The Stack Overflow Podcast

Big Tech is getting cozy with computer science departments

Episode Summary

Wealthy individuals and corporations have been giving gifts and getting libraries named after themselves for decades. But there is a new trend emerging - big tech companies are contributing everything from supplies to curriculum.

Episode Notes

You can read more about the operating systems and business principles schools are adopting from their corporate sponsors here.

You can read about the latest version of Tailwind and what it has to offer here


Episode Transcription

Paul Ford There's also, just I think as I'm getting older, I just go like, maybe we just need to tear this all down and start over. I say that about my society, my own career. [Ben & Sara laugh]

Sara Chipps I only say that about CSS. [Paul laughs]


Ben Popper Are you struggling to deploy cloud native applications to a hybrid cloud? Do you want to become familiar with Kubernetes and Istio? IBM cloud has a set of free, hands on training, ebooks, and an always on free tier of services to help you learn. Visit to learn more. That's Overflow. 

BP Good morning, everybody!

SC Good morning!

BP Welcome to the Stack Overflow Podcast.

PF Swiggilty-diggilty-doooooo!

BP Eh, Paul's here, hooray! [Sara laughs] I'm Ben Popper, host of this podcast, Paul and Sarah as my co hosts. We're here today to chat a little bit about software, a little bit about code things that came up. Paul, I wanted to talk to you about an article I saw and Sara, I know you have thoughts on this. The headline of the story was This College Degree is Brought to You by Amazon, and the subhead is, university budgets are squeezed and student debt loads rise, an era of close knit relationships between companies and universities is getting underway. So let's start with the academic portion of this and then maybe go a little broader. You know, what are the pros and cons of having tech companies, I mean, really corporations, but in this case, tech companies, you know, sort of tightly embedded with what students are learning? Providing the resources, maybe the platform, the tools, obviously, maybe it gives students resources they wouldn't otherwise have. But are they then beholden to the company in some way? So students at Cal Poly is University, walk through door proclaiming the department is powered by AWS. Here, students participate in the AWS Cloud Innovation Center.

PF First of all, every university that has Poly needs to kind of rebrand you can't be anything poly anymore. I'm sorry.

SC Yeah. Also, if I'm playing $100,000 a year, the school is powered by me. [Paul & Ben laughs] I don't want to see any brands like, yeah, yeah, you have enough of you have enough money. This can't get you sponsorship? What's going on? 

BP But yeah, I mean, you know, this is a strategy that I'm familiar with, from my time working at a tech company, when I was at DJI, you know, they talked a lot about building platforms that students would learn on and how that would then in the future, lead to these people choosing DJI as their robotics platform of choice coming to work for them. And so they gave a lot of stuff away to schools, and they built you know, STEM toys and STEM platforms specifically to try to, you know, get mindshare early on it and engineering schools and stuff like that. So, you know, it's--

PF All we're here is saying that the quiet part loud, right, like, educational discounts have always been a part of the software industry. They, they love to get in there. And they they want people in their tools. I think, like, corporate funding of research has always been a thing. I think, you know, it just always gets dystopian, when it's like 'Harvard, brought to you by Microsoft' you know, you know, but then, yeah, is that really different than like the Bill Gates Center for Aquatics and--

SC That's a great point.

PF I mean, it's it's that money is always been there, I think what's what's happening is just, there's more of it. And well, Sara, you have a you have a comp sci degree, right? No, Oh, you didn't go? Well, what was your degree?

SC I studied computer science, but I didn't graduate--

PF But you you did study. So like, you have a little big O notation in your in your soul. I don't, I learned all that stuff as I went along, often just completely in the wrong way. Like, 'Hey, have you read these books by Donald Knuth?' And they're like, well, I'm an idiot. You know, it's just like, there's, it turns out that, you know, a lot of it is pretty learnable. Nonetheless, there's a great divide in the industry between, do you need computer science? Or should you learn practical programming skills? Because there's actually a large delta between the two, right?

SC Yeah, that's what I was wondering, too, if it's Amazon, like, what is it like MATLAB brought to you by Amazon? Like, they're not teaching AWS in most curriculums?

PF Well, I guess then then there becomes the argument, which is like, ''Well, if college is to prepare you for a career in industry, you should be learning AWS'' or, you know, 10 years ago, it was Java or Java. But there was a huge conflict when MIT changed its intro to CS class, which, of course, in MIT world is like full blast brain melding from scheme to Python, because they were trying to react to the fact that that students are going to spend a lot of time managing and working with lots of libraries, as opposed to building compilers from first principles. And you know, that was, that was heartbreak for the MIT scheme community.

