Ten years ago, the department graduated roughly 400 students with Bachelor’s degrees in EECS or computer science. This year, the number will probably be 1400, or roughly 15%, of all undergraduates. Almost all of that growth is in the College of Letters and Science CS major.
and more specifically:
The campus provides departments $50 per student-credit per hour to pay for both lecturers and TAs. A student in a 4-unit class contributes roughly $200 out of the student’s overall tuition for these expenses. Yet the actual cost to a department is more than $350 per student for that 4-unit class, forcing departments to find additional funds to cover the upward of $150 per student gap. [...] The EECS department is running an annual deficit of nearly $5 million.
At a high level, this problem is "growing pains" rather than "falling backwards". Computer science is booming at Berkeley, and they can't hire enough TAs to keep up when they are paying so badly. I hope they simply start paying CS TAs better.
Impossible. Students definitely don't pay more than $50 per credit hour in tuition, so there's nowhere for the university to get the additional funds from.
Tuition is $14,000 for in-state and and a laughable $44,000 for out-of-state. And they can’t find an extra $150 per student? Fire one worthless administrator and the $150 gap is completely covered. Voila!
Aha, that makes more sense now. I was almost going to say is this really news? I thought the concept of an "impacted" major was a pretty old definition--it basically means there are more applicants than there are spaces. But this other bit of information about a growing slice of the student body pie electing for a hot and lucrative major.
This reminds me of my alma mater. The provincial government previously had funded $X/student. Then they changed it to a fixed grant. Meanwhile, enrollment continued to expand; the institution saw the expansion as serving the public. Soon enough, the government cut back the university's funding, which cut the operating budget. Eventually the Faculty of Science was looking at a $10 million shortfall; the equivalent of an entire department's budget. So they cut back on student spaces to the old $X/student allocation. One would hope that the public would notice and speak up in response, asking their politicians to properly fund education.
Correction— the author of the op-ed, Nicholas Weaver, is a former lecturer in the EECS department at UC Berkeley.
And a personal insight, as a former uGSI (TA) in the department, the problem isn’t paying TAs enough. The pay is actually quite high for a student job (especially when you factor in the fact that in-state tuition is paid for). But the course budgets simply aren’t high enough to hire the sheer quantity of TAs that would be required to handle all of the students who want to take the classes.
I was involved in hiring some semesters, and there were always more people we wanted to hire than the budget would allow. I believe the situation has only gotten worse since I left.
Can I ask what it is that costs $350 per hour per student? There is one professor lecturing and a few TAs who work 20 hours a week at max. There are 60-100 students in all the major classes, 20-30 in more advanced ones. What exactly is the cost that leads to $350 per hour per student? Compute and infrastructure costs amortized over many students per year can't be it. Then what?
maybe he meant "credit hours" which is a different thing, I think a 4 credit course means that classes meet 4 hrs a week, but the $350 is for the entire semester or the quarter.
The largest classes (CS 61A and CS 61B) have roughly 1800 and 1600 students respectively during peak semesters.
A 20 hour TA is paid ~$2250/mo for approximately 4.5 months per semester, plus ~$7000/semester in tuition remission (so ~$17k). $350 is the total cost for the whole semester per student. 17k/350 is about 50 students per TA, which sounds about right (and even that is a higher than ideal ratio). There are other costs too, but these are the some of the highest.
Things being what they are in the USA, the cost of a grad TA may also include the cost of healthcare. Plus most grad students would be charged out-of-state tuition, which can be quite a bit higher than the in-state rate.
Grading was actually pretty fun. I kind of miss it! Exams are scanned and then graded electronically, which makes it less of a hassle. It’s still a lot of work, but you have an army of TAs to do it.
It's $350 per student per semester, rather than hour. The course mentioned in the op-end is a 4 credit-hour course which typically means the class gets $200 per student per semester to host it
It takes a lot of TAs to handle student support, which doesn't scale well in practice. The class I teach publishes lots of videos and documentation about how to do the labs, projects, and homeworks, and we also have a forum where students can (to a reasonable extent) help each other out with common questions. However, the classes teach a lot of stuff really fast and you need a lot of TAs doing things like office hours to help students who didn't understand the speedy lecture, or have a particularly nasty time trying to debug an error.
One particularly difficult thing about this scaling is that students typically need TA help all at once. Office hours right before midterms and project deadlines have many students and more bodies are needed just to have good handling of "peak" times
Another aspect of TA cost is that TAs are paid significant amounts of tuition remission for their work, and the department has somewhat of a mandate to ensure any graduate student who wants to TA can get full tuition remission from a 10 hour position or higher (I'm guessing this is somewhat of a symptom of skyrocketing tuition and inadequate aid being foisted on the department to deal with). The department used to get around this by hiring tons of undergrads at an 8 hour level so they wouldn't get remission, but they were sued and paid out millions in backpay as well as paying all TAs a scaled remission based on their hour count. (https://www.insidehighered.com/news/2020/01/16/arbitrator-sa...). Generally an 8-hour TA gets paid around $40 an hour after remission and a 20-hour TA gets paid around $50 an hour. [very rough math read the other comment for harder numbers]
Wait $350 is for the whole semester? So, if a student takes 5 courses per semester it will cost $3500 per year. Why can't the $15000 instate and $44000 out of state pay for this? If even 20% of the tuition is not going for class expenses there is something messed up somewhere
The $350 is the cost to the department in terms of things that scale per student (specifically faculty for teaching, grading homework, etc). The university as a whole has many other expenses.
If my tax dollars go anywhere, I would rather encourage students study multiple high-demand majors than provide loan forgiveness to societally unproductive majors (gender studies, literature, etc.). I don’t see why we should use tax money to prop up anything other than proven productive fields of study. Society should get something back, and I’d rather get back a smart physics/econ double major.
Regardless of your feelings about tax dollars, I hate to break it to you, but the UC system hasn’t being subsidized by California in decades. Something like 15-20% comes from the state.
