Caltech Awards 10,000th PhD Degree | Hacker News Reader

Caltech Awards 10,000th PhD Degree

caltech.edu

220 points0xedb135 comments

I wound up at Caltech by pure luck, finding an institute particularly well suited to my personality and what I needed. For one example, Caltech has an honor system where the default is for the professors and students to be collaborators and trust each other rather than be adversaries.

Thanks, Caltech!

> the default is for the professors and students to be collaborators and trust each other rather than be adversaries.

Can you explain why professors and students would be adversaries?

I knew a professor who was sued by a student over her grade on an exam. Going through that made him keep the originals of each student exam and only return photocopies so that he would have your original work if it ever came to that in court.

Toxic event breeding more toxicity.

How very sad.

I had a professor brag about a legal battle where he got GitHub to take down a student's submission for the class project (which was of course the same every year) because it contained his "copyrighted material" (read: a mediocre, incomplete MIPS emulator and hacked-together toolchain). Some professors are just not very nice people about these things.

> For one example, Caltech has an honor system where the default is for the professors and students to be collaborators and trust each other rather than be adversaries.

What does this mean in practice? Students and staff collaborate in every institution I've ever been part of.

Maybe professors leave the room during exams. I think a number of universities do this. Outright "adversaries" sounds insane and toxic though. I don't know how an undergrad institution could even function if students and professors were actively trying to undermine each other.

Thirty years ago, the exams were all take-home.

At Caltech, they were normally take-home. Institute policy was they were not proctored. The exams had a time limit from opening the exam to pencils down. The conditions, like open-note or open-book, were specified.

The students were totally on their own with this and nobody would know if they chose to cheat or not. Cheating was so trivial to do there was no cred for bragging about it. If other students found out someone was cheating, that someone would be ostracized by the students, which was much worse than anything the institute could do to them.

Of course, people are people and I'm sure some cheating went on. But I bet it was at a much lower rate than at places with proctored exams.

As for me, it was the first time in my life adults trusted me and treated me like an adult rather than a child. I much prefer that to the extended childhood offered by other universities. I was motivated by simple pride to earn the degree rather than cheat and get a tainted one.

I graduated from Caltech last year, and the exams are all still take-home. You're expected to time yourself and not consult course materials, and basically nobody cheats. It's a rare sort of equilibrium where nobody's cheating, so it's very hard to start -- you won't find any co-conspirators, and if you're found out it's social death.

Incredible! What an environment that must be! I hope my kids will find themselves in a university like that.

I'm a bit saddened that other universities have not attempted to emulate this. Maybe it only works because Caltech is such a small place, small enough that you get to know most of the people there, and the social interconnectedness is strong.

10 years after Caltech, I went to law school. I was surprised to find that they did not have take home exams. Law is a profession that actually has ethical requirements, but apparently we can't trust law students to behave ethically?

This annoyed me not only because of the lack of trust, but because law school exam questions are almost always essay questions that take a few pages to answer. No way did I want to try handwriting that much.

Fortunately, they did allow you to at least bring a typewriter to the exam (they set aside a separate room for this so it would not disturb the people who were handwriting).

There were limits, though. The typewriter could not have more than 2 lines of storage. This was around 1992, and already by then microprocessors and LCDs had become cheap enough that even most entry level typewriters exceeded the limit. It took a fair bit of effort to find one that was acceptable.

All my exams (but one) were take home with well understand rules of engagement. The one exam that wasn't take hime was a visiting professor who made us take the exam at the same time in our normal meeting room and then the left the room for 3 hours. I don't think any of us cared that it was in that particular room, but it was weird to all take the test at 2pm.

One example of an adversarial university environment is how fraternities and sororities keep copies of exams and assignments from prior years. Professors know cheating is rampant, so have to change the questions every semester.

Some courses at Caltech had almost identical exams for at least a decade when I went through. The professors knew cheating like the above simply would not be tolerated by undergrads.

I sat on and helped run the Board of Control, which handled academic honor code violations, for several years and professors who had been at other universities would absolutely rave about how much more they could trust Caltech students. And that was while reporting a suspected cheating case to me.

This honestly sounds ridiculous to me. Every course I ever did we had access to years and years of past papers. Doing past papers is one of the best ways to study. Teaching or learning to the test is a good thing if the test is good.

