The Stack Overflow Podcast

Developers believe AI will soon be everywhere, but aren't sure how to feel about it

Episode Summary

This week we sit down with some fellow Stackers, Erin Yepis and Joy Cicman Liuzzo, to discuss the results of our latest Pulse Survey. Developers told us what emergent technology they expect to become mainstream, what they think is a fad that will pass, and how they feel about everything from AI to open source to blockchain.

Episode Notes

You can dive deeper into the research, including some lovely matrix charts, on our blog.

Erin has also explored tag trends among our most loved languages and job insights from our community.

Learn more about Joy on her LinkedIn.

Thanks to our Lifeboat badge winner of the week, russbishop, for helping to answer the question: Where is the app content folder in the simulator of Xcode?


Episode Transcription

[intro music plays]

Ben Popper Today's episode is brought to you by ReleaseHub. Release automates production-like ephemeral environments of any kind. Terraform, IAC, Kubernetes, databases are just a few dependencies they solve for to allow developers to spin up, update, and spin down ephemeral environments at will. Get started for free at

BP Hello, everybody. Welcome back to the Stack Overflow Podcast. I'm your host, Ben Popper, joined as I often am by my colleague and collaborator, Ryan Donovan. Ryan, how are you doing today?

Ryan Donovan Oh, I'm pretty good. How are you doing?

RD I'm good. So today we're going to be doing an inside baseball episode. We're going to be chatting with a couple of our colleagues from the marketing team about some of the latest results from our pulse surveys. So for the last year or so we've been running these on a regular basis on Stack Overflow, and it's a great chance for us to get to query the broad audience of developers and technologists who visit our sites about everything from the blockchain and their thoughts on that technology, to career and how folks are thinking about development or upskilling. So the most recent one was focusing a bit on emergent technologies. Which technologies among all the stuff you might categorize as the tech or software industry people feel either positive about, they think it's proven, or to the opposite, maybe are a little bit unsure whether it's proven out or still a bit of hype. So without further ado, I'd like to welcome Joy and Erin to the show. Hi, y'all.

Joy Liuzzo Hey, everyone. It's good to be here.

Erin Yepis Hello!

BP So for folks who are listening, why don't you tell them a little bit about who you are and what you do on our team, kind of give them a sense of how you interact with the community and what you know about this data from where you sit.

EY So I'm a Senior Analyst on the Product Marketing team. I work with Joy and a couple of our other colleagues specifically looking at the products that Stack Overflow has, which there are a couple of. And one of the new things that I've been working on also is survey research, so a decent amount of my job is going to be looking at not only these pulse surveys, but the upcoming developer survey too.

BP Ooh, yes. Your first developer survey, so a trial by fire. Very exciting.

JL We always plug that as a benefit and then you get your hands in it and go, “Wow, this is a lot.” This is Joy Liuzzo. I'm the Vice President of Product Marketing here at Stack Overflow. And like Erin was talking, our team handles product marketing for both our community public platform as well as all of our paid products. So we are a lot of coverage, we have our hands on a lot of different things, and one of the things that I try and encourage the team to do is tell stories with data. So in this case, Erin's telling stories about those hidden things that developers and technologists are thinking about or feeling about, and whether that's with tag analysis that she's done or these pulse surveys or the dev survey, we're really trying to help folks get a better understanding of developers, what they want, what they need, what they're excited about, frankly because if we can let folks know what they want, then people can create better products, better services, things like that. The more educated other people are, the better developers' lives are.

BP All right, let's start from the top here. This was a survey about emerging technologies, and perhaps no surprise to anyone, especially if you've been listening to the podcast over the last few months, the big thing here that stood out was AI-assisted tech. It was in the middle of the list on sort of the proven to emergent scale. Erin, Joy, do you want to walk us through a little bit of what the data was telling you here?

