtailed should the President or his underlings wish to do so.

This occurs mostly in private student loans vs. federal ones. Private offers even fewer options than the federally mandated ones provide.

> But there was no stipulation that the job needed to be in tech or anything. So the whole thing might be a total waste of time and you would still have to fork away a cut of your salary when you find a job as a kitchen hand or whatever.

But isn't it objectively less "scummy" than charging money up front, which most universities do? Then if you fail to find a good job after graduation, you are still responsible for the full amount regardless of your income.

Some minimum threshold before the percentage is taken off would make sense - the "tax" would be insignificant income to the university but a high cost to someone working for minimum wage as a kitchen hand.

But restricting it to tech is not that obvious. Every office job in the modern world benefits from a) having some sort of degree paper, and b) having some tech skills.

> But isn't it objectively less "scummy" than charging money up front, which most universities do?

It entirely depends on the price and what you're getting for that price. I generally agree with you that the model is 1) interesting and 2) not at all scummy.

But the cost-benefit on these coding schools always confuses me.

The total cost of one of this certificate is probably something in the 30k-60k+ range assuming you're successful (location is SF -- and marketing print specifically mentions Google et al). That's just tuition, not cost of living. And that's for two 9 month project-based sessions.

For reference, the average student loan debt is 37k for four years and an accredited diploma. Which, okay, might not be the full cost. But to be fair, you can DEFINITELY pay for college in 100% debt + weekend jobs/internships, and come out in the 20k-25k range. And that's total cost, with no scholarships, assuming you go to a second-tier state school or can live at home (ok, that skews total cost, but it's a very real option for many, many students!).

IMO universities are the clear winner in this cost-benefit calculation. Hands down, no questions asked. You can double or triple major and have one or two very tenable backup careers if a tech bubble bursts. In addition to that, if you want, you'll have everything you would learn at a coding school. Plus some extra theory/systems knowledge that opens up the non-web-dev job market. Graduate school is an option if you want to go the research route. And, very likely, you'll have a much larger network of close-knit peers that includes not only coders but also managers, engineers, business types, doctors, lawyers, etc. Immigration becomes easier. You'll have some physics/chemistry/calculus knowledge as well as some econ knowledge if you choose the right courses. And so on.

The problem is that a lot of people don't take full advantage of university while they're there, or choose universities that don't align with their goals. I think coding schools are weird in this respect -- IMO they tend to be pretty expensive for what you're getting, but maybe the target clientele wouldn't thrive in a university environment or don't want to take 4 years to get a holistic education.

So again, not at all scummy! But also IMO definitely not a better deal than the traditional route.

In terms of scumminess I think it is relevant that the student loans (mostly) come from the federal government. Neither the for-profit or nominally non-profit universities bear any risk of default. They make sure to get all their cash up front. This means that their incentives are as well aligned as they might be.

One interesting proposal I've seen is to create a 'skin in the game' rule akin to the one that was supposed to go into place for mortgage originators under Dodd-Frank (but mostly didn't). That is require educational institutions themselves to finance and hold part of the student loans used to pay for their programs.

Agreed, 100%!

> That is require educational institutions themselves to finance and hold part of the student loans used to pay for their programs.

Wouldn't this requirement make "17%" institutions illegal? I think the 17% approach is reasonable. IMO coding schools should be allowed to offer bona fide diplomas. Maybe not Bachelor's Degrees, but they should be able to get accredited and be eligible for GI bill etc.

The last thing higher ed needs is even larger moats.

So, skin in the game -- absolutely! But not at the expense of making it even harder for new players to enter the market.

I would think that the skin in the game requirement would be tied into eligibility for federal student loans. Since a 17% institution wouldn't be using those it wouldn't impact them. I agree that model is even better at aligning incentives and shouldn't be forbidden.

What would a bona fide diploma be? A diploma is just a certificate that you have completed a course. It's not legally protected in any way is it?

It's not the certification of the student that matters; it's the accreditation of the company providing the education! You can't use your GI bill benefits or subsidized student loans at most coding schools. That's a shame in cases where the coding school is high quality!

If the coding school is an accredited degree-granting institution, then federal/state assistance can be used to pay for the education.

