Copy of Sacha Grief - AUDIO EDIT
===

Noel: [00:00:00] Hello and welcome to Pod Rocket Web Development Podcast, brought to you by Log Rocket. I'm Noel, and today I am joined by Sasha Grief,~ uh,~ online creator at demographics. Here to talk about state of Dev 2025, a survey you put together. How's it going, Sasha?

Sacha: Hey,~ I,~ I'm doing good, ~uh, still pretty early here in Japan where I live, so hopefully I don't sound, uh, too sleepy, uh, on the podcast.~

Noel: ~these are always tricky. We did our best to like coordinate time zones, but you know, I appreciate, I appreciate the flexibility here. Um, yeah, let's just jump right in.~ Tell us a little bit about,~ uh,~ state of devs in 2025. ~Kinda~ what is it,~ like, what's,~ what was your goal? What motivated you?

Sacha: ~Sure. So, I mean,~ Sure. ~So, I mean, ~I don't know how familiar listeners are with the other surveys I've been doing, but ~there's the, ~the first and largest one is,~ uh,~ state of JavaScript, state of js, and I've been doing that since,~ uh,~ I think 2016, so quite a while now. And then after that,~ um, you know, ~state of CSS was like a natural extension.

~Um, ~and we started doing state of HTML to ~kind of ~catch all the other aspects of web development that weren't covered. And, ~you know, ~I've been expanding and doing more surveys about ai, ~uh,~ react, react ecosystem and so on. But then I realized one thing I wasn't really doing was asking people about things, ~you know, ~other than code, because.

We might be developers and we are all passionate, of course, but there's still, ~you know, ~other ~acts,~ aspects [00:01:00] to our lives,~ uh,~ besides code. ~So, um, I actually worked with a company, uh, here in Japan called Tokyo Deaf to help them with their survey. And their survey was very focused on, you know, these, these other issues like career, uh, workplace and so on.~

~And that kinda made me realize, uh, hey, that hasn't been my focus, just because. It's not something I have a lot of experience with. I've always worked mostly by myself, so I haven't really had to deal with these kinda issues, you know, uh, workplace issues, colleagues and, and so on. Uh, but this made me realize, hey, there's actually a lot more things I could be asking about.~

~So maybe it makes sense to have a survey, uh, dedicated to these more like, you know. Human, uh, centric issues rather than always only asking about features, JavaScript features, CSS properties, uh, front end frameworks, and so on. Uh, year after year,~

Noel: ~Nice. Was there,~ did you kind of do some early,~ um,~ like pre-survey ~kind of ~questionnaires to figure out ~what kind of, ~what kind of questions you should be asking? Was there any early feedback you got that kind of steered you one way or the other?

Sacha: ~well, um. ~It was a mix of a couple things, ~I guess. So, ~like I said, that this,~ uh,~ Tokyo Dev survey,~ uh,~ informed a lot of the questions ~because, um,~ so ~the, ~the person I worked with, they already did some work for a couple years now, coming up with good questions around,~ uh,~ the workplace. I also knew there were questions that I was interested in personally, like things around,~ uh, you know, ~sleep and health and hobbies,~ uh,~ physical activity, ~um.~

Even just,~ uh,~ movies, music, ~you know, ~basically the things I care about outside of programming. I wanted to know what other people, ~you know, ~think about those. And,~ uh,~ also another,~ uh,~ factor was I wanted to try and find questions that would appeal to the people who usually don't fill out,~ uh,~ the other surveys, ~you know, ~so it's no secret that,~ uh, you know, ~women,~ uh,~ people of color.

Minorities are sometimes harder to reach online because they might not feel [00:02:00] as welcome in the public spaces. ~You know, ~X is the ultimate example where,~ uh, it can, it has~ traditionally, it has had its issue and even lately, even more, where it's becoming very toxic. So you can understand why if a survey is,~ uh,~ spread through online means.

~Uh, ~it might not, ~you know, have, uh, it'd~ be very representative. So I also wanted to think about, okay, how can I find questions that are gonna target these groups? Maybe,~ uh,~ things around,~ uh,~ discrimination things around,~ uh, you know, ~whatever mental health issues,~ um, any or~ disabilities,~ uh,~ any issue basically, ~that's ~where.

