Hacker News Reader

Comment by smcf | Hacker News Reader

When I went to college, having already done full-time software work and countless hours of programming in my spare time, I went to a departmental advisor confidently requesting to test out of the introductory CS classes aimed at first-time programmers. I nearly got laughed out of the room. I pushed the department on this, but it was clear: they simply did not do this. Everyone takes intro to CS. Everyone.

It sucks, but there are people out there no smarter than you yet more powerful, and sometimes they impose a speed limit.

If you get the opportunity though, I'd still suggest doing what the author did. No harm in learning something twice, particularly from two different perspectives.

View on HN

When I started college, I thought I could fly through the intro CS courses. Within the first week the professor had completely reset my understanding of computer science, rebuilding it from the ground up using purely functional concepts. I’m incredibly grateful I could not test out of that class. I would have done so in a heartbeat and missed out on the most important lessons I’d need for my career.

Our “week 6” assignment was to implement optical character recognition (OCR) in Haskell, a lazy pure functional language. That assignment blew my mind!

Is that online somewhere? Sounds like a fun thing to try!

You might want to have a look at the detexify source. https://github.com/kirel/detexify-hs-backend

This was decades ago, so I doubt it!

The lecturer pre-generated small black & white square bitmaps of single characters from a sample set of fonts to use as aids during development. (Just the lower case letters a to m.) Then your mark was the percentage your OCR could recognise from a much larger set generated from about a thousand fonts.

In the same course the next major project was to make an AI for the card game Hearts. This is a four-player game, and the project materials included an automated tester that could be given any 4 AIs and would score them against each other. In the labs we would battle our AIs while developing them.

I ended up getting the second highest score out of a class of three hundred students and I still don't know how to play that game well. (I got the AI to teach itself the probability tables for optimal play, which I didn't memorise.)

I'm just struck by how cool those assignments are — I feel like this might be one of the biggest differences in school quality, most of the stuff I've had to do at an unremarkable school felt a lot more like "fake problems", but this OTOH sounds like something that would actually get me a lot more motivated even if I had no clue where to begin at first

That lecturer was spectacularly good at teaching, and stood out from the rest of the pack so much that it was almost hilarious.

He used a digital projector in an era where that was still rare in even computer science courses so that he could demonstrate the basics of the edit-build-debug cycle. He'd deliberately leave small errors in the code to "fix" and he'd get the class involved with suggestions on where to look and what to try.

All of his projects were competitive, because it got the students much more engaged if they were racing each other. I believe the year after mine had to play a monopoly-style game where trucks were going around a "game board" and could buy and sell stuff for profit, could run out of fuel, borrow money, etc...

Meanwhile all the other lecturers would drone on and scribble unreadable gibberish on a blackboard. We had one lecturer give us a final exam where half the marks depended on converting 7 lines of C to assembly, except that 7 lines had 5 errors in it. Two of the lines were just the braces, which means effectively every line was faulty.

Pretty much the same situation for me. I was already writing C++ professionally and had embarrassingly high thoughts of myself.

The first language we started with was SML and for a good couple of months I just laughed at this stupid language with its silly and unnecessary limitations. It wasn't until we had to implement a parser in it, something I coincidentally had struggled with on my own, that I started to realize that I was dealing with a different paradigm, and a very powerful one at that.

I was less cocky after that.

Similar not same situation, but different outcome. Had done hobby programming in high school, read a bunch of books, took a junior college course on C.

Got to undergrad. Followed the speed limit, took intro to programming course. 450 person class, 1000 person wait-list. 75% of the class did not pass. Only attended two lectures, hand written programming finals. Never lost a point.

It wasn't until the 3rd year programming languages class that we wrote minimal lisp interpreters in scheme and did crazy stuff in prolog that I had that rewiring sensation.

Not trying to brag, just putting this here for others reading to say that not every intro to programming course speed limit is worth it.

Same here. I came from the demo scene and was disgusted by the thought of having to sit along total beginners, because hey, I had demos ranked in the charts.

CS is a totally different beast and I am glad about the humble experiences around CS. I might not totally agree with all the educators tell you, especially around OOP, however especially the mathematics around CS was impressive and very helpful.

One of of the TA's when I did CS was a member of the Crusaders [1]. We had a lot of interesting discussions of the difference between demo-style coding and CS...

