314: Resolutions Without Hubris

Resolutions Without Hubris

[00:00:00] Intro: Tired. So tired, Overtired.

[00:00:04] Jeffrey: Hello, everybody out there. I see you. I see you in your kitchen. I see you in your car. I see you walking your dog. I don’t really see you. That’s the government you’re thinking of. Um, this is Jeff Severance and this is the Overtired podcast, and I’m here with Christina Warren. Hello, Christina.

[00:00:21] Christina: Hello, Jeff. Happy New Year again. How

[00:00:24] Jeffrey: I thank you. I thank you. It is still a happy new Year. See, I celebrate New Year officially on my birthday, which is January 16th,

[00:00:31] Christina: Yes, which I, I, I knew that this was coming up cuz I was looking at my, um, calendar and, uh, it, it populated in my calendar and I was like, oh, I gotta remember to, to, to wish a Jeff happy birthday. Um, so ha, happy, happy birthday.

[00:00:45] Jeffrey: Thank you. It’s a great thing because I have usually completely bailed on my initial idea of resolutions for the year. By my birthday and my birthday allows me a chance to declare that this is really [00:01:00] the beginning of the year and, uh, and make a, a less, um, I, I make some kind of resolution with less hubris,

[00:01:07] Christina: I like

[00:01:07] Jeffrey: maybe Yeah.

Something I can keep too.

[00:01:10] Christina: Less, less, less hubris and uh, um, keep keeping things, um, uh, more honest, which I think is good cuz that’s, I I was curious. Like I, I gave up on New Year’s resolutions, like in my teens, I think probably. Um, do you ever really like, try to have them or, or is that something that, that you like.

[00:01:36] Jeffrey: I, I, I do try to have them, um, but they don’t really go anywhere. I just, what I’ve come to believe is that, It’s worth it just to have the thoughts about what is it I would like to change and then to just let ’em go and maybe it’ll, maybe it’ll, you know, take care of itself over the course of the year. My, my, uh, my wife has the most amazing, [00:02:00] um, uh, resolutions.

One year it was to park better

[00:02:04] Christina: Oh, I love that.

[00:02:05] Jeffrey: just generally to park closer to the curb. And another year it was to watch national, uh, nationally televised events like a normal American where she watched every single thing from the Golden Globes to the Super Bowl. Um, and I’m trying to, I always try to take inspiration cause I think those are just genius resolutions, but I can never find mine.

[00:02:25] Christina: see. Okay. Those are awesome. So I love that actually. And so, cuz I’m like you, like I, I like to have things that I think about and I have goals. I stopped probably when I was around like 15 or so, like actually writing out like. Resolutions because I realized I’m not doing a great job keeping these and then it would just make me feel bad.

And so I was like, eh, don’t do things that make you feel bad. But I do still in my mind like, oh, these are my goals for the new year and this and that. And there are things that I, that I, I have like goals from three years ago, like start a newsletter and I’m like still working on that. Um, but, uh, [00:03:00] I love, I love your wife’s cuz those seem like things that are actually actionable and that could also have some sort of impact on your life.

I mean, I don’t know if like watching live events does except like then like, you know more about the, the water cooler conversations and the topical things that, you know what I mean? Like, so, but, but those are, those seem like low, low stakes things. I wonder too, you’d have to ask her about this. I would love to ask her myself, but I you, you can ask her like do, or, or even based on your observations, the fact that she has kind of these really interesting resolutions, do you think that that helps her?

Achieved some of her other goals that might be bigger, broader, like the fact that she can say, my resolution is to park better. She does. She works on it. Do you think that helps maybe with follow through on maybe some of the bigger goals she wants to achieve?

[00:03:49] Jeffrey: That’s a good question. I’m not sure I will say. Even a, um, sort of silly, uh, resolution, like I wanna park better, which is perfectly reasonable. I don’t, I [00:04:00] don’t mean silly, like

[00:04:00] Christina: No. No. Totally. But,

[00:04:01] Jeffrey: silly. It’s fun,

[00:04:02] Christina: right, right. We just, we j we just mean in Sava as

[00:04:04] Jeffrey: it’s whimsical, you know? Um, by the end of the year when you realize she’s still doing it, it’s like, for me, I’m always like, man, I gotta think of something.

Because at the beginning I thought, well, that’s not enough, you know? But at by the end of the year I’m going, no, that actually feels like a

[00:04:19] Christina: No, I

[00:04:20] Jeffrey: I would’ve been glad to be there.

