Episode 78
· 33:39
Don't live inside your head. Just listen. It's there. Like, people are telling you, users are telling you you you are wrong. The signals are there.
Igor Zalutski:You don't even need to dig it up. It's all in the surface. Just listen.
Jack Bridger:Hi, Hi, everyone. You're listening to Scaling DevTools. I'm joined today by my friends, Igor and Upal from digger.dev. Digger.dev is an open source Terraform cloud alternative. And, yeah, welcome, guys.
Utpal Nadiger:Thanks. Thanks, Jack. Nice. Super happy
Igor Zalutski:to be here. Thanks. Excited to be here.
Jack Bridger:Okay. So I was, like, very briefly kinda working with you guys a little bit, like, probably 2 years ago, something like that. And at that time, you guys were, like, or at the very first point, you were kind of, like, a the power of AWS, but a Heroku like experience at that time. And what I remember, like, very distinctly is that you were very good at kind of shipping ideas and experiments and kind of using that to tweak the messaging, and now things are, like, you're doing something, like, very differently, and there's 3 of you and your but now with, like, a smaller team, you're growing, and I'm seeing, like, a ton of traction, and it seems like it's going really well. So I have personally a lot of questions about, like, how you got from this, like, quite, I think, quite difficult message to get across to something that really resonates with people and is growing, and, I've seen, like, your community Slack support requests are like it's like a that classic, like, exponential graph.
Jack Bridger:So, yeah, like, that probably a big question, but I wonder if you could kinda tell me a little bit about that.
Igor Zalutski:Thanks. So first of all, it's it's great to, to hear these, you know, your assessment of of how we iterated. Certainly lots of, lots of learning along the way. And, well, that's probably the reason, you know, we're still around and and and and have some some exciting signs of of of growth and traction. Well, back then, we we had this, what we thought, can be called an idea.
Igor Zalutski:Probably the biggest learning is that, well, if it's just in your head, that's probably not not a real thing. And and and and you can you can actually check if it's a real thing by trying to go out, sell it, get people to use it. And if they don't use it, then maybe that's not the right thing. I mean, that sounds obvious. Right?
Igor Zalutski:But, it took us a while to sort of get into that actual honesty with ourselves. Like, is it working or not? And if if you spend a month pushing something out and it's not working, as in, like, not growing quickly, then it's probably not working. Well, these kind of things. So we went from making AWS great again, which is, I mean, okay.
Igor Zalutski:Cool. But, but in in one start up with, like, 7 people, can can we really make all of AWS great again? The answer is probably no. Probably need to, to do something smaller. And so we started shaving our features.
Igor Zalutski:And, well, okay, well, we we thought application deployment, we're not gonna touch that. And then we ended up with, like, kind of infrastructure automation thing. Okay. For some reason, which is fun, we built it, on top of Terraform, which is, one could say, you know, a bit of overengineering. So we ended up with a with a tool that you can sort of connect your Git, connect your Git repository and get it deployed, and then customize the underlying Terraform.
Igor Zalutski:Well, this is cool, but still too big. So then we realized, well, why would we even touch the application thing? Why don't we just deploy people's Terraform? And then we realized that we've just built an inferior version of Terraform Cloud, which is I mean, it it worked, but, like, on all accounts, it was worse, than than Terraform Cloud. Okay.
Igor Zalutski:And then we thought, could we actually make it better? And turns out, yes, we could. And so we made it open source, and the key insight we had was, like, why are we running it on our servers? It's gonna be expensive. Right?
Igor Zalutski:So we thought we just reuse the CI that that people already have, and we put together, like, a scrappy GitHub action. Literally, it took a week, just a week, put it out on Reddit, and people are like, oh, shit. I'm gonna use that. Like, a couple 100 people engaged, started using. That was a year ago.
Igor Zalutski:That gave us some confidence to kind of double down, and, yeah, that's that's our product now rewritten in Go in in a couple weeks. But, yeah, that's still the same.