These ~underrepresented, ~underrepresented group are gonna be more likely to care about them and maybe more likely to spread the survey. So that was also,~ uh,~ something that informed my survey design process.

Noel: ~Yeah. Yeah. ~Were there any,~ um, kind of ~trends ~I guess,~ that stood out in particular as you were going through? 

Sacha: ~Um, yeah, I mean, ~it's kinda hard,~ hard, ~to answer because ~I, ~I did ask about a lot of different things. ~Um. ~One thing that I was actually pretty happy with is that,~ uh,~ women participation increased compared to our other surveys. So traditionally,~ um,~ for the state of JS and so on, it's always around six, seven, [00:03:00] 8%.

~Uh, ~before this one it went up to 15%. So that kinda showed me that,~ um,~ the strategy of shifting the focus of a survey towards issues that are more relevant to different groups, like I said. Does seem to be working. ~Um, ~what else? I'm just randomly,~ uh,~ coming up with things here. But ~like, um, ~in the health section, ~you know, uh, ~poor sleep came in, number one, which at least matches with my own experience because that's something I always struggle with.

~Um, ~and so that was interesting to me personally as well. Then there's lots of ~like ~SubT trends or micro trends when you can,~ uh,~ correlate different,~ um,~ variables. ~You know? ~So maybe one thing we can talk about later is,~ uh,~ job titles that was interesting to correlate that with income. And then even,~ uh,~ also in looking at things,~ uh,~ by gender,~ like there's often, um, you know, ~you can see different people experience things,~ uh,~ differently.

~So, ~for example,~ um,~ when you look at how people deal with conflict,~ uh,~ women,~ uh,~ or non-binary respondents were much more likely to avoid colleagues,~ uh,~ on, in the workplace if they have an issue, ~you know, um, ~as opposed ~to, ~to more, ~you know, ~active ~or, or, you know, this is if, uh,~ recourse,~ like, uh, you know, uh, ~getting legal counsel ~or, ~[00:04:00] or,~ uh,~ requesting ~a, ~a transfer.

So they're also more likely to change their own behavior compared ~to ~to men, for example. ~So, um, ~it's always a bit tricky to talk about these things because these are send the sensitive topics. And also I'm not,~ uh,~ an expert in this field. ~Um, ~at this point at least I'm mostly just collecting data and seeing what patterns,~ uh,~ I can find.

~Um, ~but taken as ~kind of ~the starting point for deeper research,~ uh,~ I think it's actually,~ uh,~ pretty interesting.

Noel: ~ you mentioned health was kind of something that you stepped into, into this thinking about quite a bit. Why was there a particular back, do you have a particular background there that kinda led to your interest on the health side?~

Sacha: ~Uh, not really. I guess I should add this disclaimer to every answer. I don't have any. Um, special background. Uh, my, my background is as a developer, so I, I kind of fell into doing these surveys and I, I'm learning, you know, on the go about the, the whole, uh, data, uh, data science aspect of it, and I'm trying to do, not to make too many mistakes.~

~Uh, but, so yeah. Anything I say here, well, what I can say is I definitely trust in the data, but anything I say as far as interpreting the, the data as far as coming up with. Conclusions based on that data. That's my own personal opinion and should not be taken as anything more than that. Uh, but regarding health, I think, you know, there's, uh, a bit of a movement.~

~It's not new. There's always these, uh, health fads and trends, but there's people like, uh, Brian Johnson, right? Uh, who say doesn't want to die and. It's not so much that I think, oh, he has all the answers, or This other health guru has all the answers. It's more that I wanted to see if these kind of, uh, preoccupations are reflected in the developer community or not, and maybe, uh, be able to measure that, uh, over time.~

~Right? Because as we do the survey again, uh, the following years, we can see if maybe. People are starting to get better sleep or maybe care more or less about their health, uh, things like that. So, um, and you know, plus I think health is important for all of us, and that's something, uh, one thing the survey does show is that health becomes more important as you age know, under 20, uh, respondents.~

~Weren't that preoccupied about their health, which makes sense. But, uh, I'm, I turned 40 a couple years ago, and so speaking personally, it's something I'm thinking about more.~

Noel: ~Yeah. Yeah. Were there any,~ were there any,~ um,~ I guess ~kind of ~particular. Takeaways on the health side,~ um,~ that you think just might be interesting to listeners? I know ~you talked about, or you kind of surveyed, um,~ you had several questions about sleep. Was there anything else that was ~kind of particular~ particularly interesting there? 

