236: Trunk or Treat
In this week's episode the gather round and share what they've been up to for trunk or treat. Adam shares his waning motivation for his Jump Run side project, we explore sustainable motivation, the rewrite temptation, and whether it's okay to just... do the fun thing sometimes. Meanwhile, Tim provides a reality check on AI coding tools—he spent real hours comparing GitHub Copilot and Codex on actual work, and the results are messier than the hype suggests.
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 Wednesday.
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.
[00:00:00] Intro
[00:00:00] Adam: Okay, here we go. It is show number 236. And on today's show we're going to, do a little trunk or treat, you know, we're recording.
[00:00:06] Adam: This is pretty close to Halloween, so we're gonna celebrate. I hope you enjoy as well. We've we each brought something sweet or maybe raisins, to share and.
[00:00:16] Tim: bars all around.
[00:00:17] Adam: Sweet.
[00:00:18] Carol: For you.
[00:00:19] Adam: and, yeah, you know, we're just gonna do that. But first, as usual, we'll start with thes and fails and Benjamin, Mr. Ws, US East. One breaker.
[00:00:28] Ben:
[00:00:28] Ben's Fail: AI Personalized Content Concerns
[00:00:28] Ben: I'll, uh, I'll kick it off here with a, a failure, and it's not really a failure, let's call it growth opportunity, which is that I am, it feels like in the last year or two, just being continually reminded of how differently. People see the world and not just see it, but experience it and, and so much of everything that I felt skewed on the objective side of the spectrum is clearly just highly subjective and.
[00:00:56] Ben: For reasons. You know, the, for all I know are just how people are hardwired. And this, has come to the top of my mind just recently. 'cause I was listening to, I think it was the Pivot podcast with, Scott Galloway and Kara Swisher, which is a, it's a fun, it's like a, a guilty pleasure. and they were talking about the.
[00:01:16] Ben: AI in media and specifically this future fantasy where all of your shows will just be generated on demand for you, and all of your music will be generated on demand for you, and all of your books will be generated on demand for you, and everything will be tailor made to the things that you specifically like.
[00:01:37] Ben: And I've heard several people talk about this kind of future and they talk about it with so much excitement like, oh, how great is all the content gonna be? It's built just for you and all the things you like. And I look at that and I think that sounds terrifying because we're already in a world, we're in these weird little echo chambers, and when you can start to have content generated specifically for you, it kind of hyper.
[00:02:03] Ben: Minimizes those echo chambers. And now like you're really only in these tiny, tiny, tiny echo chambers. It's for a very niche vertical of content. And like, what do we even talk about at that point,
[00:02:13] Adam: Have we learned nothing from Twitter?
[00:02:15] Tim: Right.
[00:02:16] Ben: right? Like you, like, you know, imagine this reality is real and project Hail Mary comes out and, I haven't read the book, so I don't know anything about this. I go, oh man, how crazy was it when he was on the spaceship and like, and, and Donald Duck shows up and you're like, what are you talking about that is not in Project Hail Mary?
[00:02:34] Ben: And you're like, well, it wasn't the project Hail Mary I saw because I love Disney. And forgive me if Donald Duck is not Disney. I have no idea. But I don't know, I just, I'm, I'm so blown away at the things that get people excited are sometimes the things that I think are just absolutely terrifying.
[00:02:50] Adam: I think you're right on the money about that. At least that one piece right there. Like so much of, the reason to enjoy media is to have a shared experience, right? Like, that's why book clubs exist. That's why people go see movies together, right? Like. To be able to talk about it and, and share that experience and,
[00:03:08] Tim: it's kind of like you ever had someone who's like really excited about a dream they had
[00:03:12] Ben: Yeah.
[00:03:15] Tim: and then they try explaining the dream to you. They realize it makes no sense. 'cause it was really just for them. That's kinda like, I think the mo, if I don't believe this future is gonna happen, that they're talking about this just AI hype.
[00:03:27] Tim: But if it were to be that way, that's the way it would be. Oh yeah. AI made a movie for me, it was about this story about this dog that had hepatitis but then fell in love with this, with a skunk that had magical powers, but he didn't like the smell of the SC mean. And you'd be
[00:03:41] Ben: You are
[00:03:41] Carol: Read
[00:03:41] Ben: are you talking
[00:03:41] Ben: about?
[00:03:43] Tim: what are you smoking?
[00:03:45] Carol: Vampire Got it all day.
[00:03:47] Adam: This is like, the AI version of like, you, you realize you've changed your Instagram agro algorithm all wrong.
[00:03:53] Tim: Yeah.
[00:03:54] Adam: It's only feeding you weird stuff.
[00:03:55] Tim: Yeah.
[00:03:56] Ben: Oh man. Oh, the stuff I get in my Facebook reels. It's just, it's so strange.anyway, that, that's, that's, yeah, that's a totally different topic, but, all right. So I'm, I'm gonna call that my soft failure slash opportunity for growth in how people see the world. So I'll kick it over to Carol. What do you got going on?
[00:04:15] Carol's Win: Everyone's Back & Sourdough Journey
[00:04:15] Carol: Well, I'm gonna go with a couple wins 'cause I think we all deserve them. I don't know if Adam said it, but if you haven't heard, we're all back like it's been a little while. Like, that's a big win for us. and I have a personal win that has nothing to do with technology at all. But, about a month ago I decided I'm gonna start making sourdough bread because every Karen should be making sourdough bread or so I've been told.
[00:04:38] Carol: And so I decided to do it except. It's awful. It's terrible. It's not easy to do. It's just 50 grams. A hundred grams, a hundred grams. It should just work. Except no one tells you that in the desert because it drops to 30 degrees at night. Your kitchen gets very cold and you don't realize this. So until you buy a heater.
[00:04:58] Carol: Suddenly nothing works. So the beginning of the week I was like, I'm just throwing this out. I'm quitting. I'm not gonna get up at four o'clock every morning to make this this sourdough starter feeder. And it's like having a kid. So then I, I get the, I get the mat and it finally is working and I'm going to make sourdough bread tomorrow we made two loaves and they turned out like crap.
[00:05:19] Carol: I mean like so hard and the inside was like gum. You couldn't eat it. It was terrible. So yeah. So hopefully this one actually goes well tomorrow. So I think that's a.