JL So I'm going to start with the reason why we actually looked at the data this way because I think it's interesting. So there's been, as you've mentioned, a lot of focus on AI technology and what it can do, but a lot of that has been looked at in a silo, so there's no reference point to kind of how folks are thinking about it in context of other technologies. So that was the first thing– we need to know where this lives in the giant environment or ecosystem of technology that's out there. So that was the first thing. And then the second thing is looking at it more of in a matrix perspective. And Erin's going to dive into the matrix piece because I think that's really cool because you don't just evaluate things on one criteria. You're looking at it on multiple different fronts to try and figure this out. And we know that developers are both curious and skeptical– my favorite combination. It takes a lot to prove to them that this is something interesting, so by diving into these and looking at the technologies in multiple different factions, we actually get a really good insight into what is going on and how they actually perceive it. So that's the setup. Erin, over to you.

EY Our colleague, Matt, had a good comment last time I talked to him about how not only are we trying to see this ecosystem, but I think where these pulse surveys become particularly interesting is that we're also getting insight to something that's not just an echo chamber. So rather than this community of technologists and developers being told, “Hey, this is what you think,” it's their chance to say, “This is what I think unprovoked or unprompted.” So that's really a big part of the ecosystem story and the bigger picture. But the coup de grâce, the really interesting and cool part of this data that we were able to collect is putting together this quad chart, this matrix, these four quadrants of the two main sentiments questions that we had for, “Do you think that this technology is there yet,” and, “Do you think this technology is going to have a positive impact in the world?” Two great questions. Not too specific, but also specific enough that you would have an opinion on it.

RD I thought it was interesting that the two questions can be related in their answers, but not necessarily. You can say, “This one's proven, but it's proven to be terrible.” Did you have any tech in there where you were like, “Oh, this is either proven and proven to be terrible,” or people are hopeful about it but thinking it's not there yet?

EY The one to me, and I think just because I've been looking at this earlier with a couple of other folks too, the bad rep that low-code/no-code gets on the negative to positive spectrum, to me, I definitely feel neutral about it and, no, developers are not into it. It’s very low on the negative scores there.

RD Yeah, I mean that makes sense. Any time we have a low-code/no-code article, we get very controversial comments either on our blog or on Hacker News. Some people are for it, some people like the tools, and other people are like, “Ugh, you're going to give me so much code I have to clean up.”

BP Yeah. We had the CEO of Retool on the podcast recently and they sell stuff to companies who want to make internal tools in a simpler way, and they've trained their sales staff for a long time to never say those two things– low-code/no-code. You're going to immediately turn off the customer, so interesting to know people's feelings about this stuff. Have you been thinking about AI-assisted technology? Everybody would say yes, but that doesn't tell us whether the sentiment is positive or negative. AI-assisted tech on this survey did score highest for, “Which tech do you believe will be used by everyone?” Can we dig a little bit into that? I guess from our perspective here at Stack Overflow, it kind of makes sense. There's already been autocomplete in your code editor. We know that folks who are writing this stuff now are using sort of a fancier version of that that's AI-assisted, and I think we have a perspective that this is going to be an increasingly powerful resource for getting things done quickly. But what did the survey have to tell us about this? And I guess it's a good time to reflect a little bit I think on where we still sit in the ecosystem and the role we have to play because I think Stack Overflow continues to see enormous traffic and engagement hasn't changed since these AI tools came on board, but we know we'll have to evolve alongside them. And so think a little bit about how we can maybe tap into some of this stuff given all the data and knowledge we have on our platform.

EY Yeah, so our survey results do reflect the amount of news that's currently going on in the AI space, but in a different way. I think the developer community that we have, they are able to see a little bit more of that hands-on aspect of, “What does this mean? What does it do,” which has been coming out a little bit more lately. But yeah, I would say, it looks like for AI-assisted technologies, smack dab in the middle as far as all of our results for the emergent to proven scale. So just as many developers feel like it's there versus those that feel like it's not there yet. On the negative to positive scale– “Do you think this is going to have a positive impact in the world or not,” it's a little bit lower. It's a little bit lower than half, so there's definitely people who are skewing more towards that they don't think this is going to help, don't think this is going to be what saves the world. However though, I think it's still up in the air. One of the big things that I was looking at as far as the tag questions that we have on Stack Overflow, machine learning is big for us. Machine learning was also part of this survey, and machine learning is one of the little concentric circles inside of the bigger AI universe. Machine learning is a type of AI. So there's definitely more positivity for machine learning reflected in these survey results. They definitely feel that it's more proven than AI in general, so I think that's interesting. Again, my thought is that it's just a little bit more tangible. It's a little bit more specific, a little bit more tangible, and it's something that the developers also came in and said overwhelmingly, “This is something I want more hands-on training with.” Along with AI in general, machine learning was one of their top choices.