The overall budget gets weird with medical centers and something called “Sales and Support”, but in any case California tax payers put in more money then students. I don’t think that’s a problem, but it’s definitely subsidized.
The strange thing is I as a SWE interviewer generally find an engineering degree to not be indicative of engineering talent, over say, a masters in underwater basket weaving. I’m not sure dumping money into stem education does anything, or if any field of study is “proven productive.”
> But for subjects like Civil, Electrical, sciences like Physics, Biology, I suspect the story is quite different.
I can speak to Biology; most undergrads are left with little opportunity as most roles require a graduate degree to do anything meaningful when the economy is actually favorable, and often seek roles outside of their field when it's not. As a undergrad that was lucky enough to enter the field in the wake of the GFC long enough to just pay off my student debt I've met so many biologists/biochemists who either couldn't find anything in their field, or simply gave up trying and sought the highest paid and somewhat stable role they could get 'that required a degree.' It's quite clear that the system has failed, and the system is designed to onboard more STEM grads as it sounds as their are ever-growing holders of degrees chasing so n amount of open roles.
Tech in unique in that it is one of those few Industries where the barrier of entry is rather low, but the attrition and competence level vets for viable candidates almost entirely on it's own--the gatekeeping interview process is still an unnecessary hurdle.
Personally, for all the controversy Eric Weinstein has made directly and indirectly of his own actions, the concept of Embedded Growth Obligations was incredibly well researched and consistent with what we have seen as outlined since 2005 and then solidified itself in the financial collapse of 2008 [0]:
> Embedded Growth Obligations are the way in which institutions plan their future predicated on legacies of growth. And since the period between the end of World War II in 1945 and the early 70s had such an unusually beautiful growth regime, many of our institutions became predicated upon low-variance technology-led, stable, broadly distributed growth. Now, this is a world we have not seen in an organic way since the early 1970s, and yet, because it was embedded in our institutions, what we have is a world in which the expectation is still present in the form of an embedded growth obligation. That is, the pension plans, the corporate ladders, are all still built very much around a world that has long since vanished.
We have effectively become a Growth Cargo Cult. That is, once upon a time, planes used to land in the Pacific, let's say, during World War II, and Indigenous people looked at the air strips and the behavior of the air traffic controllers, and they've been mimicking those behaviors in the years since as ritual, but the planes no longer land. Well, in large measure, our institutions are built for a world in which growth doesn't happen in the same way anymore.
Re: the “hard” stems, maybe? They have a clearer function as a certificate pipeline, but in terms of prep I find it difficult to quantify. There’s also limited positions at the end of this pipeline (medicines a killer example, academia, research). Are we creating better candidates or more positions with stem programs? Or just more candidates who 1) don’t find positions or 2) don’t end up with relevant skills for the roles they take.
(Incidentally, I think CS is a very good place to invest, precisely because there’s a lot of demand, start up cost for self employment is low, and applicability is super broad)
Especially when it comes to fundamental research a lot of work is "useless" until it isn't. This idea of picking winners before people even start at a college is not the way to go.
Many people would argue that linguistics is one of those useless studies compared to comp sci, but it shares deep connections to research in logic and by extension type theory (see handbook for logic and languages)
what I understood parent to be saying is that the picked winners were picked by students, not some central arbiter of what should be studied.
unless youn disagree with the laws of supply and demand, it seems the market consensus at this moment is that compsci majors are far more useful than any fundamental research (which still remains undefined).
Even if the market was wrong and fundamental research could somehow be valuable, why would you not trust the primary beneficiaries oj making that decision, vs some bureaucrat that has no stake in the outcome?
You're right, the search space is deceptive and going directly towards the goal might lead to a dead end. Only in hindsight we can tell which stepping stones were essential. Immediate results are the best, but in the meantime we need to focus on novelty and interestingness.
A great video about open-endedness (warning - very long) with Kenneth Stanley, author of the book "Why Greatness Cannot be Planned"
Gender Studies and Literature will never not be useless professionally. That said, there is immense value in having some number of scholars study those fields in great detail and we would suffer as a species if we let our knowledge in those fields rot.
Would it perhaps be prudent to cap the number of participants in those departments to match the number of real-world vacancies available?
People are really out here thinking that arts and critical theory haven't advanced human civilization... and that science and technology exist in a cultural vacuum.
According to 'The Master Plan', UCs are designed for research and post-grad work and some professional careers. Only the UCs give out PhDs (with a few exceptions). CSUs are designed for professional and teacher education and cannot give out PhDs (with a few exceptions). So, as an undergrad thinking of grad school, a UC is a better place to go to as they have the grad labs and classes on campus, while the CSUs mostly do not. Your chances of getting into grad school are a bit higher.
Berkeley, being a UC, is more geared toward research than Chico, a CSU. I mean, Cal even has a nobel prize parking lot.
I'd like to make sure I understand your position here.
You seem to be saying there is a body of ideas that has gained such a widespread foothold that they hold a significant place in the culture such that they can unfavorably affect social dynamics at civilization-wide levels.
If this is true, it would imply that we should be carefully -- some might even say critically -- be examining civilization at all levels from ideas (and their implications and consequences) through practices so we can understand those dynamics and perhaps help civilizations be more healthy.
Does that sound about right?
Because it also sounds like some kind of critical theory...
Your argument appears to be circular. He said he believes critical theory to have done social harm. You observe that he says there are ideas present in society that cause social harm. Then you say this means we need critical theory to examine such damage(as OP said is the result of critical theory).
That's actually not what I said, unless any socially critical examination does in fact meet the definition of "critical theory"...
And in fact, no one here knows what the commentor means by that phrase, much less how it's ostensibly done damage, which would be necessary in order to change their mind as invited.
Psych is an "easy A" batchelors program for people who don't know what they want to study. It's a useless degree without moving on to the severely placement limited graduate programs.