Your $37k debt figure sounds low...really low. I graduated 20 years ago with a (then average) debt load of ~$20k. Tuition rates and debt loads have skyrocketed since then. $37k could represent average debt load including those who have been paying for years already but I'd love to see your source for that figure.

Here's where I got the number; no citation but they do stipulate 2016 at time of graduation: https://studentloanhero.com/student-loan-debt-statistics/

Here's a breakdown by state for 2015 from a more legit looking source and a report I don't have time to read; they actually put it at 30k: http://ticas.org/posd/map-state-data

> Tuition rates and debt loads have skyrocketed since then

Total debt load across the entire population has skyrocketed in part because there are a lot more people attending college and during the recession a lot of people weren't paying off their loans, so that number could be misleading for this reason. Could be that debt load at graduation per pupil has not increased, but total debt load has skyrocketed.

Tuition rates have certainly increased, but it's really hard to find data on what people are actually paying outside of borrowing rates. I graduated high school with high ACTs but an extremely low GPA and still had >66% "scholarship"... Tuition != amount paid. If you have "actual amount paid" data I'd love to see it.

FYI we (Holberton) have a minimum threshold: if you don't find a job or if it under 40k you don't pay us back because we failed at either training you or selecting you as a student.

Interestingly enough, that's around the same number I came up with as a minimum wage for students with big loans to pay back a University. People would tell me that $15 an hour or so for IT work like we get out here was decent pay for new people. I said a lot of these people have Bachelor's degrees with minimum payments of $2,000 a month. Lowest cost-of-living out here (poor people life) + that payment = you need at least $40,000 a year just to scrape by. People just don't get how the traditional model of education plus companies not paying people anything puts such a financial burden on graduates that they're damn-near better off taking a no-degree approach.

Did you go to a private school? Federal loans (~90% of all student loans are Federal) have income based repayment plans.

Under the income based repayment plan, a single person making $15/hour 40 hours/week would only need to pay about $100 per month.

Yep. A choice I regret now.

> ...or selecting you as a student

IMO this is primarily where many universities are failing, which is an enormous disservice to their students.

Any reasonably intelligent and dedicated student can go to university and come out the other end with a well-paying job.

> Any reasonably intelligent and dedicated student can go to university and come out the other end with a well-paying job.

That's a pretty restrictive filter. Currently you have to be at least two of smart, goal-directed, and curious to make it through the university system with a decent education and good job prosepcts. Many other students make it through with good job prospects but no real education because they were able to bull their way through an in-demand major program. But a lot of them slip through the cracks and end up with neither job nor education.

> That's a pretty restrictive filter

I agree. Ideally, tuition would be extremely subsidized and higher ed wouldn't constitute an enormous financial gamble. But in the current US system, the restrictive filter really is in the students' best interest.

That's a rather self-serving criteria, isn't it? Oh, you failed to get a job? Must be because you weren't reasonably intelligent enough!

It's not self-serving to tell someone "really, no, you shouldn't take out a big loan to pay for this because you're probably unprepared and will fail".

On the contrary, taking literally every student who can get a federal student loan -- regardless of your faith in their ability to get something meaningful out of the degree -- is extremely selfish!

90%! of my childhood friends succeeded at getting a degree but failed at that "well paying job" part.

Graduating is the easy part.

Tbf 17% of a kitchen hand wage vs 17% of a software wage is a huge difference. The school is incentivised to find their students good work and therefore train them all in order to build their own reputation. That doesn't sound scummy at all.

I'm still not sure that's bad. With this plan, the school would much rather you land a job getting $50/hour instead of $7.50 per hour because they make more money. I can think of a few ways that could make this better (such as a sliding scale that goes lower if you are making less), but overall the incentives seem to be aligned with them landing you a good job, and getting you a good pay scale early on (since they only get paid from your first 3 years).

Sounds like what happens in New Zealand in regards to student loans. Basically, 8% of my pay goes to paying off my student loan (but only on earnings above ~$20,000). I can't pay less than this, it's taken from my payroll like tax.

Not really scummy at all in my opinion.

The US has basically this exact same program. For a single person, it's currently 10% of your income over $18k.