So why not just provide students with officially sanctioned practice papers that are not past papers and also guaranteed to not share questions with the actual exam?

If you write a new exam paper every year that is also guaranteed not to share questions with the actual exam. From a professor’s point of view many questions that are superficially different are similar enough to practically be the same question. There are many, many questions that ask about what the real cause of the French Revolution was, or test to see if you will recognize that this problem is basically a red black tree.

Some professors would provide past exams for study aids or use them as homework problems.

However, the explicit default was that you should not look at solutions from prior years. Professors would announce at the beginning of the course that they reuse questions and looking at prior years solutions was an honor code violation. I think it's pretty clear it's cheating when the expectations are clearly outlined.

If you had inadvertently come across the problem before and independently solved it, you were expected to disclose that as part of your answer. I personally had to do this several times, and never suffered any negative consequences for it, but the expectation for honesty was there.

In college where I was at last year exams were public knowledge and anyone had access to them. No one seen it as cheating to try last year exam before going on this year.

Yes, it means teachers have to vary tests, but then again it gives you repository of exercises to learn from and to train on. It is just win for learning.

This is an explicit rule introduced at the beginning of most Caltech courses, so the norm is that that behavior is cheating due to being warned in advance.

An adversarial system is where, for example, exams are proctored because the students are not trusted. It's where students actually brag about cheating to their peers.

At Caltech, I never heard anyone brag about cheating. I know of one student who willingly took an F because he accidentally went over the time limit for the take-home exam, and reported it. Nobody would have known otherwise if he hadn't reported it.

For some more context, I've seen many threads on Reddit that typify an adversarial system. The posters would brag about cheating, and would justify it by saying the course material was bullshit and the professors were unreasonable and the degree was just gatekeeping. They'd describe the arms race between students figuring out new ways to cheat and the effort expended by the professors to thwart cheating.

At Caltech, it was effortless to cheat and professors made no attempt to thwart it. (That said, if two exam papers came back with the same mistakes, the students would still have some 'splainin' to do.)

The students all knew each other (being a small university) and the A students were often asked to help out the struggling students (I got a lot of help this way), and so it was pretty obvious to spot an A student who didn't actually master the material. You knew they earned it.

> An adversarial system is where, for example, exams are proctored because the students are not trusted.

Much is made of unproctored exams, but at least 30 years ago (and I haven't heard of this changing since) students were given master keys which opened almost every part of campus, and had official 24/7 access to almost everything. Also, lockpicking was a widely known and broadly accepted student hobby, and there was apparently a fairly well-known heirarchy of which places to which students did not have official access were merely nominally off limits and would result in token punishment for the sake of form (provided no damage to persons, property, etc., was done) and which were really, seriously off-limits.

That is, the culture went beyond just unproctored exams.

> That is, the culture went beyond just unproctored exams.

Indeed. Students also left their dorm rooms unlocked. I never heard of a case of one student stealing from another. A friend of mine at another university had his locked dorm room looted within his first week. Ugh.

We had coed dorms (apartment-style) and always left our door unlocked because a party might arrive. No theft problems.

This reminds me of my time at Virginia Tech (recent grad here). Those of us on upper quad never locked our doors. This had apparently been a thing on upper quad all the way back to the start of the university and that only changed when we moved to the new buildings with electronic locks. I'd say once we moved into the new dorms 75% or more of us were jamming the doors/locks so that they wouldn't lock. This only really started to change after "the adults" started reprimanding people for refusing to lock their doors (There were concerns about ability to compartmentalise/hide in an active shooter situation as well as potential fire safety issues).

It was a really unique and pleasant experience. So much less stressful being able to completely trust the people around you. It's a shame that the tradition has started to die out.

Of course the upper quad is made up of the cadets. That might have something to do with it. I would never leave my door unlocked when I lived on campus. These were the days when Vawter was all-male and before they razed Thomas Hall and got rid of the best breakfast dining hall.

They are no longer given masters. The steam tunnels are patrolled by security and alarmed now.