[00:04:21] Christina: well, no, well, well this is kinda my, my, my, my thought, right? Like, is that, that is like enough, like especially if you, cuz at that point it’s no longer even a resolution. It’s like that has been a behavior change, which I think, uh, is, is even better than a resolution. Cuz I think a lot of times, , I, and, and I’m including myself in this, I’ve viewed resolutions as like things to do and like things to like seeing like, okay, I need to check this off, right?

Like, I need to, I need to, to do this, be one and done. Whereas what she’s doing is actually saying, I want to change my behavior in a

[00:04:52] Jeffrey: Yes.

[00:04:53] Christina: I think is actually much more important, harder, and like much better. Right? Like, I think that’s [00:05:00] actually a way better way of, of viewing things,

[00:05:02] Jeffrey: Well, and I think, I mean, at least as an observer, like it’s, it’s absent all the like, because I will be better if I X. Right. They’re usually not things that like are going to, are going to carry the weight of, will I be a better person? How

[00:05:20] Christina: No, totally.

[00:05:21] Jeffrey: You know? And I love that.

[00:05:22] Christina: except by being able to successfully change your behavior. You are right.

[00:05:28] Jeffrey: Yes, yes. Every time you do that, it’s data

[00:05:31] Christina: It, it is at, and, and every time you do that is also, I think, and this is why I’d be curious to know, like it’s helped her. Um, and I guess resolutions is probably the wrong word, but I would be curious to know if that has helped her make behavior changes in other ways.

Right? Because if you could do it for these things that seem small, right? But you can do it and really carry through, right? Like, I, and I failed at this, but I’m gonna get back on it. When I had my nails painted for GitHub universe, I didn’t bite my nails for like, the entire time that I [00:06:00] had those, those, those gel nails on.

And, but, but as soon as it shipped off, I then immediately bit my nails again. And it was awful. And I was really upset with myself, but I was like, okay, well this is just, I’m going to start getting my nails done and keeping them up. And, um, I just, I, and I’d wanted to actually get them done, um, before it was, uh, chipping, but I couldn’t find, um, a, a place, uh, close by.

Soon enough for, for when you know it was happening. And then by the time it happened, I was like, all right, I’ll just, then I got sick and everything, but, but I’m gonna, like, this is actually, okay. So this is one of my behavior changes. I don’t think that I will reasonably probably, at least for a long time anyway, I don’t know if I will ever be able to stop biting my nails.

However, when I had the gel tips, which, you know, and, and I’ve had painted nails before, but it hadn’t really worked this well, like, I really was able to at least for a month change my behavior. And I do wonder, like this would be a thing like, okay, I might not ever be able to completely prevent myself from like, biting my nails.

It’s a nervous habit. It’s something [00:07:00] I’ve been doing literally my entire life. Um, like I, before I bit my nails, I picked my nails and, um, and so I, I don’t, I don’t know if that’s ever anything that’ll go away, but I do feel like I could, you know, by going every three weeks to get my nails done like that, that could be part of it, right?

Because if they’re painted, they look a certain way, then I’m not So, um, I like that. I like that. Rather than my resolution of being, I’m going to bite my nails, uh, I’m not gonna stop biting my nails like I’m going to, you know, commit to getting my nails done every so

[00:07:32] Jeffrey: yeah, that’s, that sounds way more fun too. Right. Like you handle it, but you also have fun handling it. I mean, that’s like, it’s also similar to what is the thing that Judo wrote? Atomic Habits, his thing about habit stacking, but also his thing about the envi, the, the, your environment that like he, he would say, or he says, there’s not like a type of person who’s good at holding their resolutions, right?

There’s a type of person who’s good at creating the environment wherein [00:08:00] they can hold their, um, resolutions. And in a way you’re doing that by just making sure you get those nails, , those nails done. And that’s, and that’s like way more fun cuz it’s not loaded with like, oh God, I don’t tried to stop biting my nails.

[00:08:13] Christina: It’s like, no, I’m not even, I’m not even gonna attach that. I’m just gonna say, okay, I’m gonna get my nails done every three weeks, and maybe that will, if I do it long enough, there will be a real behavior change. Right? Like, maybe that would be the real goal. Maybe, maybe at a certain point, if I didn’t go every three weeks and, and something chip that maybe I would finally be past that point where like, I wouldn’t be doing exactly what I’m

[00:08:35] Jeffrey: yeah, sure.

[00:08:36] Christina: which listeners, uh, can’t see.

But I’m like literally putting, I’m

[00:08:39] Jeffrey: right now live.