BP But that, you know, in that example, you're saying they're moving towards something that's more sort of universally accepted. I give you an example here from the story, and it actually has some history. So they're working on this, this big campus donated by the heiress to the Dodge automaking fortune. Okay, so that was what sold the kid on going to the school. He's in systems engineering, but the German industrial giant Siemens provided software instruction curriculum technical support and gadgetry. And so this student decided he should be learning or is learning tech pneumatics which is the software language that you need to work on Siemens equipment, you know, so like it's very specific.

PG Yeah, no, absolutely like one of the best Kraftwerk albums that has ever come. [Sara laughs] Ah, so good. After that, 45 minutes of that same blooping drone when you hear that voice go 'tech pneumatic' Ah, I love Kraftwerk. Very emotional, important song for me. Go ahead. Go ahead.

SC I think the whole conversation brings up this there's a question on Stack Exchange on the computer science educators Stack Exchange of why is computer science hard. And it's talking about how how high the dropout rates are from computer science students. I think that one of the reasons why, so when you know, like, in your first month in the computer science major, you go in a big auditorium, there's all these stories, and someone stands in the front of the room and they say, ''Look to your left, and now look to your right. In four years, only one of you will be here.'' And then everyone's like, ''Oh, my God!''

PF Like in four years, one of you will be replaced by a low code solution. And the other one will be like, you know, like, you don't even ask, why is it why because it's math. Like, it's hard. And it's abstract. And kinda like, it's, it's not cool. Like, you're not learning how to make games or do the things that you like with computers.

SC But does math have this--if I was a math major, do math majors have as high dropout rates?

PF Yeah, they do. And it's all to the benefit of the liberal arts. So as an English major--

SC A lot of philosophy degrees.

PF Oh, yeah. You'd come in like that, you know, junior year in undergrad, you just see these new faces. You'd be like, yeah, welcome. Yeah, we're gonna make you feel at home. They're like, ''All you got to do is read books?'' And like, it's amazing.

BP Yeah. I mean, I think one of the things that's in this first answer, Sara, from Buffy, thank you, Buffy, is that a lot of people go into this nowadays. And I think we've talked about this before, because they think it's going to lead to a lucrative career, right. And so they're not going to necessarily because they're like, I love, you know, programming, and I did it as a teenager. They're saying, like, you know, I see what's going on with big tech companies, this is my way in, which I think is a little different. Nobody's saying like, if I, if I get a mathematics degree, you know, I'll become a Silicon Valley billionaire. So the mathematics a little bit more self selecting people actually love math. So I think that that could be part of it is that there's a lot of people who start at who just like, maybe this will be good for me. And they learn pretty quickly it's not.

PF You know, I have a friend who runs and it's enough for profit in he works with a lot of coders who are kind of coming from underrepresented backgrounds, getting them into the industry. And I was like, so you know, I talk with them pretty regularly. And I'm just like, ''Hey, you know, what's everybody excited about? Is it JavaScript?'' whatever, I might have mentioned this on this show before, I can't remember. But, um, you know, what are the big languages? What are the big libraries and frameworks? He's like, ''Oh, no, AWS certification'' people who are trying to break into this industry, the signal that has gotten out is that if you want to break into tech and make tech money, which you know, I think in mentally is like, above $70,000 a year, like I want to make a decent salary and live in America. Your number one skill is to really understand all the tools and frameworks inside of AWS, that'll give you the most Career Mobility, they'll help you get certified, etc, etc. And so like, to me the answer to the question, what's the cool new programming language? Was AWS certification, right, like that, that kept me up a little bit late. Because, yeah, that's a new world. And I mentioned this to other people who work around cloud services. And they're like, ''Yeah, yeah, sounds about right'' right. It's just, yeah, world is taking over.

SC I talk to beginners a lot. And usually what I observe when we teach them things, they want to know why and what it does, right? Like, they want to know, why am I learning about this concept? And what is it good for? And I think the way we teach computer science, you don't get to there until year three, right? You're just like learning all this random stuff. And it's hard to get, you're not really, you're not seeing the real world application very quickly. And so that's hard to be patient through. 

BP Definitely, yeah, a lot of people are gonna back out because it's so theoretical, it's so abstract. And that is where the, you know, the void is being filled by the boot camps, right. And like the startups that are doing computer education, it's like, if you want to become an AWS or a Salesforce administrator, and start making 75K a year, like these are the tools and Paul I think, to your point, which is really interesting. It's like, that's a good starting place. Because from inside of a company with a salary and some insurance, you can start to learn other things and get some you know, data science knowledge from your colleagues and all this other stuff, because people are gonna be asking you to do jobs and you're gonna have to figure it out.