[00:05:29] Adam: you said you have a mat. I, I was thinking like maybe you should put a, like a heating pad underneath of it. Is that what you did?
[00:05:35] Carol: Yeah, well, no, actually I bought a sourdough warmer. It kind of looks like a, candle warmer. but it has a timer on it for like 24 hours. So what I did was I took a pan, like a baked, a baking glassware thing and put water in it, and then I set my sourdough in it. So then it keeps the water at a temperature of like 84 degrees.
[00:05:53] Carol: And then it helps keep the, the sourdough starter at the right temperature. So
[00:05:57] Carol: yeah,
[00:05:58] Carol: I'm excited.
[00:05:58] Adam: one of the things that I like occasionally rotate through when I'm going to sleep, 'cause I, I can't sleep without something on to watch, but, is, good Eats, you know, the TV show
[00:06:07] Carol: Yep.
[00:06:08] Adam: Brown and the, the last two seasons that he did of Good Eats, there's each, each of 'em have an episode like during COV.
[00:06:16] Adam: like where he kind of acknowledges COVID and he is like in this like post apocalyptic world. He pretends the whole crew is dead and everything. it's really good. But he, one of 'em is about sourdough and I immediately thought of that and it was like, if you haven't seen it, you should watch that episode.
[00:06:29] Adam: It's good.
[00:06:30] Carol: Yeah, I would check it out for sure, but the only thing to add to the sourdough is. I also started working out again, so just in time to start making bread. I shouldn't be eating bread, so I'm like, I'm having to juggle, like how dedicated am I now to actually making the bread after this first batch, because I really don't need all the carbs, but you need food to work out for energy.
[00:06:52] Carol: So I don't know. I'll balance it, I'm sure.
[00:06:54] Adam: It's gotta substitute it in for something.
[00:06:56] Carol: Yeah.
[00:06:57] Tim: Yeah, if you can actually make good sourdough more power to you. 'cause I tried and tried and tried during COVID and it did not go well.
[00:07:04] Carol: Well, I'll, I'll post some pictures and let you know if it turns out well, what about you, Tim? What you got going on?
[00:07:11] Tim's Triumph: Portugal Family Trip
[00:07:11] Tim: Well, I'm back. You miss
[00:07:13] Carol: Hey, welcome back. We did.
[00:07:15] Tim: yeah, yeah. So Portugal is a great time. Sorry, missed you guys for the past few weeks, but, you're having a nice time with the family. So we rented a, a big house in, in, little town called Abido. So if you look at Abido, Portugal, it's a, like a medieval castle town. So it's got a wall and people still live inside the ancient city.
[00:07:35] Tim: There's a castle and a fortress up there. It's pretty cool. but the house we were at was further down, wasn't in that part. We were down by the, down by the beach and yeah, like seven bedrooms, four baths. And we had,
[00:07:49] Adam: That's a lot to fit in a van down by the river.
[00:07:52] Tim: it, and I did, we rented a very big van, a 10 seater van that we had.
[00:07:56] Tim: But yeah, we had grandparents and cousins and aunts and uncles, and. My kids and yeah. So yeah, the house was full, which was, it was awesome. So it was a good family reunion, good to see, good to see our family and nice to see them somewhere other than Ireland where it's
[00:08:13] Ben: Nobody brought back, COVID this time.
[00:08:15] Tim: no, no one gave.
[00:08:17] Tim: Oh. 'cause we had COVID like right before in September, so I think we were inoculated and nobody got sick,
[00:08:22] Ben: That's fantastic.
[00:08:23] Tim: during the time. So yeah, it was, it was nice 'cause it was three there for three weeks kind of laid back. We'd do something. Take a day to recover, do something, take a day to recover. I only had two goals.
[00:08:32] Tim: I wanted to see Centra Portugal, which was absolutely beautiful. And I wanted to buy some new shoes. 'cause we're, we were flying, carry on. I took some crappy shoes with me and I'm like, I'm, I'm not coming back with these shoes. I'm leaving these here. And like, and like when I was in Centra, we're walking along and, and I see a store and they had some leather shoes and they were handmade leather shoes, leather, you know.
[00:08:54] Tim: Made in Portugal. That's why I bought those. And they weren't, they weren't terribly expensive but nice. So I got those and Centra was absolutely stunning. If you, if you ever get a chance to go to Portugal, definitely go to Centra. It's, it's like the whole town is just beautiful. You turn a corner and it's like a, it's a castle.
[00:09:12] Tim: It's a beautiful landscape, and it's like, oh my God, this place is magic. So highly recommend it.
[00:09:18] Ben: Very cool. were there any work emergencies that happened while you were there and you were like,
[00:09:22] Ben: not my
[00:09:23] Tim: the AWS outage that you caused? Yeah. Thanks for that, Ben. Yeah. That ruined my day.
[00:09:28] Ben: Oh, no.
[00:09:29] Tim: Yeah. There were a few work, there were a few work things. We, we were in the process of transferring some folks to different backend processors and they did not go well. I mean, I don't know if me being the states would've would've, I don't really do hands on keyboard a whole lot with that kind of stuff.
[00:09:44] Tim: I'm more kind of quarterback the direction, so maybe it would've gone better, would be there. But I, I don't know. I'm on a plane trying to manage one of these, conversions, you know, in the plane. So that was, that was stressful. So I'm, I'm glad to be back in our time zone 'cause by the time everyone got started working, it was like 5:00 PM for me and I'm ready to knock it off.
[00:10:05] Tim: And,
[00:10:06] Carol: Yeah, that's hard.
[00:10:07] Tim: been working all day, so. But anyway, I'm glad to be back and I missed you guys, so
[00:10:12] Carol: Oh, we missed you too.
[00:10:13] Adam: Yeah. Glad to have you back.
[00:10:15] Tim: I bet you, Adam, what's, what's your triumph or fail?
[00:10:17] Adam's Fail: Hand Injury Struggles
[00:10:17] Adam: I'm gonna go with a fail. so I've been working out for, I don't know, 4, 5, 6, 9 months now, whatever. and things have been going relatively well. and about two, three weeks ago I started noticing some like forearm pain, mostly up toward my elbow, just in different. Like situations, picking something up at a odd angle or something like that.