My personal take is that humanities departments should be aggressively courting 2nd majors (somewhat egotistical, I was a double major). As much as we might want to believe that the purpose of education is to give person a fuller life, increased earnings potential is an undeniable part of there reason people go to university. Selling a line like “the CS department will teach you how to make money. We’ll help you live the rest of your life” or “engineering got me in on the ground floor. Clarity of thought and speech got me to the top floor” seems like it could open up more demand than trying to go it alone.
My N of one experience suggests it might be effective. I thought I was just getting a vanity politics degree on top of my CS degree because I liked it. After 10 years in career I think I refer to lessons and skills learned in those humanities classes as much as the ones in my CS classes.
I would figure that the whole point of getting a psychology degree (as opposed to taking some psychology courses as part of your main degree) is to work as a psychologist? Why are all these people failing to get their psychologist qualifications?
Where I live, I've never met or heard of anyone who went for a psychology degree without the intention of becoming a psychologist. Sure, not everybody makes it to the end, and few end up doing research instead, but the whole goal of the degree is to produce qualified psychologists.
Because in some countries there are different branches of psychology studies. Clinical psychology that will earn you a license to practice and psychology which is really just I’m interested in psychology.
Where I live, there just aren't enough spots in psychologist training programs. There are 8 spots for clinical psychology grad programs where I graduated from with 1200-1300 psych undergrads. Psych enrollment increased by 20% in the last 5 years and the clinical psych program's training capacity stayed flat. Of those 8 spots, between 1 and 6 spots are filled by people from outside of the province including international students.
(I'm oversimplifying a little. There are 6 accredited programs, but there are 10 undergrad universities that produce psych majors, so the numbers probably look worse overall.)
You're right that the goal is to become a qualified psychologist, but the training pipeline doesn't match that goal. Crudely, there are training spots for less than 1% of psych undergrads (maybe 2% if you include research psychologists). There are alternative programs for counseling, but it's not the same license. The cost of these counseling programs is also astronomical for Canada. I know people who went that route and have $150k loans to pay back. If you're competitive enough to get into one of the clinical programs, they are funded (with students given stipends ranging from $30-$40k/year), but even if they aren't tuition is under $7k/year.
I want none of this. I want a whole bunch of apprenticeships and tradesmen from our colleges. Technicians and specialists. Subject matter experts. Manufacturing engineers.
I have a math undergrad. I imagine someone who values physics & econ wouldn't think a math degree is "societally unproductive", but the decided majority of people I meet including other college graduates have absolutely no idea what I studied/learned in the course of doing that.
What reason do you have to believe that you aren't doing the same thing with the majors you're judging societally unproductive?
Are we really going to rehash the same, tired argument that only STEM majors are useful to society? Having the narrow view of social benefits being solely centered around money is incredibly short-sighted, and I’m not even sure I want to respond to physics/Econ majors being “smart”.
Dumb, or courageous? Do you not benefit from the people around you who studied liberal arts? Who is going to make the Netflix shows you watch after you get home from work after spending 8 hours slinging JSON around and making bank? No, seriously, these people are going into fields that we benefit from and yet refuse to pay for, and if they didn’t our society would be worse off. Why do you consider this a dumb decision rather than a sacrifice they shouldn’t have to make?
I was going to post something similar but parent got flagged and I couldn't.
Gender studies is important. I have to think about non-gender issues for a living and this means that I don't have the time to think about the thousand ways in which women are disadvantaged in society.[1] But gender studies majors do, and I am better off for reading the arguments written or summarized by them than not learning about many of these issues at all.
I doubt the people making shows for netflix are not being paid well. Even the ones pouring out absolute junk lately are being paid bank. a Better example would be english teachers who absolutely are being underpaid while providing value to society.
gender studies is still bullshit. I took a few mandated classes back in university. 90% of getting an A is just knowing what the teacher wants to hear. I have a hard time believing any of them can understand critical thinking giving how much of their doing well in that major is just going with groupthink.
> Are we really going to rehash the same, tired argument that only STEM majors are useful to society?
It seems so, since it's hard for people to admit that an average STEM graduate is much more useful than an average humanities graduate. It's not that the humanities aren't useful, they are, but to succeed and actually have an impact on society, you need to be one of the best humanities graduates.
We don't need more psych bachelors grads who took on 10-20x their future annual income in student loans, only to file paperwork in an office for the rest of their lives. There are much better ways for them and for society to spend that money than subsidizing/bailing them out.
> It's not that the humanities aren't useful, they are, but to succeed and actually have an impact on society, you need to be one of the best humanities graduates.
Your average STEM grad is not having the impact on society like what you describe here.
We also don’t need more CS bachelor grads churning out CRUD apps for adtech companies.
It doesn’t really matter because most of those psych grads aren’t and would never be capable of administering mental healthcare beyond what any of us can already find on Google.
> societally unproductive majors (gender studies, literature, etc.). I don’t see why we should use tax money to prop up anything other than proven productive fields of study.
Some majors are actually unproductive and a waste of both the student's and taxpayer's money, and should not be subsidized.
But many humanities that often get binned this way are actually civically productive, and they're only economically unproductive because of significant market failure. The benefits to society aren't priced properly. Such majors, and that is the social sciences, should arguably be subsidized by the government, because they are technically public goods.
What the last few years has taught us is how much damage an electorate can do who isn't educated. If they don't understand anything about economic systems, political systems, sociology, history, psychology, then their mind can be taken over too easily by propaganda, which can have catastrophic consequences, all the way up to the death of democracy.
The only problem that's cropped up is the decline of the education system due to continued undervaluing of our countries bedrock of teachers. With an increase of business majors during a continuing recession.
I think it’s more of a matching problem. There are lots of students who get nothing out of CS classes. Ideally we would match the available slots with the students with the highest analytical aptitude.
University of Washington does a full on application with nearly all engineering majors, including Computer Science and Engineering, which is weirdly split between college engineering and college of arts and sciences. It is awful for everyone who experienced it. Hopefully it is different now, it really limited peoples choices.