Sacha: ~Yeah, so the,~ the average,~ um,~ number of ~h~ hours of sleep per night was 6.9, and that's about what I get myself,~ um,~ on a good night, I guess. And what was interesting is. There weren't that many factors that seemed to affect it. So for example, I thought that,~ uh,~ having kids would really hurt your sleep, but in terms of sleep quantity at least,~ uh,~ it doesn't seem to have that ~big, ~big of an impact.

~Now, of course, sleep quantity and sleep quality, that's two different things. Um, but all age as well. People get seven hours, um, about seven hours no matter their age, although it does go down as you get older. But not by that much. So it was fairly stable.~ And even when looking at strategies that [00:05:00] people can take to get more sleep,~ uh,~ it didn't really make a big impact.

It was all always around seven, 6.8 hours of sleep or not. ~that was interesting to me because I feel if you, you know, if you have trouble sleeping and you Google, uh, that or GPT. The, the answer you're gonna get will be pretty, you know, encouraging in the sense that, oh, here's what you can do.~

But I think it's interesting to see an answer where ~it's like, well, ~there's not much you can do, right? ~Because maybe then you can feel a bit less guilty about your poor sleep. Or maybe you can, you know, get used to the fact that, well, that's just my natural, uh, sleep rhythm. And so, uh, especially talking about.~

~These health issues. I feel like all the messaging we always got is always got, is very, uh, proactive, which good in a way, but there's not much discussion of, oh, maybe those, those supplements don't actually do anything. Or maybe, you know, being in the gym, uh, four times a week, of course it's great, but maybe there's a diminishing returns past a certain point.~

~And so, you know. I think it's a fine line between being very health conscious and promoting a positive message, but also then at some point you're just selling people something that's not gonna have a big impact on their life. And so, um, I thought it was interesting to see that the results were going a bit more in that direction of being like, Hey, we're all getting about seven hours of sleep drug doesn't seem to be much, that much that can, uh, affect that.~

Noel: was that kind of a common ~trend, a ~trend that ~there was kind of more like~ the data was a little bit more homogenized than you expected it to be Like, ~you know, there ~there was less differentiation across demographic factors than you expected, or was that kind of unique to sleep?

Sacha: No, I think it's a common trend,~ uh,~ for surveys maybe in general. Like I, I think it really depends. ~I, ~I wanna be careful here, but,~ um, you know, we are, ~we like pattern matching as humans, and I think we're very quick to latch onto small differences. And be like, oh, that, that's meaningful. ~Uh, ~but I think often when you look at raw data,~ uh,~ we have a lot more in common than we think,~ uh,~ between each other.

~And, ~and I think a lot of the times, ~you know, ~yeah, I do think differences can get a little bit played up. ~Uh, ~sometimes it's something you see with scientific research like. It can get, ~you know, ~overhyped a bit, especially when it comes to that ~health,~ whole health industry fit,~ uh,~ supplements, like I said.

~So, um, ~it wasn't something that surprised me. And [00:06:00] also, of course, there's always like ~a, ~a rush when you do find a correlation that seems meaningful. But even then,~ I,~ I wanna be careful about not putting too much, ~you know, ~of my own,~ uh,~ subjectivity or my own. Experience into the data ~and, ~and coming up with conclusions that might not actually be supported.

So I think that's probably,~ um,~ a way my attitude has changed over the years where when I first started those surveys, I was all about, Hey, let's find an insight. Let's find ~a, ~a, some, a secret that revealed by the data. And over time I've realized, ~you know, the data, if,~ if your methodology is right, the data is probably gonna.

Make sense ~and, ~and can be what you expect, that there's not gonna be that many surprises. ~Um, ~and if there are, that may be actually be an indication that you ask a question in a leading way or you're not considering something at some point.

