244: Ben vs 2026
It's a new year and you've probably got a mental list of things you want to learn. But how do you decide what's worth the investment? Ben explores the difference between "just-in-case" learning and "just-in-time" learning, while grappling with AI anxiety and the fear of falling behind. Along the way, Tim shares his own struggle—turns out, saying goodbye to something you built hits different.
Follow the show and be sure to join the discussion on Discord! Our website is workingcode.dev and we're @workingcode.dev on Bluesky. New episodes drop weekly on Thursday.
And, if you're feeling the love, support us on Patreon.
With audio editing and engineering by ZCross Media.
Transcript
Spot an error? Send a pull request on GitHub.
[Transcript to be added]
[00:00:00] Cold Open
[00:00:00] Ben: I was talking to ChatGPT today
[00:00:02] Ben: And, I argued with it for like, I don't know, probably like an embarrassing amount of time, probably like half an hour.
[00:00:09] Ben: 'cause it really kept pushing back. I mean, I was really arguing
[00:00:13] Adam: have been really wrong.
[00:00:17] Carol: Oh man.
[00:00:19]
[00:00:39] Intro
[00:00:39] Carol: Okay, here we go. It is show number 244 for the first time in 2026.
[00:00:42] Tim: Wow.
[00:00:44] Adam: on this episode we're gonna talk about Ben versus 2026
[00:00:49] Ben: do this.
[00:00:49] Tim: ding, ding, ding. Round one.
[00:00:52] Ben: Oh, you, sorry.
[00:00:55] Carol: that's Kool-Aid, right?
[00:00:57] Adam: there's, there were
[00:00:58] Ben: I, I, I I was trying to get my macho man Randy
[00:01:02] Adam: Savage.
[00:01:02] Carol: Oh, I only heard Kool-Aid men busting through a wall there.
[00:01:08] Adam: we're off to a
[00:01:09] Carol: yeah.
[00:01:09] Adam: for the year,
[00:01:10] Tim: Yeah, yeah.
[00:01:11] Adam: but first, as usual, we'll start with our triumphs and fails. And the, episode eponymous. Mr. Ben Nadel, we're coming to you first. What do you got going on my friend?
[00:01:22] Ben's Triumph
[00:01:22] Ben: I'm gonna go with a small triumph. the triumph is not immediately obvious, but it's more of a personal triumph, which is that, I was talking to ChatGPT today about. had some SQL query that worked in MySQL 'cause that's what I'm most used to, and I wanted to get it ready for Microsoft SQL Server.
[00:01:40] Adam: So I just put in the MySQL Server, version and it spit out the SQL Server, but it wasn't quite formatted the way that I would've formatted it. So I, I said, why did you do it this way? I would've done it this way. And it started to tell me why my way was wrong. And, I argued with it for like, I don't know, probably like an embarrassing amount of time, probably like half an hour.wait, what is it? Uh, old man yells at
[00:02:05] Ben: yeah.
[00:02:06] Adam: There you go.
[00:02:06] Ben: End. I'll tell you, at the end of,let's say maybe it was more like 20 minutes. At the end of 20 minutes, I feel like ChatGPT had changed my mind that I found its arguments compelling enough, and I don't think that I could back up my arguments sufficiently to counteract their arguments. And at the end of the day.
[00:02:26] Ben: I think I'm gonna change my mind and I'm calling that a triumph because I feel like maybe historically I would've really wanted to stick my ground because I felt strong in my convictions. But I feel good that when faced with sufficient evidence that my approach was wrong, I am going to accept that and, and try and change my ways.
[00:02:47] Ben: So I'm calling that a personal triumph.
[00:02:49] Tim: Was that about testing?
[00:02:50] Ben: No, no. Please, sir.
[00:02:52] Tim: Oh, oh.
[00:02:53] Adam: Strong beliefs loosely held, except that one.
[00:02:57] Ben: no, it was pretty good. Such odds, you know, like the elevator pitch on, on the conversation was happening. It was a, an update that used a join. So it was joining across two tables, essentially, where a relationship existed. I historically, in my joins, I include both the.
[00:03:15] Ben: kind of relational condition, like join where these two rows, where this ID equals this foreign key kind of a thing. But in my on clause, I will typically also include any additional filters that would limit the cross product.because to me, when I think about sql, I think about it as this top down join creation machine.
[00:03:37] Ben: So if I have two tables and I want to join them, part of the filtering there is both the, the straight relational, but then also the what might also exclude this particular cross product from existing. 'cause I think of it in a top down ChatGPT came out hard against that, which I think to be fair, 95% of SQL developers would also have come out hard against my approach.
[00:04:00] Ben: I think I'm the only person who I, I say 95 to be generous to myself.
[00:04:06] Adam: mean, you easily got three quarters of this podcast, so
[00:04:09] Ben: Yeah. There you go. So, but it, its argument was that the, the on clause is strictly about defining the relationships between the tables and the, WHERE clause is about filtering the cross product down to the target rows. And I, I pushed back and I said like, for 20 minutes I pushed back and at the end I was like, you know what?
[00:04:30] Ben: ChatGPT I think your argument makes more sense than mine.
[00:04:33] Adam: So you, this was with your, you were converting from MySQL to SQL Server, right.
[00:04:37] Ben: Correct.
[00:04:38] Adam: Does SQL Server have CTEs common table expressions.
[00:04:41] Carol: Yes.
[00:04:41] Ben: I think they were like the first ones
[00:04:43] Adam: Okay. Yeah. I, I only, I knew it from Postgres and brought it over to MySQL when they started supporting it. That, so that would be a, a place where I would take a similar approach using CTEs, like generate a bunch of these like little inline views of tables and then join those, right.
[00:05:02] Adam: So I would do the filtering in advance and join those together.
[00:05:05] Ben: Oh, that's interesting. I, I like that. I, I, I don't think to reach for CTEs that often 'cause I, they're relatively new for me.
[00:05:14] Carol: So I, I go to them often when I'm writing something for myself. If it's something being published, I don't, because I feel like it adds a layer of complexity that someone else has to figure out. And not everyone on my team understands, use like CTE, and I'm like, do you understand the CTE? There is just an alias.
[00:05:32] Carol: You can call it whatever you want. You don't have to call it CTE. And they're like, no, what's an alias? And they're like, oh, all right. No more CTEs.
[00:05:40] Adam: Ugh. It's just like a little inline view. It's a
[00:05:44] Carol: It is, it's nice. It's, they're wonderful. Yeah.
[00:05:46] Ben: It's so funny. I don't, I don't know if everyone else has the same type of experience, I assume you do, where it's like you're learning, a subject that's relatively new or new adjacent and there are things in it that seem. Okay, that makes sense. It's new, it's a little confusing, but I get it enough and I think it's a simple enough mental model.
[00:06:05] Ben: I'm gonna incorporate it and then you take two more steps and you learn something also new and you're like, no, that's bad. Crazy. I'm never gonna put that in my programming. Even though, like probably the difference is just the familiarity level. So the CTEs, when I was learning up on the newer MySQL stuff, when we upgraded at previous work, CTEs was one thing that came in.
[00:06:27] Ben: And then also were the window, like the windowing functions where you could like group over subsets of rows and then get like first and second and last and previous. And then the next, I like, I don't even remember what they were, but I was like, CTEs made so much sense. I'm like, oh, this is what I've wanted to do forever.