[00:08:41] Christina: I’m like, Ooh, I see a little bit of white. That’s because I, I, I bite them to the quick and I’m like, Ooh, I see a tiny bit of white. I could totally bite this now. And yeah.

[00:08:51] Jeffrey: Oh man. You’re reminding me I need to put a pedicure on my birthday list.

[00:08:55] Christina: Ooh. Yeah.

[00:08:56] Jeffrey: had two now in the last year and a half. [00:09:00] I had never had a pedicure before, and I don’t know what toxic masculinity was keeping me from doing that, but what an amazing thing.

[00:09:09] Christina: I.

[00:09:09] Jeffrey: it tickles

[00:09:10] Christina: Yeah, I, I, I wouldn’t mind getting that done just in terms of, uh, I guess like having like, them like clean stuff up. Um, my toes are terrible. My feet in general are terrible, which is, uh, we talked about this cuz I discovered this cuz this was a horrific discovery that I did not make that one of my friends in informed me about.

But, um, uh, we talked about this on a, a addition of the, the podcast like a year and a half ago. I, I’m on Wiki feet. Um, yeah, which is exactly what it sounds like.

[00:09:42] Jeffrey: Yeah.

[00:09:43] Christina: And, um, it’s hilarious because

[00:09:46] Jeffrey: How are you on Wiki feet?

[00:09:47] Christina: somebody submitted my photos and thinks I’m notable enough. I don’t know,

[00:09:51] Jeffrey: That’s amazing.

[00:09:52] Christina: somebody went through all my Instagram photos and all of my like Twitter things and submitted photo, any photos of my toes.

It’s bizarre

[00:09:59] Jeffrey: that is? [00:10:00] Wow.

[00:10:01] Christina: very disturbing. I’m not gonna lie. Like that was one of the most disturbing things I’ve ever learned. And I was not at all like, happy to learn that. But also I was, I was a little bit, I was amused because my feet are absolutely my worst feature. Like my toes overlap.

Like I don’t have attractive feet. Like there’s nothing attractive about them. Like, you know, like if, if I had longer nails and painted nails, like I, I don’t have ugly hands, um, or ugly fingers like that, that’s, that’s fine. But like my feet are not attractive. Um, and, uh, so why anybody would wanna look at my feet is just, uh,

[00:10:33] Jeffrey: Wiki feet, . Oh my God. That is, it’s just like what my brain is doing is, um, going through the, the de decision patterns of whoever was responsible for that. And then the action, the workflow itself, um, and then the horrible conclusion,

[00:10:50] Christina: 100%. And then it always reminds me of the Great King of the Hill episode, um, where, uh, Peggy, um, uh, becomes a foot model, um, unexpected, like unknowingly, [00:11:00] um, and starts making fetish films, um, because of her giant feet, not realizing that, that, um, thinking that people like them because they’re beautiful, not realizing people get off on them because they’re terrible.

Um, and, uh, yeah, um, it’s a great episode. Um,

[00:11:16] Jeffrey: Okay. So we’ve, we’ve entered a total cultural corner there I, that I had forgotten existed,

[00:11:23] Christina: yeah, then I, I, I, I’d forgotten existed too until you said pedicure. But you should get a pedicure for sure, cuz treat, treat, treat. Treat yourself. Treat yourself for your birthday. Treat yourself.

[00:11:30] Jeffrey: wanna get myself up on Wiki feet.

[00:11:32] Christina: absolutely. See, uh, listeners, uh, if, uh, Jeff posts photos of his feet, please submit.

[00:11:38] Jeffrey: Mm-hmm. . Mm-hmm. . Yep. That’s all

[00:11:40] Christina: actually, no, no.

Don’t actually do that because it, when you find that out about it’s

[00:11:45] Jeffrey: Yeah. It’s not, no, that’s

[00:11:47] Christina: not at all like, yeah.

[00:11:51] Jeffrey: I’m just seeing if, if both the other co-hosts were also on Wiki feet, you know, maybe it takes some of the sting in the weird out, or it adds much more [00:12:00] weird.

[00:12:00] Christina: Yep.

[00:12:01] Jeffrey: You’d be the judge audience.

[00:12:02] Christina: agree.

[00:12:03] Jeffrey: All right, well, good. Check in. I’m, I’m gonna still put pedicure on my,

[00:12:07] Christina: You definitely should. Yeah.

[00:12:09] Jeffrey: I haven’t found the place I love yet.