[1] https://demozoo.org/groups/12/

But as long as you know the math it is easy to self learn the other parts of computer science. I just studied physics and math in college and self learned the rest and learned algorithms, compilers, distributed systems etc on my own and have worked on those at Google and never had a time where someone knew something they learned in CS that I didn't know. It was the opposite, since when I self learned I went through every CS topic you can take I often knew a lot of CS things that the CS grads didn't know since they were limited in how many credits they could take.

I know many who took a similar route as me, but I've never met anyone who self learned math to a reasonable level. And since math is a very small part of a CS degree you could learn all of it in 1-2 semesters, so the optimal way to learn computer science would be to take 1-2 semesters of college math and spend the rest self learning. I bet you'd get a better understanding that way in much less time than it takes to go through a CS degree.

Things are different now. I spent my intro class constantly correcting the instructor.

The approach when I did CS seems pretty reasonable: Lectures were attendance entirely optional for most courses (exceptions were some courses with small number of places; e.g. I did a "French for sciences" class where attendance was compulsory), and for the intro course I showed up to the first and last lecture, but you had some compulsory lessons in smaller groups and a couple of compulsory group projects.

I got to save plenty of time by opting out of most of the lectures, but the groups and projects gave you feedback if you started skipping too much (something I learned in the introductory maths course that I also first thought I could get away with not going to every lecture for) and forced you to read at least the main set books so you got a good idea of whether it was material you understood.

It's indeed easy to think you know more than you do, but at the same time, sometimes you do know exactly what you know - been in both situations.

Like why bother going to college to skip the in-person classes? Might as well just buy books and work through them and save money. Make yourself a bookcase. That's cool. Not "well-educated", a term I hate, well-read. Though you do at some point have to interact with the medium for real, can't just read about martial arts. At that point hire a tutor--in-person classes.

Matt Damon as the protagonist of Good Will Hunting: "You wasted $150,000 on an education you coulda got for $1.50 in late fees at the public library."

That wasn't true before the pandemic but without in-person classes it is absolutely true.

Because there were other classes more worth my time, and so by skipping the in-person classes of subjects I needed to take but where I knew the material well I could take a higher course load and get more of the things I actually learned new material from.

By the time I started university, I'd been programming for 13 years, I'd done paid freelance development jobs, and written my first compiler, and I'd done more advanced stuff than the first several semesters worth of classes. I had gaps, and I filled them with the group work and books, but there was no value to me in the in-person lectures for the introductory computer science courses.

That's a pretty good answer but there's a few gaps in it. Like I get there's still "taking a higher course load" ie score points by jumping through hoops to impress the future. Like I did a lot of this in high school, endless jumping through flaming hoops, and good exercise in hindsight, but come on. Well I guess everyone has little choice.

Yeah and you did pay some attention, got a sampler plate, makes sense.

Plus it's fun acing courses with no effort every once in a while. It's a good sign.

Taking a higher course load was to enable me to actually learn things I wanted to. It was not about jumping hoops - didn't even finish my degree, because I had better things to do (a startup). I did finish it up years later, and that was about jumping hoops because the job market in the aftermath of the dot com crash in the UK where I moved to in 2000 meant having a degree was a lot more important.

It was also not about enjoying acing courses, but about needing to take courses that were compulsory for the programme and that you could not test out of. As I said, the option to skip most lectures felt like a reasonable middle group - if I'd run into problems with the group sessions, I could have easily gone back to the lectures at any time.

Very very good answer. No gaps.

I had the opposite for the intro to programming class. Sure, I learned a language I didn’t know (but that took like a week tops) but all the actually programming topics were things I learned years ago. It made me very bored and gave me a tendency to skip class, which is a bad habit to fall into.

The other intro courses (eg computer architecture, which taught assembler) were great though, even if I found them easy due to prior knowledge.

Unfortunately that experiences doesn’t give a clear line for what to test out of and what not. I bet I could have passed the computer architecture one too, but I would have missed out, while the programming one would have been best if I used the time for something else (I mostly did, but there were lab sessions that were graded and couldn’t be easily skipped)

As someone who teaches in university I can tell you by experience that the worst students are those who think they already know what you are going to tell them.