[00:10:38] Adam: I'm not exactly sure what that was. Like, maybe some tendonitis or something. I figured, whatever, it'll go away. And then over the weekend, this last weekend I, on like a Saturday night basically, I started having some really. Interesting pain in inside my hand, like they call it the snuff box. I have since learned it's this kind of like area down very close to your wrist, but sort of between your thumb and your first finger.
[00:11:04] Adam: and, it, it can be related to arthritis. she said, like a lot of people will injure it, if you like fall and you kind of think like I'm falling straight down, so I can put my hands down beside me or behind me and press down on the ground to, to catch myself.
[00:11:16] Adam: A lot of people do that. I don't know of any time that I particularly fell really hard on it or anything, but that part of my hand started hurting. They did an x-ray and there's like, it looks fine on the x-ray. But injuries to this particular part of the hand can, take some time to be visible on x-ray for whatever reason.
[00:11:34] Adam: But I'm also worried that my autoimmune disorder is related to arthritis stuff, and this is alsorelated to arthritis stuff. So I'm just like, oh God, please don't let this be like the next phase of my autoimmune thing. 'cause this kind of sucks. It's
[00:11:46] Adam: Not completely, but has kind of like killed my, my grip strength. Right. I'm the person in the house that everybody yells at for putting lids on jars too tight. 'cause I'm just like, okay, well I want it to be well sealed, so I just do it until it's tight to me and everybody's like, dad, can you open this?and, and so, you know, I, I have traditionally had a very good grip strength, but now I'm a little worried.
[00:12:06] Tim: I hate to hear that. 'cause you gotta keep your pimp hands strong.
[00:12:08] Ben: Yeah. In case Wayne Brady has to lay down the law. Yeah.what, what you can't see here because it's not quite in frame, but I actually have a, it's an athletic board, like, like a mat that you would put down in a gym and I've
[00:12:22] Ben: cut out, uh,
[00:12:23] Ben: a hole for my tummy so that I can fit it around my tummy. And, I'm using, to like put some padding under my arms.
[00:12:31] Ben: 'cause I'm also been having some left hand discomfort and I feel like maybe it's where my. Wrist is hitting the edge of my, my desk or
[00:12:40] Adam: Ah, desk or, or like the arms of your chair, or
[00:12:43] Ben: Uh, this chair doesn't have any arms. The chair behind me has arms, but my arms are very long and those arm rests are very high. And I'm like this at the lowest setting. so I feel your pain. It sucks. I, I've had, my right hand will occasionally just hurt. And, like I won't be able to pick up anything heavy or, or like, or like
[00:13:05] Carol: That
[00:13:05] Ben: towards me.
[00:13:06] Ben: And so the fun part of that is that my left hand gets stronger, which is normally my weak hand, but it sucks. and, you know, and then I catastrophize, and I'm like, oh, like what if I can never type again? And actually I have some pain, not, not to make this like the everybody hurts show, but like, I have pain in my bicep here.
[00:13:25] Ben: I'm going, to get an MRI, in November to see if there's anything like nefarious going on,
[00:13:32] Adam: Yeah, they did the, they did. I went to urgent care on Sunday morning to get the, the hand looked at and they did x-rays of that. And then like Sunday evening, I was like, you know, I should have had 'em do like the whole arm so that we could see if there's anything wrong, like. What if there's just like some hairline fracture up there near my elbow or whatever, and that's been causing other pain.
[00:13:49] Adam: And this is from like, you know, compensating or something. So,
[00:13:52] Tim: I mean, you do jump outta planes. I'm surprised you haven't broke more.
[00:13:55] Ben: That is true.
[00:13:56] Adam: yeah, I've never, never had a serious injury. I mean, I've had a couple of minor injuries from that, but nothing, you know, nothing bad. Anyway, so that's it for me. just I'm, I'm frustrated that I'm not able to work out 'cause like a, the grip strength and BI just don't wanna make it even worse, right? If, if there has been once or twice where like I push really hard through a particular set or whatever and it's like, okay, now my arm hurts worse.
[00:14:17] Adam: So maybe let's not do that more.
[00:14:19] Carol: Just go do a bunch of calf phrases and make your calfs really big.
[00:14:23] Adam: I should, I should do like leg pressing calf raises. 'cause those don't require the arms at all. Most of what I do, Even on leg day, a lot of what I do requires like my hands, IR DLS deadlifts and stuff. So,
[00:14:33] Ben: Hands are important.
[00:14:35] Adam: yeah. So anyway, let's move on to today's topic. it's probably gonna end up being a little bit of a short show just because we got a hard stop coming up.
[00:14:42] Adam: Ben's real excited.
[00:14:44] Adam's Trunk: Jump Run Motivation Waning
[00:14:44] Adam: Yeah.so, first stop on the trunk or treat here. You're, you're here at my trunk. and I'm just gonna go with this. I got raisins. So,
[00:14:51] Carol: Oh no.
[00:14:53] Adam: basically, yeah, basically my thing is just like I'm feeling the motivation to keep working on jump run my side project kind. Sapping. It's like, maybe it's a kind of a seasonal thing, and I'm just like, it's outta season for me right now.
[00:15:07] Adam: It'll come back later. But I'm just like, there's, there's things that feel like they should be within reach. I've kind of start, like I talked before about the instructor change I was making to help with payroll changes. Like I've got that sort of half done in a branch and I'm like, hit a little bit of roadblock and I'm like, eh, I'd rather just sit here and watch the Eagles game.
[00:15:26] Adam: I don't feel like trying to figure this out. You know? Like, eh, you know, eh. it makes me sad 'cause I want to keep pushing on it. I feel like if I had a bunch of customers and they were like, clamoring for stuff and, and paying me, then I would feel motivated. There would be like that, that sort of fire under me to, to, oh, these people are paying me.
[00:15:42] Adam: I have to provide the thing for them. But there's not, and so it very much feels like volunteer work at this point, even though there's potential for making money off of it later.
[00:15:52] Tim: That's what I was thinking. I think you got to the point where you're like, all right, I've made this. It's m, it's at MVP. If anybody wants anything else, they're gonna have to pay me to do it.
[00:16:01] Adam: yeah, it's, it's tough 'cause like I, I, you know, I've been donating the software usage stuff to the club, you know, that I'm letting them use it for free. It doesn't, it's not costing me that much. I think I paid $12 for a year of email hosting, and I've been on the free tier of everything so far. The one thing that might sort of reignite the interest for me, the, of course you're always gonna wanna do this is, I found out about a database I'd like to try just migrate the whole database over.