Jack Bridger:This episode is brought to you by WorkOS. At some point, you're gonna land a big customer and they're gonna ask you for enterprise features. That's where WorkOS comes in because they give you these features out the box. Features like skin provisioning, SAML authentication, and audit logs. They have an easy to use API and they're trusted by big dev tools like Vercel as well as smaller fast growing dev tools like Nock.
Jack Bridger:So if you're looking to cross the enterprise chasm and make yourself enterprise ready, check out WorkOS. We've also done an episode with Michael, the founder of Work OS where he shares a lot of tips around crossing the enterprise chasm, landing your first enterprise deals, and making sure that you're ready for them. Thanks, Work OS, for sponsoring the podcast and back to the show. That that's amazing. And, like, did you was that, like, a fundamental, like, mindset shift before that that, like, was it just you were doing the same things that when I was there, which was, like, there was, like, lots of kind of, like, can we can we ship something in a week?
Jack Bridger:Can we make, like, a big splash in a week with an experiment? Was that, like, also that same style of doing things, or or was it, like
Igor Zalutski:Improved, I'd say. Yeah. That that motto is is probably is the reason we are still around because, that's what keeps you accountable. Right? Like, if you can't make enough of traction in in a week or 2, you better acknowledge that all your beliefs are bullshit.
Igor Zalutski:Like, there's no way around that. Like, there's no way around that. But doing that repeatedly led us to this sort of place of desperation when we basically proved by shipping all possible things within the original idea that the original idea is not the great idea. And so that but that was the first step of, like, we proved ourselves wrong. Now what?
Igor Zalutski:I mean, it it it was a pretty miserable place when, like, you know, you you just crossed all boxes. It's almost like a checklist. Could we try this, this, this, this, this? And then the fullest is crossed off, and nothing's working. Like, it kinda sucks.
Igor Zalutski:But then, I I remembered clearly, we're sitting in in this national theater, cafe, and they're, like, eating some chicken and, like, this elephant in the room is there, like, like, well. And, like, we're just sitting there, and, you know, everyone's eating lunch, and there's this awkward silence, and, like, like, everyone knows that that it's not working. And then we had this random debate that had happened, like, oh, is it like Terraform Cloud? Could it be better? Why it's no open source?
Igor Zalutski:And it all took, like, maybe 30 minutes to kind of arrive at this, let's try this, because that that was pretty much the only thing that hadn't been tried. And and then the next week, we shipped it, scrappy thing in Python, like like, the worst possible way to build a DevOps tool. We cut all the corners just to ship it in a week, but then it actually took off. And then we rewrote it in Go, then then someone someone said us a bunch of users told us, like, why is it not in Go? Like, literally.
Igor Zalutski:And and and no one in the team knew Go. Well, so we picked it up, and, like, we wrote it in Go anyway, and that got us a bit further. But, yeah, same kind of approach, I guess.
Jack Bridger:Yeah. Why do you think this one was different? Like, do you know?
Igor Zalutski:Well, it takes it takes a good deal of learning, to to see the signs when they're there. Like, if if I were to do it again, on all the launches that that you've seen and and been part of and and actually, I mean, you remember these lemon and AWS Bootstrap and all of this. Like, you can see when it's not working. If you trace the funnel, you get all the way down to retention. And if people are not sticking around, that's your answer.
Igor Zalutski:There's nothing you can try top of the funnel that will fix it. People are not sticking around. That's your answer. It doesn't work.
Jack Bridger:I think that's, like, a really key key point about, like, if it's, like just because because I think, like like, digger and all the early iterations got, like, a ton of, like, buzz. Right? Like, it was, like, product of the day and, like, it's quite big on, like, Hacker News and everything. And, like, a lot of people were coming through, but I guess they weren't sticking around.
Igor Zalutski:Yeah. And it it can be known in a couple days. So you don't need months of, like, more launches, more traffic, more anything. Like, it's not gonna fix it. People not sticking around, you built the wrong thing.
Igor Zalutski:Move on. Like, that that is just hard to acknowledge.
Jack Bridger:Yeah. And, like, if you had to, like, kind of, like, step back and analyze, like, why why do people stick around for, like, the current iteration versus, like, the first iterations.