[00:06:45] Ben: I wanted to just take a table and make it into a variable basically that I could query against that. Like cognitively, that didn't seem to have a lot of load, just like some new syntax, but the windowing functions, I'm like, even after I've learned it. I've tried it and I can see what it does. I'm like, I'm never gonna use this.
[00:07:01] Ben: It's far too complicated and like, if I don't use it every single day, you know, I'm not like a data analyst. I'm, I'm like, it's never gonna be comfortable enough for me to ever want to actually incorporate this.
[00:07:13] Carol: I tend to find myself looking at what the costs or I think it's called, like the barrier to entry, right? Like I take a look at what that looks like, and if it's a lot for me to even get into it, I know I'm not gonna be able to propagate it. So then I kind of pull back and go. If it's quick to understand and I find a use case for it, then I'm good to go.
[00:07:31] Carol: When it's more complex and it takes a lot to learn, I have a feeling it's not gonna be adopted well, or it's not gonna be adopted correctly. So I really have to consider is it worth the spend on it?
[00:07:41] Ben: Yeah, a hundred percent. I'm always trying to find the story in my head that I can tell myself that it's not just like a hundred percent personal bias. Like I, I, I feel like there's some, like you're saying, like a barrier to entry versus cost versus usefulness. you're, you're, you're more articulated on it than me.
[00:08:00] Ben: I'm just like, Hmm, stupid.
[00:08:02] Carol: Stupid. Oh, get it.
[00:08:04] Ben: All right, so that's me. Personal triumph. Carol, going back over to you. What do you got going on?
[00:08:09] Carol's Triumph
[00:08:09] Carol: Oh man, guys, I am kicking this year off with some amazing news.
[00:08:13] Adam: Okay.
[00:08:15] Carol: the first, the actual first we drove over to Dallas, Texas, which is like a 10 hour drive from where we live, from one side of Texas to the other is not easy. and we picked up our newest family edition. So we have a new Labradoodle puppy, actually.
[00:08:33] Carol: Sorry, she's a golden doodle. Take that back. Her sister's a Labradoodle, so she's a Golden Doodle and her name is Rde Array because, you know, RDE for her dad racing me because array is my favorite data type. that's how I store everything is an array and it just flows off the tongue easy. So Rde Ray is what she gets called when she poops in the house and Rde just, you know, every other day.
[00:08:56] Carol: But she's all of, uh, 10 weeks. She is like six and a half pounds right now, and she's just the biggest fluff ball ever
[00:09:03] Tim: As she is cute as can be.
[00:09:04] Ben: Yeah,
[00:09:05] Carol: So perfect with our family. Yeah. I'll post a picture in Discord now that we've mentioned it, but don't tell my sister. She's coming out next week and it's a surprise she's gonna walk into the house with two dogs.
[00:09:19] Ben: Super adorable. I am somewhat jealous. I
[00:09:22] Tim: mm-hmm.
[00:09:22] Carol: I know. when I introduced her earlier, I looked at her and I told her, I was like, don't make Ben sad. Okay. This
[00:09:29] Ben: no. Puppies never make people sad. They only bring
[00:09:31] Tim: How did, how'd your other dog accept
[00:09:35] Carol: Rube Rube loves her. at first she was just kind of like, get outta my space. Like, why are you taking my seat in the car? Because we put her on the side that Rube usually lays on, and she wasn't super thrilled about that.
[00:09:47] Carol: But now they're fine. She climbs all over her. earlier we were outside and Rube went to go stand up and I yelled at Rube 'cause I thought she had the puppy bit by the head and was lifting her. And as I realized what's happening, Ru's eyes are very, very sad because RDA has latched onto her neck and she's just dragging Rhonda around now.
[00:10:08] Carol: And she's like, how do I get this, this lesion off me, mom, come help.
[00:10:13] Tim: Oh,
[00:10:14] Carol: So
[00:10:15] Carol: she,
[00:10:15] Tim: well that's lucky. 'cause when we first, when we got, so we have two pugs and Kaiju is the, male pug. And we got, Momo, who's five years younger than him, and he absolutely hated her for like three weeks.
[00:10:31] Carol: Oh,
[00:10:32] Tim: I mean, he just, just barked at her and really just hated her. And eventually now he loves her.
[00:10:37] Tim: They, they're inseparable, but it's like it took three weeks of keeping them separated in the very slowly, kind of introducing them together until they got along. But that was very stressful, three weeks.
[00:10:50] Tim: So I'm very glad that, your doodles got along.
[00:10:53] Carol: they're happy.
[00:10:54] Ben: when I was a kid, we had an English setter. That was a loving, loving dog, but did not take change well, and we had a bunch of dogs and every time we got a new animal, we would have to leave him in the crate for like two days straight and just let the other animals walk around the cage as he freaked out.
[00:11:14] Ben: And he just sort
[00:11:14] Ben: of like drained himself over time. But it was, yeah. It's funny how some dogs are totally fine and some are just highly offended.
[00:11:23] Carol: Well, we had the feeling Rube was gonna be fine because she's very social. Anytime she's around another dog, she goes from like her normal attitude with us to this just playful teddy bear who is alpha, but will also be submissive. So she's always in control of things, but she is just I'll play at your level.
[00:11:41] Carol: So we had a feeling it was gonna go pretty well, but we didn't know it was gonna go this well. So, yeah, 2026 is gonna be, thank you. Thank you. Alright. That's me. What about you, Tim?
[00:11:51] Tim's Fail
[00:11:51] Tim: unlike you two winners, I'm going for a failure
[00:11:55] Carol: No.
[00:11:56] Tim: and it's, it's just a general affairs. It's like, I hate, I really, really super hate the Christmas New Year's time. It's like this liminal space where you can't really get anything done, but at the same time you gotta work. It's like,so, I mean, we were off a week for Christmas, but then the week after Christmas, like the week of New Year's we're off like a day.
[00:12:19] Tim: It's like nothing gets done in December, it feels And I was like, can we just shut the whole month down? Why, why do I have to, why do I have to do this theater while I'm pretending to work? 'cause everyone's pretending Everyone I need something from is not available.
[00:12:32] Tim: It's like other companies I'm dealing with, I can't get responses from them. we talked to you a week ago and I'm still waiting. And they're like, yeah, we'll get to you next. It's yeah, it's so frustrating and just annoying and just, I'm just glad that we're done with, it's all behind us now.
[00:12:51] Carol: This what's the next big holiday? Valentine's Day.
[00:12:55] Tim: that, yeah. That's not even a real holiday, so it's like,yeah. I was just so glad to be back in a normal routine of things I hated, I hated these past two weeks. It's just awful.
[00:13:04] Ben: So it's so sad, but I also totally get it.
[00:13:07] Adam: There are those of us there. I would say there are dozens of us, who actually work in December. On December 23rd, I worked a half day. It was my last day for the year. I shipped three features that in the past would've taken me at least a day each, if not longer. and I shipped all three in one day.
[00:13:26] Adam: happy about that. Yeah,
[00:13:27] Tim: Good for you.
[00:13:28] Adam: thanks ai.
[00:13:30] Carol: I feel like,as a development team, from December the 15th to January the eighth, should just be one giant hardening sprint where all you do is just clean up bugs and there's no deadlines, there's no commitments. You just make everything better. You fix any processes that are broken, like that should be just hardening.