I'm 1.5 years out of a coding bootcamp I paid about $12,000 for.

At a 17% rate, I would have paid $11,500 already, and would be on track to pay $22,950 more if I stay at my current job for 1.5 more years with no salary increases (I expect salary increases).

There were some students in my cohort who reverted back to their previous careers after bootcamp, but most became developers. I would bet the Holberton plan is far, far, far more lucrative than an upfront fee, as long as they are discerning in who they accept.

Hey pharrlax, current holberton student here. So happy to see that a bootcamp worked for you! I have a few friends who are going through or just completed galvanize's gschool and hackreactor. About your comment, you totally right! 17% for 3 years seems like a lot! However the program is for 2 years. I view it as a college replacement rather than a bootcamp. Therefore I feel like what i will be paying is 150% worth it!

> I'm 1.5 years out of a coding bootcamp I paid about $12,000 for. At a 17% rate, I would have paid $11,500 already, and would be on track to pay $22,950 more if I stay at my current job for 1.5 more years

But you needed that $12,000 upfront, which means you spent time in crappy jobs to save up for it. Time that could have been spent in your new job making more money, so you're not accounting for that wage difference, ie. if it took you 6 months to save that $12,000, how much faster could you have made it at the new job had you started code camp earlier?

> But you needed that $12,000 upfront, which means you spent time in crappy jobs to save up for it.

That's what loans are for. The question is, do you want to take the risk or do you want to offload that risk?

Maybe the factor is more than just risk. By giving someone a vested interest in your success, they might also be there with other support after the class is done, such as job hunting.

I know it's not a fair comparison anymore since I realize that Holberton is 2 years and not a "bootcamp" but most bootcamps offer a certain period of time for you to repay the tuition fee.

For example, at the bootcamp I did, I think there was a 1000 fee due before 1st day of class, a couple more during the bootcamp, then 5k due 6 months after the last day of class.

"At a 17% rate, I would have paid $11,500 already, and would be on track to pay $22,950 more if I stay at my current job for 1.5 more years with no salary increases (I expect salary increases)."

If the bootcamp is going to shoulder that much risk, expect them to get paid for it. If you can afford to just pay for it up front, expect to pay less net money, but then, you're shouldering more risk that it won't be worthwhile. If you consider the distribution of both money and risk it looks more balanced. Whether it's precisely balanced or something doesn't have a unique answer because people have different valuations on the risk vs. the money. In particular, someone with no money to speak of at all is obviously going to prefer the single option that is available to them for the chance to come out wildly better off than they started, even at a 17% salary cut for 3 years.

(Also, I note you were not making wild accusations or anything, so I'm not being accusatory either, just elaborating.)

That was my exact thought as well. I paid $8000 for my bootcamp. I'm 18 months into an average of 65k over that time. So in that 18 months I'd already have paid in the area of $16k, and assuming I don't get a raise double that over the 3 years.

Also, I'm curious the arrangement for contract work that the developer does on his own time. If they charge a percentage of that, I'd have paid them more than a downpayment on a house over the course of 3 years.

My first reaction was that 17%x3 seems like it's on the high side. But then I read Sylvain's (who I've met a few times over the years back when he was at LNKD) profile and saw that Holberton is a 2 year program, which is something like 8x as long as any other "coding school" (I refrain from using "bootcamp" since a 2 year school is clearly not a bootcamp) that I can think of.

That puts an even more interesting twists on how their financials work out.

That works out to $54,000 for two years of schooling. Community colleges offer the same for under $10,000 total.

Would have been $28k for me, working mostly in the Midwest.

For someone who graduated this without an internship, then worked 3 years full time at minimum wage (I used $7/hr for this), it would be around $7k.

Still seems high, but variably so.

There should hopefully be a significant difference in scope and quality between the material offered at a coding school and a community college's CS program.

A CC programming degree was enough to get me a professional job at 19. It was taught by the same professors that taught CS at the 4 year affiliate, they just focused on application over theory.

I can't imagine boot camps are any better. We built applications using technology stacks that were popular at the time (.NET/MSSQL, LAMP) and our capstone project involved writing actual production code for a company. Everyone who graduated the program got a job or went on to a four-year university.