JL Yeah. And I think the other thing too that's interesting about this specific topic as a whole, is when you're talking about AI-assisted technologies, it's a broad category. You've got everything from image generation to code generation to marketing aspects where you're writing emails and things like that. So there's a broad use case within this entire thing, and I think that is actually why folks are saying that somewhere along the lines, everybody's going to be tapping into this technology, whether it's a developer using it specifically for their role or if it's something that's involved in their hobby or something else like that. I think that's really where it's getting the highest marks. People can see it bringing value to them in some way, shape, or form, and bringing value across the board.

EY Yeah, and also just to add to that, I always do this and I apologize, but TikTok had this really great little video on the difference between AI and machine learning as far as their beautify filters go. And I just thought it was great. It was basically that there's this new filter that you get all your makeup and stuff showing up on your face, but one uses AI to take every single pixel that's showing up on your camera phone and switch it, and the other one is using machine learning, as in it can tell how big your face is, how tall, how wide, and then it overlays something on top of it, which I think also speaks to, again, if you didn't have that hands-on experience and you want it, or you don't even know what it is that you want, your sentiment is going to kind of go in that direction too. I, as a general lay person, didn't know those were two different things working for what I perceive as the same thing.

RD I thought it was interesting– on the sentiment scale, open source is the top there and it's also the top of the proven. But at the bottom of the proven category is the stuff that makes superheroes in movies. It's nanotech and quantum computing. And in the sentiment scale, those are somewhere in the middle. How do you look at those differences in sentiment?

EY For the experimental to proven scale, I would say, again, maybe you have brought up a really good point here –I didn't think about it– that sort of connection to superhero movies. In general, I feel like some of the tech that we decided to focus on, as it got those higher emergent scores farther on the emergent side of the scale, I feel like there was also with it a greater proportion of people that responded just in the neutral area, in the middle area, in the ‘I could go either way.’ I group them all and call them ‘passive’, so not necessarily saying ‘emergent’, not necessarily saying ‘proven’ either. And to me, again, my hypothesis is that it just has no tangible something, that there's just less experience that these developers have with it, so it just seems more far away, more obscure, more ethereal.

BP Interesting here that open source, both positive sentiment and folks felt it was proven, and the direct opposite was InnerSource. Developers were not having it. InnerSource I guess means trying to work within your own company, but sharing knowledge–

RD In an open source manner.

BP Yeah, not actually open source to the world, but open source between all the silos in your company. Joy, I'd love to hear your thoughts because I know you and I have sort of looked into this and talked about this as one strategy around knowledge management or collaboration within the context of Stack Overflow for Teams.

JL Yeah, and I think what's interesting about this is there's a group that understands those connective tissues very, very well. Open source was built upon knowledge sharing and knowledge reuse. You're contributing to the collective codebase, you're sharing knowledge, you're collaborating asynchronously, you've got all these pieces, so there's a lot of connective tissues between open source and Stack Overflow for Teams in terms of what the culture is underneath it. Those same culture points are what InnerSource is fostering. And for whatever reason, each time we talk about it, we've written blog posts about it, we've talked about it, it's not quite resonating with developers and I don't know if it's that skepticism. Can we really be InnerSource within my organization? Can we really be highly collaborative within my organization? Do we have the culture? So I'm thinking that that's where that low rating comes in– that folks don't quite see it happening within their organization yet. Either barriers from culture, barriers from tooling, whatever it is, but I think that's the piece. It's not that they don't believe in the components of it, they're just not seeing it play out.

RD I wonder if it's almost like the hate that Agile gets, where it's meant to be a sort of way of working in software development, but in practice it's a way of management so people are like, “My boss is not going to let me do this.”