[00:13:47] Carol: 'cause a deadline isn't gonna be met. Like you have to hand off all your projects. Like I had to hand off so many work items to only come back to find out someone thought they were already done and didn't even work on it. And I'm like, what the heck's going on now? And yeah.
[00:14:01] Adam: Yeah. We don't do like a code freeze in December or even like last week or two of the year, but we do have like a. all of our customers have like year end for tax reasons. hey, do make some donations, you know, make the most of your, you know, whatever you have left in your tax write-offs.
[00:14:16] Adam: and so it's a big time for all of our customers to, to collect payments. And so it's do not screw up, you know, our online giving or membership applications and stuff. So, the big important stuff doesn't get touched, but like you're saying, we do, you know, we still ship stuff.
[00:14:30] Adam: Like I, I ship those features and that was, you know, improving things, hardening, you know, adding, making things more robust, but mm-hmm.
[00:14:37] Tim: But it's so back to normal routine. This week, it's like I'm spending like all my meetings starting with, okay, where were we? Where, where did we leave off?
[00:14:46] Carol: Yep.
[00:14:47] Ben: totally.
[00:14:47] Tim: into a meeting, I'm like, okay, so last time we talked was seriously was like three weeks ago. Where are we at? Oh, we, nothing's moved.
[00:14:55] Tim: Okay. All right. I guess we're, I guess we're still okay. We're still doing that thing that we were doing three weeks ago, so, but yeah, just, it just, yeah, just that, that liminal space is, it just feels like theater for me.
[00:15:07] Carol: Yeah. I'm telling you. I'm telling you January 8th, things should pick back up on the eighth, but not a day
[00:15:11] Carol: before
[00:15:12] Adam: I wouldn't be opposed to like, all of December is just like a nobody work. You know, it's a free month off, like free space on a bingo board sort of thing. But you, you know, as soon as we did that, then the last two weeks of November would become this liminal space you're talking about.
[00:15:25] Carol: again. Yeah. Yeah.
[00:15:27] Tim: And we got customers that have so the customers are like, oh, everything's a freeze. So they like starting from December 15th to the end of the year. it's a freeze. They don't do anything. They refuse to like, they're like, no, I'm sorry. You can't, even though it's like they're are customers, but they're like, no, you can't post any changes from December 15th till the end of the year.
[00:15:47] Tim: I was like, code freeze. So,
[00:15:49] Adam: The hell I can't watch me.
[00:15:51] Tim: alright. So anyway, that's me. How about you, Adam?
[00:15:55] Adam's Failiumph
[00:15:55] Adam: I'm gonna balance it out. I had two triumphs of fail and now I'm gonna go with a emph. Nice.
[00:16:00] Adam: which is my emph, is that I didn't get nearly as much reading done as I wanted to over my Christmas break. You know, I had kind of picked out a whole bunch of different blog posts, articles, books and stuff I wanted to spend some time with, during my time off this year.
[00:16:15] Adam: And I didn't really get even like half of that done. And mostly it's because I ended up distracting myself with the part of this. That's a triumph. Which is, Convex is awesome. I dunno if you've ever heard of this. So it's a, it's like a database. So you've probably heard of Firebase, right? There's a old school Google product, which is like kind of a, everything all in one does authentication.
[00:16:37] Adam: I think Firebase even had like sessions. It had, you know, web sockets for real time data stuff, and it had database storage type stuff. But so Convex is like, if you reinvented that this year, 20 24, 20 25 ish, right? So it's built on top of, I think MySQL. I think it's on top of PlanetScale. But, you don't have to do any manual web socket stuff at all.
[00:17:00] Adam: You just use their library for, for connecting to your data. And it is automatically real time. So if I, you, I pull up a page that has some data on it, I go open the database on, on their dashboard, edit the database directly, and I go and look at the pages already updated on the page sort of.
[00:17:16] Adam: So it totally takes out the need for doing like manual custom web sockets to push data around.
[00:17:21] Adam: You just put it in the database and whoever's looking at it has the latest version.
[00:17:25] Carol: That's really cool.
[00:17:26] Adam: yeah. And it has all stuff baked in. It has sessions, it has file storage, so you can do like
[00:17:31] Tim: Is that like a, is that like a pub/sub or what's
[00:17:34] Adam: up. honestly, I'm not sure, I haven't gotten to that part of it yet. so what I'm the, I don't wanna spend too much time talking about this week and maybe dig into it later as another episode, but,
[00:17:44] Adam: Jump Run, right? So I was like, okay, I've kind of got this like perfect use case except for the fact that I don't have a whole lot of tests yet. but, I've kind of got this perfect use case where it's a relatively small application, very, I would say cleanly written code, and. so it, it was pretty straightforward to be able to say, okay, here's the current tech stack.
[00:18:03] Adam: Here's this new tool I want you to use. Here's some documentation about it. Go Right. Sort of thing. And, and that didn't it, it tried really hard to make that work, but it's, it struggled and I've been trying to like, sort of like iterate on that. Like starting from scratch. I've started from scratch like four or five times trying to like, figure out the best way to get through this process.
[00:18:22] Adam: and I'm really starting to refine it. And I feel like this last one that I've started may actually end up being the, the last time that I do this because I'm, I'm really, liking the way this is going, so I'm happy with it and, and Convex seems so far Great. I'm really happy with it.
[00:18:38] Carol: It seems nice.
[00:18:40] Ben: Sounds very cool. And I like the idea that it provides the, web socket stuff because
[00:18:45] Adam: Oh yeah.
[00:18:46] Ben: I always feel like that's something that people just sort of hand wave over the complexity of web sockets, like when GraphQL was really kind of cresting in popularity. There were, I've never used GraphQL, so I, I'm hope I'm not misremembering here, but there are a lot of people who are like, oh, and it's so great.
[00:19:03] Ben: 'cause you can just push data, like you can just subscribe to changes in the GraphQL graph and it'll just push data to you. And I'm like, I, I feel like that's not as easy as you're making it sound. I feel like that's probably gonna be really complicated, really fast.
[00:19:15] Adam: That was the story for everything. GraphQL,
[00:19:18] Ben: Yeah.
[00:19:18] Adam: it sounds really simple on the surface until you get down into the details of any one feature of that. Yeah.
[00:19:25] Ben:
[00:19:25] Ben: Firebase, I think when people talk about agentic coding IDEs like Cursor and Anthropic, I think Firebase also has like a Firebase code console or something that's supposed to be agentic code
[00:19:42] Adam: So I, I, I would not be surprised at all to hear that. I, I actually do happen to know Convex has their thing. I don't like the name that they've chose for it. They call it Chef, it's their own little AI agent. where you can just go in and say, okay, write me an app that does this, whatever, like a v0, bolt.new or any, you know, lovable any of these tools, but however they've trained it or their system prompt or whatever, like it has Convex tooling down pat, right?
[00:20:06] Adam: You know, so many, you know, you go to Lovable or v0 or whatever and you're like, okay, write me an app that does this. And usually the first couple of things that it trips on are like, okay, well I need to create a, a database. So it's like, okay, I know Supabase, so go create yourself a Supabase database and pasting your keys here and we'll work with that.
[00:20:22] Adam: Or, you know, whatever off library it's doing. And all of that is built into Convex. So it just like plows through it. It doesn't stop. There's no tripping over trying to get it configured and struggling with it or anything like that. It just goes. So that part is really interesting to me. I think if I find myself starting up another side project and I'm like, I might think very hard about using Convex as the starting point or the, they call it chef, Convex chef to start for that.