Utpal Nadiger:I think this one's points of failure. Right? Like, I think by by by removing different parts of the product, we limited points of failure. Earlier, again, small team, we used to go on demos where we used to just sit with them and and try try to walk through the entire journey. And each time, the point of failure was different.
Utpal Nadiger:Right? Like, obviously, you know, small team and, you know, we're shipping fast, so corners have been cut in different places. And then we sort of realized that, hey. Hey. Look.
Utpal Nadiger:There's a corner cut there. Hey. Look. There's a corner cut there and stuff like that. But shrinking those points of failure, I think, actually helped, helped in in terms of in terms of retention, if that makes sense.
Utpal Nadiger:Because, well Mhmm. Well, lesser points of failure means, so it was sort of if if you wanna plot a Venn diagram, it's limiting points of failure and actually something people wanted. And I think we had that intersection of sorts that we, with with this equation.
Jack Bridger:So it's, like, the smallest possible thing that people want in a sense.
Igor Zalutski:Yep. It's it's do one thing well. Like, another learning, which is hard hard one, and and so many people, including us, refuse to believe it. It's it's written everywhere, y c and, like and and we're like, no. No.
Igor Zalutski:No. We're special. But, like, if you try to solve many problems and and build a product which is defined by, like, doing many things, which we have this AWS do it well, like functions, containers, databases, deployment, infrastructure, like, all of these things, different like, we thought of it as features, but these were, like, different problems which different people would have at different times. And you actually just need to imagine it. One person, single person with a name, find their LinkedIn profile and see, do they have a problem, when they have a problem, and then build a product for more people like that with the exact same problem.
Igor Zalutski:Sounds obvious, but our original product had all was all over the place. Different types of people with different types of problems. And so the definition of a product was, like, many different things in one place. And that's not a product, that's a bag of features. And you can't just launch a bag of features and hope for success.
Igor Zalutski:One thing well, as soon as we did one thing well, it actually could be any other thing, I think. It just just pick 1 and do it well, and then maybe something good happens.
Jack Bridger:That week where you launched this iteration, like, what what did you do, in a kind of in a very almost, like, boring way, like, what exactly did you do, when you launched this?
Utpal Nadiger:What happened was, like, there were 4 of us or 5 of us, and each one was tasked with coming up with a version of so there's there's a there's an open source competitor competition or competitor tool. We see them as inspiration. It's called Atlantis, and it's very well known used by some very large organizations. And so we we actually thought of, you know, how does Atlantis do stuff? How do the other commercial vendors do stuff?
Utpal Nadiger:And how can we come up with the smallest possible, implementation of it in the week in in in that week? And then, of of course, there was there were 4 people tasked with doing it. In terms of task, it was, like, almost like a, you know, according to challenges. So it's but then, but then, yeah, we we arrived at one implementation after the end of that week, and that implementation was running it within actions. And it was called TF run back then.
Utpal Nadiger:And then and then actually, what followed from that was that, you know, why are people using Terraform Cloud if if this can achieve what they want to do? And that is the exact same question we went and asked for it. And then we said, hey, look. We have a very lightweight implementation of this. We've cut corners.
Utpal Nadiger:It's it's not perfect, but look at this. And then we linked TF run that we we'd sort of gone over in that week and then said, look, do you wanna, you know, do you wanna tell us why you use Terraform Cloud? Do you wanna, go and, you know, give TF run a spin? TF run became bigger later on, but initially, it was called TF run because, well, we didn't know if that's the, you know, pivot that we're gonna be doing. So it was an experiment.
Utpal Nadiger:But yeah. And then people end up commenting, and obviously people have opinions. Right? So they said, here's what they can start from cloud. So here's what they use competing vendors for.
Utpal Nadiger:Here's what they use Atlantis for. Here's what Atlantis does well. Here's what from cloud doesn't do well. And then, you know, that actually, acts as a source of truth for us even today because people keep commenting even today. I think it's it's a year plus in terms of how old that post is already, but, like, we still keep getting people commenting there today.