Security almost certainly doesn't patrol the tunnels with any regularity (they're uncomfortable to be in, after all), especially after the recent switch to contracted security. Maybe they'd patrol the entrances on occasion. The majority of doors are silent-alarmed, but plenty of students know which ones :)

My recollection is that you had to actually have a reasonable justification to be given an official master key, but if you acquired an unofficial master (such as by borrowing someone's official master and copying it) nobody would bother you about it.

When I was there (late 70s, early 80s) the best masters were actually not official masters. The best were Whitehead masters, made by a student named John Whitehead. He hand filed a master with intentional deviations from ideal designed so that when you copied it, the copying errors would make it closer to ideal. Whitehead masters were copies of that hand filed key.

Several people from campus security used Whitehead masters instead of their official masters because the Whitehead ones worked better.

I had an accidental master for a while. My room key happened to function as a master when inserted halfway. I don't remember how I discovered that. I'm assuming that they have long since changed the lock systems, but on the off chance they have not, it was room 21 of Ricketts.

> if you acquired an unofficial master (such as by borrowing someone's official master and copying it) nobody would bother you about it

How do they justify this when the keys may be locking doors that for example protect extremely sensitive personal information? Like for example financial information for support or allegations of misconduct or criminal activity? Don’t they have a legal obligation to protect these things?

Personal observation - assuming everyone is out to get everyone else is a cultural thing that wasnt such a concern until quite recently. It's a bit like looking at certain phrases in common use 70 years ago and wondering why anyone would use them.

You don’t have to assume everyone is out to get everyone, just that someone might possibly be out to get someone.

Also, whatever you assume, there are legal and moral obligations to protect information in any case.

Because once upon a time you could trust and at the same time people didn't feel like that they had that much to hide. In modern society secrecy is the only new 'experience'.

Presumably they multiplied the probability of this occurring with the damage caused and decided that it was lower than the benefit of the high-trust society.

I don’t understand - don’t they have legal obligations to protect? Saying ‘but we trusted’ won’t work in a court.

Sure, but the court won't make them pay unbounded damages.

I don’t know how it is handled now, but when I was there things that needed higher security like that were in rooms that had locks not on the master system.

You had most older buildings on one brand of ordinary locks with a master system, and the newer buildings with better ordinary locks on a different master system, and then a few rooms with Medico high security locks.

> How do they justify this when the keys may be locking doors that for example protect extremely sensitive personal information?

The locks for rooms that had extremely sensitive information, personal or otherwise, did not use the same master as was available to students, and used a different brand of lock that was more resistant to lockpicking. And, perhaps more important in practice than the actual resistance to lockpicking, the brand was widely understood as a signal that picking the lock was not acceptable. Presumably, they also had electronic security.

As sibling commenters have commented, yeah, the higher-security rooms used Medecos instead -- they had not only up * down for the pins but also directional slant (don't really know how to describe this, but pretty and much harder to copy or pick). But also, think through student motivations: I used my master to get into the steam tunnels, to find cool things, to gain access to the old wind tunnel, to find space to make art, to win at capture the flag. Who wants personal info or allegations of misconduct? And moreover, isn't the computer a much lazier vehicle for finding such info?

> My recollection is that you had to actually have a reasonable justification to be given an official master key

My recollection is that that was strictly correct but that the scope of acceptable justifications (ca. 1990) was broad enough that that was all but a formality.

I think an adversarial system be one where professors and students have opposing interests like if professors were rewarded for giving out the lowest grades possible. Proctored exams is just keeping people honest and the playing field fair.

> I know of one student who willingly took an F because he accidentally went over the time limit for the take-home exam, and reported it. Nobody would have known otherwise if he hadn't reported it.

Well done. Somehow I feel that this individual will do very well in life.

Exams are generally take home.

Students are expected to follow the exam instructions on the honor system.

It's a refreshing change of pace to take exams when and where it's convenient.

The flip side of take-home exams was that they could be really long. Exams were generally four hours apiece, but some of mine were eight hours long. And you needed all the time they gave you! (At least I did anyway....)

Of course I wouldn't change anything. :)

I looked it up a while back and apparently the new SAT is only 4-5 hours long, which really put it in perspective for me how nuts[0] Caltech final exams are (especially when a typical finals week for me has several at once) :-P

[0] I mean this humorously, not insultingly—I also don't regret choosing Caltech in the slightest.

Except for the Math 108 exam for which I had six hours and finished in 3 because I did not know the meaning of the remaining 2-3 questions. Ugh.