Noel: ~I guess let's kind of go, go to the other end. Were, were there any, were there any, um, uh, I guess, yeah, ~despite your efforts not to ask leading questions and things like that, were there any~ kind of~ findings in particular ~that that,~ that surprised you? Just ~like~ in how differentiated they were across different demographics,~ was,~ was there anything that you expected to be more ~kind of.~

Standardized across respondents, but ~it, ~it [00:07:00] was more,~ uh,~ varied than you thought it would be.

Sacha: ~Well, ~I can give you an example,~ uh,~ but then I have to ~kind of ~deflate my own example because,~ uh,~ I was looking at,~ uh,~ job titles. And one thing that was very clear is that job titles with,~ uh,~ engineer in them,~ uh,~ pay much higher than those with developer in them. So if you just compare front-end engineer versus front-end developer, there's already quite a stark difference.

And so ~I, ~I started looking into why, because, ~you know, the, if,~ if it turned out that you can just change your resume from frontend developer to frontend engineer. And then start earning like 50% more. That would definitely be one of those surprising insights that, you know, that the survey has revealed.

~And, and, ~and that would be pretty cool. But, so I consider different explanations. So one thing I looked at was,~ uh,~ countries because maybe,~ uh,~ different countries use different job titles, right? ~So if. And ~if the US uses frontend engineer, but ~uh, ~Poland uses frontend developer, then of course,~ uh, because~ just because Poland pays lower salaries on average, that's gonna drag the average for developer down. And that did happen a bit. ~Uh, ~[00:08:00] but even just looking at the US there was still a discrepancy between developers and engineers. And so ~I, ~I looked at other factors, ~um. ~And then it turns out what is probably going on is ~it's~ due to company size. So again, that's where I kind,~ uh,~ add ~my own, you know, ~my own ideas, my own opinion to the data, because it's very hard to prove something like that without further research.

~Uh, ~but it does seem like larger companies are more likely to recruit engineers while smaller ones are more likely to advertise for developer positions. Because larger companies have more resources, they also pay larger salaries, and that's what results in,~ um,~ the discrepancy between the job titles.

Again, probably that's my working theory for now, and so I thought that was interesting because it's both an interesting insight, like you might not know, ~you know, ~that ~an,~ a front end engineer position is more likely to be in a larger company and maybe have ~a more. ~A role where you're ~a bit like ~charged with overseeing a more complex architecture, maybe while front end developer might be a smaller scope.

So there, there are [00:09:00] like small nuances between those job titles. ~Uh, ~but at the same time, of course it's not the job title itself. ~Uh, ~it's not calling yourself a different job title that's gonna have an impact on your income. It's actually going to work for different companies that have different,~ uh,~ pay structures.

Noel: ~ did you find much, uh, like WW ~was the data granular to the level of ~like, kind of, um, ~technical background experience to devs that had ~that role?~ The role distinction there? ~Or, ~or ~was it, ~was it, is it ~kind of ~hard to piece together,~ like, you know, ~like how much of a salary can this dev actually demand?

Sacha: Yeah, so that's where ~the, the, ~the data kinda stops giving us answer because~ I didn't ask~ in this survey at least. I didn't ask about specific job descriptions or job responsibilities, but that would be something I could ask about next year. And so that's what's so cool about the doing the surveys on a yearly basis where I can ~kind of ~course correct over time and build a more granular, more detailed picture of a specific topic.

Noel: ~ I want to eventually kind of connect this circle, talk about, talk about what you're gonna do going forward, but I'm, I'm curious leading up to this point, um, is there,~ is there anything that, ~like ~you found, stood out now,~ uh,~ about devs ~kind of ~around community, open source, that kind of thing,~ uh,~ versus surveys you've run in the past?

Sacha: ~Uh, ~one [00:10:00] thing that I thought was interesting, and again, with the. Caveat that this is my opinion, and don't take this as like ~a, ~a fact, ~I guess, ~but it seemed ~like, um, ~when looking ~at ~at gender,~ um,~ there was a difference between men on one side ~and, ~and women and non-binary respondents on the other where,~ uh,~ women and non-binary were a bit more likely to participate ~in, ~in-person events and maybe a bit more likely to join,~ uh,~ discord,~ uh,~ forums, ~you know, ~even social media. Another thing that I felt was going in the same direction,~ um,~ was concerning open source,~ uh,~ participation. So that was a little bit lower,~ uh, for, ~for women and non-binary respondents. ~So, ~I don't know. It seems to me ~you, ~you could maybe hypothesis that,~ um,~ a lot of our online spaces kind discourage participation from minorities. And I had the experience recently of,~ uh,~ seeing a comment. ~So, you know. ~Something we all do. I left a comment on a GitHub issue ~asking,~ saying, Hey, it doesn't work for me. I don't understand. And got a reply that was a bit, ~you know, not, ~not mean or anything, but it was basically, Hey, we addressed that in the thread, or please read the comments.