[00:20:48] Carol: Wasn't chef already a thing though,
[00:20:50] Adam: It is. And that's why I don't like the name. it's a, it's like
[00:20:54] Carol: right?
[00:20:55] Ben: and puppeteer
[00:20:56] Carol: puppets here. Yeah. They're like, IC like infrastructure is code, right? Like you,
[00:21:00] Adam: Sort of. Yeah. It's, it's like orchestration
[00:21:03] Ben: It was like pre docker and
[00:21:05] Carol: Oh man. Yeah, I think people still use it. I mean, somebody has to use it. 'cause once it's in there, some people can't get
[00:21:11] Carol: rid of it.
[00:21:13] Ben: I just wanna say thank you, chef.
[00:21:14] Adam: Yes, chef.
[00:21:15] Ben: Yeah,
[00:21:15] Carol: Yes, Jeff.
[00:21:16] Adam: Well, so that's it for me. so that brings us back to, you know, ring the bell. Ben's getting in the ring here with 2026, trying to knock it out.
[00:21:24] Ben versus 2026
[00:21:24] Ben: to set the scene, 2025 was a hard year for me. I think it was probably a hard year for everybody, so I don't wanna go too deep into issues, but yeah, I don't, I don't want to be a, you know, sad violin, sad clown of life kind of person. but what I'm looking forward to is having, a better sense of control over how 2026.
[00:21:44] Ben: Plays out. And specifically in this show, I wanna talk about the professional aspects of that and the kind of academic and personal growth learning opportunities. And, obviously AI is gonna be a big part of this because I am typically a laggard on most things. I fear change and, I'm slow to adopt new technologies, sublime tech's lover over here still to this day.
[00:22:09] Ben: And, just been doing a bunch of thinking about how I want to best allocate the limited amount of time that I have
[00:22:17] Ben: And I was listening to a, the, I'm pretty sure I've mentioned the podcast People I Mostly Admire.
[00:22:24] Ben: It's been running for about five years and it just had its last episode. Steven Levitt, who was the host, he's going on to do some other stuff. But he was talking, about his experience in this podcast and over his career. And he was talking about how when he had to learn mathematics in college, he hated it.
[00:22:44] Ben: he just tried to muscle through it as best as he could. He didn't do so well and he forgot all of it when he was done. And then he goes on to become a very successful economist and he got so excited about economics that he had to go and learn a lot of math to understand how the economics plays out.
[00:23:00] Ben: And he was so excited about economics at the time that he found the math thrilling and he found, being able to learn the math concepts. Very effective. And he was able to retain a lot of it because he was so excited about it. And he kind of referred to these two different contexts of learning as just in case learning and just in time learning.
[00:23:19] Ben: Just in case learning. Being that you're learning something. if you're in college because they're forcing you to or because you think one day it might be helpful, you're learning it just in case you need it at some point in the future. Right? I'm learning long division 'cause maybe I have to do, dosing calculations in a hospital and we've lost electricity and I don't have my phone.
[00:23:39] Ben: just in time learning is learning just in the moment that you actually need to be learning that thing. You've come up against a problem, you don't have the information that you need and you have to go learn about it, you know? And I think this is how a lot of learning that I do happens. if I look at the things that I write about on my blog, I don't know, a solid 80% of them are the, oh, I had this problem yesterday, I solved it, and here's how I solved it. And that's sort of how I concrete it in my head. So I don't, I think just in time is the most effective way for me to learn personally, but I don't think that negates the, just in case learning, if I think back on my career over the long term, there have definitely been times where I'm curious about a topic like ColdFusion, like HTMX, like angular, like MySQL, like CSS, like JavaScript.
[00:24:34] Ben: Well, I'll just go and I'll get a book and I will read the book, or I'll go to the documentation site and I'll just read through the whole documentation. And maybe it takes me a couple of hours, maybe it takes me a couple of weeks. I had a, ColdFusion MX Bible. It was like three inches thick and I literally woke up every morning and I read until work started for weeks.
[00:24:52] Ben: that's just in case, you know, I don't need all the stuff in that book, but it helps paint the landscape of possibilities in a particular area. I might forget a lot of details, but I can remember that things may have existed and it kind of gives me a foothold into to
[00:25:06] Conferences for Exposure
[00:25:06] Adam: Yeah. I'm gonna, I'm gonna jump in here.
[00:25:08] Adam: Uh, so I, as you've been describing this whole, like, just in case learning paradigm, it was making me think of another example that I think rings true for me, which is conferences, right? Like, I go to learn, but I'm not going to memorize, you know, stuff and be able to come back and whip out that code and, and just put it to good use.
[00:25:28] Adam: It's more about learning that something exists or I, you know, I meet somebody there and I'm like, okay, Ben is an expert on CFML, so I'm gonna make sure if I have a CFML question, I'll look at his blog, right? That sort of thing.
[00:25:40] Ben: Yeah, a hundred percent. And, and if I could say that, that was actually a very transformative understanding for me when it comes to conferences because. At first when I went to conferences, I was approaching it as a, how do I take a skill that I already have and harden it by going to particular talks. And it's not like I was always disappointed, but I'd go to a talk and be like, okay, I knew most of this stuff already, and when I flipped it to what you're saying, where actually what I'm really supposed to be doing is going and exposing myself to stuff that I don't know about, then I'll skip over the talks now for things that I feel like I probably am comfortable with.
[00:26:18] Ben: And I'll go to the, I don't have any comfort here, or I've never looked at this subject, or, you know, I don't know anything about running a consulting business or you know, something where I know that it's about the exposure and not about the hardening of an existing skill. that was pivotal for me, which, embarrassingly enough took me many more years to figure out than I would like.
[00:26:39] Adam: We're all on our
[00:26:39] Ben: Yeah, on our own journey. Exactly. Okay. So, so I have these two types of learning just in case, just in time. the, just in case stuff I think is more just about catch up and I, I like the idea of putting some effort into it, but I, I don't wanna put too much effort into it because again, I don't think it's the most effective type of learning.
[00:26:57] Ben: But one of the big things that I want to do is catch up on the web platform. The web platform itself. CSS and JavaScript have just moved forward pretty aggressively in the last two years.and at InVision, I, not until the very end necessarily, but for a large portion of that, I had to. IE 11 compatible in the code that I wrote.
[00:27:23] Ben: So I knew that there was stuff that existed, but I was almost comforted in that I just couldn't learn about it yet because I couldn't apply it. So I knew it wouldn't be valuable use of my time. And I'm no longer constrained by that. There is new, wonderful stuff that I can theoretically be learning about and using.
[00:27:40] Ben: So what I wanna do is Google has,the, what I think they call it, the baseline project, where they have their modern browsers, which I think is, you know, Firefox, Chrome, safari and Edge. Safari being both mobile and desktop. And they have their baseline newly available, which is it's just landed in every single browser.
[00:28:00] Ben: and then they have their baseline widely available, which is that it's been in every browser for 30 months, giving it, you know, time to really propagate and. I don't know if there's a list. Maybe I'm, I'm hoping there is, or maybe there's something ChatGPT can give me. But I, what I would love to do is look at a baseline widely available.