Utpal Nadiger:And and well, yeah. That that sort of give us confidence and also give us some some some, you know, knowledge in terms of how how the tool is used in in in the current state.
Jack Bridger:And did they use it as well, these people? Did they try it out and, like, use it on Reddit?
Utpal Nadiger:Yeah. There was there was usage for sure because it was, it was super lightweight and it was it was a GitHub action. So there was usage. Back then, you know, we didn't know analytics, no telemetry. There was nothing.
Utpal Nadiger:But, yeah, there was usage and there was issues flagged. There was people joining Slack and commenting. Mhmm. That's when so another sort of community validation for us was, and how we knew people are using because we didn't have anything any guardrail set up or anything set up back then was people got running Slack and shouting when things didn't work or commenting and and letting us know that, hey, look. This doesn't work.
Utpal Nadiger:Hey, look. I think this is what you need. It or just, you know, contributing because they want something that doesn't currently exist, and then they started contributing. None of this was known to us yet because we had never maintained an open source project before. And then, yeah, we knew that people started using because they came and said, hey.
Utpal Nadiger:Look. We're using this. We're considering it using it in in production. Can it have feature x or feature y? Can it have drift detection?
Utpal Nadiger:And can it have I don't know. You know? Are you thinking about state management? You know, these questions came up, but, yeah, that's how we initially knew that people were using it. Because even if you look in our Slack members' graphs, like, there's an inflection point of sorts in Feb 2023.
Utpal Nadiger:That's when TF run was launched. So at late February, March 2023, and then from then, the growth of the Slack, Slack community in general was, you know, it was growing at a much faster rate.
Jack Bridger:Yeah. That's that's really interesting. It's, like, do people because because it's, like, it's not just, like, go to the website or, like, oh, this is cool. It's, like, actually dig in, hit a roadblock, ask a question. It's, like and you care enough.
Jack Bridger:You don't just give up. You, like, actually care enough to, like, go and find out the answer because you believe in this thing, I guess.
Igor Zalutski:There there is there's one more thing that that told this experiment was different from any any other. For all the time before this, there was a need to maintain this sort of, you know, they call it vision in your head. This is how it needs to be, the product, the the the hypothetical sort of, you know, end state of this thing. And it needs to be like you need to keep it in your head and you need to, like, see it. Right?
Igor Zalutski:Because because, otherwise, how do you know what to do? With this one, like, no. It's all there. People tell you, like, exactly what's missing, and and and they tell you, like, I would use it if and there's, like, a list of problems. That's it.
Igor Zalutski:There's no need to, like, invent some kind of new thing. It's just, like, there. And that that's that felt refreshing.
Jack Bridger:I love it so much. I I'm gonna apply this to my my own experiments. And, like, Igor, one question I had is, like, I remember, like, there was some point, like, we were talking about, like, the analytics matter and stuff like that and measuring, like, does did, like I'm talking about, like, at the earlier stages, like, do you think do you still worry about, like, you know, having, like, analytics in place or this sort of thing?
Igor Zalutski:I I I'm glad you asked. I guess, I'll I'll hand over to Tal because he he can do a better better better summary of that, I think.
Utpal Nadiger:I think I think yeah. Initially, we weren't very very, you know, keen on measuring stuff or, it it's also in terms of shipping fast, you you ship the most important stuff first. So I think in speed, we initially, we were not measuring enough. But that's something we've tried to correct as much as possible with with this situation. Given that it's stuck, like, you know, how many docker downloads do we have?
Utpal Nadiger:Obviously, you know, we've we've experimented with telemetry. Sometimes there's been backslash for the community as well. Right? Like, how much telemetry is enough in open source and not? Like, that's the age old question as well.
Utpal Nadiger:But, yeah, we've arrived at a certain, you know, the community has actually shown us how telemetry should work. And now I think we're measuring it in a manner where we know how much weekly active usage growth is. And, you know, if it's not growing, we are aware that it's not growing. We know how much, let's say, a launch or a particular growth activity contributes to weekly active usage growth. We sort of can understand how retention works and stuff like that.