The thing is: it is hard to tell them apart quickly from those who actually already know, without having them take some test.

And even if you have them take the test, you cannot be sure they miss some fundamental core concept that will be crucial to understand later.

I myself had to go through introductions more than once that I thought will not offer me anything new, but going through the basics once in a while with a better understanding can be of incredible value

> And even if you have them take the test, you cannot be sure they miss some fundamental core concept that will be crucial to understand later.

That seems like a bad reason to err on the side of unconditional speed limits. Opportunity cost is real.

In practise your university will not provide you with the time/resources to do extras like these.

On top of that as an educator you have a real responsibility not to send someone on a hail mary without fundamentals, which might ultimately drive them out of university because it will not get easier after that. This is a real risk.

Another aspect: What so you think how many students will try to skip classes if word gets around it is actually possible to do so? How many of those will cheat to do so? And bow many of those who made it through will be able to face the next exam without cheating?

I sympathize with this and have definitely been horribly wrong about basics before, but it‘s also crazy how many teachers don‘t manage to set up courses and asssignments such that they really force you to understand the basics.

This is sadly true. I have suffered under this as a student myself, so I try to not forgot how it felt and constantly check my material for unexplained aspects, that I just assume blindly.

In school, you for some reason take the lessons first and then the test. In real world, on so many levels, you start with a test and then proceed to lessons as needed, if at all.

If you pass an earlier test while missing some fundamental concept, then you'll simply spend the time learning it when you do need it. If you never need it, maybe it wasn't that fundamental after all.

I think it's a similar situation as if you only have a hammer everything looks like a nail.

You can probably find a solution with what you already now (as a junior programmer) as most problems programmers have to solve are not that hard but you may completely miss a better solution because you had no idea it was possible. I may be fine but you may also lose a lot of time later because it wasn't.

I agree that it's possible to self teach almost everything in CS but the point of university is to speed up the discovery of CS from scratch and have solid foundations. You certainly don't know everything graduating university but should now where to look when you have a problem imo.

> If you pass an earlier test while missing some fundamental concept, then you'll simply spend the time learning it when you do need it.

Only if one knows what they don't know. And fundamental concepts have a habit of hiding in plain sight. It's easy to waste lots of resources because one doesn't know about a fundamental concept.

Sounds a bit like 2nd-graders who wish to skip the multiplication because they don't need 2*3 - they can pass the test just by calculating 3+3 after all...

When learning, unlike the person teaching you, you don't know what's coming in the next lesson or the one after it, so it's sometimes a good idea to trust them on a curriculum. It's also the one of the biggest downsides of being self-thought (as I am in big part), you always have holes in your perspective that you're not even aware of and it takes years to stumble on something that someone else was told in the 1st year of being a dev.

Your response neatly encompasses everything I disklike about academia. It's easier for the institution to bucket everyone into a group and disregard the individual, so that's what inevitably happens.

It’s not only easier, it’s the only way you can teach a large number of people at the same time.

College wouldn’t exist without it.

Well instead you could get one-to-one tuition sensitive to and tailored for your unique strengths, brought by the very best in the field. That would be pretty resource-intensive, but maybe you are worth it. But now look at from the providers' perspective: they can only afford to do this for a small fraction of the students that will think themselves deserving of it. What approach will let your special gifts shine through all the competing candidates?

This is, more or less, how graduate school works.

Not sure you are projecting a little bit here, what I do and how I do it is very untypical for academia (I teach at an art university, so it just counts to give people skills not to teach them some fixed curriculum), I also get the chance to do one on one teaching sessions where students will profit because I can do things at their level. Also we don't have tests and students are free to choose their courses.

Given that environment if someone comes to "introduction into analog sound synthesis" I can expect them to want to hear just that (or they need the credits). If people don't come to learn, I will not force them. They are grown ups, it is their decision if they want to learn.

Please also consider that unless you have experience in teaching groups you might have a rose tinted view of what an instructor can practically achieve. If you have a group of 20 people with 10 of those having no clue what you are talking about, 2 that are very advanced and the rest with mixed levels inbetween you must find a way of teaching the 10 that have no clue while also not boring the 2 who are more advanced. This can be a hard problem to solve in a good way as you cannot split yourself. If your group is bigger it gets harder even (and at some point you have to stop worrying).