[00:12:11] Christina: okay. Yeah, no, that, well, that’s the thing too. Okay. So you’re gonna be looking for product here? I’m gonna be trying to find, um, a good manicurist. I’m probably gonna have to go to Bellevue, um, which is like the bougie, uh, uh, town, like over the, um, over the lake, but

[00:12:25] Jeffrey: Bougie. Bougie. Yeah. All right.

[00:12:29] Christina: you know, I, I, I’ve had to come to terms with the fact that I’m becoming just like a, a rich Bellevue mom, and, and I don’t like that cuz I like to live in the city and whatnot.

But there’s just a lot of

[00:12:37] Jeffrey: Get a minivan.

[00:12:38] Christina: Oh, fuck no.

[00:12:40] Jeffrey: Oh, they’re the best

[00:12:41] Christina: Oh, absolutely not. Under, under

[00:12:42] Jeffrey: I love. I am long past needing one cuz my boys are both teenagers and one of ’em is practically out the door. But I fucking love Minivans . Thanks for asking.

[00:12:57] Podcast Swap: Mac Observer’s Daily Observations Podcast

[00:12:57] PodSwap: Let’s take a little break to do a podcast swap. [00:13:00] Swap is with the Mac Observer’s Daily Observations podcast. Do you follow Apple News? Do you listen to podcasts? Well, I already know the answer to that one. Well, there’s a podcast all about Apple News that we would like you to check out. It’s the Mac Observer’s Daily Observations podcast.

And when they say daily, they, they mean it for the most part Monday through Friday. That’s reasonable. Each of those days, they hit you with 20 minutes of talk. The most interesting Apple stories, they’ve been doing it since 2014. Um, and that’s like . They had the announcement of the Apple Watch. Remember the free U2 album?

Nobody Wanted. Get that thing off of my computer. Uh, the announcement of Apple TV plus the transition to Apple silicon. I mean, all these things that have happened have been covered by the Mac Observers Daily Observations podcast, it’s ripped from the virtual pages of Mac observer.com host Kent Ray brings in Mac Observer staffers and other tech types for quick, informative and entertaining talk [00:14:00] centered on the apple and related stories of the day.

So if you follow Apple News and you listen to podcasts, again, you’re already busted on that one. Put this in your ear, the Mac Observers Daily Observations podcast [email protected] wherever you get podcasts.

[00:14:17] AI-Assisted Coding

[00:14:17] Jeffrey: what do you wanna talk about?

[00:14:19] Christina: Uh, so, um, I guess, well, I guess that was kind of our mental health corner,

[00:14:24] Jeffrey: I think that, I love the idea of that standing in for mental

[00:14:27] Christina: Uh, honestly, I do too. I do too. Um, okay, so on our topic list, you have chat and G P T. And this I wanna hear about because, because you’ve been going, you’ve, uh, you said that this is like, taken over your life over the last week.

And, uh, and I, I don’t wanna hear more about.

[00:14:43] Jeffrey: Yeah. Okay, so here’s what, and, and just the footnote is this is not one of the existential conversations about chat g p t. Um, I like kind of had that experience in my own head initially and then wanted to kind of start trying it. And um, [00:15:00] and what I’ve been using it for, uh, for the last week is like tons of coding help basically.

Um, and so like, here’s, here’s my deal, my sort of level of literacy when it comes to programming. Um, I have a comfort level with Python. I couldn’t just write you a Python script, I mean, That was 10 or 15 lines. I could write you one that was two or three. Um, I have a comfort with Bash scripting. I’ve usually, I may learn that mostly through Stack Overflow, um, and borrowing code from people that were, you know, kind enough to, to offer it to me.

Um, and I generally know what a computer is capable of doing and how you talk. Right. Like, what, what can you ask a computer to do reasonably, right? I know these things, but I always need help to actually do them. And so until recently, that would be, you know, maybe I had a friend like Brett, and you know, Brett and I have worked together on a project.

So maybe I’d just be like, I’d [00:16:00] call Brett. I’d be like, run the clock. You’re on the clock. I want you to help me figure out the last bit of this script. Um, or I don’t understand, uh, how this particular function works. It’s, I’m just having all this trouble. Um, and Brett would very kindly help me out. Um, if I didn’t have Brett, I have Stack Overflow.

Um, but I’ve always thought of myself as someone who, if I just had someone next to me all the time who could code to my specifications, I could do so much for myself and for my workflow that I cannot otherwise do cause I’m too slow or because it’s too much of a pain in the ass to search out the answer.