PF I mean, it's very William Gibson Cyberpunk, right because it's like I've aligned with my boss, or my boss, or I don't know and then and you're like, I'm in my giant mega corp, and now I can can You to explore options that feed and support the mega corp. But it's real, Amazon's always going to be glad if you learn a little more computer science. Look, here's what's tricky is that the academic computer science, right? Like, like, what, why do people learn and do things to get more power and control over their life. You go to college, because you exchange the money in some time, because you think you're gonna have more authority and control later, right? And comp sci has, to Sara's point that's very real, it has a slow payoff. And in fact, the comp sci pay off, the true comp sci pay off is that maybe you're gonna get go get a PhD and teach work computer science.

SC And that's never as lucrative as the other options. It's like the ultimate marshmallow test. It's like the marshmallow test, but three years long.

PF That's right. And it's just the whole time you're chewing computers. I have met really, really smart comp sci people, and they are not necessarily they're not economically motivated individuals. They are like, I really, sometimes they actually laugh like this. ''Haha!'' I'm not kidding. Like, it's just, you know, they're, you know, they're just very--

SC I know that I know a few of those as well.

PF I love them. I love them. But their relationship with capitalism is that it gets in the way of them understanding interesting optimization problems.

SC Yeah. And they get and even the idea that that's why they might be doing it is annoying to them. Yeah, what I've observed with new coders, when you're coming out of a boot camp, you need to spend your first you know, 5, 10 years figure learning those computer science concepts. And when you're coming out of university, you need to spend those first 5, 10 years learning those practical concepts that you would have learned in a boot camp, you know, like, how do I how do I Ruby on Rails? Or how do I scaffold an app? Or, you know.

PF Okay, so let's be real, you kind of you don't need a comp sci degree unless you doing fundamental work. What's fundamental work, I'm going to be fixing the database, because I would like it to, I'm going to be, you know, thinking about how data is striped to disk, because I would like to crease the retrieval speed, like those are comp sci related problems, where analysis of algorithms and optimization is key. Building a website, in Ruby on Rails requires you to almost learn rituals and object oriented thinking. The whole point is, you don't need that stuff with that. You know, that's actually why they made it.

SC Yeah, I think one thing we're seeing in universities and boot camps, is specialization. That's something we really didn't see before. It was always right in the 80s it was like IIS, wasn't it Information Systems? And then you'd computer science, that there was like a big umbrella that everything's on but now if you look at places like Carnegie Mellon, you have like 10 different degrees you can get and different things like you know, data science and different things.

PF Oh, let's go to Carnegie Mellon. Which first of all, what even school is it these days? Computer Science? Like, is it the School of informatics Carnegie Mellon School of computer science? Let's look at their let's look at their undergraduate programs. So minor in computer science major. Yeah, that's all fine. Very nice. Okay, good. To hold on a minute, BS in artificial intelligence.

SC Yeaahh!

PF BS in computer science, BS and computational biology, BS in human community computer interaction. They've had that one for a long time at Carnegie Mellon, BS in statistics and machine learning, additionally, and then, and then robotics, so minor in neural computation, and minor in language technology.

BP I want to hire all these people.

PF I mean, that's Carnegie Mellon, right? Like, we haven't even gotten to Stanford. And yeah, maybe we never will. That's hard to get in there. But yeah, no, I mean, this is and that is, you know, sort of a big classic program. There's, there's a million other, like, once you get into the more trade oriented programs, you're talking about cybersecurity, and sort of all those things. And so like, it's just a vast field these days, even as an undergrad, you're specializing quite a bit. But now let's let's play for a minute. Let's let's think, okay, what is Amazon college going to be look like? First of all, there's going to be Amazon college and Amazon college Prime.

SC Yeaahhh!

PF That's the Ivy league right there.

SC Is it gonna stay $100 a year, no?

PF Oh, no, it'll be $100 a year, and then they'll just get you to take more classes. Yeah, no, no, your course load is 300 times because you're just like, I got to get the value out of my Amazon college prime. 

BP And like, yeah, Paul, I think you're right. You know, they were talking about how the schools are infiltrating or, you know, making big inroads into colleges. But the reality that I read about after it seemed it became clear that people wouldn't necessarily get to go back on campus this year, was that big companies the Amazons, the Google's the Facebook's, Microsoft's the world started offering a lot more high level paid internships to people that would almost sort of directly transition into a career. So it's like, exactly like you're saying you could go to college or you could get paid to work for Google this summer and then maybe come work?