I might change a few things. Caltech faculty were singularly unhelpful in getting me into grad school.

I didn’t go to caltech, but at Harvey Mudd we had a similar vibe. I once had a take home exam that stipulated I had to take a 5 minute break half way through the test. During the break I was not allowed to think about the exam problems. That was probably the most difficult rule I was ever asked to follow during a take-home. I tried to follow it as best I was able.

On one take home math exam I decided to TeX up my answers. This was permitted as long as you stayed within the time limits, but I forgot the command for a specific symbol. I wouldn’t be able to print my exam until after my time had elapsed, and so I wouldn’t even be able to draw the symbol on the printed exam without violating the letter of the law. Naturally using the internet to look up the command would also be a violation of the rules. I chose to just include an apology in my proof that explained the situation and then introduced a replacement symbol that I used instead.

I got the impression that most students were similarly strict about adhering to take home exam rules.

What was the point of the exam break? Were they trying to encourage self-care, or is there some other lesson they're trying to convey?

I made a huge mistake of not applying there and to MIT. I was concerned about the heat in Pasadena and the piles of snow in Boston (silly me. why don't Ivy's relocate someplace nice like Hawaii?). I aced the math SAT-I section without studying and was in GATE and NMS. Instead, I went to a top 50 public state research uni (that became a party school just as I got there) because it was a sure thing. I think I learned more and developed more initiative because it was so sink-or-swim, and the standards were high and the assistance was low. I even hustled an undergrad paid research gig in a top security lab to fix grad students' and postdocs' network code that looked like spaghetti after being puréed in a blender.

> I was concerned about the heat in Pasadena and the piles of snow in Boston

As a native Californian living in Pasadena, I also hate the heat -- but, unlike Boston (in my experience), air conditioning is standard, and makes the outside situation matter far less. (It's also not usually that bad, and it's almost always a dry heat.)

The one time I visited Boston (for an undergraduate research program), it was frosty the first few days, and then hellishly humid every day thereafter. The dorm I was staying in didn't really have temperature control, and I got the sense that AC wasn't as common.

Your mileage may vary!

It's not even that hot in Pasadena. Like, it gets warm in the summer, you may want to stay inside and turn on the AC, but it's typical southern California weather…

> I made a huge mistake of not applying there and to MIT. I was concerned about the heat in Pasadena and the piles of snow in Boston (silly me. why don't Ivy's relocate someplace nice like Hawaii?).

Neither MIT nor Caltech are Ivy League schools.

Would you qualify Caltech as a competitive environment?

Personally, I absolutely would not—collaboration on sets among undergrads[0], for example, is very common (and even officially encouraged) in my experience.

[0] I can't speak for the graduate experience, but I'd assume it's also non-competitive?

Not really. Institute policy disallowed curve grading. I also found the better students to be generous in helping me understand the stuff. Without them I would have flunked out.

The opening words of ACM95a my year are seared into my mind:

"I would like to apologize to the students who took this course last year. I always aim for a mean exam score of 50 and a standard deviation of 10. Last year's mean was 29 and I will attempt to not repeat the mistake"

Haha. I remember AMA95's first lecture where Prof Cohen said "the course catalog describes this class as introductory. It is, but make no mistake, it is not elementary."

AMA95 had a reputation for being a trial by fire. Doing well in it meant you were going to graduate.

Tough as he was, I liked Prof Cohen. He was a no nonsense kind of guy, and obviously enjoyed his subject. I was sorry to read he passed on recently.

At least I didn't hear about people being haunted by 95 afterwards. For us math majors, who instead of 95 had to take Ma108, it tended to haunt us.

For years or even decades afterwards the number 108 would show up. Call a busy tech support line and get put in queue, and you are told "The average wait time right now is 108 minutes". Check after lunch to see if it is time to go back to work...it's 1:08. That number would just show up way more often then it should have.

I know one person who fought back. She took the intro to digital electronics class and the intro digital electronics lab class in her senior year, instead of in her freshman year like most people did, which means she took them after 108.

For her electronics lab project she built herself a digital alarm clock (the traditional project). But her clock was special. It skipped 1:08, instead holding 1:07 for two minutes than going to 1:09.