And so it is, it is in that sort of context that I turned to chat g p t last week first with a, with a script I needed but didn’t have. Um, it was very simple. I was actually doing some budgeting and I wanted to be able to quickly throw a number into my terminal and have that number, [00:17:00] uh, be turned into the total with sales tax.

So, you know, looking at some of this 1599, I put it in, comes out and it adds it to my clipboard the whole bit, right? And so I wrote this paragraph about what I wanted it to do, ends with it putting, uh, the amount onto my clipboard so I can then paste it into, you need a budget, why a, my, my wide app software?

And it did it in about five minutes. And then I was able to kind of do some finessing, add a few features, kind of do a little bit of yes, ending ala Brett Terpstra. Um, and pretty soon I had this like, you know, pretty complex script for what I’m normally able to produce. I had to run it and get errors and run the errors by chat G P T.

So it’s like I still needed to know how the process works, right? And I needed to be able to read an error enough to know that like, oh, this error is something I should run by chat G P T, or this error is just a stupid thing I did and I can fix it myself. And so I started with this like sales tax [00:18:00] script.

Then I took an old script I had from my journalism days, like a bash script. And it was just something that used this, there’s this amazing tool called CSV Kit. It’s created by Christopher Gross. Um, who does a lot of like news nerd programming and it’s just a suite of tools to work with CSV files or without opening an Excel file to convert it to csv, all this different stuff.

And I used to always, when I got a new data set, I always did this like audit using a few of the CSV kit tools that would kind of just give me like, here’s like, um, you know, the most common values for each column. Here’s it is like everything. It was like a two or three page like audit report, right? And, um, and that was written in Bash.

I did it, wrote it totally using Stack Overflow. And by the way, one of those questions per our conversation last week turned into one of that dude’s weirds videos of my Stack Overflow questions. Um, And what you do is you just, you know, you type an input file, you type an output file, and it creates a text file data audit of the, um, of the data [00:19:00] you just got.

Super useful, super simple. Even today though, I can’t read that bash file. Like if I try to read it, I’m just like, I don’t know what the fuck this stuff is. I basically treated Stack overflow by chat g like chat G p T, and went like line by line

[00:19:15] Christina: Right.

[00:19:17] Jeffrey: And so I was looking at, I’m like, you know what, you know what this script doesn’t have and I’ve never done.

It’s one of those things where if I type, you know, the name of the script and then uh, you know, uh, dash dash help, it tells me what this thing is and, and what the correct usage is. And so I was like, I don’t, I don’t really know how to do those things. So I was like, chant. How do you add a help function? So you, I said, well, it turns out you add a help function.

I was like, cool, what does that look like? They gave me an example and I’m like, cool, what would it look like for this script? And I just plugged the whole script in, right? And it gave me text for a help function that is now part of my script. Um, and so that’s, that’s some of how I’ve been using it. Now the [00:20:00] other way that I’ve been using it is APIs are something where it’s like, I thousand percent understand how an API works, why it exists, what you do when you’re calling an api.

I mean, I get it, like conceptually, I totally get it and I can usually make simple requests just off the documentation cuz one would hope the documentation was good enough to help you make simple requests. But what I did is I took a list of like 10 APIs that I’ve used in the past, and I just said, show me a complex, a p I call for the Airtable, a p i.

Show me a complex call for the Twitter API for the, you know, I just like went down the road and I learned so much by looking at, cuz they, it’ll show me a complex, um, a complex call and then it will, it will tell me in plain language what it’s doing. And so I was able to do that with a bunch of APIs, which allowed me to kind of tighten up a few API calls I make, uh, as part of my workflow.[00:21:00]

It was just this loving friend . And, and the fact that you could Yes. And that fucking thing to death. I mean, I can continue. I remember one of the things I tend to do is like, if there’s an API I want to know a lot about, I’ll say, I’ll start with, show me a complex api. I’ll say, now show me 10 features you didn’t show me in the last one.

Right. And they’ll be like, now show me five features you didn’t show me in the last one. Um, and you can just keep going in it. And I, and what’s funny is I started to think it was human because. When I’m working with someone like Brett, I, there’s a point pretty early on where you just feel bad. Like, I’m sorry I’m bugging you with all these remedial fucking questions that like are a little bit beyond what I could have Googled, but not too far beyond.

Right? So I found myself wanting to ask nicely, but then I was like, that’s fucked up. And so I just started being really straight, like it would’ve been as a text, if this were a text conversation, I would, I would have [00:22:00] way too little affect and everyone would wonder what I meant. But with this thing I’m just like, yeah, now show me five more.