[00:28:18] Ben: So this is stuff that's truly been saturated over the last 30 months, and then almost look at a time-based release. You know, what is the widely available stuff that came out in 20 25, 20 24. And I really want to go and just learn that stuff because I know that I can start using that at work today and mostly get away with it.
[00:28:37] Ben: things like the dialog element, which, you know, people have been talking about for years, but it's actually only been in browsers widely available for 36 months. So it's actually just barely over the, the widely available, but widely available enough to embrace it. And I think, you know, that'll just have compounding effects because, not compounding effects.
[00:28:55] Ben: That's like a compounding value add because then I can use that in every project going forward. It's not even technology specific per se. The other, just in case stuff I want to learn about, I think is, I'm on Microsoft SQL Server now and like I can skate by because MySQL and SQL Server is a huge Venn diagram of overlap.
[00:29:15] Ben: But there's probably some SQL Server stuff that I should know about that that isn't MySQL specific. And this feels like one of those things that I'm just not gonna know it exists until I actually go and read a book about SQL Server and figure it out.
[00:29:26] Adam: So you, you've been rattling on about a lot of stuff that you wanna learn for. I don't even know how many minutes now.
[00:29:33] Ben: Sorry. Yeah, jump in
[00:29:34] Adam: No, no, no.
[00:29:34] Learning and Marketability
[00:29:49] Adam: It occurs to me. It occurs to me. Ben, this is a lot of homework you've assigned yourself. And I, I just wanna ask you, is learning like the, the types of stuff that you're talking about here, is this something that you make you feel, makes you,energized?
[00:29:50] Adam: Does this, does this make you feel good? Or does this feel like homework that you're like not exactly excited to do?
[00:29:55] Ben: learning energizes me when it feels like it's going to have a value add. And that's where some of the contention comes from with, with the AI stuff, which we'll get to in a second. which is where it's like the baseline widely available stuff for CSS and JavaScript. To me that is of an obvious value add because that will almost certainly simplify the way that I build web applications, which is what I do all day. You know, if I can replace some JavaScript with CSS or if I can replace some JavaScript with native HTML attributes, like that's just less to do, less to maintain, more ability to lean on the platform. I'm excited about that. And you know, for the last couple of years I have felt guilty's not quite like the right word, sheepish maybe.
[00:30:45] Ben: Like, I think I have hidden behind the constraints of the products that I was maintaining and allowed myself to not really care about the most up-to-date standards. But now that I feel like I can't hide behind that, I have to face my demons a little bit. And not to say that learning is negative, but it's you know, there's
[00:31:04] Adam: it was an easy excuse to not, yeah.
[00:31:07] Ben: exactly. So, I I, I do get excited about that stuff where, where the, you know, connecting the dots on effort and return on investment feels. It's a fairly straight line, you know, I learned something new and I can sort of immediately apply it. It's not a lot of hemming and hawing about how to best use it, that kind of stuff.
[00:31:28] Ben: so that's the just in case stuff,
[00:31:30] Carol: I wanted to follow up to kind of what Adam had said. Do you, so, you know, he was asking about like your passion here, right? So do you feel like your, just in case learning has any play into your marketability as an employee? is that something that even goes into your head?
[00:31:44] Carol: oh, if the market does this, is this gonna make me more marketable? Or is this just for you and what you're doing now?
[00:31:50] Ben: I think, yes, and, I, I, I think it's, I am a web developer and learning about the web is just part of the process of becoming better at, you know, becoming more of a craftsman. But I also am very pragmatic in that, there, there's a marketability to it. I mean, I think I'll be able to do a better job where I am right now, but my attractiveness is also going to hopefully increase.
[00:32:21] Ben: Or let's say I'm talking about it like it's a sure thing. I wanna be, you know, the, the hot girl at the ball. And I feel like I can't do that unless I learn the latest stuff. So, so it's, it's, you know, a little bit of it is vanity and wanting to be attractive, and a little bit of it is, I know that this is actually just gonna make me a better development.
[00:32:41] Carol: You are my hot girl, Ben.
[00:32:43] Ben: Thank you. I appreciate that. I'm blushing. All right. So that, that the just in case stuff, I feel like that's something I can just sort of dribble in effort when I feel like I have the time, or almost like a pallet cleanser, I can probably just come up with a checklist, right? I mean, that's the nice thing about.
[00:33:00] Ben: Some of the just in time stuff is, it's, it, it can be codified. here's the list of features that have been added to the web platform in the last two years. just go learn. That's not, that's not super complicated. I like that. I like it. It, it's, it's codifiable.
[00:33:14] The AI Elephant in the Room
[00:33:14] Ben: The, the elephant in the room for me is ai.
[00:33:18] Ben: I use AI in a kind of conversational respect. I have a question. I go to ChatGPT. I type it in. I argue with it for 20 minutes about formatting SQL statements and you know, like that's where I get my value. apparently I'm missing out tremendously by not having agentic coding in my life. Sean Corfield has, has raved about it.
[00:33:39] Ben: I know Adam Cameron has raved about it. I've seen a bunch of people rave about it. I've heard,
[00:33:44] Adam: I, I don't remember Adam raving about it per se. I would imagine he uses it, but
[00:33:48] Ben: you know, like Adam Cameron level
[00:33:51] Ben: raving.
[00:33:52] Adam: Well, yeah. Yeah. So him, him saying, actually, I don't completely hate it. That's him raving about it.
[00:33:56] Ben: It's not like beer level raving, but it's this person didn't irritate me level ravings.
[00:34:01] Adam: Yeah.
[00:34:02] Ben:
[00:34:02] Tim: if he doesn't actively trash on it, he likes it.
[00:34:10] Ben: but you know, I, as much as I have my whole life hated on people for watching reality television and like wanting to be like the Kardashians, I feel like I have almost become my own victim of tech podcast listening. I listen to these podcasts, these people are talking about how great AI is and how it's revolutionizing everything.
[00:34:32] Ben: And I feel like I have now fallen to the cliche of what is like comparison becoming the root of all evil and all I feel like I'm
[00:34:39] Ben: doing is com The thief of joy.
[00:34:41] Ben: Thank you. And it and it, and it 100% has. Stolen so much of my joy out of this last year, I just constantly am concerned that I'm falling behind, that I'm becoming irrelevant, that I'm gonna be replaced, that I'm not as effective as other people.
[00:34:56] Ben: And whether or not that's true, I do want to do something about it this year. and I think that's where the just in time learning is gonna come into play, because I don't think I'm going to be able to effectively learn about agentic coding from, you know, one page demos and little white page, you know, isolated proofs of concept.
[00:35:19] Ben: Like I might with a JavaScript, you know, if I'm gonna learn about the dialog element, I can probably do that. You know, just like this nice little isolated, here's a button. It opens a dialogue, it makes the rest of the page a nerd traps mouse focus. okay, that, that, that can be small and it can be tiny and isolated and comprehensive, but I can't.
[00:35:36] Ben: I can't, I don't think, use that same kind of a strategy to understand how to take an agentic coding product and help me become more efficient. Because every project I'm working on is gonna be a little different. They're all gonna be in various states of disrepair. Some of these code bases are gonna be 10, 15, 20 years old.
[00:35:57] Ben: Some of them are gonna be greenfield, some of them gonna gonna use MySQL. Some of them are gonna use JSON files as data persistence. You know, some of them are heavily tags, some of them are heavily scripted. Like it's, I, I think in order to actually get value out of a agentic coding, I'm going to have to do it in a real world scenario, on a real world project.