> It skipped 1:08, instead holding 1:07 for two minutes than going to 1:09.

Haha, love that story. Reminds me of another student who was going to build a digital clock that only displayed the time, very accurately, in 15 minute intervals. Because, he reasoned, nobody should need time more accurate than that!

I, too, built a CMOS digital clock for a freshman lab project out of about 40 chips. It did not work, it just blinked the display LEDs erratically. I still have it, it still does not work and I still don't know why.

My senior EE91 lab project (single board computer) did work, though, but I misplaced it somewhere in the last 40 years :-(

AMA95 is one of the few Caltech classes that I still think about, mostly because it was not useful. Almost all of my other classes have been useful over the years.

I wish that the math sequence for non-math majors had included linear algebra instead.

Similar anecdote: I deeply believe that Caltech is the best[0] university for me (and that's even while I'm struggling with major burnout due to an insanely rough sophomore year[1]—even though my current situation is decidedly suboptimal, I don't regret a thing).

My particular favorite aspects of Caltech are three-fold (in no specific order):

1) The Honor Code and collaborative nature, for the reasons you've mentioned :-)

2) The breadth of the Core and CS major requirements. e.g. I really doubt that I would have found a fascination for physics, without Caltech having required me to take two terms of special relativity and electrodynamics as a freshman.

3) The theoretical focus. Even though I'm a CS major, my classes are often proof-based. One (non-CS-specific) example was Ma 1a, the mandatory first-quarter freshman calculus class: It spent the first two weeks building up the real numbers (i.e. starting from the Peano axioms). Caltech's theoretical-ness has a particularly special place in my heart, because it's one of the big reasons I recently cared to learn about interactive theorem proving and formal verification (which I love so much that I'm working on a undergraduate research project in Coq).

[0] (among the universities that I know enough about)

[1] I'm honestly a bit hesitant to share about my burnout publicly (since mental health issues can have stigmas and I don't want to shoot myself in the foot re: grad schools or employers), but HN is sufficiently anonymous (at a casual glance) that this feels safe enough.

> hesitant to share about my burnout publicly

Wish this weren’t the case! The more open and okay we are with it the less perceived stigma there will be. We should consider it as typical and normal as physical conditions like asthma or diabetes that many of us have experienced or live with today.

> sufficiently anonymous

your name is in your profile

I'm not an MIT alumnus, but if I was one I feel like this joke would write itself.

maybe BalinKing IS from MIT and the joke's on all the caltech readers.

I didn't realise clicking on the name lead to the profile for 2 months after I started lurking on HN. It's not that obvious...

Security in obscurity.

I can’t gloat, after 10 years I recently discovered I can favourite threads. It’s changed everything.

Hacker News has a number of little hidden features that you find as you keep using it. It's like an easter egg hunt of sorts, I guess. You could blame part of it on poor UI or whatnot, but I personally find it kind of fun to discover new things about the site (having a general understanding of how it works, so it's not like I am missing some critical functionality).

Yeah, but that’s an extra click that would require more effort than a quick Google search... I’m not worried about a determined attacker here ;-)

Also, the intersection of people thinking (burnout at Caltech = tautology) and (reading HN is a good way to flick) is sufficiently large that I really think there will be no negative repercussions if you otherwise conduct yourself honorably online.

Hi from an '03 mole whose is mildly surprised you're hitting that wall in sophomore year -- back when I was there it was junior year. But hey, I didn't do CS :)

I guess he thinks of it as a trapdoor function - it's trivial to find his name given the comment, but just given his name it's probably hard to turn up this comment.

TIL the term "trapdoor function"—and also, you're correct[0] :-)

[0] with the added protection that someone non-HN-familiar might not know to click on the profile name to get more information

I did not go to Caltech. But my advisor did, and he is a mathematician who attended Feynman's lectures on computation. (Again the breadth : a math Ph.D taking a course in theory of computation from a Physicist.)

Feynman's requirement for an A grade in the course was "prove something original". Someone timidly asked, "Well, what if we can't?" He replied: "Then I feel sorry for you." I wish my history were slightly different.

But you must be more discreet, I think. It's not wise to give intimate medical information to strangers who might not be medical experts. All the best for your future.

So did the course result in 20+ something new proofs for computational theory?

alternatively, not every participant got an A.