Tangent rant - Some time ago I was rejected 4 times from CSE, ended up getting admitted to EE. Which hilariously, now has changed its name to… ECE, electrical and computer engineering. They offer an different Computer Engineering degree now!!! The buildings are literally linked floor by floor and they won’t work together, haha. Got a great education in computer engineering even without the name.
It's not. HS students are direct admitted to the CoE but not their major. Allen School is a separate path and the only reasonable path to the Allen School CS is a direct admit from HS that is incredibly competitive. We are OOS, so he would have had an easier time getting into MIT than into Allen as a HS applicant. I think they took 4 OOS/Int'l students out of 400 total admits.
After you accept that fact, 96% of rising sophomores in the College of Engineering got their first or second choice of major and only 4 majors had to reject anyone who wanted it as their first choice. If you took your first year seriously, you are almost certainly exactly where you hoped to be when you first applied coming out of HS.
My son just got accepted into the new ECE program. He plans to follow a path that gets him closest to the Allen school degree while adding in some additional hardware knowledge that those students don't necessarily get.
That is sad to hear about the Allen school, but it sounds like the CoE has figured out a good way forward.
Goodluck to your son though with the new ECE program! It was a great program even before the name change with a lot of very good professors and lecturers. I especially recommend any class by Scott Hauck.
Thanks. I'll let him know. He's had a great experience one year in and I wouldn't hesitate to recommend UW to anyone considering it but understand that you basically get to apply twice for any desirable major. Once to get into the school and again for whatever major you want if it is capacity constrained. Ironically, my cousin's daughter opted for Berkeley over UW this fall. She's on a non-STEM track though.
I graduated Purdue in 04, and then ECE was still basically EE with a few computing related electives.
A CS major couldn't take any engineering classes; a ECE student might be allowed to take some CS classes, but there weren't any in mine, and realistically they were probably too busy anyways.
This was a long time coming. I graduated in 2021, and while the UC Berkeley computer science program is still absolutely stellar, it was starting to collapse under its own weight. While it scaled admirably, there are limits. I TA’ed operating systems (CS162) my last two semesters, and the office hours queue would fill up immediately upon starting, and due to an ever-shrinking number of course staff due to budget limitations, we were never able to get to everyone who wanted help. We’re talking several hundred students with less than a dozen TAs. This situation was not uncommon across other courses. The absolute smallest non-graduate-level CS course I took had around 100-150 students.
Its unfortunate, but almost inevitable. Eric Roberts wrote about CS' latest rise 6 years ago [1]. Looking at its prior "peaks", its always come down to a resource issue - there's simply not enough space for everyone. As a consequence, schools started to do several things: A) Limit enrollment to the better prepared; B) Increase retraining for other STEM instructors/teachers to teach CS; or C) increase enrollment to unmanageable standards (my Intro to AI course has grown in the past 3 years from 45, to 80, to 120 next spring).
Each has their own downfalls - higher enrollment requirements limit lower socioeconomic students, retraining programs cost money and can produce subpar instructors (2 month retraining PD vs a 2 year), and increased teaching loads make becoming an instructor unattractive. It doesn't help that CS PhDs are highly sought after by large tech corporations, pay MUCH more, and with less work (maybe).
We haven't really solved how to teach CS, problem solving is still a difficult thing for humans to learn and we don't really have a better way to teach it. Even if we include all the current "soft skills", at the end of the day problem solving is the trait many companies want.
If history repeats itself, the CS bubble will burst. Not necessarily plummeting, but new students seeking STEM careers will pick programs without as many barriers of entry or low quality instruction. I don't really have a solution to the problem, just acknowledging that the issues the domain will face as enrollment continues to increase.
Because it is hard to judge who will make a 'good' computer scientist based solely on SATs and high school grades. Letting everybody take a real computer science course and exam and then judging future potential based on those results should give a much stronger signal
When I was there 20+ years ago, the intro course (CS61A) was definitely known to knock the unprepared/uncommitted out of the running. Back then it was taught with Scheme out of SICP.
When I went (early dotcom to not-berkeley) they would milk the poor saps for a couple of years knowing they would never make the cut.
I think weeding earlier would be kinder.
The previous CS major declaration standard was roughly the median grade (B+) over three intro courses. The intro courses have standard "grade bins" to prevent sections from turning into the Hunger Games. In recent years, far more students were getting into L&S CS than into EECS.
This especially sucks for CS since a nontrivial portion of incoming CS students have programming experience but a large number have none. Using a 101 course as a weed out has the effect of stratifying these two groups. As one would expect, men are disproportionately likely to have prior programming experience so this process contributes to gender imbalances. CMU found that reorienting its curriculum to not place so much emphasis on prior programming experience in intro classes had a significant effect in reducing the number of women who tried CS and then left.
As a mid-90's graduate who started in chemistry and accidentally found my way to computer science... Oof. The promise of a well rounded education where you can explore and find out for yourself what you enjoy and where you want to study no longer exists.
I took CS classes because I wanted something fun to do outside of what I was officially studying. About the time I finished up, UCB was already limiting upper division classes to CS majors only. And this was thirty years ago.
I ended up in my career because I married the thing I thought I wanted to do (chem) with the thing I really enjoyed and was good at (programming). I met a lot of grad students who wanted help programming and I got to do a lot of cool work with different research groups in chem/phys/bio that did protein folding, data visualization, etc. I ended up getting a job at Industrial Light and Magic because I had a lot of experience programming OpenGL on Irix(SGI).
Where is the room for that kind of experience now? You need to test into the major before you even enter campus? I guess the college experience is getting a piece of paper stating that you can jump through a set of hoops.
This is disappointing. I've been fortunate enough to work in an industry that combines art and tech and know the benefits of working with people who come from a variety of backgrounds and expertise. I want to live and work in a world where you don't need to match an algorithm to have a chance. And unfortunately, many people conflate a university degree with that chance.
> You need to test into the major before you even enter campus?
That's how it works in Germany and I believe a majority of other European countries as well. You don't apply to a given school, you apply to get into a specific program at a given school.