And it was a bit dismissive. ~I, I, ~I would say,

and, ~you know, ~I don't have any ~like, um. ~[00:11:00] Imposter syndrome. I've been in the industry for a long time. I've had my fair share of dismissive responses. ~Um, ~so I didn't think much of it, but in the light of the results, I feel like maybe if I were ~a woman, ~a woman, if I were a person of color, if I had any kind of characteristic that puts me in a subgroup like that, I might think,~ well, well.~

Is that person acting like that because of my identity, right? I might feel more targeted or ~maybe, ~maybe I just more,~ uh,~ sensitized to these kind of microaggressions, right? So I feel like even things that are not,~ uh,~ discriminate, discriminative in nature, even things that to us, ~you know, ~or just speaking for myself, to me as a.

A person who's fairly close to the average,~ uh,~ developer in terms of gender, race, and so on. To me, it's nothing but maybe to somebody who's different than me. ~Uh, ~it might slowly, it might be, ~you know, ~the lid that, the straw that breaks the camel's back and makes them be like,~ well, uh, ~open source is not for me.

People are just not welcoming, coming. And so when you put those two things in relation, I think it paints a picture that maybe, yeah, we need to do an even better job of,~ um,~ creating these spaces [00:12:00] for. ~Uh, ~minoritized groups. And,~ uh,~ another thing that the survey,~ uh,~ reveals is that one common reason for ~discrim~ discrimination is just,~ uh,~ age, right?

~Uh, ~as you get older, you're not one of the cool kids anymore and you're not part of the,~ like,~ the cool group. And so that's something that's gonna happen to every single one of us. And I think keeping that in mind is also a good way to remember that, ~you know, ~discrimination is not something that. Happens only to others, even for people like me who are squarely in the middle of the developer demographics. And so, yeah, it's kind just of ~a, ~a general message of, I guess it sounds maybe a bit corny, but inclusivity and empathy. ~Um, ~which I think when you look at demographics data ~and, ~and just remember that there's so many different ways of being a human. You kinda have to, ~uh. ~Consider those factors a bit more.

Noel: ~Yeah, I guess, did you,~ did you feel like ~any of ~ those trends on inclusivity or at least like the feeling of it, did you find that there, there were trends over time? ~Like, ~did you get the impression that ~there was,~ that was the perception? In the developer community was getting better or [00:13:00] worse in that regards?

Or~ is this just,~ was it just surprising ~kind of for this, ~for this survey in particular? 

Sacha: I'm glad you bring that up because that was another reason ~I, ~I forgot to mention, for doing the survey. ~Um, I mean, ~we've all noticed that there's been a big of a bit of a seed change in the tech community in terms of values ~and, ~and and, and ethos, ~I guess, uh, ~where for the past decade maybe,~ uh,~ the tech community has presented itself as very progressive, very, ~you know.~

I guess woke to use ~a, ~a trendy term, and now there's a backlash a bit against that. I think people are more open about being conservative, which, ~you know, ~if you are conservative on a, ~you know, ~policy front, like economically or whatever, ~I mean, ~that's your business. But a lot of it spills out into being a bit less tolerant and being, I think having values that, ~you know.~

Aren't ideal in my own personal opinion. ~Um, ~and so yeah, ~that was also,~ I also wanted to measure that in terms of, ~well ~there's been, talk about the backlash again, the I initiatives. Is that something that's actually being perceived,~ uh,~ by people? Is that something that people maybe are happy about or scared about?

And ~so, ~because this was the first year, I [00:14:00] don't really have data about that yet, but definitely something I wanna. Keep an eye out. ~Uh, and, and maybe, you know, maybe start, I, I can, uh,~ I would say I dipped my toes into politics. We have one question about worldview,~ um,~ in there, but going forward, actually, even though it can be a bit controversial, I wanna ask about those topics even more because I think they're only gonna be becoming more important,~ um,~ as we go forward.