I am totally for people being able to skip classes where they can demonstrate they already know everything thought in it. But practically it might be a lot of work for anybody working at university to create such an test for e.g. just one person. From the perspective of a student it all looks a lot simpler than it might be. For example even if someone could test a student to make sure they are not sending them into a hail mary be letting them skip fundamentals, maybe that someone has so many other tasks on their shelve that even if they wanted they cannot do that?

Also: for every student where this might make sense you will get 3 or 4 that really overestimate themselves with a nearly narcissistic inability of judging their own ability. You know, the type that would like to construct the equivalent of an iPhone in circuit form while not being able to explain ohms law.

> Also: for every student where this might make sense you will get 3 or 4 that really overestimate themselves with a nearly narcissistic inability of judging their own ability. You know, the type that would like to construct the equivalent of an iPhone in circuit form while not being able to explain ohms law.

But of course they could. And indeed, considering them narcissists is what many hate about the academia. Maybe they'd be better off taking a less linear path. Sure, in the context of the academia skipping classes is not justified and there's no problem with anything you're doing. But when one is frustrated with schooling, the first instinct is often to try to go faster. Quitting school is supposed to be a very bad thing, that's the one acceptable way to do something different.

For the most part I had the nicest and most accommodating teachers, had no problem with them, they had no problem with myself, can't blame them for anything. But I do strongly dislike the nearly universal borderline religious belief in traditional schooling being the right way. It's okay to not know some thing and not want to know them.

I think in 10 years of teaching I met one student who genuinely might have skipped through the intro programming course without too much harm.

Being able to pass a test is indeed not a sign of mastery of the material (this might come as a shock to some! Just because you got good grades doesn’t mean you’ve mastered the subject! It just means you’re good at the test! But wait, why use tests then? Well, sit down, get comfortable and let me tell you a tale of how we got here…)

The fact this conversation exists highlights that our education system is silly. Without evidence, I subjectively like the ideas in “One World Schoolhouse” as an alternative.

The way I see it people who just like to get a degree will complain about having to take classes with material in it they already think they know.

People who are really interested in the subject will not complain. I for example have been programming for the past 15 years and I will still read something like an introduction to C programming, just because there might be some ideas, explainations or examples in it that will help me deepen the fundamentals of what I already know.

Arguably there are topics where this works better and topics where this works worse. In electronics or mechanics groking the fundamentals is so essential to the whole field that you should know them by hard. And the only way to know them by hard is spaced repetition. So having some overlap in the things you learn is not only necessary here, it is crucial.

You mean “by heart”, right?

I know you typed it twice, but I just never heard that expression.

autocorrect, perhaps.

>Being able to pass a test is indeed not a sign of mastery of the material

I think/hope you mean "guarantee" rather than "sign", because, yes, someone who has mastered the material is indeed much more likely to pass the test than someone who has not. The existence of non-representative counterexamples does not refute that.

A great point! (Sorry for a drive-by comment!)

When I was in college I needed to take two classes, one of which was a prerequisite for the other, within semester to graduate on time. The CS advisors adamantly refused to add me to higher level class because I was missing the prerequisite class. Furthermore, the classes were taught at the same time to prevent people from taking them in the same semester, for some reason.

I talked to the more friendly advisors in the Philosophy department, who also had the ability to add and drop students from classes, and got them to add me to both classes I needed. I persuaded them not to worry about the overlap. I completed the semester by attending the higher level class except for exam days (I had to get a special dispensation to take one final early because their final exams were scheduled at the same time).

My point with this anecdote is that sometimes more powerful people try to impose a speed limit and sometimes you can get around them.

I think the scheduling both at the same time might just be an "optimal" solution if you already know that people can't take them at the same time. That frees up other time slots for courses people taking either one might take.

This happened to me twice. I'm actually endlessly annoyed at the whole college system being structured like this.

So I get that college in the US can be expensive, hence the rush, but for me, one of the main benefits of university was time.

My CS class very quickly divided into "those who could already program" and "first time programmers". (testing out was not a "thing" then - at least I never heard of anyone doing it for any class.)

Interestingly it was more classification than groups. I spent time helping friends who were stuck, but I also spent a lot of time with other advanced students, in the lab, pushing each other. The assignments were trivially simply, so we just spent time making them more interesting.