Works like that in Sweden as well. However once you are in, there is nothing stopping you from applying to courses outside of your program. For example I was in the mathematics program, but took 'extracurricular' courses at the philosophy, economics, computer science and electrical engineering departments. It is also possible to apply to change program once you are in.
>The promise of a well rounded education where you can explore and find out for yourself what you enjoy and where you want to study no longer exists.
Ideally, you want to figure out what you want to do with your life /before/ you saddle yourself with debt going into college (assuming your dreams take you that way).
If the tides are turning such that you need to have a decent idea about your life's prospects before entering university, I say that is a very good thing.
Expecting people to confidently pick their life course by 18 is pretty silly. The ability to flex and explore as you grow into adulthood in college is very valuable. And you can do it while learning very useful skills and topics (as the above commenter did).
The time to "grow into adulthood" is during your school years. If you don't have a decently clear idea of where you want to go in life, it's probably not a good idea to take on a 5- or even 6-digit debt.
Once you figure out what your life goal is, /then/ see about going to college if that's on the way to achieving your goal.
The widespread notion that everyone must go to college, including for no good reason, is nonsense. College isn't the end all be all to life.
I went to berkeley in the 90's. Tuition was ~$2k (usd) per semester. I lived in a shared house for $300/month. I paid for everything out of pocket by working for a campus job (which paid $8/hour vs $5/hour minimum wage). So I was lucky and walked away with no debt.
I feel for the current generation that will need to take on debt if they feel they need a college education, and I know the amount I've been saving for my kids.
I can't change anything about the monetary policy that allowed student loans to be so large. But I do try my best to do outreach and recruiting from candidates with non-standard backgrounds and who didn't come from one of the big CS schools.
could be that freedom still exists, just not in the hyper competitive realms. I'm starting to contemplate exits when things get too competitive--that it may be time to shift focus to some other area that gets more return for hours poured into something. I'm not entirely sure about that however.
Initially I felt similar to you, but then I considered the potential impact on accumulating student loan debt both from the perspective of paying for classes that don’t contribute to a major and rising tuition costs in the name of giving every student a reasonable chance to explore all available classes (not saying either is true for UCB).
Also I feel like it creates a new opportunity for some other entities to help students “test their major” without incurring the same costs.
Things are different than the mid 90s. There are plenty of schools teaching computer science well. I think a school has a responsibility to its current/incoming students to give a quality education, and the UC system does that at a greater scale than any other top universities.
I graduated this past May, and let me tell you that TAs are struggling, pulling 40-60 hour weeks of pure TA work plus whatever classes they need to be taking to handle the sheer scale of these CS classes (800-2000 students). 15% of the undergraduate population (32,000 students) is graduating CS or EECS. You can't just keep scaling up class sizes and undergraduate populations forever, and the fact that Cal is getting this criticism for doing it at a level far above any other top CS university is surprising honestly.
> You need to test into the major before you even enter campus?
As an EECS major of roughly the same period...yes? The Cal engineering majors have been direct admit for ages, precisely because demand is always higher than capacity. In our day, the CS path was neglected because most incoming CS kids wanted the EECS guarantee. Once demand rose and the L&S admission advantage obviously outweighed the value of the EECS major guarantee, it was all over. The "backdoor" IEOR major (ORMS) has the same issue and is also going "high demand."
That said, the last plan-of-record I have seen involves a "discovery" path that will take a few undeclared L&S majors each year. So today's you would still have a chance. But you would be ineligible for the "discovery" path if you had declared "intended CS" during admission (i.e., were actually rejected from the direct admit path).
* High school GPA of 2.5, so obviously university was out of the question :)
* Enroll into local Junior college
* Couldn't balance jr. college classes with no car, full time work (+ freelancing when primary job didn't cover rent) -- also didn't make sense to pay for material I already knew, and the CS department wouldn't let me test out [1].
* Already knew programming & CS fundamentals from pure curiosity
* Enrolled into Hatchways :) [2]
* Now work as full time SWE
[1] From my understanding, California actually made required prerequisites illegal (limited to Math and English only), not for bogus CS prerequisites which is a shame. Prerequisites are courses that are required but do not count towards major/degree/program satisfaction.
How times have changed, in 2006 when I enrolled into Undergrad CS everyone was saying all coding jobs would be offshored to Bangalore. All the cool kids went to Finance or Quant.
It's really nuts, computer programming has this gigantic language translation component to it, so asking somebody to translate from English, to another human language, then to a computer language, is fraught full of peril and basically fails most of the time.
The only teams that can pull it off have people who are really talented in the company's native language, and computer programming, and communicating with non-native language people. Those people are super rare and are not paid nearly enough to encourage more people to seek out that niche.
Yeah, but then 2008 happened, finance got sketchy, FAANG CS jobs started to pay more than quant, and all the salary-minded undergrads realized they wanted absolutely nothing more than to write software for a living.
How will the UC Berkeley administration prevent its head from exploding from cognitive dissonance, if they start requiring people to successfully complete lower courses to get into the program? Think of all the injustice and inequities this will produce. The privileged just get more privileged.
123 comments
[ 3.8 ms ] story [ 294 ms ] threadhttps://www.dailycal.org/2022/04/08/a-personal-view-of-uc-be...
Ten years ago, the department graduated roughly 400 students with Bachelor’s degrees in EECS or computer science. This year, the number will probably be 1400, or roughly 15%, of all undergraduates. Almost all of that growth is in the College of Letters and Science CS major.
and more specifically:
The campus provides departments $50 per student-credit per hour to pay for both lecturers and TAs. A student in a 4-unit class contributes roughly $200 out of the student’s overall tuition for these expenses. Yet the actual cost to a department is more than $350 per student for that 4-unit class, forcing departments to find additional funds to cover the upward of $150 per student gap. [...] The EECS department is running an annual deficit of nearly $5 million.