Noel: ~Yeah, yeah. You, you,~ you'd found that,~ um,~ younger devs are more likely to prioritize traditional values or like free speech concerns while older devs lean more. ~Kind of ~in that like traditional progressivism. ~Um, did you, did you find that that was kind of like, like, was that, ~was that a pretty solid trend that you noticed across age demographics and is that going to inform your questions going forward?

Sacha: So yeah, that's something that the survey did find,~ uh,~ a bit. It wasn't,~ it,~ it was pretty solid, I would say, but it only concerned like under 20 respondents. And another thing ~I, ~I should say is that a lot of respondents come from the US so sometimes trends can be a bit US centric. And I think in this case,~ um,~ it's true that in the US at least, if concerns about free speech are often coded as a right wing [00:15:00] issue.

Whereas of course, if you're in a. Iran, let's say. ~Um, ~if you wanna have more free speech, that's not gonna make you right wing, right?

Noel: Yeah.

Sacha: the opposite. So ~it's, ~it's really dependent on the country and the cultural context. ~So, ~and again, I'm not a political scientist, I'm not a data scientist, so always refer back to the data if you want to be sure.

Don't take my word for it. But yeah, it's something I did notice and I think it would be interesting to, to dig more into it because, ~um. ~I think the next generation, of course,~ is,~ is gonna replace me,~ uh,~ eventually. And ~so, uh, ~I'm curious to know where they're coming from. ~Uh, ~same with things like ai, which we don't ask about in this survey because we have another survey,~ uh,~ specifically about,~ uh,~ AI and web development.

But if we are raising a generation of, ~you know, ~right-wing web coders,~ uh,~ I would wanna know because I think that's pretty scary.

Noel: Yeah. Yeah. if you do continue to run this survey,~ uh,~ in the future,~ you've,~ you've mentioned a couple things, but ~is there, ~is there anything ~kind of in, ~in, in that regard that's like at this kind of cross section of demographic information and behavioral [00:16:00] characteristics?

~Like, is, ~is there anything in particular you wanna focus on there? Are you ~kind of, ~are you ideating about questions for follow up surveys?

Sacha: ~Um,~ yeah, like one idea I had that I had haven't, didn't have time to properly implement would be maybe asking people what they think about key figures in the tech industry. Elon Musk,~ um,~ Tim Cook, whoever. ~Um. ~Yeah, I think that's something that's changed. Also, I wish I had this data from 10 years ago, but I'm sure you could chart public perception of Elon Musk and it would just fall off a cliff.

~Right. ~And I think generally public perception of actors in the tech scene has gone a bit more negative, maybe deservedly ~so, ~I would say. And so I can go back in time, but maybe going forward that, that would be interesting to track. ~Um. ~Trying to chart if people are more left-wing or right wing, even though that's ~kind of a, a, ~a simplified, ~you know, ~spectrum.

I know it doesn't encompass the entirety of ~a, ~a person's ideology, but I still think it could be interesting to, to know that, and especially track it over time. ~Uh, ~and then, ~you know, ~just looking at the feedback,~ uh,~ we got on this survey there, there's actually at the end of every survey,~ I,~ I [00:17:00] ask people,~ uh,~ what question they would like me to ask next time.

So that always gives me like a big,~ uh,~ like hundred, literally hundreds of,~ um,~ comments to, to look at for inspiration.

Noel: Yeah.

Sacha: then beyond that, it's just a matter of,~ um,~ going back section by section and trying to. See how you can flesh out,~ um, the, ~the questions, maybe how you can ask more questions that can support a hypothesis that you have,~ uh,~ things like that.

Noel: ~Yeah. Do you think, has your kind of, has your,~ has your perception of the community changed at all as a result ~of, ~of what you got here or were a lot of these,~ um,~ responses ~kind of ~what you expected? 

Sacha: I wouldn't say it's changed that much, but I think ~what's. ~What, maybe not changed, but is being revealed is that the community is not like a monolith, ~you know? And especially ~because when you ask people about JavaScript frameworks, ~it's kinda, ~no matter who you ask, you can pick 100 developers at random and you can tweak their, ~you know, ~geographic location, race, gender, whatever.

