It says when trying to figure out what to study, think about what might give you most options for the future, a future that you might have almost NO way of predicting now. He calls this "staying upwind." He gives an example:

Suppose you're a college freshman deciding whether to major in math or economics. Well, math will give you more options: you can go into almost any field from math. If you major in math it will be easy to get into grad school in economics, but if you major in economics it will be hard to get into grad school in math.

And given a choice between studying two subjects that interest you, but one is easier for you and one is a bit harder (but still within your abilities), opt for the harder one:

The best protection is always to be working on hard problems. Writing novels is hard. Reading novels isn't. Hard means worry: if you're not worrying that something you're making will come out badly, or that you won't be able to understand something you're studying, then it isn't hard enough. There has to be suspense.

You don't have to automatically agree with everything in his essays as Great Truths, but it's worth considering his ideas.

https://www.joelonsoftware.com/2005/01/02/advice-for-compute...

The moral of the story is that computer science is not the same as software development. If you’re really really lucky, your school might have a decent software development curriculum, although, they might not, because elite schools think that teaching practical skills is better left to the technical-vocational institutes and the prison rehabilitation programs. You can learn mere programming anywhere. We are Yale University, and we Mold Future World Leaders. You think your $160,000 tuition entititles you to learn about while loops? What do you think this is, some fly-by-night Java seminar at the Airport Marriott? Pshaw.

The trouble is, we don’t really have professional schools in software development, so if you want to be a programmer, you probably majored in Computer Science. Which is a fine subject to major in, but it’s a different subject than software development.

I don't know. Some people will do better if they follow the established path. Others blaze their own trail and don't do well in rigid environments. I know I did not.

Ten years later, the best part of my studies are the ones that felt like a distraction and a waste of time back then: mandatory engineering and liberal arts classes. I would have learned programming on my own either way, but these other things added so much depth to my life!

I think that there are so many ways to live your twenties and grow as a human and as a professional. I strongly doubt that saddling yourself with debt is the best way to do it. Given four years of your full attention, you can achieve so much more, provided that you have the curiosity and discipline to try. I just wonder where the strength to explore the world beyond your main interest would come from.

It's a fairly recent phenomenon. (I'm old, so recent may mean different things depending on you frame of reference.)

Not amazed. I presume it used to be easier to get a job back in the day.

I am in a similar position as OP's son right now, and I am trying to get into a university to do CS because I believe the only way to enter the field right now is through internships.

I believe this because I don't even see any alternatives. Like, if you are not going to go to university, what are you going to do instead? Contribute to some open source project and hope for the best?

https://www.linuxfoundation.org/about/mentorship-programs/

It's nice to have, but ROI on that is terrible. Most of the time people don't even read your resume.

> https://www.linuxfoundation.org/about/mentorship-programs/

Huh, didn't know this existed. Looks promising.

Yes, my open source contributions got me a job the year before last as the second highest contributor to a popular in its niche official “AWS Solution”. But no one cared about my other open source projects that I highlighted that were also well documented, solved a unique need and were on the official “AWS Samples” GitHub repository.

I’m sure it’s far from the most common modality, but this is obviously false. I’ve personally referred people to apply at my company based on their PRs to F/OSS stuff, and at least a couple off the top of my head have been hired.

And “a couple” is statistically so close to zero, it might as well be zero.

> And “a couple” is statistically so close to zero, it might as well be zero.

No, it’s three that got hired (me and two others I hired), with a self-selected sample of one. Certainly not statistically significant, but it’s also far from nothing.

I wouldn’t recommend anyone grind leetcode. There are plenty of jobs that done’t require it, and IMO those are the ones you likely want anyhow.

Who wants $160K+ straight out of school anyway?

Shouldn’t that tell me something about the WLB at the company?

It’s not necessary to have one, nor to maintain it. But for a recent graduate, it’s one way to demonstrate competence both technically and socially.

I got my start before university by building websites for people, then eventually web apps. I had a successful little business by my first semester.

I git my start making websites for small businesses. Then I got more complex projects, and eventually contracts as a web developer. Then I started university and had to pause all that.

My internship in Germany was a typical one. I found it through a job board. At the end, I dropped out of school and stayed, albeit with a proper salary.

I think that the small business world is massively underserved. You have the skills to solve a lot of real problems, and to learn good practices the hard way, as well as a host of soft skills. It's a lot of fun because the impact is immediately visible, and you can't bullshit your way through like in a large corporation.

So I think the problem is that you are looking for a well-marked path. There are so many people who have unmet needs, and they don't advertise it as an internship on a job board.