PF Well, that's the whole like, do I go, you know, do I go straight into the MBA? Or do I go to college first, right? Look, we're in a funny world where purely corporate funded undergraduate education feels like the slippery-est of slopes.

SC It feels a little smarmy.

PF Well, and just like, I'm good on Amazon teaching me computer science, it's gonna be around cloud services, but God knows they know it. They've done the fundamental work. What bothers me is the Amazon economics course. Econ 101. You know, Amazon literature is gonna be great, except it's all self published Kindle unlimited ebooks.

SC Yeah, from like someone you've never heard of.

PF Are we reading Shakespeare? Or watching Shakespeare in Love? You know?

SC I would accept it's like Amazon. Okay, let me talk this route before I decide I would accept it. I think if Amazon wasn't sponsoring the building, but like was sponsoring my education and was giving me a job afterwards, would I think that was okay? I think that's okay.

PF Sara Chipps. Well, you of course, you're Sara Chipps. You're a Bezos scholar. Or you got that you wrote the the Bezos thesis. Look, it's all in there. Again, it's the quiet part loud, which is just like Siemens and you know, so far saying like, now we're gonna bring you your whole education. The reality is that a corporation, even with the best intentions, is going to align educational goals with what they see as their needs, which they see as primary, like, why wouldn't you build a career along what exactly we need, that doesn't create a resilient population. 

BP And there's more beyond just the technical stuff, right? Like you're saying, in this case, at the cloud Innovation Center at Caltech students are taught things like the company's leadership principles, customer obsession, and, you know, as principles of formal education in a college workshop, so that's more like, the HR motto becomes part of your curriculum.

PF It's the libertarian seasteading fantasy, right, which is like we're gonna we're gonna raise a family of little VCs. You know, we're gonna have, I'm gonna send my children to venture school, and they're going to often have startups by the time they're seven. And you know, it's just, we're back to remember when we talked to Matt Cutts. And he was like, there's a reason the fence is there. Right. There's a reason there's a fence between higher education and corporations and boy, it's impermeable, that money is a great fence jumper. But, but there is an actual reason that education exists as it exists. Now, professors are often the worst advocates for this, because they just have no idea how industry works. And so they yell about a lot of things, many do. But our goal in life, you know, the goal of education is not to teach people to be good Amazon citizens and make Amazon more money, it's to think as broadly as possible about a problem domain and come up with all kinds of solutions. And will 99.5% of people probably go get jobs and live their lives? Sure. But that .5% is critical to the health of the society, you got to have them out there doing things doing and it's not just doing startups, they need to run the NGOs, and they need to help the, you know, they need to make sure the food supply is well regulated, and so on and so forth. And they need to think a little bit bigger. And if we, you learn to think you do you don't just get born, you know, on Twitter running a VC firm, you actually have to think.

SC Yeah, that makes sense. Yeah, it's really, uh, you know, I feel like this I don't want to descend into, like talking about universities, but I feel like, it really does make you question the neutrality of the university. And how much does it take? Like, how much does it take to change the curriculum into something that is in your benefit?

PF Well, that's the thing is, is the force of money. I mean, the disciplines are so very carefully defined. I mean, they're territory, right, people protect their territory. There's a fundamental difference between history and English, let's say and like, you know, that the English folks go to the the Modern Language Association meeting and the history folks go to the history conference, and the computer folks belong to the ACM. And those curricula are nationally discussed and sort of syllabi are shared. And yeah, so on and so forth. And so there's a real set of boundaries, right. And so then there's Interdisciplinary Studies, which, you know, when they tell you to do that, that just means they don't know what to do with you, as a student, like, just leave college at that point. Come back a couple years later, when you can accept. The power is all in the disciplines. And it's just you know, it's so tricky because it's a self sustaining system. And there's also no academic job. So it's not like the old days where like, ''You're smart and talented, stick around.'' It's like ''You're smart and talented. Be an adjunct for 35 years while you watch your your life erode.'' And there's a part of this that just definitely does feel like a weird, capitalistic end game. We're just, we're all right, just riding that dragon.

SC Yeah well imagine if like your doctor went to school, like sponsored by Pfizer.

PF I mean, they did though. 

SC Did they? 

PF Yeah. Yeah, they did. 

BP After they got out there, they went on a lot of speaking cruises, speaking engagements and cruises. 

SC I'm down for that. 

BP That taught them about what what type of medicine--

PF I mean, if you if you go to a doctor's conference, it's I think someone was described that, like, you know, even the stairs were branded with Pfizer products, like, yeah, yeah. I mean, we're gonna, they're gonna get that brand in there. And they're gonna start with college students. So there's going to be like, you know, we just need to put this one poster up.