Because the course work was done quickly, we had a lot of time to go the extra mile. Time we wouldn't get later in life.

The important things though, the theory, stuck with me, and even the early classes were valuable. Intro to programming only lasted a few months, then there was a lot more theory, and we could meld that into our code a lot more.

I went into college knowing how to code. I came out knowing how to program, and there's a big difference.

If you are in college today, funding aside, I'd say - don't be in such a rush to finish. Fill your spare time by sucking in riches, seek out every nugget, help others, be challenged, see the time as an opportunity, not a class to simply pass.

This is my experience as well. I was doing well in CS courses, but struggling in others, and didn't graduate early (well, didn't try to either). However, finishing whatever they threw at me allowed me to go ask for more, or try more harder things.

As a result, I learned way more than curriculum offered, got way harder assignments and projects (like designing a compression algorithm from ground up as a graduation project), and satisfied much more overall.

I continued my M.Sc. and Ph.D. at the same university (because I already had a job, and I just wanted to learn), and this head start (and being in good terms with the professors) allowed me to do similarly heavier curricula during these studies, too.

It's worth it.

I'm in the UK and this studied in a system where you can't just skip ahead, but I also got so much out of sucking in the riches as you say and helping other students. Because I already knew the basics I could spend time really mastering every last part of the courses and it was wonderful (and also gave me space and time for lots of non-academic activities!)

This isn't possible anymore. The system has been fully automated.

I tested out of multivariable calculus in college and to this day, it is one of my biggest regrets. My peers who took the course have an intuitive understanding of the material that I never developed. Sometimes there's a difference between learning on your own and working through the material with others.

There are loads of information available on this. For example, Khan Academy has a very nice multivar glass, done by Grant Sanderson (3Blue1Brown). Why are you limited by what happened (presumably) years ago?

The lack of homework, I think. It's not a natural thing to do. Apart from an explicitly pedagogical environment I have found the best thing to suggest people is to try to work their learning into a project. For math this can be hard unless you also have relevant engineering experience to make something with your math you are learning.

Introductory multivariable calculus courses are seldom amazing and often mediocre.

If you want to work through this material yourself, you may enjoy Hubbard & Hubbard’s book, https://matrixeditions.com/5thUnifiedApproach.html

For 99% of multivariable calc students, once the course is over and the social environment moved on, there is no reason or motivation to leaen it. Like most of college coursework, it's just a ritual.

But if you're on HN, you might have an interest in machine learning, in which case it can be extremely helpful.

Seconded, IMHO the most underappreciated modern mathematics introductory textbook.

I'm still struggling to find the right blend of self study and group study. With peers you can coast along through the help of others to pinpoint you where you missed something, what's the "right" path to a solution. Which means .. you didn't really understand it. But sometimes I found solutions on my own, but then I realized that there were still some issues in them. I'd say it's important to self study but to reality check regularly.

The first part of 18.02x (Multivariable Calculus) is open at the moment, not for certificate, but the autograder works. The second part starts next month

https://www.edx.org/course/multivariable-calculus-1-vectors-...

https://www.edx.org/course/multivariable-calculus-2-surfaces...

After learning programming and CS on my own, I thought I'd go back and pick up on areas I missed by taking some courses at a university (I already had a degree). It was a fairly highly ranked university for CS (in the top 20), but the whole thing was honestly a waste of time and money and I was shocked by how little I learned.

I did test out of some classes, though they only allowed it for a handful of classes. And you needed to be able to get a higher score than students who had just finished the class, while doing so with much less material (students going through the class get told specifically what's on the test and are given materials accordingly; people trying to test out aren't given either). It was also surprising that on the Computer Science II test they had a large amount of the score based on memorizing default Java methods.

Because I was doing this strictly for educational purposes, I got really interested in how much students were retaining between semesters. It seemed to be very little, and there was little that they needed to retain. Most would remember big O, but no one remembered little O, big theta, Big omega, little omega despite the time they were required to memorize it. As I mentioned, CSII was very Java focused where students had to memorize specifics of default methods and Java inheritance edge cases, but after the class was over that information was almost all forgotten. You might have a class where you do a few weeks of stack based programming in a toy language, but it's not enough to actually do anything with and, again, is wiped from the students mind as soon as they finish their finals.