Now this is where I will sound like a boomer. Go around and ask. Literally just introduce yourself and ask if there's anything you can help people with. Maybe you can throw a quick form together or write a script or two. You might have to fix someone's printer every once in a while, out there in the real world, not sheltered by an org chart.

That's what I used to do, and it was a fantastic learning experience. I remember those days fondly.

I agree with you 100%. One thing I am unsure about though, and which I believe is very important, is, how much should I ask for? I had an experience where somebody reached out to me through relatives to make a website, so I quoted them $1000, but they just never responded. I believe two primary ways of pricing are per hour and per project, but there are issues with both. If you choose to price per hour, the client doesn't know in advance exactly how much your services will cost, and I believe the client has no way to verify that you actually spent those hours working on their project? And the number of hours you spend largely depends on how you choose to approach the problem, so I would expect people to feel reluctant towards this kind of model. On the other hand, if you price per project, you might end up doing way more work than you expected at first, and isn't it also very common for client needs to change all the time? How did you approach this? Thanks.

My policy was to never compete on price. The cheapest clients are also the most demanding. I priced per project because the scope was small and it let me charge much more per hour. It was fun earning 50-100 CAD per hour right out of high school. It's not greedy; not every hour is billable.

That being said, you're also learning, so there's a discount there. If you work for free or cheap, you should do it for people who deserve it. My first clients were non-profits and companies that were just getting started.

A much smarter way to work is to approach people who need websites or dev work for their clients. After pairing up with a graphic designer, work just kept coming. You only need to be trusted once, not for every project.

If it's not immediately obvious, you will learn a lot about salesmanship and customer service. These are valuable lessons.

You learn an awful lot selling your skills as a freelancer.

How long ago did this happen?

Most people have an addiction to food and shelter and can’t wander through life aimlessly finding themselves trying to be a better citizen of the world.

Learning programming and being employable directly out of school is entirely different.

So that's another option to consider: do the CS degree part-time while working on breaking into the industry. You can adjust your workload each semester depending on circumstances, so it's flexible for someone job hunting.

[1] https://www.coursera.org/degrees/bachelor-of-science-compute...

I have considered enrolling in a university precisely for the same reason, but I never would have thought that something like that would be considered "valid".

I guess I'm full of questions for you, but in short:

- Where are you coming from and what is your target destination?

- Have you made 101% sure that this way of studying would be acceptable to the employers in your prospective destination country?

- Does one need $20-25K up front, or is there a payment plan?

Thanks if you decide to answer, and I wish you a good luck.

Happy to answer. Feel free to get in touch on one of the socials in my profile to discuss further.

Q1. I'm from Australia, and the target destination is USA. Although I have no firm plans, I want to have the option in future. Australian's have access to a particular visa called E-3, which I have failed to obtain in the past due to lack of degree.

Q2. Yes. Upon completing this, I will have a BSc in Computer Science from the University of London. It's a bachelor's degree as good as anywhere. The exams are proctored, and all the exams/assignments are graded via Goldsmiths University. It's 100% legit.

Q3. You pay per module, so you don't need to pay up front. You just pick how many modules you want to do that semester and pay for them. The only requirement is that you must complete enough modules per semester to finish within 6 years. Since the education is directly related to my work, it is also tax-deductible in my country.

I realize that you are doing this to check off the degree box on your visa application, but if your degree will be from Goldsmiths (that is, "Goldsmiths, University of London" on a résumé, as opposed to the generic "University of London", that would be especially helpful.

> "... having registered with the University of London and passed the approved examinations conducted by Goldsmiths College, has this day been admitted by the University of London to the Degree of BACHELOR OF SCIENCE".

So I don't think it would be unreasonable to mention Goldsmith's College alongside UoL as your education history.

> The exams are proctored, and all the exams/assignments are graded via Goldsmiths University. It's 100% legit.

- Are all exams done remotely?

- Is it possible to skip classes and go straight to the exams in case I feel confident I can pass?

The "classes" are just videos on Coursera - totally optional. You have mandatory midterm assignments, and either a final assignment or an exam, depending on the class.

Looking at Inspera proctoring requirements, it only mentions windows/mac. Is it possible to use it inside a VM on Linux?

So I think the real question is less "should you get a CS degree" (I would have said no in 2015 too), and more "should you plan for a career in software development".

I did end up with a CIS degree (computers in the business college), rather than a CS degree (engineering college). I’m not sure how the required coursework might influence the development of various soft skill. It was never something explicitly taught, just things I was forced to develop in order to get through to graduation.