Utpal Nadiger:I think over time, that's a learning that I think we should measure more, and and make measuring a nonnegotiate. I think that's that's a learning from iterations prior that, you know, there are some corners that you cannot cut if you're if if that makes sense. And, and and I think measuring, measuring usage and measuring retention is is is something that, we've learned the hard way, but now I think we're we're doing a good job. And that's helping in terms of, you know, what features are used the most and and and and that sort of stuff. Basically, how traditionally analytics is work.
Utpal Nadiger:We we we took a while to realize best practices, I guess.
Igor Zalutski:I'd like to add to this that it's it's, it's like very often people are, like, oh, we need a full suite of analytics integrator, like all sorts of segments, Junes, and whatever, and that's just a good idea if you can. But what's needed, really, all you need, and that's the one nonnegotiable, You need a way to know that you are wrong in the super fundamental way, like this whole thing is bullshit kind of wrong, and you need a way to know it quickly. There always is a way. That's the the truth. There always is a way to know it in a week or a month.
Igor Zalutski:Like, months is too much. If if it takes a month to learn that you are wrong, like, come on, that there are faster ways. But but, fundamentally, you know, what we started with was was, like, recording, user actions from, like, this this action and posting it to to Slack. And then it was, like, just just activity, just like someone did something. And then from there, we would manually compile monthly, weekly, whatever numbers.
Igor Zalutski:No analytics whatsoever in the sense, we just recorded events and just went over it 1 by 1 by hand. And that gave the answer that the user is just growing. And if it didn't grow, that would be the answer that we created another piece of bullshit. We were lucky enough to, you know, have the opposite, but but we had this this core of analytics, and that's that that's the nonnegotiable, I guess. You need a way to prove yourself wrong really quickly.
Jack Bridger:So could you so it may so the thing that totally makes sense is that, like, you launch and then you see, like, do people care about it? Like, do they make an effort, or do they just say that's cool? But then once you kind of get that initial pass that initial point of, like, okay. People are trying to use this thing. People are joining Slack and stuff.
Jack Bridger:But then you mentioned, like, if it if it's not growing, then it's also not, like in your words, like, it's also a piece of bullshit. Like, if it's, is that also, like could could I mean, because that hypothetically could just be, like, okay. We just didn't market it that well that week or, like, yeah. How did you how did did, like were you kind of okay. This is correct, or is it, like, a constant process of
Igor Zalutski:It takes, like you don't need to market a lot. It's more like you got some some people from a launch. Like, was it 10 people? Well, most launches, even the worst ones, get get 10 people. And then, like, what happened to these 10 people?
Igor Zalutski:Like, are they still around? Yeah. That that kind of question. Right? And then are we still getting usage from them?
Igor Zalutski:Even if it's synonymous, let's say. Right? Same ID, like, anonymized hash. Right? Is it still present?
Igor Zalutski:And then, like, if the answer is no, and it it's no for all 10 of the first 10 users, you don't really need a big splash to know that something's not right. They they just don't come back. Right? Like, that's that's it. That's the answer.
Igor Zalutski:Yeah. And then, like, top funnel thing, as soon as you start as soon as you fix the bottom funnel, then any small activity at the top funnel just blows up. You just do a little thing, Like, you just put something out there, and, oh, shit. It goes all the way down because people are sticking around.
Jack Bridger:Yeah. Tell me about that. Because, like, you guys have had a few I mean, there's, like, there's the big, like, HashiCorp thing and the OpenTofu and stuff like that. That must have been, like, a big kind of boost. Right?
Utpal Nadiger:I think there was a bigger boost tried before. I think about a month or so before before that, just before, the open TOFU, the open TOFU fork happened. It actually started because Igor, wrote an article on our Golang Cool. Saying, we moved to Golang in a week. And that that's when, actually, our sort of growth journey with this and this integration of bigger started.
Utpal Nadiger:And and Prime Agent picked it up. Right? Like, so, Prime Time Prime Agent, the dev and transfer. He picked it up. Yeah.