> they just focused on application over theory.

That would be exactly the criterion I had in mind. Colleges and universities should teach computer science, not coding. Dijkstra was an accomplished computer scientist and never owned a computer; there is a whole academic discipline there which should be taught in its own right. Whereas a coding school should probably lead off with source control and the command line, the university should start with lambda calculus.

It's not that these things can't be just as useful, it's that they are distinct topics and neither should be used as a replacement for the other. (The unfortunate corollary to this is that it takes 6-10 years to become competent as a programmer, and a number of those years must be in the industry. We should really get a guild structure together.)

You should be comparing them to top universities, not community college.

Top universities can cost twice as much per year.

Why would I compare a two-year program nobody's ever heard of to a four-year program at a top university? Community college is a much more accurate match, as another two-year program nobody's ever heard of.

If Holberton wants to publish standardized test scores for their students, we'd have better information to go on.

The reason is because the starting salaries of people coming out of the boot camps is higher than the average starting salary of people coming out of top colleges.

Tech Bootcamps and tech college replacements have a starting salary around 80-100K whereas top schools for an average graduate(including ALL graduates, not just CS majors) is less than 80K.

Test scores are dumb. They don't matter. What matters is OUTCOME. And the best way to measure outcome is via salary.

> Tech Bootcamps and tech college replacements have a starting salary around 80-100K

Not according to the comments in this very thread from satisfied bootcamp customers.

But even if it were true, the appropriate comparison would indeed be starting salary of top school graduates who took a job similar to what the bootcamps are targeting, not starting salary of all graduates of top schools.

MissionU does something very similar (charge nothing up front and take X% of your salary for X years after you graduate) http://www.missionu.com/ It is really a great model, the "school" doesn't get paid if you don't get a job after graduating.

Higher Education in it is current form has limited time left. Universities will have to evolve or die. Student loan debt is completely out of control and University's aren't preparing kids for real jobs.

in some ways this is similar to the australian university funding system where loans to cover education are managed by government - not the private sector - there is no interest beyond inflation, and repayments are taken automatically out of salary after your income exceeds a certain threshold.

i would imagine that Holberton School is much more focused on only accepting candidates for training where there is a good probability that the person will end up in the workforce afterwards. unlike perhaps some things one can study at university, or people who are studying but aren't thinking about a career in the workforce.

They're also incentivized to provide a high quality program and have their grads land high paying jobs. Essentially they're invested in their students success in a way most educational programs aren't.

The US loan system works almost the exact same way. Federal loans (90% of student loans) are paid directly by the government, and come with income based repayment plans capped at 10% of disposable income.

For reference, I plugged in my own early career (I graduated fall 2006).

I went to school in the Midwest, worked 2 full years as an intern, then two years as a full-salary employee, then moved to Maryland. So 17% of all my earnings in the Midwest + my first year in Maryland comes out to about $28k.

If I cut the internship down from 2 years to 1 summer, it's around $24k.

That's roughly the same as my 4-year degrees at the state university I attended from 2002-2006 (double major: BS in Computer Information Science, BA in German).

edit: As cableshaft mentioned, I did not account for the cost of interest -- my ~$28k was just for tuition + books/fees.

> That's roughly the same as my 4-year degrees at the state university

Tuition or total cost of attendance? Notice that the 17% for 3 years is equivalent to tuition, not full cost.

> double major: BS in Computer Information Science, BA in German

The way I think about this is that you got a free BA in German, some free theory/systems CS courses, and some free gen eds out of the deal.

Tuition + books/fees. No student loan interest, housing, meals, etc, accounted for.

And that's probably a good way to think about it. I also came out with fewer math courses -- the specific CIS track I took had a lot of electives, as did the German major, so I just married them together. A pure CIS track would have included a math minor instead.

edit: Also, assuming the 2 years for the course in the article is roughly equal to the computer science education I got, I could see being willing to pay the same amount to get out 2 years earlier and start earning graduate salaries more quickly. I certainly don't use the gen eds or German knowledge in any money-making way that I have noticed.

Assuming you took out student loans, are you factoring in the interest accumulated into your total costs? There's no interest in the "17% over 3 years" model.