Yes, it's possible to make it without a degree, but it makes things a lot more difficult. Don't second guess it. Do it!

Also, no reason to dread the intro classes IMO. Given his experience, it shouldn't be hard for him to ace them and race on ahead to bigger and better things. I learned some interesting things in intro CS, despite also coming in with prior programming experience.

Feel free to send me a message if you have any questions.

On the practical end, your degree is still a big filter for how companies screen applicants. A lack of a CS degree makes it way more difficult to get in the door. I know there are counterexamples, but 99% of my coworkers had CS degrees. Unless the money is a huge burden, or the school you would go to really is crap, then yeah, the degree makes the rest of the journey much easier.

As advice, try to get evaluated out of the intro courses and skip to higher level stuff. Often if you can show a prof that you can program (before the school year begins) you can place right into a course that is at your level.

The AI arguments are also facile. A CS degree is still the best training to be ready to use AI effectively. You understand more about how it is built than everyone else, plus you know way more about every tool surrounding the AI, and would have an easier time making your own.

I know some people have success with alternative paths. But the main road is the main road for a reason here.

Edit: Please also look at the best schools for CS, and don't just discard them as options. My coworkers out of these schools got even more out of their classes and costudents than I did. Your son sounds talented, and the best schools can help talent go even further.

Not to discourage choosing math over CS, of course.

You just have to do it for the right reason: Because math is friggin' cool, and programming is easier to learn on your own than math, and there is a skill transfer that goes from math to programming that doesn't go the other way.

I don't really mind it either, if I want to do something novel I do things on my own.

The cost has been that I don’t have deep domain knowledge that could lead to making a startup myself.

And it’s the kind of stuff with few or no examples of anywhere, so no LLM help. For example, building cryptographic primitives for a custom zero-knowledge VM in the VM’s own esoteric assembler language. Nothing closely resembles it.

I’ve found myself drawn to areas of CS such as graph machine learning and programming language theory where I wish I had stronger math chops beyond calculus, linear algebra, and introductory discrete math.

I don't know if AI will reduce demands for SWEs, but it seems pretty risky not to get a degree IMHO.

Even more so as demand reduces.

If he's actually interested in CS, then a CS degree (or minor) makes sense. If he's interested in programming, study just about anything else and minor in CS or double major. Being an AE or MechE that has a much better than average (for the discipline) grasp on programming is a better edge career wise.

A late addition. Your son should test out of those classes given your description of his skills. Alternatively, he should talk to his advisor and seek to skip lower level courses. He'll still need the same number of credits, but I did this to get out of three lower level math and CS courses that I already had down. Some advisors are more amenable to this than others, though.

The competition is fiercer now, but good devs will still get hired. Average devs have a lot more trouble these days.

Does your son need to go to school? Maybe he's hireable as-is.

Finally, has he looked at financial aid? It can take a bite out of that debt.

My personal belief is that not hiring entry level engineers is unsustainable unless we get AGI capable of both outputting and _maintaining_ software better than humans, and if that happens all bets are off anyway.

But can you come up with a good argument why a company should hire an entry level developer out of school - especially out of a tier 2 school in 2025 let alone in 2029? As I said I’m not being elitist, I graduated from a college that no one has heard of outside of the state.

My perspective though is going to come from someone who worked at Amazon (AWS) in 2020-2023 and mentored a couple of interns who got return offers the next year.

Companies aren’t recruiting for the most part from tier 2 schools and it’s hard to get an internship.

I was also looking for a job in late 2023 as someone who was experienced, credentialed, networked, etc.

The market for developers sucks right now. I would have probably had a hard time finding a job if it weren’t for my network and a couple of niches where I’m one of the industry experts.

When I submitted my job to hundreds of jobs as a Plan b, I heard crickets there are a lot of people chasing few jobs and to a first approximation, no one wants entry level developers. It’s getting worse as AI improves.

Anecdotally, there are projects I lead now that I would have needed at least one entry level developer to do the grunt work that I can just do myself using ChatGPT.

I would even go as far as saying if he doesn’t go to one of the top schools that companies recruit from, don’t even bother with a CS degree.

Get a degree in something else and maybe take some CS electives if he can’t get a degree from a top school.

What that something else is? I have no idea.

Life is already difficult for young people right now and 6 figures of college debt is just putting another obstacle in their path.

For what it's worth, I don't know that the open source route is particularly fruitful either. I hear people recommending that as a way to get hired, but I never hear much about it making a difference in the hiring process.