Now show me five more sucker. Come on motherfucker, show me five more. I bet you’re not done. You know, like that’s essentially what, that became my , my sort of attitude in using chat G p T And the, the one thing I regret is that I had, I had put a Firefox extension on that allows you to download your different chats and um, and as markdown files, And I had been doing that pretty faithfully, but I had built up quite a bit of questions and today, uh, they’ve hopefully temporarily, uh, made it so you could not see or interact with those previous chats because of the overload that they’re dealing with.

[00:22:45] Christina: Mm-hmm.

[00:22:46] Jeffrey: And, uh, and I’m a little scared cuz I, I really got, I’ve gotten quite good and clean at essentially creating for myself documentation on an API or on a certain, you know, feature or whatever by asking [00:23:00] very specific questions that it allows you to build on it, you know, you can build on previous questions.

And so anyway, I have just found and Unselfconsciously, I don’t care that I can now write scripts that are mostly existent because of chat G P T. That’s fine. I knew that I wanted him and I, I know how to think in that way. Uh, hold on, that’s my mom calling.

[00:23:21] Christina: Sure.

[00:23:22] Jeffrey: I like, I feel, I feel good about myself because I know what it is I need to ask, and that itself is, is half the work, and so I’m just, I’m putting that out to you.

It’s a beautiful, beautiful new life I’m living. If they can just keep this thing online, which seems to be a struggle right now.

[00:23:39] Christina: Yeah. Yeah. Um, I, uh, I know some stuff about that that I can’t talk about, uh, in terms of like the, the, the, the, um, the strain that things are under. But yes, um, uh, they’ve said this publicly, so, so this isn’t me, uh, sharing anything outta school, but, uh, I, the success of this, uh, took everyone by [00:24:00] surprise because it completely, I think, surpassed everyone’s expectations.

[00:24:04] Jeffrey: Because they didn’t realize they were changing society and how it functions.

[00:24:08] Christina: Well, I, no, I, I, I think they did. I think that for a lot of people, and this is what Sam Altman, uh, mentioned this, and, and, uh, I think I’m stealing from him. Uh, someone else had mentioned this too, like the, the, the functionality already kind of existed, like copilot already did a lot of what you’re talking about.

And in fact, copilot Labs, which is a a a Vs code extension, which has been out for a few months now, has had a, um, which basically does like experimental features, um, which, uh, has had a feature like explain for a while that’ll let you, you know, highlight a block of code and then it’ll explain it to you very much like exactly what you were using co-pilot for.

And, um, you know, has, uh, and hey, GitHub is, is a thing that we announced at GitHub Universe, which will also be very similar to, to kind of this thing where you can kind of talk to, to copilot and, and have it, um, you know, give you responses and whatnot. [00:25:00] Um, so. , a lot of these things were, these models were already kind of available.

Granted, uh, you know, co-pilot costs money. Um, but, you know, these things were possible. The interface was different. And, and I think that’s what has changed things with this is by making it a chatbot, which is really smart. The interface is, I think was the, was the perfect thing to kind of go mainstream in a way that I, I don’t think I, I certainly didn’t anticipate it taking, I, I knew it would be big and I knew all this stuff would be big.

I’ve been really, really bullish on AI for the last couple of years, actually, largely because of G P T three and, and the things that I’ve seen. Um, you know, companies building also things like stable diffusion. Like I, I’ve, uh, I’m somebody who tends to be skeptical a lot of times, but this has been an area where I’ve absolutely not been.

Um, because like you, when I’ve played around with it and I’ve, I’ve seen the results, I’ve just. Okay. This is, this is like an iPhone level type of thing, right? This is, this is the

[00:25:56] Jeffrey: Yeah, yeah,

[00:25:57] Christina: I, I, I’m not saying that they’re the same, but I’m just saying, [00:26:00] like, in my mind when you think of technological advancements like this, this feels ai, this, this next generation of, uh, you know, uh, natural language AI feels like an iPhone type of moment, right?

Um, but I feel like chat G P T, which is still very, very primitive, which is what makes it so e exciting to me, I think is the, such a great interface for. Showing what the power of these models are, because everything else has required either you know, you to know a little bit of something, you know, be, you know, have, have a a, a code editor and an extension up or, or be using, you know, something like, like Dolly too, which, you know, they give you free credits, but it’s also, you know, takes a little bit of time.

It feels, it just, it feels different. This, I think just the, the chat nature of it and, and the, uh, instantaneous nature of it and the fact that, like you said, you can get it to do so many different things is really, really impressive, uh, to the point that, yeah, you know, people are now building so many things on top of [00:27:00] it.