All this was driven home even more when I tried to discuss topics with colleagues who had CS degrees. They retained a very small amount of the things they studied in college, with the vast majority was met by "Oh yeah...I think we studied that? Isn't that the thing where [insert some vague broken memory]."

The whole thing felt mostly like a waste of time in order to justify four years of teaching. Of course, someone who never studied programming or CS would get _something_ out of it, but even there it seemed to be in the most inefficient way possible.

I don't think anybody is expected to have perfect recall of all those small details, but it will all come back easily when you need it and go back to skim over the material to refresh your memory.

Ugh. I was told by the administration that I couldn’t test out of my intro to programming class in college. Then I had several professors ask me why I hadn’t tested out.

I’ve wondered if the professors didn’t know or if the admins just didn’t believe me I would’ve been able to. (And this was with a passing AP test score from high school)

I feel like Accreditation is to blame. It’s such a racket.

I was told I couldn't test out of calculus (university didn't recognize IB calc), but course prerequisites were easy to get around so I just started registering for higher level math classes. My junior year I went to my advisor and he happily waived the calc requirements.

i am unfamiliar with the accreditation racket, but on the face of it if Berklee lets you test out of 2 years of music theory, surely you can test out of computer science. I know I tested out of 2 years of math.

its probably a more boring answer like university bureaucrats refusing to budge because they just didnt feel like it, which is a rather more tractable problem (in the large; individual students are powerless of course and i have been there too) than the accreditation system.

we need university entrepreneurs to disrupt lazy universities who arent serving their customers well. sadly lambda school has left a black taste on all trying to do this. but i have hopes that freecodecamp will do it the "right" way - patiently.

I spent my childhood learning classical violin and it’s a completely parallel education system that’s unaffiliated with public education and goes back centuries (With freelance teachers, dedicated schools, Church run programs, and so on) so I’m not surprised it’s possible to test out.

The same thing doesn’t happen with computer science.. i.e we don’t send young children to CS tutors 2-3 times a week and make them practice on a daily basis. Perhaps society has yet to master creating virtuoso programmers in the way it has spent hundreds of years perfecting the education of young musicians.

They lied. Highschools do this too. They will try everything to prevent you from testing, because they lose the tuition.

On my first day in university the intro CS professor started his slide deck with "this is what a keyboard looks like", so I walked out of there, into the department office, demanded to see the chair, talked my way out of the first year courses, and demanded to take all of the required 2nd year courses in parallel that first semester. Not sure if the department chair was bemused, or just wanted to drop me in at the deep end, but he agreed to it, and I got out of there with a double-major undergrad and a masters in 5 years...

Where there's a will there's usually a way, but you need to either know how to manipulate the powers that be, or have enough confidence to bypass them.

If they don’t another option is to build a startup, or start on your phd research while ducking out of the the first year classes.

I'm sympathetic to having to take courses on material you already know, but I do think there is some justification to making everybody take an intro CS course. Specifically if it's a course that is about teaching the ways of thinking about programming, such as abstraction, composition, state, rather than a rote course on how to write code, then it provides students with a foundation that is rarely seen in AP courses or self-study. It can be an opportunity for a professor to provide their views and experiences on programming. It can also be a chance to talk about other essential, but rarely discussed topics such as ethics, the qualities of a good programmer, programmer culture, etc.

Of course an intro course that is this well thought out is rather rare, in which case experienced people should by all means skip the rote nonsense.

Why don't advanced courses also pcbet that, if it's important at all?

Counter story, when I was a senior I found I missed a freshman requirement. The teacher saw me in the class, I was fine with having the blow off class but the teacher recognized me as a CS student and said I shouldn't be there since I was a senior. Had me schedule and take the final for the class, I passed, and I was able to skip it for the semester.

Doesn't hurt to ask either.

The difference is that you still paid for the class. The university wants the money.

I believe at most 4 year universities, in the US at least, you pay by the semester or quarter and not on the number of classes you take. So the university won't generally make more money if you test out.

They will make more money if you do NOT test out because you have to pay for the credits, not pay by the semester.