[00:36:19] Ben: Whether or not it's a professional project versus this high project, I think that's less relevant, but that, that's gonna be, I think my big struggle this year is trying to find the ways to. Make AI part of my life, and I'm very nervous about it, to be frank. I, I'm like, I can't quite decide if it makes more sense to start with a greenfield project where I don't quite know what the code is gonna look like and I don't know what the architecture's gonna look like and maybe involve some technologies that I don't know yet, and kind of let AI help me through that journey.
[00:36:55] Ben: Or if it makes more sense to apply it to a Brownfield project where it has a tremendous amount of prior art that it can lean on. And there's less ambiguity about the architecture, less ambiguity about the approach. But my fear there is I'm gonna be so heavily married to my preconceived notions about what that project should be, that I'm gonna spend so much time just fighting the ai, trying to get it to do the thing that I want, and I'm gonna end up.
[00:37:29] Ben: Talking myself out of using it because it's not gonna get the line breaks in the right place and it's not gonna get the quotes
[00:37:35] Ben: in the right place. But
[00:37:36] Ben: you know, when you have a project that's 20 years old, like that's, that's part of the ethos of that project. And then I'm also doubly worried that I'm going to be forced to take on solutions to solve problems that I don't have.
[00:37:52] Ben: And what I mean by that is I care deeply about the way I write the code. That's the joy of writing code to me, is about evolving the mental model and the strategies and the, and the approaches. So things like linting, things like testing, things like, I guess linting and testing is really the big ones there, but it's like,like I don't, I feel like I don't need those as much, and that's not a, that's not like a humble brag.
[00:38:17] Ben: I don't make mistakes. It's, it's more like a, I think so much about every line of code that that stuff is part of that process. And I'm worried that if I go to someone and I say, Hey, I can't get the AI to format the code the way that I want it, and their answer is gonna be, well just run the code formatter afterwards.
[00:38:37] Ben: And I'm gonna be like, but I don't have a code formatter because I've never needed it. And they're like, well, now you need that because the AI sucks at formatting. So now you have to add all these solutions to the problems that you don't have until you took on the, you know, the journey of AI to maybe solve the problems that you didn't have in the first place.
[00:38:54] Ben: And so that, so obviously I'm getting very angry right now, even just talking about it.
[00:38:58] Carol: Yeah, I mean, you're starting to scare me away from AI right now.
[00:39:02] Ben: And, and, and maybe none of this is true. You know, this is all the, the, the fear that I have in my head. This is the story that I've told in my head, not based on my personal experience, but this is the, this is the concern. I don't know if this is a true story or not, but apparently Frank Lloyd Wright, you know, a famous architect, had built this house, designed it, built it, and had sold it and zero unit desk.
[00:39:29] Carol:
[00:39:29] Ben: and the person who owned it was having a dinner party and it started to rain outside and water collected on the roof. 'cause it was a flat roof and it didn't drain properly or something. And it started to leak. And when the guy complained to Frank Lloyd Wright, Hey, my roof is leaking.
[00:39:45] Ben: Wright's response was, can't you just move your dining table? And, and, and like.I know this seems like a non-sequitur there, I don't ever want to be told that the reason this isn't working for me is because I'm holding it wrong. You know, like I've, you know, I've developed a way of holding this hammer over 20 years.
[00:40:03] Ben: Not because I am holding it wrong, but because it has solved all of the problems that I've needed. And now you want to introduce a new type of nail and you're like, oh, well you can no longer hold the hammer that way. And I'm like, well, can't I just use her? The old nails? That, that, okay. That was a terrible analogy.
[00:40:19] Ben: No, it made sense. It made sense.
[00:40:20] Ben: I don't think anybody's telling you, you can't continue to code the way that you currently are. I think the, the, the value proposition that's in front of you is that if you code in this particular way, right? If you have tests that are really, you have good coverage and they're that test the right things and you use a linter and formatter, then you can.
[00:40:43] Adam: Put those tools to good use to, increase your velocity, right? You can get more done in a day because you've got tools that benefit from these investments you've made in linting and testing, et cetera. And so you basically, to a some extent, we, we'll, we'll hand wave this part away for the purposes of this conversation, but you can just say, okay, here's the feature I want you to build, you know, run the tests and, and run the code format or between every, you know, feature, between every iteration that you attempt to make or whatever.
[00:41:17] Adam: and it will do those things. And you can, you, you, you again have to be very careful. You have to be very specific. don't change the tests to make the tests pass, change the code, but that sort of thing. yeah. So it's not, and I guess the value proposition is that if you code this way, potentially you can go.
[00:41:36] Adam: Significantly faster than you already are.
[00:41:39] Adam: So,
[00:41:39] Adam: you know, you could still hit the, hit the old nails with the old hammer the old way. But meanwhile, you know, the, the new kid on the block is gonna be using the new hammer, the new way, and he and the new nails, and he is gonna be putting up five walls for every wall you put up.
[00:41:54] Carol: So one thing you said, several syntheses ago now was that trying to figure out where you put this right between your new and your old project, I hit that roadblock to you. It felt so scary to be like, Hey, I don't even know where I wanna take this, and I don't want you to drive my decision.
[00:42:10] Carol: Like I'm smart, I know what I'm capable of and I wanna do things the way that makes sense to me and I wanna do the research. So I started with old projects. I started going with, first I just want you to create me two things, generate the agent, oh, I guess three things. Generate the agent. I wanna read me about this project, and then I also want you to write me an architecture like guideline for this.
[00:42:31] Carol: Like just generate these things so I can see how you are understanding these projects. Then from that point, I was able to give it very like granular level features to go design, and then I handheld it a lot and once I started feeling more comfortable with the small things, I've just slowly increased to larger things.
[00:42:53] Carol: oh, we have this new app we've spun up. There actually isn't any auth between the two environments. Like I need there to be actual auth. I understand it's just internal, but to pass security, I need to use like our identity provider to do this stuff. Go in and start doing it. And then once I have some like steps, I'm like, cool, I like a few of these.
[00:43:11] Carol: Let's just start with one. Let's start with two. Let me see what you're doing. And I kind of got to a point where I can trust a little better, but I couldn't trust anything until I saw what it thought a read would be and until what it thought the architecture of the system was. And once I kind of was in agreement, I'm like, all right, I guess you can do a little more work.
[00:43:31] Carol: Like I'll treat you like you know a little bit more, but I was in the same boat as you of, I was like, there's no way they're gonna get it right. There's no way I can trust this. Like I know what I'm doing. It doesn't know what I want to have, like what I want the outcome to be. So I think you can get there.
[00:43:47] Carol: I think you're just gonna need to take some baby steps and be okay with baby steps.
[00:43:52] Tim: Yeah, I, I, I agree with what Carol said. I, I think once you start experimenting with it. You talked about Greenfield versus Brownfield and what I actually found so Greenfield, it was great. I took the same approach Carol did. It was like, alright, explain to me, here's the project. Explain to me how you're gonna do this.
[00:44:11] Tim: And I'm like, okay. And I correct a few things that, right, that makes sense. I like your approach. Go do it. And it did it, it was fine. But, but with Brownfield, what was amazing to me was, it was very, AI was very good. And I'm using Claude Code in this example. Very good at looking the existing project, figuring out this is a project that I, I'm the only person who's ever worked on it.
[00:44:36] Tim: So I have a very, what's idiosyncratic?