[00:16:28] Adam: It is a, it's called convex. It's kind of like a database as a platform, Almost like Firebase, but just, just different,
[00:16:34] Ben: If you can get back to a point, it seems like it started as a, an idea that you wanted to scratch your own itch to some degree, and then it got some traction, and now you're in the mindset of building it specifically for the people who want to use it. But can you, if you can find a way like with this convex database to kind of indulge in the scratching your own itch.
[00:16:56] Ben: Phase again, you know, just do stuff that is really only because you think it's interesting and you think it's cool and let that sort of reignite the fire. And then when you have the, the spoons, so to speak, to devote to other people, then, then you get back in that mode again.
[00:17:12] Adam: Maybe. I mean, the whole thing about it though is that it's, it is such a pitfall, right? Like I'm just gonna, instead of spending time and effort building new features, I'm gonna rewrite everything I already have over to some new language, some new, database. So it's a easy trap to fall into and it's right there, it looks so pretty.
[00:17:33] Ben: Well, can you find small things that you can do that are still. Incrementally improving the application, but are just much smaller. And you are like, ah, normally I wouldn't even waste time on this. But it's kind of that, like that palate cleanser work I talk
[00:17:46] Ben: about sometimes.
[00:17:47] Adam: Maybe, Yeah. I mean, the, the thing, it's funny 'cause like there's a bunch of features that people request and I'm like, yeah, yeah, it's on the list, I'll get to it. And these are things that would be relatively easy to do, I don't currently show. Each person a list of all the jumps that they've made at the drop zone, that'd be a relatively easy thing to add.
[00:18:04] Adam: but that, and other, just other similar, small asks. But I, I've kind of like said, okay, I have two or three things that are more high priority, like, oh, this is important. This will be useful to the people running the business, the people that are paying the bill in theory to me.
[00:18:17] Adam: I should prioritize that work first. And it's like, I, I kind of have mentally said I have to eat my vegetables before I can do the fun thing. and, and it's hard to choke 'em down. But maybe, maybe I'll, I'll do a pallet cleanser, put the, put that stuff aside in a branch and
[00:18:33] Ben: Yeah. You know, we talk about working out sometimes and the best workout plan, as they say, is the one you'll actually do.
[00:18:41] Adam: Yeah.
[00:18:41] Carol: Yeah.
[00:18:42] Ben: So you, you gotta be kind to yourself sometimes.
[00:18:45] Adam: Okay. let's, let's move along to the next trunk.
[00:18:47] Carol's Trunk: Making Customers Happy Through Empathy
[00:18:47] Ben: Sure
[00:18:47] Carol: I can go next.
[00:18:49] Carol: I have lots of really good candy. Nothing chewy. 'cause I'm still getting my mouth surgery stuff done.
[00:18:54] Adam: No candy, corn. I'm outta here.
[00:18:56] Carol: No, no candy, corn. You can have like three musketeers. There you go. That one's safe. So, it works crazy, right? It's always crazy for me right now. There's no other word than chaos usually is what I use.
[00:19:08] Carol: But, yesterday I made my customer happy and for them to say like, oh, this makes me very happy. Just, it made my day, it made me smile, it made me go write in my journal. Like everything just felt good and the wind wasn't even that big, to me. Like it wasn't a big deal, but I didn't even know they had this pain.
[00:19:29] Carol: And the pain is that, if you've ever tried filling out any federal forms. They're massive. They're huge, and someone has to go in and say, there's a bunch of conditional logic for these forms to then determine like what you're able to put in. And these things will be like 28 pages long. The one I looked at today had like over 400.
[00:19:51] Carol: It was like 410 response options available, like as items. I had no idea that that we developed these things in the dev environment. Then when it's ready to go to the next three environments, which is test stage training, prod, they manually have to log into the interface and put them in form number next screen question, next
[00:20:15] Carol: screen item option, next screen.
[00:20:18] Carol: I had no idea. I had no idea that, that no one had ever built anything for them to be able to like export up or to migrate this data or keep it in sync and it just. Like showing empathy to the problem that they hadn't seen empathy on before. Because previously it, it's just been a, oh, it's a, it's your time.
[00:20:38] Carol: It's not our time. Right? Like, go deal with it. And now that someone is showing empathy and I built them like, it's so silly, but I just build an Excel sheet so they can compare everything and figure out where their differences are between systems. And show them how to like merge tables and power query, like just did some little things and gave it to 'em and then put the spike in.
[00:20:58] Carol: That said okay. I see your problem. I know we can solve it. I'm going to get it written up. We're gonna figure out how to get it on the backlog because you shouldn't be wasting so much time trying to get these things up. And also you introduce errors like humans cause problems. So when you're coughing and pasting things, one, forget to save, and now you've lost questions and you don't even realize it until.
[00:21:22] Carol: These government forms are coming back in, missing things and someone's going, why is that not there? And it's just 'cause it's manual. So, the fact that I feel like I made my customer happy and I did it in a way that didn't require a ton of code, it didn't require complex solutions. It just required showing 'em data.
[00:21:41] Carol: Having empathy and then showing that you're willing to work on in the future is huge.
[00:21:45] Adam: It sounds like you haven't even done the work yet. Like you said, you did the Excel thing, but you haven't like changed the interface or anything yet and they're just
[00:21:53] Carol: changes There,
[00:21:54] Carol: There,
[00:21:55] Adam: just happy. You're saying like, I'm gonna do it.
[00:21:57] Carol: yeah. I got three messages and then like in a big channel, the one of the updates was, I'm just gonna go review the data that Carol got me. And it was like all cheers and stuff. I was like, all so sweet. But yeah, no, I mean,
[00:22:10] Adam: underserved for a while.
[00:22:11] Tim: Mm-hmm.
[00:22:11] Carol: Yeah, and I mean, it's just one of those, it was never a priority and I think they asked for it a while back and no one took it serious and they just didn't ask for it again.
[00:22:21] Carol: They thought that other things are more important, and I'm like, oh, your day is important. Like if you're gonna spend all day doing this, that's a really job. Like I wouldn't wanna coffee and paste all day long in a system that logs you out after 10 minutes fast.