I wound up at Caltech by pure luck too. Walked there from PCC, which I went to for 2 years (I didn't attend Caltech).

One of the most insane stats I've heard: Roughly 1 in every thousand Caltech alumni has a nobel prize

For context, there are only 603 prize winners total. That's extraordinary.

603 prizes*, 934 unique recipients.

I'd love figures on how many PhDs are issued broken down by school in timeseries format. JDs don't count. Would be helpful to know who is debasing the currency. Caltech definitely isn't. One of the few actually elite institutions left. No fake subjects either.

I'm guessing places like Harvard, UCB and MIT each issued close to 1000 last year. Lower ranked schools, maybe 500 each; probably cumsum to significantly higher than CalTech even with short histories.

What currency? Are PhDs fungible? I doubt it.

Disclosure: Physics PhD, working in industry for 25 years.

Yeah, whatever: I have one too; only 16 years out.

PhD is definitely a currency, and no they're not fungible: I'm trying to figure out which ones are actually worth something. In my experience it's a fairly lousy signal. Every Caltech PhD I've met has been sharp as a tack though.

Does CalTech have some institutional advantage driving better quality? I would normally guess that more 'local' factors have bigger effects: personal characteristics, then advisor effects, then department effects, and finally some vanishingly small institutional effect.

(or, more pessimistically, larger universities tend to have disadvantages that individual profs and departments might or might not overcome...)

They have a merit-based admissions reputation at least, and their small numbers keeps it exclusive.

On the other hand, most Stanford guys I've met have also been sharp, and they graduate 700 PhDs a year according to someone below.

[1] Purely merit-based admissions [2] Very selective, with a total enrollment under 3000 [3] They focus much more on the hard sciences. They are primarily a technical institution.

I didn't notice any legacy admissions when I was there, and only a very small number of students who couldn't hack it and quickly washed out.

(Many students left before graduating, maybe half, but it was only rarely because they couldn't do the work. My good friend had to leave for medical reasons, it was very sad. Another left as a sophomore. 10 years later, he asked if he could come back, and the Institute said "sure", and he graduated with straight A's. I asked him if he'd gotten smarter in the preceding decade, he laughed and said no, it just took him 10 years to get ready to work hard.)

Admission is fairly selective and focuses on interest, ability, and accomplishment in math and science.

There certainly are significant differences between individuals, but nearly everyone is smart and scientifically motivated.

Being one of, if not the most prestigious technical university in the world tends to be one of those self-fulfilling things. Since you're known as top of the line, you get your choice of top tier applicants, who in turn propagate the institution's reputation for prestige.

It's a horrible signal. Signal-to-noise ratio: Zero. Two reasons. The first is the huge variety among people with the degree. It's supposed to give you a chance to create your own thing, but this also means it's not a signal.

The other is the hate. Especially in some industries, people are widely aware that higher education is desperately in need of reform, have had genuinely bad experiences in school, or are just plain anti-academic. In fact, the notion of "signaling" is a symptom of these problems.

> Especially in some industries, people are widely aware that higher education is desperately in need of reform, have had genuinely bad experiences in school, or are just plain anti-academic. In fact, the notion of "signaling" is a symptom of these problems.

Can you speak a bit more to this, and what your experience with this has been like?

Graduate education is quite risky. You spend many years, at the mercy of one advisor, many things can knock you off your horse, and the process has minimal oversight. There is very high attrition, and unlike a regular job, you're left with little or nothing to show for your time if you drop out. There are a lot of horror stories. This is the bit about needing reform. A lot has been written about this -- I'm barely scratching the surface, but giving you some flavor.

It's almost like a business incubator, where they provide you with basic infrastructure, and perhaps some training, but you end up with a growing business or you don't. And in fact a lot of PhDs, at least in my field, end up becoming entrepreneurs.

You're supposed to be able to turn yourself into an independent researcher, but this is far from guaranteed. While you assume these risks, there's also the risk of getting the degree and regretting it, if the expectations (being in a leadership role, possibly directing other people) don't match with what you actually want to do, and the people skills needed for those things. And then, in some fields, you're competing with people who went straight to work after college, and have developed their careers on the job while also putting away some nice money.