[00:44:40] Tim: A, a very, a very distinct way of writing code. Right? The way I do things is, is it's very
[00:44:45] Tim: specific to me.
[00:44:46] Tim: You people look at it, they go, Tim must have wrote this. It was extremely good. Like figuring that out and extrapolating that. Doing. I'm like, 'cause it was building some API stuff and it's, it looked at previous examples of how I did things in the project and then built code that kind of matched the same format, which I was super impressed with.
[00:45:09] Tim: So I, I think really just rather than fearing it, just play around with it. Right. Just, just experiment with it. And then once you do that, you're gonna get more and more comfortable with it. And the AI you're using today is going to be the worst AI you ever use. 'cause
[00:45:25] Tim: tomorrow's AI is gonna be gonna be better every three months.
[00:45:27] Tim: Yeah,
[00:45:28] Carol: Every three months is new. Yeah. Or, or you're gonna be like our normal Ben and you're gonna hate it for the rest of your life. So that's okay too. We still love you. We
[00:45:36] Tim: but I don't, but I, I don't have as huge hangup about, you know, code formatting as you do. I don't really care about that. I, I just care if it works. I know that you're very, you, you're very specific about that.
[00:45:49] Ben: I just
[00:45:50] Carol: like for me, I'm not as picky as Ben, right? But I need to be able to read it. And it's things like if I have four parameters, they're all new lines. If there's three, it could be one line. That's totally fine. You have four new lines for each one. I can't handle it, just change it. But I was able to tell it like, I need four is your max.
[00:46:10] Carol: You can't do more than three. Because I do everything in small windows and I wanna be able to see everything and it quickly adapted.
[00:46:17] Tim: Yeah.
[00:46:18] Ben: well, I know when Adam was talking about his Jump Run stuff, 'cause I know early on he was, you were doing some agentic coding stuff and we had talked about this I think just a couple of weeks ago that. A lot of the success that people are seeing now with the agentic coding is this sort of two phased approach where phase one is come up with a plan and then phase two is start to attack the parts of the plan when we were talking about that the other week, I was saying that part of my concern with that approach is that oftentimes when I start to write the code, I don't know what the code's gonna look like. It's like I'm exploring and identifying patterns as they're happening and I'm doing like just in time refactorings where they make sense.
[00:46:58] Ben: And one thing that you said, which I, I have not internalized yet, but I think it's gonna be part of this process, is that forcing yourself to actually think about a plan instead of just YOLOing it, and a assuming that I'm gonna find the patterns as they emerge is helpful. And I, and I think part of my fear is that that's gonna be a lot harder than, than.
[00:47:22] Ben: I think it might be,
[00:47:24] Tim: I, I actually find that I think AI does a good job of very quickly and very cheaply YOLOing for you, right? Because that, that's the biggest problem for me is like just the blank page problem. If, if, if I see a good example of something, I can quickly say yes or no, that that's not right. This is right.
[00:47:44] Tim: It's, you know, but just that first step of creating something, sometimes it, it's just a huge amount of friction in my mind. So I, I think that is very valuable. Just the ability for AI just to throw out something there, it's like, it's reasonable. You're like, okay, I can work with this.
[00:48:05] Ben: hmm.
[00:48:05] Adam: The other thing I'm thinking about here too is you know, if your goal is to learn, especially with the like just in case thing, but just in case learning, for me, I try to have a very brief list of takeaways, right? these are the topics I need to remember so that if it ever comes up, I go, oh yeah, I learned about that in that thing, and I can go back to that.
[00:48:22] Adam: And I think you can leverage LLMs to. Kind of make that easier for you. You can like, okay, here's a blog post. give me the three key takeaways of this, you know, one sentence each. Right. And, try to kind of commit that to memory or, or whatever. that sort of thing. like we're talking about in AI conversations before, you know, instead of having the AI do the work for you, have it teach you so that you can do it.
[00:48:47] Ben: Absolutely. I mean, that's basically the conversation I was having with ChatGPT about the SQL stuff. It's like I knew how to do it in my sql. Show me how to do it in SQL Server.and it was, and I think the arguing with it was fun, but it was, you know, mostly it was like,
[00:49:01] Adam: Something we didn't comment on when, when you were talking about that too, was that it seems like that's very different from what we would've expected toward the beginning of the year. Would've been like, oh, you're so right. You know, the
[00:49:11] Ben: a hundred percent. I was totally thinking that even as I was having the conversation. 'cause it really kept pushing back. I mean, I was really arguing
[00:49:19] Adam: have been really wrong.
[00:49:23] Carol: Oh man.
[00:49:26] Tim: That's funny.
[00:49:27] Carol: Breaking of the year.
[00:49:29] Tim: There we go. We broke 'em. Good job, madam.
[00:49:33] Ben: Yo, for real.
[00:49:35] Tim: so
[00:49:37] The Emotional Toll of Shutting Down
[00:49:37] Tim: uh, a kind of slight topic change but on still on brand. You talked about how bad 2025 was for you and I, I get that it's a struggle that I'm, I'm kind of having now as well. So you basically, last year you, the company shut down, you were
[00:49:57] Tim: working for, right? Yep,
[00:49:59] Ben: dog died. New job.
[00:50:01] Tim: yep. And I'm kind of going through that now.
[00:50:04] Tim: So our payments side of things, I, I'm moving away from that 'cause we're moving that over to another company, so we're not actually gonna be doing payments anymore. and I gotta tell you, it's, I've realized something about my personality, and maybe this is true with you, Ben, but I've always been in the business of creating things. And the second you transform into, okay, we're gonna shut this thing down, it's been a real emotional toll on me.because that's not my mode. My mode is I wanna make new cool stuff, but to just manage something while you take it off life support and just slowly kill it.
[00:50:46] Adam: But is it making way for you to spend more effort on something that is rewarding?
[00:50:51] Tim: So, I mean, I have another, like, so basically I'm wearing two hats. I'm in two different companies. So one company, we're killing the other company. I'm director of development. Right. But it's still, I feel like every day I'm dying by a thousand cuts. You know what I mean? it's really hard to be invested in unplugging something and disconnecting something and handing it off to someone else.
[00:51:15] Tim: Right. And so it's it, it's been depressing actually. And so I can't even imagine what it was like for you, at InVision where a company, I mean, you've been for, for decades,
[00:51:26] Tim: and, and just seeing that go away, that that's, that's gotta take a, a pull on your soul.
[00:51:33] Ben: it was a lot.and to be completely transparent, the, the whole AI slash, you know, downturn in the tech sector slash ColdFusion is legacy technology. It's like there was a whole confluence of things, not only was it the closing of InVision, which was an emotional toll to your point, like it was a thing I had created.
[00:51:57] Ben: I'm a creator, I was a subject matter expert, and then all of that was taken away. And then combined with, am I even relevant anymore? Will I ever find a new job? Et cetera, et cetera. Will it be using things that I enjoy? Will it be building stuff that I want to build? You know, that, that was, that was a huge, adjustment and it's still an adjustment for me right now.
[00:52:22] Ben: I'm still navigating.
[00:52:25] Tim: Yeah,
[00:52:25] Carol: I'm gonna remind you, you're my girl.
[00:52:28] Tim: You're hot. You're the hot
[00:52:29] Carol: you, you're, you're the hot girl. You're my girl.