Utpal Nadiger:So he went over the entire article, and he nodded and agreed with every bit of it. And and that's when we first saw the power of YouTube, by the way, because we'd never experimented with YouTube as a growth strategy at all. And so you can post that on Reddit, and and then we go out for a weekend. Right? Like, okay.
Utpal Nadiger:We'll we'll off, and we're not doing anything. And I start seeing these stars. So I have these I have a notification that come back and somebody start, and I see get we're getting a lot of these stars, and I'm like, where is this coming from? Is, you know, is this spam? Is this is this, you know, where is this what's happening?
Utpal Nadiger:And then I realized that Igor's post on Reddit has gone viral. So I thought, okay. Cool. This Reddit post has gone viral, and it'll it'll sort of Reddit post has a life cycle, right, about 48 hours, and then it'll probably subside. And I thought, okay.
Utpal Nadiger:Cool. Amazing. We got a bunch of 100, 150 stars. But then it kept continuing so that so on on Monday, we returned and, like, stars are going by, like, 600 or or so. And I'm like, what's what's happened?
Utpal Nadiger:And then What
Jack Bridger:was the baseline at that point? It was, like, roughly, like
Utpal Nadiger:About 400, 500 ish. Oh my god. And then we at the end of it, we came out, like, 1500 ish. And in in a matter of a week or so. It's it's hard.
Utpal Nadiger:Right? Like, the the TerraForms in some sense are is is smaller. Like, Atlantis has, like, 7 k after 6, 7 years of of of existing, and and people know of it. Like, impression share is pretty good, but, like, the number of people who who star or the entire, you know, space is is not as big as, let's say, what it would be for an open source notion alternative or an open source calendar. Yeah.
Utpal Nadiger:So, yeah, so we we had this entire burst of activity from from that video. It's it still brings us users, by the way. If somebody joins Slack and says, hey. I I watched private video and, like, you know, Right. You know, it's it I joined because I I want to I wanna know how he moved from, from from Python to Golang in a week.
Utpal Nadiger:How what did that involve? And then yeah. So, basically, that that's a conversation starter. But then what we did or and over the weekend, I was like, okay. We've tried to show a channel a couple of times.
Utpal Nadiger:Let me just try to show a channel. And then the show a channel goes to the front page as well. So it it was it was like a cascading effect of the the Reddit post test and then the Prime Agent video, and then, and then the Hype News post. And then cumulatively, it sort of give us a boost in in-depth.
Jack Bridger:What was the Sure Sure HMs? Just like, his tag, sort of thing.
Utpal Nadiger:No. Sure HN was Sure HN open source star from Cloud All done it. That's it.
Jack Bridger:Yeah. Okay.
Utpal Nadiger:And that's it. Literally, that's it. And it was a Saturday, and I was like, cool. Let me try because people seem to be liking it. We should ideally get to get help trending with these many number of stars in such a small amount of period, like, such a small time period.
Utpal Nadiger:And I was like, cool. Let me just try. Why not? And then and then yeah. The I think I I read Charlie Munger's book and he says Lollapalooza a lot.
Utpal Nadiger:And I think this is a Lollapalooza. Like like, there was Converging. Yeah. The Reddit post and then the prime agent video and then the show. So yeah.
Utpal Nadiger:That that was the first spike of sorts that that we had, and and we were trending for about a week after that. And a lot of users who started using then still use bigger, like, they they're attached, and they keep they've they've very strong opinions, keep sharing super valuable feedback. But the users we acquired in that cohort, a large section of them still use Digger.
Jack Bridger:Wow. That's amazing. And just, like, kind of, like stepping back for one second. Like, so, Igor, like, when you wrote that article, like, what were you what were you thinking on that? Is it, like, replicatable?
Igor Zalutski:It's, it's very different, you know, looking back at it versus what was going on in in my head back then because it it's almost like it almost always feels like you don't know what you're doing. You're just trying some random stuff, borderline desperation, and then boom, 1 of the 100 attempts, you know, blows up. You see, you're always at this, you know, kind of let's try because because what else? And this was one of those. And and and that like, why did we rewrite this thing in Go?