It’ll be interesting to see what Open AI decides to do in terms of, you know, I think that they. probably will have always have some sort of free tier, but it’ll be interesting to see like how they start to rate limit stuff. Cuz you do, you’re already seeing businesses starting to basically be like, yeah, we’re building off of this.

Which I think it’s fine to do the model, but it’s like the compute resources have a very real cost. Uh, that’s the thing that people don’t understand enough about, um, AI stuff is that. Uh, and one of the reasons why, uh, frankly, it’s harder to do these things locally when we talked about, um, one of the, um, the iOS apps that, that does stable diffusion locally on the iPhone last week as, uh, was my, my, uh, couple weeks ago rather, uh, was, was my pick of the week.

Um, draw things, which is a great app and you can definitely, if, if you were to, you know, massage things enough, you could get some of those models and sizes that you could run locally, like on your, on your local machine. The problem is, is that a, that takes up a lot of resources and time and whatnot, and then to update it would take bandwidth, but b.

So it’s just not gonna be feasible in a lot of situations. But for these things [00:28:00] to work, what makes ’em so magical is that they are doing all this stuff, you know, in the cloud and, um, and there’s, that’s a finite resource and there’s like a hard cost to that stuff. So as this stuff grows, that’s, uh, I, I think that not just, uh, the open AI stuff, but any of these services, we’ll, we’ll start to see them struggle a little bit until people can figure out, you know, how to, how to scale them.

But, but it’s super exciting. But I love hearing about this and I I love that you were using these for these reasons. If you haven’t played with copilot, um, definitely

[00:28:31] Jeffrey: I should get there

[00:28:32] Christina: no, you sh you should do it because I think that that would be, cuz it, it, it lives in vs code, but you can also use it with JetBrains and with um, uh, uh, Neo Vem and.

It’ll, it’ll do very similar things, um, to what chat, chat g p t is doing. Uh, in that case, it’s more predictive in, in terms of like if you’re, if you’re typing something, it’ll kind of auto complete for you, um, in, in some ways. But you can also have it explain things and you can even kind of like ask [00:29:00] it specifically to do stuff for you like that.

So it’s got some similarities. Um, like, like I said, um, the, uh, like the explain feature in the labs, you know, you can highlight a block of code and then it’ll give you like the real, you know, language reaction is, is great. And, and hey, GitHub is, is gonna bring some of that stuff to it too. But, um, I love how you’re using chat G P T and I love, honestly, like this is, it’s, it’s funny because, um, internally, uh, you know, we never wanted to, to draw illusions to, to stack overflow cuz we wanna be respectful of them and everything.

But you’re not the first person who’s drawn kind of comparisons between things that you have been finding out from. You know, uh, chat g p t and, and, and using it the same way that many of us have used Stack Overflow and things like that. Um, which, which is great, but I, I, I, I really, the fact a, it’s great for Brett not to be bother as much, but also like you’ve gotten a lot out of it and you, and you’ve learned, I mean,

[00:29:58] Jeffrey: a lot. Oh, I’ve [00:30:00] learned a ton.

[00:30:00] Christina: that, that’s the thing I think that enough people don’t talk about when we talk about like AI stuff, we focus on the jobs that we’ll displace.

We focus on the plagiarism and the other stuff that it will do, and we, we focus on all these negatives of like all these instances where people are not going to have to learn. And there is truth to that. I’m not discounting that at all, but I also think that there’s truth and we don’t talk enough about it.

Scenarios like yours where you’ve learned a ton now and you’ve been able to do things that you were not doing beforehand, right? It’s not like you were learning that stuff beforehand. It’s it, you were asking Brett, you were doing other things. It’s not like you’re really rocking it. Right. So, The net positive one.

This a, is that stuff has done faster and and better, and you’ve got things that are gonna working the way you want it to, but also like you’ve learned a lot too, you know, and, and, and it have, it sounds like, have a much better comprehension of what you’ve actually done than before. You know, when you were just putting together, you know, these, these complicated batch scripts that you’re like, I know it works, but I don’t [00:31:00] know what it says.

[00:31:01] Jeffrey: Yeah, in fact, I, I have to, I was like, oh, I should finally write comments for this bash script and I’ll go through each thing here in Chat J pt, so I understand it, and then I’ll see if I feel like I could put my own words to it. I mean, there, there’s certainly, like, I certainly have instances where I would, you know, I would pause before using what was there, and I’m always testing it.