[00:22:36] Tim: Wow.
[00:22:37] Carol: No, thank you. So, yeah, some big wins.
[00:22:39] Carol: Big wins.
[00:22:40] Ben: it makes me think about a statement that I hear sometimes, and this, this is not word for word, but it's the sentiment that basically all code represents technical debt. And, I will oftentimes hear that in the context of doing little things for customers that, oh, you know, if this isn't gonna service all of our customers, if it's only gonna work for a handful of customers, then we're just taking on unnecessary technical debt and we're gonna have to maintain this forever.
[00:23:09] Ben: And it's only gonna be used by a couple people and it's just not worth our time. But in my experience, which is, you know, granted very narrow and very limited. What you're saying makes me think of this, that a lot of times when we build these little things for customers, and I, I know you haven't actually built anything in the application yet, but you could, there's a lot of code that never accrues debt, in my opinion.
[00:23:32] Ben: Like there's a lot of code that you can build that you'll probably never have to touch again because it does something very small. It does it for like one or two customers. They're happy with the way it works, and that's it. And as long as it's not like gonna be breaking when you upgrade some framework version, or library version, it's not debt.
[00:23:51] Ben: It's just part of the application. And I think we overemphasize sometimes the, the landscape of the application and underemphasize the value add for customers.
[00:24:02] Carol: Agree and like talking about the technical debt. Side, like my idea isn't even like, I mean a super quick thought of how to do this. It's just create a sort procedure that basically publishes the data. The database saves the IDs as they get inserted, so then all your links are there. And we don't even need IDs in sync between the systems.
[00:24:21] Carol: Like IDs don't matter because everything can be found based off of names of forms and if it's been published yet, like there are a few things that you can link together. So I'm like, it's not even like a package that needs to be maintained. It's not.
[00:24:34] Ben: Right.
[00:24:35] Carol: It's not anything. It's literally just a so procedure that says go do work and tell them.
[00:24:43] Ben: Yeah. And considering how often we can, as people complain about the terrible customer service that we get when we go into a store, we have to call a support line anytime we have someone who sees people in need. And feels an instinct that I have to go and fix this because there is a wrong in the world that I know how to right.
[00:25:03] Ben: And I have the tools and I have the time, and I have the skills. The, the fact that we ever work at organizations that prevent that from happening, it's just boggles my mind.
[00:25:15] Customer Service Empathy Story
[00:25:15] Adam: Yeah. like for the last week I've been trying to get, get through customer service on a thing, and it's just been an awful experience. So,
[00:25:22] Adam: Yeah.
[00:25:22] Ben: It, it's crazy. And, and you know, going back to the idea, what Carol said about just showing people empathy, it's so freaking powerful. I was just telling someone this story over the weekend, months ago, me and the misses were just having like a little Saturday road trip. We went to a town that's a little farther north of us and went to this little cafe market.
[00:25:43] Ben: There was a coffee bar, and all we wanted to do was get chocolate chip cookies. That was like, that was in our mind. That's why we went in here and we go up to the coffee bar where the cookies are, and the woman behind the counter is making espressos or something, and literally, I, I mean, it, it must've been at least five, seven minutes.
[00:26:02] Ben: She's making these drinks, making these drinks, and we're just standing there with her back to us, and if all she did was turn around and be like, I'll just be with you in a minute. I'm, hold on. This will just take me a couple minutes. I'll be right with you. If all she did was that happiest people in the world like, oh yeah, no problem.
[00:26:19] Ben: We just want those cookies. But the fact that she didn't even acknowledge that we were standing there waiting to be serviced was like, how do you live in that world? Oh, so yes. Empathy all day, please.
[00:26:33] Carol: That's me. I brought good stuff.
[00:26:35] Adam: Sorry about the raisins again.
[00:26:37] Carol: Who's going next? Tim or Ben?
[00:26:39] Ben's Trunk: How We Evangelize Tools
[00:26:39] Ben: I'll go, I guess. I've been doing a lot of overthinking of stuff in my, in my little Ben needs a minute channel. It's true, it's true. Guilty. and I did come up with something, something popped in my mind. I was, I was walking around town the other night and I was thinking about unit testing or test driven development in general.
[00:26:57] Ben: Just something that popped in there,
[00:26:58] Adam: Which is your brand.
[00:27:00] Ben: that happens sometimes. I, I'm a, I'm a big TDD guy. And I was thinking about, when we talk about testing and we talk about people who don't test, which is usually me, there's a certain degree of not testing as negligent as a professional software engineer.and then when we had Sean on the other day, and I was talking about regular expressions and how I was using regular expressions to. Find the issues in my ColdFusion code as I was upgrading from 2021 to 2025, and how exciting I was, how exciting it was, and I was saying, oh, reg expressions are so great.
[00:27:36] Ben: Everybody should learn them. It's so fantastic. It's such a value add and this, this to me was an interesting juxtaposition that we could have several people who feel very strongly about certain things and some of them are pitched as. This is so great, you should try it. And some of them are pitched as if you don't do this, it's a problem.
[00:27:59] Ben: Adam earlier in the call, talked about eating as vegetables as part of the jump run stuff. So you could look at someone who says, you know, I've never had a vegetable in my life. And you could be like, whoa, that's a freaking problem. You need to eat vegetables. That's not okay. Or you could talk to someone else who's like, oh, you know, I've never had Indian food.
[00:28:19] Ben: You could be like, what? Indian Food's freaking amazing. You gotta try, you're gonna love it. It's, it's not, it's not positioned in a, I can't believe that. How is your worldview so limited and, and tiny I don't think it's just a personality thing. I think there is something fundamental to the thing that's being described where it must cross some tipping point.
[00:28:40] Ben: Where it's no longer, Hey, this is just a great thing you should look into. It becomes a, if you don't look into this as a problem,
[00:28:48] Adam: I'm, I'm, I mean, I understand the food metaphors, but I'm struggling to place this within the, the web dev world. Can you gimme a,
[00:28:55] The Subjectivity of TDD and RegEx
[00:28:55] Ben: well even, even the the, I mean, I know you don't, we talk about AI maybe too much, but even in the AI space, people could say, Hey, AI is actually pretty valuable. You should try looking into it. I think it can really help you.
[00:29:08] Ben: And then there are other people who say, whoa, if you're not looking into ai, that's a problem.