At a high level, this problem is "growing pains" rather than "falling backwards". Computer science is booming at Berkeley, and they can't hire enough TAs to keep up when they are paying so badly. I hope they simply start paying CS TAs better.
Impossible. Students definitely don't pay more than $50 per credit hour in tuition, so there's nowhere for the university to get the additional funds from.
Edit: This was sarcastic
Tuition is $14,000 for in-state and and a laughable $44,000 for out-of-state. And they can’t find an extra $150 per student? Fire one worthless administrator and the $150 gap is completely covered. Voila!
And a personal insight, as a former uGSI (TA) in the department, the problem isn’t paying TAs enough. The pay is actually quite high for a student job (especially when you factor in the fact that in-state tuition is paid for). But the course budgets simply aren’t high enough to hire the sheer quantity of TAs that would be required to handle all of the students who want to take the classes.
I was involved in hiring some semesters, and there were always more people we wanted to hire than the budget would allow. I believe the situation has only gotten worse since I left.
You are also correct that the author is a former lecturer. Too bad it's too late for me to edit the comment.
A 20 hour TA is paid ~$2250/mo for approximately 4.5 months per semester, plus ~$7000/semester in tuition remission (so ~$17k). $350 is the total cost for the whole semester per student. 17k/350 is about 50 students per TA, which sounds about right (and even that is a higher than ideal ratio). There are other costs too, but these are the some of the highest.
(u)GSIs at Berkeley are paid particularly well compared to other schools, I think in large part due to the TA union (UAW 2865).
For tuition remission, it’s only the in-state amount, even if you’re out of state.
Some significant costs come from the massive scales the classes are taught at. This semester, the first required CS lower division class has 1800 students (https://classes.berkeley.edu/content/2022-fall-compsci-61a-0...) and an upper division DBs class has 600 students. (https://classes.berkeley.edu/content/2022-fall-compsci-186-0...).
It takes a lot of TAs to handle student support, which doesn't scale well in practice. The class I teach publishes lots of videos and documentation about how to do the labs, projects, and homeworks, and we also have a forum where students can (to a reasonable extent) help each other out with common questions. However, the classes teach a lot of stuff really fast and you need a lot of TAs doing things like office hours to help students who didn't understand the speedy lecture, or have a particularly nasty time trying to debug an error.
One particularly difficult thing about this scaling is that students typically need TA help all at once. Office hours right before midterms and project deadlines have many students and more bodies are needed just to have good handling of "peak" times
Another aspect of TA cost is that TAs are paid significant amounts of tuition remission for their work, and the department has somewhat of a mandate to ensure any graduate student who wants to TA can get full tuition remission from a 10 hour position or higher (I'm guessing this is somewhat of a symptom of skyrocketing tuition and inadequate aid being foisted on the department to deal with). The department used to get around this by hiring tons of undergrads at an 8 hour level so they wouldn't get remission, but they were sued and paid out millions in backpay as well as paying all TAs a scaled remission based on their hour count. (https://www.insidehighered.com/news/2020/01/16/arbitrator-sa...). Generally an 8-hour TA gets paid around $40 an hour after remission and a 20-hour TA gets paid around $50 an hour. [very rough math read the other comment for harder numbers]
-Am a TA, speaking for myself
Maybe that’s part of the problem.
https://www.ucop.edu/operating-budget/_files/rbudget/2022-23...
The overall budget gets weird with medical centers and something called “Sales and Support”, but in any case California tax payers put in more money then students. I don’t think that’s a problem, but it’s definitely subsidized.
But for subjects like Civil, Electrical, sciences like Physics, Biology, I suspect the story is quite different.
I can speak to Biology; most undergrads are left with little opportunity as most roles require a graduate degree to do anything meaningful when the economy is actually favorable, and often seek roles outside of their field when it's not. As a undergrad that was lucky enough to enter the field in the wake of the GFC long enough to just pay off my student debt I've met so many biologists/biochemists who either couldn't find anything in their field, or simply gave up trying and sought the highest paid and somewhat stable role they could get 'that required a degree.' It's quite clear that the system has failed, and the system is designed to onboard more STEM grads as it sounds as their are ever-growing holders of degrees chasing so n amount of open roles.
Tech in unique in that it is one of those few Industries where the barrier of entry is rather low, but the attrition and competence level vets for viable candidates almost entirely on it's own--the gatekeeping interview process is still an unnecessary hurdle.
Personally, for all the controversy Eric Weinstein has made directly and indirectly of his own actions, the concept of Embedded Growth Obligations was incredibly well researched and consistent with what we have seen as outlined since 2005 and then solidified itself in the financial collapse of 2008 [0]:
> Embedded Growth Obligations are the way in which institutions plan their future predicated on legacies of growth. And since the period between the end of World War II in 1945 and the early 70s had such an unusually beautiful growth regime, many of our institutions became predicated upon low-variance technology-led, stable, broadly distributed growth. Now, this is a world we have not seen in an organic way since the early 1970s, and yet, because it was embedded in our institutions, what we have is a world in which the expectation is still present in the form of an embedded growth obligation. That is, the pension plans, the corporate ladders, are all still built very much around a world that has long since vanished.
We have effectively become a Growth Cargo Cult. That is, once upon a time, planes used to land in the Pacific, let's say, during World War II, and Indigenous people looked at the air strips and the behavior of the air traffic controllers, and they've been mimicking those behaviors in the years since as ritual, but the planes no longer land. Well, in large measure, our institutions are built for a world in which growth doesn't happen in the same way anymore.
0: https://theportal.wiki/wiki/Embedded_Growth_Obligations
(Incidentally, I think CS is a very good place to invest, precisely because there’s a lot of demand, start up cost for self employment is low, and applicability is super broad)
Many people would argue that linguistics is one of those useless studies compared to comp sci, but it shares deep connections to research in logic and by extension type theory (see handbook for logic and languages)
unless youn disagree with the laws of supply and demand, it seems the market consensus at this moment is that compsci majors are far more useful than any fundamental research (which still remains undefined).