> I went to a departmental advisor confidently requesting to test out of the introductory CS classes aimed at first-time programmers. I nearly got laughed out of the room. I pushed the department on this, but it was clear: they simply did not do this. Everyone takes intro to CS. Everyone.

I tested out of the maximum number of courses for my degree. It's insane. This is why I have such a low opinion of structured education. I hope to be someone's Kimo. I had my own who pushed me to build good software, not be lazy, and a lot of fundamentals on simplicity.

I started by talking to advisors and professors in the CS department, and they agreed that I should test out of intro programming classes.

The dean was a cruel man named Warren Harrison who accused me of trying to "game the system" and told me that they don't tolerate people like me here. I shared the story later--other people had similarly negative experiences with the guy.

I wonder how much of this is due to formal curriculum standards and accreditation. It seems like the only permitted method of skipping a course is having had taken an AP course and scoring high enough on the exam.

I’m guessing standardized education prevents the kind of judgment calls that would allow a department to let a knowledgeable student skip classes they don’t need to be in.

I had a 5 on each of the AP physics exams and still had to take freshman physics.

The school had a policy that you could not test out of the core courses in your major (I was a physics major; had I been a different major, I could have at least gotten placement, if not credit.

It can get even more ridiculous as a transfer student.

I had to re-take two physics courses at a California State University school that would have otherwise transferred from a junior college due to minor catalog year changes. Ironically the JC physics courses both had labs (the usual stuff in Electricity & Magnetism, and a whole bunch of optics and slit diffraction fun in Modern Physics), while only the E&M lab remained at the four-year university. The Optics/Modern Physics prof from my retake also taught astronomy, and was still harboring a minor grudge against Neil deGrasse Tyson from some shared research work in the '80s.

Also couldn't test out of or otherwise apply transfer credits to freshman biology, and the subject matter was at about the same level as what I experienced in 7th-grade science at a rural public school.

Intro to CSCI was unavoidable, as my previous programming coursework was coded under MIS. The cherry on top was the professor pulling me aside after the first midterm, telling me "you could teach this, since you can't test out I'm just giving you an A for the whole course". Despite this I still attended his rambling lectures, and found that he did actually manage to tie all of his tangents together rather like a call stack in human form.

On top of all of that, the transfer coursework that actually applied toward the degree that did surprise me were my technical training credits from the Community College of the Air Force. Ten units under law enforcement and field work actually took care of general education elective requirements - "Fundamentals of Ground Combat" was rather amusing to see on the graduation transcript.

If anything the issue with AP courses is that the bar is so low to pass one. Indeed it's a bar based on percentile and not on actual knowledge. Someone getting a 4 or 5 on the APCS test tells a department zero. Which is why some schools have the policy that APCS does not allow students to skip the intro CS course.

In contrast, Western Governors University allows students to test out of any course.

For example, I completed my CS degree in 3 months: https://miguelrochefort.com/blog/cs-degree/

I was on the competitive programming team at my college. I was able to join on one of the more advanced teams pretty early, my other teammates were college seniors while I was a freshman.

I actually was able to get our 2x weekly meetings to count as course credits for basic classes and the professors who ran the group vouched for me to skip some of the pre-requisites for higher levels classes. (e.g. I took Algorithms II while in the same semester the competitive team meetings counted as Algorithms I).

Then my family moved, I wanted to transfer schools to keep living with my parents while attending school. The new school wouldn't accept my "free" credits. So in my third year of college I had to take intro to computers, intro to programming, etc.

I did that for one semester, hated it, then managed to find a job writing Python. I still haven't graduated, heh...

There is harm. Is anything more valuable to a human than time?

I had this too. I had worked the system a bit to wait until my final semester of college to actually take the intro CS courses (I had one of two possible prereqs to the second course so just started there).

I asked to test out and the department head said no. I had almost an entire CS degree and still had to take the easiest course. We made a compromise that if I take the midterm and exam with all the other students, I can pass the class. I got a 100% on both and spent like, 1 hour learning Octave and C for each exam.

I was only in college for a brief stint, but I ran into this problem as well when I went to college. Although I hadn't been in the industry yet, I was a hobbyist for a long enough time for the intro to CS at my college to be simply pointless. After a couple weeks, I was able to convince my professor that I should be able to test out, and despite the fact that there was no procedure, I was able to just take the final exam, pass, and move up to the next course.