BP So this morning, I saw a little bit of chatter and a few links to something that I thought was interesting. It's called Tailwind CSS, and then I went to their site, it is beautiful. Obviously, it's got lots of interaction and video. Makes it look like it'd be really fun to build my own website. What do you guys know about this? Who made this? Is it? Is it cool? How come people, how come it's growing? 

PF I'll tell you what I know, is that this is gonna be really exciting to Sara Chipps, because she loves CSS. 

SC Oh, my God, I hate CSS so much. You can't even imagine. It's just the worst. 

BP Well, Sara, I'll tell you it says that the creator of Tailwind says, I've written a few thousand words on why traditional semantic class names are the reason CSRS maintain. The truth is you're never gonna believe me until you actually try it if you can suppress the initial gag reflex long enough to give it a chance.

SC This looks similar to, so I'm just like reading the page right now. This looks similar to like a design system. Yeah. Okay. an API for your design system. Got it. Yeah, yeah. Yeah, I liked I love design systems. I love that, that makes everything so much easier.

PF I mean, you know, what this is doing what I like about this. I looked at this yesterday for a while, what what they're saying here is, can everyone chill out with all the JavaScript all the time? And can you can we get the people back doing the HTML and the CSS without all the JavaScript?

SC Yeah, I don't like CSS and JavaScript.

PF Yeah. So stop all the JavaScript. Look, I mean, this all works great, until you actually have to do something. And then like, I'm sure the I mean, this is a lovely design system, this is a great place to start. You know, a lot of their examples are focused on e-commerce and you know, cards and social sharing type content, things like that. The real classic stuff of the web, and how you can style that and you can make your cards change, you know, from simple to playful, to elegant to brutalist to, you know, and do all kinds of neat things like that, real estate sites. This stuff is good for that because there are when you think about like, Zillow, right, like, Zillow is probably 18,000 separate components.

SC Oh, my gosh, I think about that all the time. Whenever I go to that website.

PF Yeah. And so like, then they're like, Oh, you know, what we need to update we then you need dark mode, Stack was like this. Like, it's now it's a year of work, right? So something like this means that the number of people who can work on that project goes way, way up, compared to ''Oh, we're gonna solve it with one JavaScript solution.'' So I like that. I think these are enabling technologies.

BP The choices for building it all over again, according to Tailwind are simple, playful, elegant, or brutalist. So you have your choices, we can pick. 

PF You know, more and more, more and more things built in, more and more opinions built in. If you're building a thing that is going to function not look like all the other things, but kind of function in a world with all the other things is really good. 

SC I mean they can look similar. 

PF Yeah, but most things should look similar. You shouldn't reinvent the wheel. If you're like, if you're building an e-commerce site, you can get to really good design without changing the very nature of e-commerce. And in fact, somebody tells you, they're going to change the very nature of e-commerce that should make you run and scream. One of my best friends in the world is an absolutely masterful banner ad designer, he gets more conversions than anybody. Literally, I can't even I won't go into how but really, when you put them up against other like he'll work on a giant political campaign, he will get more conversions than anybody. And they'll be like, wow, how did you get more conversions? And it's like, literally, imagine a one room filled with engineers and analysts looking at data and one dude in a house working alone with an infinite supply of red pixels to make buttons. The red pixels are gonna win. It's just, it's the it's the wackiest thing. Everybody tries to do data science, this and my friend just goes like, ''Um, make the button red.'' And it's sort of, it's like a taller button. And then he gets more conversions than anybody else.

BP Paul, this makes me feel so bad. I just paid for that CRO audit to optimize and now

PF Oh no, my friend is an absolute master of conversion. And what does he do? He does the same thing over and over again, very well. And actually, what I've come to accept in my 40s, is that that's really valuable. The therapist who says the same five sentences over and over again, and the, the policy that doesn't change and the clear HR directives are way more valuable than ''We're gonna change America with our whatever.'' So I look at this and I'm like, Okay, that looks like a good choice. If everyone agreed to use it and stuck with it. You'd have have a really nice website.


BP Well everybody, thanks for listening. I'm Ben Popper, Director of Content here at Stack Overflow. You can find me on Twitter or email us podcast@stackoverflow.

SC And I'm Sara Chipps, Director of Community here at Stack Overflow. If you're looking for some holiday gifts this year, check out

PF I'm Paul Ford, friend of Stack Overflow, check out my company Postlight and yeah, check out those Jewelbots. Alright friends!

BP Alright everybody, let's hit stop on our recordings. 3...2..1...stop!