I had about $25k of student loan debt when I graduated, but by the time I've paid it off I think I will have paid closer to $35k-40k, factoring in interest over several years.

> There's no interest in the "17% over 3 years" model.

There isn't but it's insignificant and the potential salary increases would overshadow charging interest anyway (at least in the current low interest environment in the US)

This is an excellent point and IMO a major benefit of the 17% model.

Entirely too much money spent on higher ed in the US goes to banks.

Federal loans (90% of all student loan disbursements) are paid from the treasury.

There's an argument to be made that the government shouldn't be charging students interest. But there's no banks involved.

Yeah, good point. I didn't think to factor in interest. Noted in the OP now.

If you don't find a good job and end up working in fast food, do they still take 17% of your salary?

In Australia I think we have a really good system, although politicians are trying to destroy it now.

You get an interest-free loan for your education, the only thing it accrues is inflation. The loan is repaid as a percentage of your taxable income if and only if that taxable income is greater than 54k or so. It means that you will potentially never be asked to repay your student loans if you never find professional work.

That seems pretty sensible to me. When the educational funding bubble pops in the U.S., I could imagine something like that taking it's place.

We have nearly the exact same system in the US.

Federal loans (90% of student loans) are paid directly by the government, and come with income based repayment plans capped at 10% of disposable income.

The debt is cancelled after 20 years, so if you don't find a decent job, you'll pay very little.

There is interest though.

54K seems like a really good deal (at least, an incentive for anyone to take on a university course). In the UK, the threshold is much, much lower: https://www.gov.uk/repaying-your-student-loan/what-you-pay

Probably speaking in australian dollars.

Which is still ~32k GBP, so almost double the UK threshold.

The UK threshold is still sensible. 9% for everything above £17k and forgiving all debt after 30 years means that paying back your debt won't significantly affect your lifestyle and that lower earners will never have to pay everything.

Minor correction: there is no real interest on HECS debt but it is indexed by inflation every year.

> This is the reverse of most for-profit education, which relies on loans and doesn't guarantee a job.

I've wondered for _years_ why the "equity" (vs debt) model of funding isn't available to private citizens, at least in narrow circumstances like wanting to mitigate the risk of massive amounts of student loans and un- or underemployment. It's nice to see people actually putting that into practice. Presumably if it gets too widespread, some hack will write an article about how it's exploitative and we'll start seeing pressure for a crackdown.

Are they selective? Kinda important to know whether they are able to pick and choose who they think spending their time on.

Yes we are: the admission process is algorithm-led. The admission process is totally blind and removes human biais.

During the admission process (please tey it :)) we try to replicate as much as we can what students will be living at Holberton. Simce we are project-based and students also collaborate a lot with each other, the admission process is a project where you have to collaborate with other candidates. Basically if you has fun doing the admission process you will love Holberton and if you don't then you will drop-out during the admission process vs dropping out from school. Most of the applicants don't finish it. It takes anywhere from 10 to 60h to finish it - depending on your current level (this is designed for ppl with 0 level in CS/programming)

Probably no more selective than a top university.

No. Taking financial risks is sharing downside also - that is paying their students if they do not find a job in the industry. They only share the upside - that is called risk transfer, not risk taking.

Risk transfer is charging tuition. They are sharing the risk - not finding employment means the school doesn't get paid for the education it provided. The only way this isn't taking on some risk is if the marginal cost of the additional student is zero (or de minimis).

There is/was a popular scam in Eastern Europe - some middlemen promises to help you with bureaucratic hardships for a bribe and returns your bribe money if he could not help. The thing is the scammer does exactly nothing - just waits if his victim solves the problem himself, and if not - just returns the money saying "sorry, I could not help with the situation".

But in this case they are doing something, right? There's an education. And money isn't changing hands prior to employment.

I know nothing about the particulars of this school, I'm just speaking to the concept, but it is entirely distinguishable from what you describe.

The job of the scammer is to make his victim think he does some very important things, which always include spending on some bells and whistles.

It would be neat to also have an option to pay up front and then get a refund if you can't find a job. Would that be cheaper for people who have the money?

Reply on news.ycombinator.com