Just my .02, though.

He could always join the Army and have college paid for partially or fully at state school depending on years of service. I did this. Aside from the money it is a great experience for a young man, something most won't have and he can get paid while doing it. It will help for hiring at least for government jobs as well.

I think it is a huge mistake to believe AI will take CS jobs to a significant degree any time soon. I think if it does it will take other jobs as much or more. So not worth optimizing for that way.

Companies no longer want people who loved programming for the joy of it and the art of it.

My thoughts (certainly not a recommendation as every situation is unique) - foundations are always going to be critical. So things like cs (not just software eng), math, physics, biology will be super important because with AI you can have the audacity to tackle 10x harder problems. Now a 150k standardized program may not be the way to go about it. (Back home education was practically free so it was not even a question but in here things are crazy so I get it).

My 2 more cents is you have a 1-2 year window where large pct of population is either wary or cautious or unaware of how to use AI properly. If your son is the curious type id encourage him to build like a maniac (heck clone a few high priced SaaS products and learn how to sell). Worst case he loses 4 years while this AI dust settles. Best case he comes out of it with Independence and confidence to do a lot more than a grad who has only done course work (and even those are established in orgs)

And the FAANGs employ them at their offices in these countries.

Your people could be saving a whole year of their lives not studying algorithms they will never use.

In my 20 years in the industry, I was only asked once in an interview about my degree.

> but also to gain a better understanding of fundamentals

I believe there are many courses online, such as Coursera and others if he wants to learn more. None of them will have the social aspects or networking, but they are far faster and cheaper than the traditional path.

My suggestion is to sit down with your son, and start looking at the job market for junior or graduate jobs and see if it will make sense to invest $130k and 4 years of his life.

If he realises that this is not the way forward, I would suggest he start looking for local hackathons where he can join, learn, network and have fun.

There's some important development that happens at traditional college student age. Some of that can be fostered by attending college. Some of it happens organically no matter where you are.

Your son seems atypical: he knows what he wants to do, he has solid skills, and he is probably already better self-educated than the typical graduating college senior would be in software development.

So at this point what he needs is not necessarily the education, but the magic piece of paper -- the diploma -- that makes him hirable in more places.

A business degree with a minor in a development-oriented course of study might serve him better than spending $160K on what essentially would be a mix of formalizing his learning / reviewing what he already knows.

Because, after all, isn't the point of an education to learn something you don't know already?

So back to that important development: your son needs to become more well rounded. He seems to have excellent aptitude for programming and data processing. How are his social skills? What is the status of his business acumen? Does he aspire to work for himself or start a company someday? To what is he going to apply those skills he's got?

The answer to that is what he should go to college to learn. A generalist degree in something like business might serve him very well. Wasting four years of his life and $160K on stuff he already knows probably won't do much except help him refine his skills in beer and girls. Not necessarily a bad thing, mind you, but maybe not six figures worth of value, either.

I would encourage him to either go to school and round out his learning, or just jump into the working world and find where his skills are deficient and let that inform a future college decision. Just my $.50

A masters degree ime is neccesaray for any of the most interesting fields of cs work from what I've noticed or will help alot.

Also 130k is unessesary. Go to community College, transfer credits to public university to finish the degree.

If the intro classes are to easy it's an opportunity to tutor for money too. This also goes on your resume

Most of the big companies are now tech companies now so we are probably at peak tech. The FAAG companies in the united states are looking at major outsourcing plans over the next 20 years in the united states. Just like manufacturing did back in the day.

That was different when I did it back in the 90s.

Assuming your son is in the United States since you mentioned “state university,” an option for reducing costs is attending a community college to get the lower-division courses in CS, math, physics, and liberal arts finished at a considerably lower cost. This also allows your son to buy some time and assess the state of the market two years from now. Two years of community college tuition is much lower than two years of university tuition, and depending on the state it might even be tuition-free.

If your son is still in high school, I also highly recommend he take Advanced Placement tests and score sufficiently well on them so that way he will get college credit for those courses. This could potentially save time in college, thus reducing costs.

I also think it's worth a real discussion about what your son hopes to gain out of a college education. If the value for him is more vocational, then this strategy above seems almost like a no-brainer. If he's hoping to get that kind of intangible university life experience, community college probably isn't going to deliver.

There’s also the possibility of applying to universities where even the intro level CS courses are quite rigorous and where there’s an expectation, tacitly or explicitly, that CS students have prior experience. Berkeley’s CS 70, for example, is ostensibly an introductory course on discrete math and is required in the EECS program, but it’s much more rigorous than the standard introductory discrete math course taught at most other universities.