Okay, enough of that. The other thing is, you can pick it up after a few months of HN, that a lot of people are really down on academia and academics. Again, this varies by field. In my field, physics, the PhD is almost the default -- there are relatively few people with masters or even bachelors degrees to be compared with, and we're also rare enough that many people have just not formed an opinion about us. ;-) Plus we do the work that most people hate, such as math intensive stuff, and we don't demand as much money. ;-)

I suggest this is an issue worth just observing for a while. There are debates among my friends about whether HN is really representative of the hive mind of the tech industry, or computer programmers, or whatever,

> *JDs don't count

CalTech doesn't have a law school.

https://www.gradoffice.caltech.edu/academics/degrees

GP meant "by university" by writing "by school"

AFAIK, JDs are awarded only by law schools. (But I take your point.)

MIT issues around 600. A little lower in 2020 perhaps due to covid slowdown.

What is a fake subject?

The social justice disciplines which are pseudo-sciences, as illustrated in the Sokal hoax.

The manager of the Starbucks next door has a PhD in Caribbean Studies.

He’s a really smart dude, total waste.

I find the attitude that only STEM is worth studying (which is not coincidentally believed most fervently by STEM grads) really pernicious.

Let me restate the common argument--there's this feeling that it's useful to measure people by the good they're doing for society, and that getting a PhD doing postmodern feminist readings of Christopher Marlowe is less useful to society than getting one doing Alzheimer's research or whatever.

I'll just say that it's worth considering whether an academic discipline's contributions may be difficult to notice but in fact substantial. Some academic disciplines like history or cultural anthropology must operate more qualitatively simply because the phenomena they aim to account for admit of no other approach, with the result that their intellectual products (especially when considered from the perspective of general society) will not often be as concrete as, say, a vaccine. But it'd be premature, of course, to dismiss a discipline just because its methods do not reach our desired level of formalization and its output is intangible.

If the issue you're raising is more narrowly one of your belief that within the scope of an _individual_'s wellbeing studying some subjects will leave them worse off than others, that's obviously true in at least financial dimensions, but that doesn't make a discipline a "fake subject".

Problem people have with non STEM is that they almost always have so low quality, not the subjects itself.

I think a first rank classics PhD is probably worth more than an average PhD in physics. The problem is, exactly as you say: the majority of "classical scholars" these days don't even read Greek or Latin; something the average high school graduate could do in the 1800s.

I'm not in that camp as I am a History/CS major. My dad had a PhD in English. My CS education was much less relevant to me than the History side. Love liberal arts. ;)

It's bullshit in the sense that if you're going to be a PhD, you generally want to move to industry, enter a career path where the rigor of the PhD process pays off or tenure track academia. If the academic prospects are so poor that you get a raise managing a Starbucks rather than being a contract visiting professor for 18-36 month stints, that's bullshit in my book.

I assume you do draw the line at some point, below which you would consider a discipline a fake subject. Where do you draw that line?

In the beginning of my career I hoped somehow my path would take me in the direction of a PhD. It's a thing that I still hold in very high esteem, pretty much regardless of the area of study. The amount of work, expertise and determination still seems incredible to me.

As I get older, I get closer and closer to thinking it'll never happen for me. Not because of life or kids or money or whatever, but because I just don't have that amount of commitment to put towards one area of study. Almost a decade (or an actual decade) devoted to just one thing seems insane to me.

It's the thing that, if I won the lottery (or my options become of any value), I'd give a shot. Not sure in what really. I'd start with a master's probably. But if I didn't have to care if what I was doing was kind of ridiculous, I would try.

I am one of the 39 PhDs mentioned in this announcement. Very unexpected to see this on HN.

Congrats

Caltech has been around for 129 years, averaging 78 PhDs a year.

I have no frame of reference for this situation, but that sounds like a _ton_

List of top 10 US places for doctorates for 2016:

Rank Institution Doctorate recipients

1. U. of Texas at Austin 849

2. U. of Wisconsin at Madison 823

3. U. of Michigan at Ann Arbor 819

4. U. of California at Berkeley 796

5. U. of Minnesota-Twin Cities 787

6. Stanford U. 763

7. U. of Florida 730

8. Purdue U. at West Lafayette 727

9. Ohio State U. 716

10. U. of California at Los Angeles 689

https://www.chronicle.com/article/universities-that-granted-...