~Um, ~they're all gonna pretty much like react, maybe use view, maybe angular. It's not gonna make that big a difference. It's not like, ~you know,~ women in Kenya use, ~you know, ~Amber [00:18:00] exclusively for whatever reason. So that doesn't have a big impact. But on the other hand, when you do a survey about,~ uh,~ these factors, career, workplace demographics, then ~uh, ~those,~ uh,~ intrinsic factors have a huge impact.

So I think what the survey shows is that the community is like. It's not a monolith, like I said, and it really highlights the fact that where you put your focus has a big impact on the data you get as a return. So for me, it's ~kind of ~a reminder of having to do a better job of being inclusive in the survey, ~you know, and, ~and reaching out to the different communities and not just on the basis of gender, but also, yeah, like I said,~ uh,~ race, also, country, also disabilities.

Also things like large companies and small companies. ~Um, ~and of course, it's a much larger challenge than just publishing a survey on Twitter and being like, Hey, fill out the survey. And not really caring who fills it out, because you're only asking about ~do you use, ~do you use,~ uh,~ the has selector in CSS or not? ~kind of the, the, ~the easy version, the easy mode. And then, ~um. ~Yeah, expanding into these more,~ uh,~ sensitive issues, [00:19:00] these more personal issues, that's definitely a challenge on a different level. But ~I, ~I'm glad I undertook the challenge because like I said, we did have more participation from those demographics and so it's been encouraging to me to see that it is possible to reach people that typically you can't reach.

And I just wanna keep going in that direction. 

Noel: is there anything in particular you hope. ~Um, like other, ~other devs reading this ~that you hope that they~ take away? ~Like ~is there anything you're ~kind of ~trying to accomplish with ~publishing, ~publishing this data?

Sacha: ~Um, well,~ one thing I would love to, for people to try is playing with the data on their own. So we actually have a query builder. Write in the survey where you can query the API and filter the data according to the factors that are interesting or relevant to you. ~So, you know, ~if you're in Japan like I am, you can filter by respondents from Japan.

If you're a person of color, you can filter by people who are like you. And,~ uh, I, ~I really want people to make the data their own and maybe find their own insights rather than, ~you know, ~I definitely don't see my role as ~telling, ~telling people. What to think. I guess it's more like I'm [00:20:00] providing them with tools,~ um,~ so that they can generate their own insights.

~Uh, ~first of all, because it's less work for me ~if I have, ~I have to be the one to,~ uh,~ find out everything,~ um,~ there's so many different correlations that it would take me,~ uh,~ literally months of work for every single survey. ~Um, ~and also because like I said, yeah, everybody is different, so you might be interested in a completely different aspect.

Of the data than I am. ~So, uh, ~that's why I really want people to be able to personalize ~their, ~their own,~ uh,~ insights, ~I guess.~

Noel: ~Cool. Cool. Um, well~ I think ~that ~that's ~kind of ~as good a note as any, I'm gonna move us into some kind of more quick form questions. if you just had to pick a favorite chart or statistic from the survey, what would it be? 

Sacha: So at one point,~ uh,~ while trying to cross reference different things, I realized one chart was saying that,~ uh,~ Minecraft players were single at a much higher proportion. ~Uh, ~turns out the reason is because of age. So Minecraft players skew younger and younger people are less likely to already be in a relationship.

But I thought it was funny that you could, if you wanted to, you could paint Minecraft players. As like [00:21:00] losers or, ~you know, ~depending on how you look at it. ~Um, ~of course, ~that ~that would, I'm not saying that this, the reality, I'm just saying it was an illustration of how you could make misleading charts and I didn't include that in the survey, but I thought it was interesting.

Noel: Correlation, causation, et cetera, et cetera. ~Yeah, yeah, ~yeah. ~Um, ~if you could add one, like weird, quirky question. To the next survey, what would it be? Or do you have any that you're thinking about, including just ~kind of ~for the novelty that you think would be fun to gather data on?

Sacha: ~Um, let me think. So we already asked about, uh, music, uh, video games, movies. ~Maybe next time we could ask about food, favorite food.