Even if the market was wrong and fundamental research could somehow be valuable, why would you not trust the primary beneficiaries oj making that decision, vs some bureaucrat that has no stake in the outcome?
A great video about open-endedness (warning - very long) with Kenneth Stanley, author of the book "Why Greatness Cannot be Planned"
https://www.youtube.com/watch?v=lhYGXYeMq_E&t=350s
Would it perhaps be prudent to cap the number of participants in those departments to match the number of real-world vacancies available?
-https://en.m.wikipedia.org/wiki/Frederick_Jelinek
According to 'The Master Plan', UCs are designed for research and post-grad work and some professional careers. Only the UCs give out PhDs (with a few exceptions). CSUs are designed for professional and teacher education and cannot give out PhDs (with a few exceptions). So, as an undergrad thinking of grad school, a UC is a better place to go to as they have the grad labs and classes on campus, while the CSUs mostly do not. Your chances of getting into grad school are a bit higher.
Berkeley, being a UC, is more geared toward research than Chico, a CSU. I mean, Cal even has a nobel prize parking lot.
https://www.ucop.edu/institutional-research-academic-plannin...
What a sad way to categorize people.
You seem to be saying there is a body of ideas that has gained such a widespread foothold that they hold a significant place in the culture such that they can unfavorably affect social dynamics at civilization-wide levels.
If this is true, it would imply that we should be carefully -- some might even say critically -- be examining civilization at all levels from ideas (and their implications and consequences) through practices so we can understand those dynamics and perhaps help civilizations be more healthy.
Does that sound about right?
Because it also sounds like some kind of critical theory...
And in fact, no one here knows what the commentor means by that phrase, much less how it's ostensibly done damage, which would be necessary in order to change their mind as invited.
What's the "theory" exactly? It just sounds like common sense.
https://twitter.com/benmschmidt/status/1562212497272279041
(Weird that psychology is growing, all the psych degree people I know work in HR and I don't think that's what they dreamed of.)
https://www.hamiltonproject.org/charts/interactive_embed/car...
My personal take is that humanities departments should be aggressively courting 2nd majors (somewhat egotistical, I was a double major). As much as we might want to believe that the purpose of education is to give person a fuller life, increased earnings potential is an undeniable part of there reason people go to university. Selling a line like “the CS department will teach you how to make money. We’ll help you live the rest of your life” or “engineering got me in on the ground floor. Clarity of thought and speech got me to the top floor” seems like it could open up more demand than trying to go it alone.
My N of one experience suggests it might be effective. I thought I was just getting a vanity politics degree on top of my CS degree because I liked it. After 10 years in career I think I refer to lessons and skills learned in those humanities classes as much as the ones in my CS classes.
I would figure that the whole point of getting a psychology degree (as opposed to taking some psychology courses as part of your main degree) is to work as a psychologist? Why are all these people failing to get their psychologist qualifications?
Where I live, I've never met or heard of anyone who went for a psychology degree without the intention of becoming a psychologist. Sure, not everybody makes it to the end, and few end up doing research instead, but the whole goal of the degree is to produce qualified psychologists.
(I'm oversimplifying a little. There are 6 accredited programs, but there are 10 undergrad universities that produce psych majors, so the numbers probably look worse overall.)
You're right that the goal is to become a qualified psychologist, but the training pipeline doesn't match that goal. Crudely, there are training spots for less than 1% of psych undergrads (maybe 2% if you include research psychologists). There are alternative programs for counseling, but it's not the same license. The cost of these counseling programs is also astronomical for Canada. I know people who went that route and have $150k loans to pay back. If you're competitive enough to get into one of the clinical programs, they are funded (with students given stipends ranging from $30-$40k/year), but even if they aren't tuition is under $7k/year.
What reason do you have to believe that you aren't doing the same thing with the majors you're judging societally unproductive?
Gender studies is important. I have to think about non-gender issues for a living and this means that I don't have the time to think about the thousand ways in which women are disadvantaged in society.[1] But gender studies majors do, and I am better off for reading the arguments written or summarized by them than not learning about many of these issues at all.
[1] And possibly men too. Not gender studies but an example: https://home.uchicago.edu/~bursztyn/Misperceived_Norms_2020_...
gender studies is still bullshit. I took a few mandated classes back in university. 90% of getting an A is just knowing what the teacher wants to hear. I have a hard time believing any of them can understand critical thinking giving how much of their doing well in that major is just going with groupthink.
It seems so, since it's hard for people to admit that an average STEM graduate is much more useful than an average humanities graduate. It's not that the humanities aren't useful, they are, but to succeed and actually have an impact on society, you need to be one of the best humanities graduates.
We don't need more psych bachelors grads who took on 10-20x their future annual income in student loans, only to file paperwork in an office for the rest of their lives. There are much better ways for them and for society to spend that money than subsidizing/bailing them out.
Your average STEM grad is not having the impact on society like what you describe here.
We also don’t need more CS bachelor grads churning out CRUD apps for adtech companies.
Some majors are actually unproductive and a waste of both the student's and taxpayer's money, and should not be subsidized.
But many humanities that often get binned this way are actually civically productive, and they're only economically unproductive because of significant market failure. The benefits to society aren't priced properly. Such majors, and that is the social sciences, should arguably be subsidized by the government, because they are technically public goods.
What the last few years has taught us is how much damage an electorate can do who isn't educated. If they don't understand anything about economic systems, political systems, sociology, history, psychology, then their mind can be taken over too easily by propaganda, which can have catastrophic consequences, all the way up to the death of democracy.
Who probably proceeds to go into zero or negative-sum trading, or negative sum ad targeting intelligence.
https://www.npr.org/sections/money/2014/05/09/310114739/what...
Tangent rant - Some time ago I was rejected 4 times from CSE, ended up getting admitted to EE. Which hilariously, now has changed its name to… ECE, electrical and computer engineering. They offer an different Computer Engineering degree now!!! The buildings are literally linked floor by floor and they won’t work together, haha. Got a great education in computer engineering even without the name.