My experience doing this I have learned there is only thing that influences career fulfillment: barrier of entry. You can make money anywhere in software as dependent upon the success of your employer regardless of level, skill, or competence.

To be happy doing the work, however, requires operating at the appropriate level of challenge. If you are the smartest guy in the room everyone around will be full of excuses and selfish retardation. If you are in over your head you will be the one inventing improper excuses doing selfish things.

So, always strive for the most exclusive and challenging job you can get.

https://aws.amazon.com/blogs/aws/customize-models-in-amazon-...

https://cloud.google.com/vertex-ai/generative-ai/docs/models...

https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/...

So now where do you think things will be in 2029 by the time her son graduates?

And you still haven’t answered the question - why would this be something that everyone isn’t pursuing, there are already people who are doing it and upskilling in it and what are the chances that he is going to be competitive coming from a tier 2 school as a new grad?

You actually have to know what you are doing in this space.

The question was answered - the field is complicated - deciding to do it and being capable aren't the same thing. The field is not going to be inundated with capable people. By your thinking, one cannot be paid millions of dollars to be a lawyer because of "competition".

Saying that he can make a lot of money by chasing after the same thing everyone else is chasing after and has a years long head start is like saying he should become a football player because of the money he can make in the NFL.

The number of companies that need any more than the abstractions available from the cloud providers is vanishly small and will grow smaller in 4 years. Those companies aren’t hiring new grads from tier 2 schools.

Many lawyers making millions are from second tier law schools.

In computer science, Jeff Dean went to the university of minnesota.

Or you can look at the median earnings of CS graduates who graduated from the top colleges vs 2nd tier colleges after 5 years.

Why look at the median? This kid seems like he's far and away above the median. There is a correlation between best school and best performers, but big tech as an example does their own recruiting and at least in my experience at google, most engineers did not attend one of the very top schools. My team had an engineer who never ever graduated college.

Drop the obsession with prestigious schools and workplaces. Plenty of people are making a very handsome living without the badges from the best places.

There are many reasons why one might want a college education and college experience, but purely from a vocational advantage POV the cost and time seem hard to justify.

I think the best part of my computer engineering program at the University of Waterloo was the integrated co-op system. (Basically, internships.) It meant my degree was 5 years instead of 4, but I got work experience and professional connections at several companies and came out with a slight profit instead of a ton of debt. I even turned my last internship into a full-time job after I graduated.

Then he'll have to treat that as an investment, either into college (and hence he'd hopefully be Very Serious about it), or use it for something else - travelling, entrepreneurship, MF/stock investment, etc. The weight of the decision and the ensuing use of the monies are both the payoff, not a degree per-se.

The job market is bad but lots of hiring for AI positions, both research and application of AI are still in demand. But this could very well be cyclical.

No matter what someone has to train the AI, control the AI, etc.

If he goes to Finland or Australia he can get a world-class education for far less.

Here in Germany I would still recommend getting a BSc in CS if you're interested in the topic.

if he's that bad into it, he'll be doing gigs and projects with students and cracks all over the world.

it's not about getting a degree in CS, it's about how bad you want to feed your curiosity. the field is still in it's teen years. there's whole human-like brains to build, hardware and infra and artificial minds that run on it.

employers? your son sounds like he'll be one in no time.

There is a strong argument that understanding the relevant math (which isn't rocket science) behind AI gives one a leg-up. Most people just can't be bothered to do it and it's really important and actually practical in the modern world. Linear algebra, calculus, probability + statistics, optimization, discrete math.

It’s a harder path, but honestly… only slightly.

Or Maryland: https://financialaid.umd.edu/resources-policies/cost-attenda...

Or Texas: https://onestop.utexas.edu/managing-costs/cost-tuition-rates...

I think you get the picture...

Let's say you get a student job on campus to help cover costs. In my state you can only work a maximum of 19.5 hours per week in a student job by law. If you earn minimum wage (which you probably will) you can earn a maximum of $14,327/yr while attending school with your student job. That isn't enough to cover housing at either of these schools, let alone tuition.

People used to joke about people who went into six figure debt for an arts degree from a private college. Now a degree from a no name state school will set you back six figures.

I was just highlighting that whats taking 10k/year in public school tuition (18k for Michigan (!) ) to ~130k for the degree is the living expenses, which exist wether or not they attend school.

I'd just study law or something incredibly hard that hasn't been tokenized yet.