Noel: ~Yeah. Yeah, yeah,~ that's a good one. ~Um, ~do you think devs in the future will generally skew happier or more anxious?

Sacha: ~Um,~ if I had to guess, I would say maybe more anxious just because the world seems to be. Becoming more anxious. But I also think this is gonna be very conditioned on people's age ~and ~and background. So that's something,~ uh,~ we should definitely try to measure.

Noel: it's probably a little bit tough to discern there,~ where,~ where it's more in ~like, are, are,~ is the younger generation skewing more anxious and then that will ~pull, pull the data, ~pull the data up over time. is there something you wish devs [00:22:00] were more honest about? 

Sacha: ~Um, ~yeah, I don't know. I honestly don't know. It's such a really tough question. ~Um,~

Noel: ~Were there, ~are there any questions ~you felt that, or ~you had just ~kind of ~an inclination that you were getting potentially like less genuine answers on, or ~is, are you, ~are you skeptical to ~kind of ~try to peer into the data that deeply?

Sacha: ~I mean, ~if I start questioning people's honesty, then it ~can, uh,~ kinda ruins the whole enterprise of them what they think. ~So, uh, ~usually I try to assume that they're,~ uh,~ filling out the survey in good faith.~ ~

Noel: ~Yeah. Are there any, ~are there any questions in particular that ~you, again, like try,~ you are really intentional about? Not to ~kind of ~prompt or I guess, ~you know, ~asking a leading question that may make ~developers or anyone, you know, ~people taking the survey more inclined to be dishonest. ~Like, is there, is that,~ does that kind of play into your,~ um, like initial, ~initial pass when you're coming up with these questions?

Sacha: ~Well. ~Honestly, not really, because ~I, ~I, the surveys are anonymous and so I feel if you lie on the survey, there's not much value. ~Like, ~I don't know what you're getting out of

Noel: Why? Yeah.

Sacha: ~Um, ~but, and the other aspect is that hopefully in aggregate,~ well, you know, ~people will always enter like how many kids you have, 99.

And that skews the data. But in aggregate,~ uh,~ you can remove outliers. And [00:23:00] also in aggregate the data cannot make sense. You're not supposed to look at single data points anyway. So maybe it's wishful thinking, but I am hoping that any lie would kinda get drowned out,~ uh,~ by ~the, ~the honest answers

Noel: ~Yeah. Yeah. Yeah. Cool. I, uh, it, it, and~ I guess it's ~g good to hear,~ reassuring to hear that in general,~ you,~ you felt that,~ uh, you know, ~you can have ~kind of, ~kind of faith in people filling out the questions ~to, ~to do their best, to reflect what's actually going on in the, in industry, ~I guess. ~Is there anything else, any other~ like kinda, um,~ call to action or takeaways ~you, ~you'd encourage people to walk away with this from or to go into the data ~kind of ~looking with this particular lens?

Sacha: ~Well this year, um. ~Usually we always have a conclusion at the end of the survey, but this year,~ uh,~ I actually asked,~ uh,~ too many people to write conclusions. So instead of having a single conclusion, I ended up having four different essays,~ uh,~ throughout the surveys. So ~I, I, ~I think they're all present different perspectives from different people with different backgrounds.

So I encourage people to go check those out. ~If, if, you know, ~if you've only seen like results on Twitter or Blue Sky or something, you didn't have the full results with dsis, I feel, I think it's worth checking out. You can of course,~ uh,~ sign up for [00:24:00] next year's survey or you can go,~ uh,~ take a survey. Right now we always have a survey ongoing, right now's probably the state of HTML 2025, so we'll add a link to that. But yeah, I mean.~ I mean. ~At this point I'm doing,~ uh,~ six or seven surveys a year. So there's always something to either,~ uh,~ fill out or look at if you're interested in that.

Noel: ~Nice. Yeah. Yeah. I'll, uh,~ I'll look forward to, to, to seeing those, and particularly ~the, ~the state of devs next year and ~kind of, kind of ~seeing where things land. Thank you for,~ uh,~ coming online and chatting with me, Sasha. It's been a pleasure.

Sacha: No problem. Thanks for having me. I.

Noel: ~Of course, ~of course. Have a good one.