[00:29:12] Adam: You're gonna be out of a job soon. And that's maybe not a great example because that's very theoretical future, not like actual reality.But I think that your whole thing here was like, it's kind of subjective whether it should be one or the other. Right.
[00:29:27] Ben: right. So and I guess maybe it's the subjective versus the objectivity of it that, is part of the fuzziness of the calculation. 'cause I could look at regular expressions, which I love, and I could be like, they make my work day better because I use them to find things in the code.
[00:29:46] Ben: I can do extended searches, I can do fines and replace over fuzzy patterns. And like, it's a very. Helpful thing to do, it actually makes me a better, more efficient programmer. but then you could look at something, not to go back to the TDD stuff, but like, TDD, like, people just feel super strongly about it and you can say, well, you know, if you're not testing, then things are gonna slip through the crack and you have to test things manually. And I'm like, yeah, okay. But it's hard to point at say like a project that I've worked on and say, Hey, if you had spent less time learning reg expressions and using 'em as a project and more time adding tests, like the project would've been better.
[00:30:22] Ben: I'm like, I can't say that for, for fact. What I can say for fact is that I've used reg expressions and they're fantastic and I think you should use them because they're great and like you should put the time in because I think you'll love them. I think it's be a value add
[00:30:33] Ben: Well, I just wonder, like why, why is that not the TDD mindset?
[00:30:37] Ben: Like, why do people not come to the conversation with, oh man, TDD is the best thing since sliced breast? You should totally try it.
[00:30:43] Carol: People do,
[00:30:45] Carol: people do come to the
[00:30:46] Ben: I feel like it's much more of a
[00:30:49] Tim: No, that's just you against it. Everyone else does.
[00:30:52] Carol: like I have, I have three people on my team that all they wanna do is just write the test before we even talk about the code anymore. Like, let's build tests and then let's talk about what the code's gonna do. I'm like, that's fast afterwards for what I wanna do. Like I wanna write some code and then test it, and then modify it and make sure everything still looks good.
[00:31:10] Carol: Then build more tests.
[00:31:11] Ben: Right, and I guess, I mean, even there, right? The Adam said the subjectivity of it all. I guess I, I guess this even goes back to my triumph and fail, that we all just experience the world in such different ways, and that includes and is not limited to the value that we get out of the types of code that we write and the types of technologies that we use.I, I struggle with this even with things like TypeScript. I've used TypeScript. I find it enjoyable when I'm writing just smaller stuff on my own. I never reach for TypeScript. And that's not 'cause I don't like it, it's just 'cause why it doesn't add value in those cases. But I'm sure there are people who would describe TypeScript as like the defacto standard.
[00:31:51] Ben: And every new project going forward should just use TypeScript by default. And you're like, I don't know. It's just like I.
[00:31:58] Adam: yeah, I think that'll be an easier pill to swallow when, some more of this like, TypeScript in node stuff lands right, where it just kind of gets erased. Because then like there's the argument against TypeScript as sort of the, the default is the effort it takes to set it up and, and, you know, have your compiler there and like to run the checks and stuff.
[00:32:21] Adam: Versus like, if you, if it was just, the way you wrote your JavaScript and your IDE gave you some, Hey, by the way, it looks like you're passing a number, but this is expecting a string, or you have too many arguments or whatever, right? Like. And then it otherwise just worked like regular JavaScript like that I think would be ideal, right?
[00:32:39] Adam: No build step. No, no futzing around nothing. And I, I don't think we're too far off from that, so I'm excited for that future. But yeah, I, and like the regex thing too, you know, you kept talking about that and all I could think about as far as regex goes is like regex is so useful that at this point I consider it to be like table stakes.
[00:33:01] Adam: Okay. There, there's probably some languages, you know, scripting languages, programming languages, whatever, that don't have a RegX implementation. And that's fine for them. But I don't think that I would, like that. For me, that's kind of a deal breaker, right? That's just like, it's at some point it's gonna come up.
[00:33:16] Adam: I'm gonna need it for something. Maybe it's possible that I wouldn't need it for the thing that I would use that language for. Okay, that's fine. You know? But like, or maybe, you know, you can, Fork out to do pearl on the command line to get your RegX or whatever, like, oh, you know, as long as there's some way I could do it, then that would be fine.
[00:33:34] Adam: But otherwise, like I just feel like it's a red flag, it's a deal breaker. I'm not, I'm not gonna consider anything that doesn't have RegX.
[00:33:42] Ben: I
[00:33:42] Ben: could dig it.
[00:33:43] Carol: think recently, pretty recently, I don't know exactly when I was just looking at this and I don't see it right now. TQL server just added Reg X pattern expression searching. So I think, I don't know what version it's in, but it's been in the last couple months that they've added that and I was super excited 'cause I don't have to continue going.
[00:34:02] Carol: replace, replace, replace, replace, replace.
[00:34:06] Tim: Nested, replaces.
[00:34:08] Carol: Yeah.
[00:34:10] Don't Yuck Anyone's Yum
[00:34:10] Ben: All right, so I'll just, I'll, I'll finish my little spiel here with, again, this is, I, I think so much of my overthinking of stuff is me trying to reconcile my view of the world with other people's view of the world because. I, I think I was just raised, not, not explicitly, but I think just implicitly by my environment to believe that there is just a common understanding of the way things work and which things are valuable and which things are not valuable.
[00:34:37] Ben: And the older I get and the more diversity within the world that I see ev everybody just, you know, everyone has their own yum and everyone has their own yucks and, I'm trying to be better about not yucking anyone's yum. And I guess I'm, I, I will leave it in that. I would appreciate that fewer people.
[00:34:59] Ben: You know the golden rule, right? Do what? Do unto others as you'd have others do unto you. Don't just don't be yucking anyone's yum.
[00:35:06] Carol: So you brought the tricks to the, to the night,
[00:35:09] Ben: I'm saying like, you know, if you have a yum, tell people why it's yummy, but don't tell them that their stuff is yucky.
[00:35:14] Adam: Okay, so he brought
[00:35:15] Adam: candy, corn.
[00:35:17] Carol: Ugh. Circus being nuts.
[00:35:19] Ben: Yo,
[00:35:19] Adam: It's a very polarizing candy. Some people love it. Some people hate it.