Igor Zalutski:They just sort of jotted down some thoughts because it was a controversial decision. It's just like, you don't rewrite products in language from one language to another because that that's that's, like, from anyone who'd been in engineering for, like, you know, several years would tell you it's a bad idea.
Jack Bridger:The Joel Spalsky.
Igor Zalutski:Yeah. Like like oh, you don't. You don't rewrite things. Tools don't really matter as most people think they do. And, like, you pick something, you know something, you build something.
Igor Zalutski:If it creates value, then, you know, later on, you'll have resources to rebuild it. That's the conventional wisdom. And and most likely, you are running all the architecture assumptions anyway. So that's why you don't, you know, pick the right tool. You just pick a tool and build it.
Igor Zalutski:That's the conventional wisdom. And now we went against it, and I felt like that needs to be explained. It wasn't even full virality. It was like, this is why we did it, and then it blew up because, well
Jack Bridger:It's just interesting. Yeah. Yeah. I think it's quite, like, you know, quite good to have, like, an opinion on something.
Igor Zalutski:It's true. It's like we that opinion was probably and that that's also we you know, I I think that that piece in large part I I learned from you that that content is about creating new kind of knowledge in a sense. Not sure. I'm not quoting probably, but I attribute that that piece of learning to you, for some reason.
Jack Bridger:Cool.
Igor Zalutski:Maybe something you said, when we worked together or something like that. But the the thought of, like, content is not like words. Right? You bring new knowledge to the world in a sense. And if it's not new knowledge, then, like, what's the point?
Igor Zalutski:And and and we we tried that a couple of times, and that kind of content actually resonated very differently compared to something that is just mechanical word generation. As soon as you start writing about a real challenge, which many people could disagree with, that's when it's Yeah.
Jack Bridger:I I can yeah. I feel like it's like that kind of, like, if it's something that was, like, a difficult decision and you made a choice, it was, like, a hard choice, just sharing that even if it's, like, it is is really yeah. It's great. And it's great that you amplified it as well, like, up while seeing, like, the okay. This is this is taking off.
Jack Bridger:Like, how do we how do we get the most out of this? It is amazing.
Utpal Nadiger:That was the plan. And we we've tried, you know so, again, read in another book that that that viality is is is poker and not roulette. Right? Like, we've tried to recreate this this this Lollapalooza effect, but it's extremely hard. Things need to fall into place on its own.
Utpal Nadiger:And it's it's so it it just happened because we didn't try. But then when you try it again, like, replication is is hard. Yeah. If that makes sense.
Jack Bridger:You should go back to Python and, like, write about, okay.
Utpal Nadiger:We rewrite it back into Python. Yeah.
Jack Bridger:I guess just like a general question, like, going back, like, if you if you were, like, going back to those guys, like, 2 years ago, like, what would you what would you be telling them to do?
Igor Zalutski:Listen. Don't live inside your head. Just listen. It's there. Like, people are telling you, you know, users or whoever are telling you that you're wrong.
Igor Zalutski:The signals are there. It's not even you don't even need to dig it up. It's all on the surface. Just listen. And, you know, that's the answer.
Igor Zalutski:Just listen more.
Jack Bridger:That's, yeah. That's amazing. That's a amazing way to to end. I I I can say, like, I have learned a lot in this. Thank you, guys.
Jack Bridger:I really appreciate it. Yeah. If you're using Terraform, like, definitely check out Digger, like, yeah. Great guys. And, where can people learn more?
Jack Bridger:Digger.dev? Digger.dev.
Utpal Nadiger:Or join Slack. Yeah. Digger.dev or our GitHub repo. But digger.dev is lot lot of work's been put in, so would love for people to check to that.
Jack Bridger:Yeah. That's awesome. Alright, guys. Thanks for joining, and thanks everyone for listening. And, yeah, we'll see you again next week.
Igor Zalutski:Thanks, Jack. Thanks, Jack.
Listen to Scaling DevTools using one of many popular podcasting apps or directories.