[00:52:32] Tim: Yeah. 'cause that's one thing I think people think programmers are like all these super analytical kind,you know, Mr. Spock on Star Trek, right. Kind of people. But really I think developers are more, they're, they're creatives, right? They,
[00:52:46] Carol: Very creative. Mm-hmm.
[00:52:47] Tim: doing code and creating code is, is a very creative process.
[00:52:50] Tim: And that's, I think what most of us get our joy out of, of creating something
[00:52:54] Tim: something that didn't exist before, something new. And if you're not in the mode of creating, and it's kind of like, you know, our finance people are super happy with what we're doing. The finance people love this.
[00:53:07] Tim: They're like, oh yeah, this makes so much sense. The balance sheets, they're, they're doing the numbers. yeah, this is, you know, we're for financial reasons, this is all gonna be great. And they're super excited. But pretty much every developer on our team is like, what are we doing? We're like, we're killing our baby.
[00:53:21] Adam: Yeah.
[00:53:22] Tim: And that's just sort of the feeling that you feel. And, and so I, I, I think people don't realize that programmers are really, they're creative, I don't wanna say artists, but at least artisans.
[00:53:36] Adam: Mm-hmm.
[00:53:37] Tim: and so when, whenever your work doesn't really, is no longer creating stuff, just either maintaining a status quo orwe're, like, you and I did, Ben just, turning the lights off as everybody leaves.
[00:53:51] Tim: it's it's kind of depressing.
[00:53:53] Ben: it is. I'm glad that, I'm not alone in the struggle.
[00:53:57] Tim: Definitely not, because I thought this what the episode would be about. What you talk to earlier about. I was like, yeah, I, I could totally feel you. 'cause it's like, yeah, I've realized that about myself now I'm like, 'cause I'm trying to disconnect from this whole process. Of doing this, but I have to do it right.
[00:54:12] Tim: I have, my boss has said, we're doing this. Okay, we're doing this. But it's at the same time, it's like my soul hurts.
[00:54:20] Creating for Someone
[00:54:20] Ben: Yo. And, and if I can, yes. And I think not just creating something, but creating something for someone is the thing that is most fulfilling for me. I like where I work right now, but I am not building something for someone yet. like we're building a product, but it has not faced a customer.
[00:54:45] Ben: Whether, you know, our customers are gonna be internal, it's gonna be other people at the company, I have not yet had a chance to put something in front of somebody and say, use it and tell me how I can make it better for you. You know, what are the problems that you're facing that I can solve for you?
[00:55:00] Ben: I have not yet had an opportunity to solve a problem for somebody. And I think that's, that has been the most draining of this whole transition. It's I have not found that place yet.
[00:55:11] Tim: Yeah, I
[00:55:12] Carol: Is it on your roadmap?
[00:55:13] Adam: it's gonna happen. We're ready to pop. But, it's just we're just not there yet in, in the team that I'm on. Oh, man. Why don't we just wrap it there.
[00:55:23] Ben: yeah. Or, or so you've, you've brought the mood from an 11 on New Year's Day down to about a 0.2.
[00:55:31] Ben: alright. But, but quick, quick consensus. So you think. Greenfield versus Brownfield. What is the consensus on agentic coding? Should I start with a totally new thing or start with something that exists and not
[00:55:43] Adam: I agree with Tim. Yes. Yes.
[00:55:45] Tim: and Yes don't listen to them. They're idiots. Okay. I completely completely think you, I love you guys. You, I think if you want an honest answer from me, I would tell you to take the thing you already know, because that's gonna be where you can iterate the most. If it's something brand new, you're not gonna trust it yet.
[00:56:05] Carol: Like you're gonna still have your own views, you're gonna wanna develop. So if I were like telling you day one to start, I'd say take something you already know. Just go at it, just see what happens, and then iterate on and then turn to the new stuff and say, okay, from what I've learned with my existing, how can I leverage it for this new,
[00:56:22] Adam: Okay, that is, I'm sure, very valid advice. Now, I'm gonna take the exact opposite.
[00:56:28] Carol: this is why you are not like my favorite girl, but go ahead.
[00:56:33] Adam: I can live with that. I'll be over here in the green dress. anyway, I, I, I don't disagree with Carol. I think everything she said is true. It's a great way to get your feet wet with AI in general. But I think that my view is that, the way you use it with Brownfield and Greenfield is two different skills and you have to learn both.
[00:56:55] Ben: Oh, oh, that's, that's an interesting take. Okay.
[00:56:58] Carol: That is actually okay. You're not so bad.
[00:57:03] Adam: I just took a step closer to the
[00:57:04] Adam: Punch bowl.
[00:57:05] Ben: All right. All right. I think let's, let's leave it there then. That's a, that's a great final thought.
[00:57:10] Patreon
[00:57:10] Adam: Okay, cool. Well then this episode of Working Code is brought you by depression. Are you even relevant anymore?
[00:57:15] Carol: Yes.
[00:57:16] Adam: And listeners like you, if you're enjoying the show and you wanna make sure that we can keep putting more of whatever this is out into the universe, then you should consider supporting us on Patreon.
[00:57:24] Adam: Our patrons cover our recording, editing and transcription costs, and we couldn't do this every week without them. Special thanks to our top patrons, Monte and Giancarlo. You guys rock.
[00:57:33] Thanks For Listening!
[00:57:33] Adam: we're gonna do go do the first after show of the year if you're new here and you don't know what the after show is yet.
[00:57:37] Adam: Basically it's part of the podcast that you don't get unless
[00:57:40] Carol: Oh,
[00:57:41] Tim: That you're paying.
[00:57:42] Adam: Unless you're a patron, patrons get early access to new episodes and the after show, it's just more of us bantering. We're gonna talk, I think it was Carol wanted to talk about Las Vegas. I wanna talk about, I, I, it's not a big thing, but I wanna talk about my, my cheap knockoff Chinese apple pencil.
[00:57:58] Adam: somebody is in here typing as I speak that they wanna talk about a Netflix CTO interview.
[00:58:03] Tim: I wanna talk about pluribus.
[00:58:05] Adam: okay. Oh, well, no, no, no, we can't. 'cause I haven't started watching it yet.
[00:58:08] Adam: no no spoilers. No
[00:58:10] Ben: There's no way to talk about that show without it being spoilers.
[00:58:13] Adam: All right, well
[00:58:15] Carol: That's a spoiler
[00:58:16] Carol: Ben.
[00:58:16] Ben: right. All right.
[00:58:17] Adam: Gimme like two weeks. But anyway, like I said, we're gonna go record the after show.
[00:58:21] Adam: If you would like to become a patron of the show and get that, and you get other perks too, then you can go to patreon.com/workingcodepod. We'd love to have you. one of those perks you can get is special recognition in the form of a role. turns your name Gold in our Discord server. If you wanna join our Discord server, which is open to the public and free, go to workingcode.dev/discord.
[00:58:43] Adam: that's gonna do it for us this week. We'll catch you next week. And until then, yeah,
[00:58:46] Tim: Hey listeners, you'll always be our hot girl at the bar. Your heart matters.
[00:58:51] Ben: Hey, can I just say, I have missed you guys and this has been Chicken Soup for my Soul and uh, I'm just so
[00:58:59] Carol: you,
[00:58:59] Tim: you too, buddy. Love you.
next episode: 245: Browser Passwords? You're Doing It Wrong
prev episode: 243: Oops, All Aftershow