After you accept that fact, 96% of rising sophomores in the College of Engineering got their first or second choice of major and only 4 majors had to reject anyone who wanted it as their first choice. If you took your first year seriously, you are almost certainly exactly where you hoped to be when you first applied coming out of HS.
My son just got accepted into the new ECE program. He plans to follow a path that gets him closest to the Allen school degree while adding in some additional hardware knowledge that those students don't necessarily get.
Goodluck to your son though with the new ECE program! It was a great program even before the name change with a lot of very good professors and lecturers. I especially recommend any class by Scott Hauck.
A CS major couldn't take any engineering classes; a ECE student might be allowed to take some CS classes, but there weren't any in mine, and realistically they were probably too busy anyways.
Each has their own downfalls - higher enrollment requirements limit lower socioeconomic students, retraining programs cost money and can produce subpar instructors (2 month retraining PD vs a 2 year), and increased teaching loads make becoming an instructor unattractive. It doesn't help that CS PhDs are highly sought after by large tech corporations, pay MUCH more, and with less work (maybe).
We haven't really solved how to teach CS, problem solving is still a difficult thing for humans to learn and we don't really have a better way to teach it. Even if we include all the current "soft skills", at the end of the day problem solving is the trait many companies want.
If history repeats itself, the CS bubble will burst. Not necessarily plummeting, but new students seeking STEM careers will pick programs without as many barriers of entry or low quality instruction. I don't really have a solution to the problem, just acknowledging that the issues the domain will face as enrollment continues to increase.
[1] https://cs.stanford.edu/people/eroberts/CSCapacity/
So they were clearly not optimizing for weed-out.
I took CS classes because I wanted something fun to do outside of what I was officially studying. About the time I finished up, UCB was already limiting upper division classes to CS majors only. And this was thirty years ago.
I ended up in my career because I married the thing I thought I wanted to do (chem) with the thing I really enjoyed and was good at (programming). I met a lot of grad students who wanted help programming and I got to do a lot of cool work with different research groups in chem/phys/bio that did protein folding, data visualization, etc. I ended up getting a job at Industrial Light and Magic because I had a lot of experience programming OpenGL on Irix(SGI).
Where is the room for that kind of experience now? You need to test into the major before you even enter campus? I guess the college experience is getting a piece of paper stating that you can jump through a set of hoops.
This is disappointing. I've been fortunate enough to work in an industry that combines art and tech and know the benefits of working with people who come from a variety of backgrounds and expertise. I want to live and work in a world where you don't need to match an algorithm to have a chance. And unfortunately, many people conflate a university degree with that chance.
That's how it works in Germany and I believe a majority of other European countries as well. You don't apply to a given school, you apply to get into a specific program at a given school.
On other hand even to get into university you probably need to know where you are generally aiming at beginning of High School.
Ideally, you want to figure out what you want to do with your life /before/ you saddle yourself with debt going into college (assuming your dreams take you that way).
If the tides are turning such that you need to have a decent idea about your life's prospects before entering university, I say that is a very good thing.
Once you figure out what your life goal is, /then/ see about going to college if that's on the way to achieving your goal.
The widespread notion that everyone must go to college, including for no good reason, is nonsense. College isn't the end all be all to life.
I went to berkeley in the 90's. Tuition was ~$2k (usd) per semester. I lived in a shared house for $300/month. I paid for everything out of pocket by working for a campus job (which paid $8/hour vs $5/hour minimum wage). So I was lucky and walked away with no debt.
I feel for the current generation that will need to take on debt if they feel they need a college education, and I know the amount I've been saving for my kids.
I can't change anything about the monetary policy that allowed student loans to be so large. But I do try my best to do outreach and recruiting from candidates with non-standard backgrounds and who didn't come from one of the big CS schools.
This is how it works in most of the world. There is no major/minor system, you just study one thing at university for ~3 years.
Also I feel like it creates a new opportunity for some other entities to help students “test their major” without incurring the same costs.
I graduated this past May, and let me tell you that TAs are struggling, pulling 40-60 hour weeks of pure TA work plus whatever classes they need to be taking to handle the sheer scale of these CS classes (800-2000 students). 15% of the undergraduate population (32,000 students) is graduating CS or EECS. You can't just keep scaling up class sizes and undergraduate populations forever, and the fact that Cal is getting this criticism for doing it at a level far above any other top CS university is surprising honestly.
As an EECS major of roughly the same period...yes? The Cal engineering majors have been direct admit for ages, precisely because demand is always higher than capacity. In our day, the CS path was neglected because most incoming CS kids wanted the EECS guarantee. Once demand rose and the L&S admission advantage obviously outweighed the value of the EECS major guarantee, it was all over. The "backdoor" IEOR major (ORMS) has the same issue and is also going "high demand."
That said, the last plan-of-record I have seen involves a "discovery" path that will take a few undeclared L&S majors each year. So today's you would still have a chance. But you would be ineligible for the "discovery" path if you had declared "intended CS" during admission (i.e., were actually rejected from the direct admit path).
* High school GPA of 2.5, so obviously university was out of the question :)
* Enroll into local Junior college
* Couldn't balance jr. college classes with no car, full time work (+ freelancing when primary job didn't cover rent) -- also didn't make sense to pay for material I already knew, and the CS department wouldn't let me test out [1].
* Already knew programming & CS fundamentals from pure curiosity
* Enrolled into Hatchways :) [2]
* Now work as full time SWE
[1] From my understanding, California actually made required prerequisites illegal (limited to Math and English only), not for bogus CS prerequisites which is a shame. Prerequisites are courses that are required but do not count towards major/degree/program satisfaction.
[2] https://www.hatchways.io/
How come they can't afford to put students in a lecture room and pay TAs a few hours of work?
The only teams that can pull it off have people who are really talented in the company's native language, and computer programming, and communicating with non-native language people. Those people are super rare and are not paid nearly enough to encourage more people to seek out that niche.