[00:35:24] Carol: It's,
[00:35:25] Ben: as an adult, I legit think that raisins are delicious. And when I was a kid, I got this little red, what is it? Like the sun kissed raisin boxes.
[00:35:32] Adam: Yeah. it reminds me of the like land O'Lakes,
[00:35:34] Ben: Yeah. Yeah. Like some, some little circle woman on it kind of a thing, and you're just like, oh, who gives freaking raisins? And now I'm like, raisins. Freaking Awesome.
[00:35:43] Carol: Mm-hmm.
[00:35:45] Ben: All
[00:35:45] Carol: Tiny little side tangent. when I was a kid, I remember like we had this old car and I put grapes in the back of the window during the summer and waited for them to turn to raisins, and I got in so much trouble. They did not bake off very well.
[00:36:03] Ben: That's awesome.
[00:36:04] Ben: All right,
[00:36:04] Adam: uh,
[00:36:05] Adam:
[00:36:05] Adam: yeah. What do you get going on, Tim?
[00:36:06] Tim's Trunk: AI Coding Tools Head-to-Head
[00:36:06] Tim: so in September I really, you know, we've talked a lot about ai, mostly Ben, but, I dunno how much he actually uses it, but he sort of talks about it a lot. It's like that, you know, I've got a girlfriend in Canada,
[00:36:18] Tim: um,
[00:36:18] Ben: just totally real.
[00:36:20] Tim: You wouldn't know her.
[00:36:21] Tim: She doesn't go to this school. but I said, all right, let me sit down and actually, you know, because I got, you know, I had some time, so I'm playing with it and.
[00:36:28] Tim: Main thing was I want to do it. We use VS. Codes. I want it had to work within VS. Code by default you get copilot. So I was playing with that in by default and then I said, well let me compare it. 'cause I really, I've really liked chat GBT five and I have a subscription and it, it seems pretty. Pretty good for what I was using it for, for code generation, but it wasn't really tied into the repo, right?
[00:36:52] Tim: So copilot, it knows about your repo, it kind of looks at it a bit. So I downloaded Codex a plugin from VS Code. And just to compare the two, I worked on the same thing. So I, the, the project
[00:37:04] Tim: was.
[00:37:05] Ben: is, is,
[00:37:06] Tim: Codex is chat GBTs plugin, right? So it, it knows about your workspace, it, you know, it looks at your repo.
[00:37:14] Tim: So it's more of a, an agent, right? So agentic coding rather than just, 'cause before I was like taking a problem pasting into chat GBT, you would generate it. I copy it back in. I'm like, this isn't really
[00:37:25] Carol: Does have the full context. Yeah. Mm-hmm.
[00:37:27] GitHub Copilot Experience
[00:37:27] Tim: Doesn't have full context. And, and to be fair, the Codex is a very much, it seems like it's kind of a version one bleeding edge kind of product for chat GBT.so what I found was like, so copilot. I really had to spend a lot of time reigning it in. I explicitly would tell it. I give it the context. I'm like, please use verbose variables. Like don't use the letter V as a variable, or, you know, truncate something down from, insured to INS. Right.
[00:38:00] Tim: Don't, don't do that kinda stuff. And copilots. Really generated a lot of bad code. I was surprised just how bad it was. It, it, I was constantly fighting it. And then Codex was pretty interesting. So Codex like, I guess it like uses power 'cause I'm on Windows, so it uses PowerShell commands and it doesn't seem to have, I, I haven't this, I think this is a bug, but it has to ask permission from you every single time it writes a file.
[00:38:26] Codex (ChatGPT Plugin) Results
[00:38:26] Tim: And so you gotta, which, you know, in the. Agentic coding. It's, it's constantly doing. So I, I could ask something in 10, 15 times. I gotta click yes, yes, yes. Allow, allow this session, allow this session, allow this session. But
[00:38:40] Carol: there is a thing you can change for that.
[00:38:42] Tim: is there, I haven't been able to find it. So after, after the show, you have to show me how to find that.
[00:38:46] Tim: But I was super impressed. The, the thing I was working on is I. I gave it the, basically the swagger info from a page for an API that I needed to consume. So I want you to consume this API and then I want you to run this query loop through the query and check to make sure that, an entry exists in the API.
[00:39:08] When AI Gets Off the Rails
[00:39:08] Tim: Right? So basically it's an iterative loop that goes through. And I could have built this probably myself, probably about two hours. And co-pilot, I eventually just gave up 'cause it kept messing things up. Like I would ask it to redo something and it would not just change the thing I asked to change, it would change other stuff too.
[00:39:27] Tim: And I'm like, why did you change that? Oh, you're completely right. I shouldn't have touched that. And, and right, because, you know, and but Conex, I was, I basically explained to her, I said, is a. It's A-T-E-A-P-I. So I, and I explained to what TAF E was restful API, and it figured out from the content I said I had to create two pages for me.
[00:39:49] Tim: And they were basically taffy endpoints, for restful API. And it figured out from the conventions that I had been using and that were also in TI itself how to do it. Copilot can never figure it out. It's always trying to just return without status. but you know, in TE you have to return with a, with a status and it figured that out all its own.
[00:40:09] Tim: So the best thing was though, that what I was doing at the end of this, after I had the the API output, I was con converting that into a report. And I do not do pretty, I do not. I don't generate pretty things. I, I think things that work, yes, things that look nice. And so I had to build a bootstrap page with a report that you could click the top headers and sort everything.
[00:40:32] Tim: And and then if you double click it, you could see like the full data payload of the thing. But it looked really, really slick. But that took probably all day to do of just with codex. Just redoing it. Redoing it, and it takes so long you're sitting there waiting for. Five, six minutes while it's regenerating.
[00:40:51] Tim: So yeah, it's like I'm still, the jury's still out. It's like sometimes the these things, one shot it, and sometimes it's like once it gets off the rails, you gotta like constantly like beat it into submission until it gets right where you want it to be.
[00:41:04] Ben: Well, it's interesting. I was listening to an interview over the weekend and they were talking about the future of all this agentic code stuff. They were talking specifically about how if you try to keep pushing it in in a different direction, it's like it, it kind of gets worse and worse and worse as it goes.