I mean, I’m not asking it to do something where the output is something really critical, you

[00:31:24] Christina: right? Well, that’s the thing, right? Yeah. Sometimes it’ll work, sometimes it won’t. And sometimes the explanations. The only tricky thing is the explanations are so accurate sounding. They can sometimes be wrong, but I think. For most of what you’re doing. I don’t think that’s a problem at all, so,

[00:31:40] Jeffrey: What I’m essentially doing is, okay, chat. G p t said I should do this. Let’s go over here and test it. No, that’s not working. Let’s go back to chat. G p T. You know, like the, the real thrill is it, it gives you code that produces an error. You give it the error and then it gets it, right? It’s like, it’s almost like that is why it’s like having someone next to you, right?

Or it’s just like a person, not a, [00:32:00] not a bot. And I definitely am, am cautious of like the like white male, uh, threat of like, you sound awful, confident. This must be true

[00:32:10] Christina: Well, but, but it’s been designed in a way that it is and it, it, it, it’s kind of a, it, it’s kind of fun in some ways, right? Because it does sound like just completely, like likewise, I found with most programming stuff, at least at the level that I’m doing things at, and I’m doing things at a very similar level, that you’re doing things.

I think that it’s, it’s fairly reliable and again, you test things, not running things in production before, you know, things have been tested and whatnot. But there, there have been some circumstances where, where it’s been wrong, but at least from my, this is anecdotal, but from my experiences, it’s a lot more accurate in terms of like explaining code blocks and whatnot than it is, uh, in terms of like, cuz it can sound just as accurate explaining, you know, like, uh, NCHE or, or Kafka,

[00:32:55] Jeffrey: Yeah. Right, right, right, right, right, right.

[00:32:56] Christina: where it can also sound very, very, uh, [00:33:00] big. But it’s, it’s, um, uh, it, it’s, it’s funny, uh, Joanna Stern from, um, wall Street Journal to a great video where she used chat G P T to write, um, to try to see if she could use it to pass AP English. And so she went to an ap, an AP English class in New York City, and she used it to write an essay and she asked the, the, the teacher, she was like, , you know, would you have caught that this was an ai?

And he said, no. Uh, he’s said, I gotta be honest with you, I probably wouldn’t have. Now, he did mark it down and gave her like a three, you know, which is a passing grade. Like, you know, like a, like, like, like like a B or a C. Um, and it, because he was wanting to, uh, the prompt was actually really great. It’s one that I wish I would’ve had in high school, which was comparing fair Thursday off, uh, with the metamorphosis and, and, and talking about like, existentialism in, in, um, uh, uh, uh, fair Thursday off, which I actually, again, I, I would’ve loved that in AP English.

Like, that would’ve been like my dream. Cause that’s one of my favorite movies ever. And, and I, and I love it. Um, but the, the essay that the, the [00:34:00] AI wrote confused the characters of Cameron and Ferris a few times and got some other

[00:34:03] Jeffrey: Mm.

[00:34:05] Christina: Um, and so, and, and those are things that are, that are gonna happen, right?

But like, the bulk of it, you know, like, he was like, yeah, I, I probably, I, I wouldn’t have, this wouldn’t have flagged to me that this was an ai, um, , which is, you know, concerning to a certain extent. Although I think it should also be maybe reaffirming that it’s like, okay, even though this would not have been caught by a plagiarism detector at least yet.

And even though this is something that, um, like, uh, what was I gonna say? So even though obviously people could use this to cheat or whatnot, and, and the plagiarism detectors didn’t catch it, it’s still, it, it might’ve given a passing grade, but it, it still wasn’t like a good essay, right? Like, it’s not like somebody’s gonna write, like, like to me, I, I care much less about if somebody’s look, cuz if somebody’s gonna use people cheat, if they’re gonna use a tool like that to cheat, they’re gonna cheat.

Right? I think I’d be much more concerned [00:35:00] if people were being able to create like Pulitzer worthy. , you know what I

[00:35:03] Jeffrey: Right, right,

[00:35:05] Christina: that would be much more concerning to me than, than, than if like, okay, you’re giving an essay that’s grammatically correct and that has some decent sin structure, but has factual errors and some other things, right?

That, that a human being who didn’t resear like that, a kid who frankly didn’t watch the movie or read the book mistakes they might have made. Right. Which, let’s be honest, that’s a very common scenario anyway. So I some, some of the, some of some of the, uh, like furor over that I, I think is completely overblown cuz I’m like, you’re maki