There were ~55K doctorates issued in the US overall.

https://www.nsf.gov/news/news_summ.jsp?cntn_id=297405

So really, 78 is a year is pretty low for a place like Caltech.

Makes sense - the top schools on the list are over an order of magnitude larger, by student population.

> 78 is a year is pretty low for a place like Caltech.

Except that Caltech is pretty small. When I attended, the student body, undergrad and grad, was about 1500. It was smaller than my high school.

Interesting, those are all somewhere in the range between elite and thoroughly respectable. There's no random crap schools you've never heard of pumping out oodles of PhDs every year, which is good news at least.

They probably filtered out those schools. According to this[1], Walden University was #7 in awarded PhDs in 2018.

[1] https://ncses.nsf.gov/pubs/nsf20301/assets/data-tables/table...

This comparison makes no sense whatsoever, I'm pretty sure if you looked at the data 100 years ago the student body was significantly smaller and with few PhD's across the board.

So you can't compare a current day number to a 100year average

In 2019 Caltech graduated 207 PhDs.

https://www.caltech.edu/about/news/caltech-celebrates-the-cl...

Fun fact: The Caltech campus is a popular location for Hollywood productions. For example, an episode of the Amazing Spider-Man was filmed there. The photo on wikipedia:

https://en.wikipedia.org/wiki/The_Amazing_Spider-Man_(TV_ser...

is one I took :-) I remember wandering to class one day in a sleep-deprived haze and saw the filming, went back and got my camera.

Well. This is a random post to be seeing on the HN front page.

Yup, Caltech is small.

More children in the academic family tree!

I would love for a well funded organization to use big data analysis to try to build as complete as possible a full family tree of every PhD:

http://nghiaho.com/?p=978

It might help show where there a blind spots :)

Anyone else Ricketts House? Caltech '04 here. I miss that place sometimes.

Current Caltech PhD student here. Happy to answer any questions.

Are the following phrases still used - trolling, flicking? how about 3AM runs to Tommy's? and The Ride? finesse vs brute force and ignorance?

BS 2001 graduate here. When I was there, Negative Time Tommy's Run was still a thing (although they did build a Tommy's like 5 minutes away). Also, I wonder how much of the undergraduate culture was visible/apparent to the graduate students.

Negative Time must have been new jargon!

And true, the only interactions I had with grad students was with TAs.

Oh Negative Time Tommy's Run was done the night when Daylight saving time ended, so students could leave campus and come back before the time they left. :)

You just made me LOL. Thank you!

I have literally never heard of any of these things. I know the undergrads have a very unique subculture here, which grad students aren't really a part of.

trolling - studying. A troll is one who studies a lot.

flicking - playing

3AM runs to Tommy's? - just fun thing to do, just because

The Ride - The Ride of the Valkyries was played, and only played, at the beginning of finals week. The KCAL radio station in Page House would play it, then everyone with a stereo would tune in so it was all synchronized all over campus, all at maximum volume. Playing it any other time meant you'd get tossed in the pond.

finesse - doing something artful and clever, like redesigning a motor to be more efficient so more torque is produced.

brute force and ignorance - the straightforward approach, like upping the voltage so more torque is produced.

> The Ride - The Ride of the Valkyries was played, and only played, at the beginning of finals week

There was a loophole. You could play it at other times if you were playing the complete opera it is from.

I did that a couple times, although to be extra safe I believe I did the whole four opera Ring cycle instead of just the one containing the Ride.

2015 BS alum

Trolling, flicking - I never heard anyone say these, but had heard they were used in the past. Along those lines, flaming or flaming out was used for failing out or needing to take terms off.

A few people still did 3AM Tommy's. Negative time Tommy's was a significant event in at least a few houses.

The Ride tradition was definitely around, with the only sanctioned playing on mornings of finals

I never heard anyone say finesse, brute force, or ignorance

> I never heard anyone say finesse, brute force, or ignorance

Not even on ditch day [1]?

[1] https://www.caltech.edu/about/news/ditch-day-today-39501

Impressive - I want to meet the recipient!

Sheldon Cooper must be proud!

Order french fries & a coke and we will add a Phd from Caltech for free!

Reply on news.ycombinator.com