BP Yeah. And I think at this point open source is mature enough and has been embraced by so many companies that if you really wanted to work with those principles, you'd be better off finding a way to open source something within the company, that way engaging with the public and kind of proving it out, whereas InnerSource maybe they feel is more like paying lip service to some of those concepts but not actually doing them.

JL Yeah, and I was actually just on a customer call where they were talking about what their culture is and how they're always helping each other, they're always there for each other, that insights can come from anywhere. That's all InnerSource, open source, but the word didn't come out of their mouth. That's not how they're describing it. So again, I think that there's the culture and there's the tools within the systems, but the words just aren't there.

BP All right, so last thing I wanted to touch on before we go: there's been a few articles recently about folks who are either part of the quiet quitting or have been caught up in the recent sort of tech layoffs who are pivoting from working at big tech companies to trying to work in climate tech. They have an option now to go do something maybe more mission-driven, and that seemed to be something that stood out in this survey, which was that sustainability was something people were thinking a lot about, but not necessarily something that they felt was happening fast enough or was proven. Erin, can you help me out a little bit with what the data told you here?

EY Yeah, definitely. So sustainable technology is kind of right there just a little bit below AI as far as the experimental to proven scale, how people are feeling about that, is it there yet? But at the same time I feel like, I love that you're bringing in this conversation about pivoting in your career, I think there's the ability to, do you see the results yet? Do you see these people with their careers out there? Do you have that tangibleness, something you can touch, something that you can talk to someone that you know? I don't know necessarily that we're quite there yet, although I think now is the time when that sort of thing would be becoming more of a focus, like you said. If I'm going to pivot, why don't I pivot towards something I feel good about? But it definitely didn't have a terrible score. I think it's doing a little bit better than AI in terms of our matrix, so that's good.

JL Well, what I like is the polar opposite of the positive impact of sustainable technologies versus blockchain, which is the exact opposite of the sustainable technology. I like seeing that that's at least coming out in the data. Again, it gives us some good insight there.

BP You could have an open source sustainable machine learning project focused on robotics. You'd be right at the top of sentiment and get to hire all the devs. On the other hand, if you're working on a VR blockchain company with a bit of low-code and InnerSource, you're in trouble when it's time to hire folks.

JL That actually needs to be called out. The do and the don't. Please don’t do all of these, yeah.

BP The do's and the don’ts. Even if it's just for marketing purposes, we're trying to help you hire here.

JL Exactly.

BP All right, very cool. So this podcast hopefully is coming out the same time as the pulse survey so we'll be sure to put the link in the show notes. You can go check out some of the beautiful graphics and the data analysis that we put together. And if you're interested in future pulse surveys, we're going to be running a bunch this year focused on different topics and you can always get in touch with us through the podcast and let us know if there's something in particular you’re interested in.

[music plays]

BP All right, everybody. Thank you so much for listening. We appreciate it. And as always, this time of the show we want to shout out a member of the community who came on and shared some of their knowledge and helped save a question from the dustbin of history. A lifeboat was awarded March 2nd to russbishop, “Where is the app content folder in the simulator of Xcode?” Russbishop has you covered. You can find the answer if it's been bothering you, and thanks, Russ, for helping out over 22,000 people with your answer. I am Ben Popper. I'm the Director of Content here at Stack Overflow. You can always find me on Twitter @BenPopper. If you have questions or suggestions about the podcast or want to talk about pulse surveys, hit us up, And if you like what you hear, leave us a rating and a review. It really helps.

RD I'm Ryan Donovan. I edit the blog here at Stack Overflow and put together the newsletter. You can find the blog at And if you want to reach out to me on Twitter, you can find me @RThorDonovan.

JL Perfect. This is Joy Liuzzo. I am not on any of the socials, but I am on LinkedIn. You can find me under Joy Liuzzo.

EY This is Erin Yepis. I am on the socials, but I don't think they're very active so I would say go over to the Stack Overflow Blog. That's where you can find me.

BP All right, everybody. Thanks for listening and we will talk to you soon.

[outro music plays]