That next course also seemed pretty basic, but I hit a brick wall, because I didn't have the required math credits to get into the next course up if I tested out of that one. And testing out of a math class, while certainly doable, was not as enticing as a prospect :)

The truth is, while it's reasonable for an intro to CS course to be kind of basic, I think it was a good sign that either college, or at least the college I was going to for computer science, was not a good option at that point in my life. Thankfully, circumstance would knock me into the industry instead in short order.

(Now I'm interested in college, but not for computer science.)

I had the opposite problem. When I was a freshman, if you took AP CS you were allowed to self choose out of CS1. And I did. It was a terrible decision for me, in retrospect. It set me down a bad path. I was a very bad student. I could keep up with the software stuff in HS but that was about it. I shifted to college where everything was in C and pointers and I was struggling to wrap my mind around them at the time. And oh by the way, I was a freshman in college, with a million other new things going on in my life.

I very much get the desire to place out of CS1 or whatever. My experience was that the overall college experience was the jam. A low level class here or there isn't going to change your life.

I had a similar experience. They said I could test out of it if I took every test and did every homework assignment. And that is why I didn't get a minor in CS. Fortunately, at the time they didn't strictly enforce course prerequisites so I was able to take a few higher level CS courses, but for 300 level classes you needed a signature from a CS advisor, which I couldn't get.

Some of the greatest coaches of all time start every season by going over the basic’s like ‘How to tie your shoe’. Drop your ego at the door.

I think the bigger point in this story is not that there will be times in life where you're confronted by authority to stop you from going over the speed limit, but that for every one of those cases there are countless other scenarios where you are the authority yet still think there is a speed limit.

> It sucks, but there are people out there no smarter than you yet more powerful, and sometimes they impose a speed limit.

College's point was never about learning anything for a while now, it's a credentials machine and the machine only works in one way: the one that extracts the most value out of students, the longest.

College’s point is to prove that you can learn a couple of dozen knowledge sets in six months each, to a testably adequate level, and also stick with something for three years (a reasonable guess at how long a technical employee takes to pay off the investment of hiring them.)

> you can learn a couple of dozen knowledge sets in six months each

You mean shallowly enough just to pass a few tests that can be easily hacked by accessing the previous year's? You can definitely fake it until you get the paper, I've seen that so many times.

That's not hacking or faking it, any more than shipping a couple of products using the some pointers from stackoverflow. If you show you're actually able to stick with a thing and finish it to a sufficient level of quality, that's kind of the point. Developing a deep understanding of esoteric topics can be fun and intellectually satisfying but is usually a tiny part of most careers, even the fun and satisfying ones.

I think the one course requirement I got waived based on my experience was an internship requirement. Everything else I had to take. I don’t regret missing out on the early courses as they provided a lot of surprisingly useful instruction. A lot of the “why”s that I’d take. For granted.

Any publicly funded uni that does not allow testing out of every class should be blacklisted. They intentionally prevent it due to quote "costs". They depend on the tuition so admins can buy lake homes. That's straight out of the words from someone whom handles it.

The best option is not to go. I had also done full time programming for clients from age 15 on. 5 years after that I got opportunities to lead small projects. 10 more years after that and I was made VP of an engineering department. I've worked a lot with fresh graduates from "good" schools with good GPAs. I've always been thoroughly unimpressed by what they know and what they think is important for the job.

You sound like someone who thinks they're better than everyone else.

Maybe they are. Better at programming and leading projects, that is.

Across virtually any measurable trait or collection of traits with a reasonably normal distribution, about half of the people will come out better than average.

Iratewizard may or may not be better than everyone else; but he/she is almost certainly better than average at a lot of things. We all are.

Your skill at a job doesn't equate to your value as a person. A job is just a way I and most others contribute back to the world. It's in everyone's best interest if there aren't unnecessary barriers to that contribution. For instance, universities' poor curriculums and the way they predatorily pretend to be the gatekeepers of "prestigious" jobs.

This is basically what I did. The only complication were "quizes" which were just completion marks but at random times. So I would skip 4-5 classes after each quiz then come to classes and so personal programming in the back until the next quiz landed.

Reply on news.ycombinator.com