[00:41:22] Ben: And one guy said, oh yeah, you know, in the future probably what we'll do is just get to a point where every change you want to make, you literally delete the whole app and just have it regenerate with a slightly modified prompt. So he's like, and he goes, you know, the trick will of course be to make sure that it regenerates the same app every single time like that.
[00:41:41] Ben: We're just gonna hand wave over that.
[00:41:42] Tim: Yeah, I, I actually did that today. It, it kept, it was, I was trying to get it to do something It was a CF script page and it was shoving, tags into it and breaking it. I could not get it to stop doing that. I specifically said, you know, everything should be CF script.
[00:41:57] Tim: There should be no tags whatsoever. Okay, got it. Yeah, sure. There's a CF loop right in the middle of a CF script. A CF tag loop. Yeah. So eventually I'm like, forget everything I said about this project. Why let's, let's start from scratch. I deleted the files and started over with the new prompt based off of what I learned it was bad at, and specifically said that, and I got, yeah, it, that actually did wind up working.
[00:42:22] Ben: Yeah, it's so fascinating. I don't know what to make of it all.
[00:42:26] Trying Claude and Cursor Next
[00:42:26] Tim: but, but my next, the next thing is I'm gonna compare Claude. 'cause I was telling my findings to some of the guys on my team. It's like, oh, you need to use Claude. 'cause Claude's really, really good at it, and it's a gent coding and download that plugin and I'll show you how to use it. I'm like, okay, I'll, I'll try Claude next.
[00:42:40] Tim: But I feel like I'm wasting more time learning, learning how to do this than it would, than just doing it.
[00:42:46] Adam: My understanding, Claude is supposed to be pretty great, but the, there's a little bit more of a learning curve because it is like a whole CLI thing. It kind of, it's its own IDE basically you have to learn the commands and stuff. It looks really interesting. if you, if you don't know who Matt Pocock is, he created like a type script course that is pretty popular.
[00:43:05] Adam: Yeah, it's, if anybody's interested, it's called Total TypeScript. I'm sure if you just Google that, it'll take you to his thing. But he's on a whole, like, AI kick now. I think he might be making an AI course, and he's been putting out videos on, and newsletters on AI stuff. And he made a video very recently about, how he's using Claude, which I found very interesting. So that might be useful for you. I'll, I'll send you the link, if you're interested in digging into Claude, however. I wanted to throw out like just as of today, I've kind of reinvigorated my interest in Cursor. so Cursor is like a fork of VS code and they have their own stuff built in. I think that it was either today or earlier this week.
[00:43:42] Adam: That they released, the Cursor 2.0 and that with it, they have released their own model specifically for coding. It's supposed to be faster, like higher tokens per second and uses less tokens to get the same amount of work done. It's supposed to be really smart, really good. Haven't really tested it all that much yet, myself.
[00:43:59] Adam: so my company, we all have, copilot subscriptions, and. I use it, but I've, I've kind of gotten out of the habit of trying to do any of the agentic coding probably. 'cause like you're saying, it's just not worth the effort, you know, it doesn't do the best job. And so, but when I started looking into this cursor stuff, like the way they've changed the ui, they've really kind of like reimagined what the IDE should be like for doing agentic coding, which looks really interesting.
[00:44:25] Adam: so I was like, well, I can't really, I, I don't want to be that guy that's like, Hey, let's. Let me have this, $20 a month subscription for this and let me have the, you know, I want seven different AI subscriptions so I can test all the tools and find the best one. So I think what I'm gonna do is just on my own dime, I'm gonna buy a month or two of cursor, see how like that, because it includes a certain amount of usage in that $20 a month.
[00:44:46] Adam: And like I'll see, okay, am I running out in a week or, or am I, you know, not even coming close to that $20 or, or, or what, and, you know, get some better understanding of how. the quota usage goes and how useful it is. And then maybe I'll like lobby to be like, actually, instead of copilot, I want mine to be this, or maybe we should all be using this instead.
[00:45:06] Adam: So, but that, that's where my interest on it is right now. And I just wanted to point that out for you, Tim, in case you're looking for something else to
[00:45:13] Tim: Yeah. And I didn't realize that, I didn't realize there was a new version, so, yeah. Of, of, cursor. So I'll check that out. 'cause that was also, someone also mentioned Cursor as well.
[00:45:20] Adam: Yeah.
[00:45:20] Tim: didn't, I, I didn't realize it was an id. I thought it was just another LLM or, or Cly.
[00:45:24] Adam: Nope. Nope. And I've never heard anybody call it a Clyde before
[00:45:27] Ben: I was just thinking that.
[00:45:30] Adam: Yeah. Yeah. I mean, hey, it makes sense. I'm not, I'm not mad about it. It's just that's the first time I've ever heard it.
[00:45:35] Ben: Tim
[00:45:35] Tim: You heard it here first folks.
[00:45:37] Ben: Yeah, he's a taste maker.
[00:45:39] Patreon
[00:45:39] Adam: Alright, well, we gotta get, going, we're coming up on our, hard stop here. So, why don't I end the show there. this episode of Working Code was brought to you by Ben's TDD framework. Big Sexy Tests.
[00:45:50] Ben: Show us your tests.
[00:45:52] Carol: Oh, that's amazing.
[00:45:55] 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. Our patrons cover recording, editing and transcription costs, and we couldn't do this every week without them.
[00:46:08] Adam: Special thanks to our top patrons, Monte and Jean Carlo. You guys rock.
[00:46:11] Thanks For Listening!
[00:46:11] Adam: we're gonna go do the after show, although I think Carol's gonna have to go. so it'll just be the gentleman's show this evening. I'm not even gonna tease what we, yeah. it's a, it's a secret club. I'm not even gonna tease what we're gonna talk about since Carol's not gonna be there.
[00:46:25] Adam: and definitely not because we don't know yet. but if you want to get, early access to new episodes, get the after show, get those warm fuzzies, then you can go to patreon.com/working code pod, throw a few dollars away through, throw a few dollars away, throw a few dollars our way every month. and, we would be very appreciative and I better just stop talking now before I mess even more up.
[00:46:48] Adam: So, that's it for us this week. We'll catch you next week. And until then,
[00:46:51] Tim: Just like Ben, we would never yuck or yum. Your heart matters.
next episode: 237: The Internet Is Eating Itself and We're Just Watching
prev episode: 235: When Romance Becomes a Database Problem with Sean Corfield