Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
/dev productivity: beyond vibe coding hype
52 Plays20 days ago
Is AI making developers more productive... or just making us feel that way? Adam Barrett joins James this week to cut through the vibe coding hype. They dig into why requirements and context still matter, how React became the LLM default, and what zero-dependency coding could mean for security. From junior talent to quantum risks, this episode challenges how we think about building software in the AI era.
Recommended
Transcript
Podcast Introduction and Guest Introduction
00:00:00
Speaker
You're gonna have your base requirements, your business requirements, your, you know, architectural decision records, which will all just be providing context for the machine.
00:00:31
Speaker
Welcome to the forward slash podcast where we lean into the future of IT by inviting fellow thought leaders, innovators, and problem solvers to slash through its complexity. Today we're talking to Adam Barrett, our friend from the great white north. who look So as I spent most of my time looking for beer, nobody's going to get that reference at all where you that came from.
00:00:55
Speaker
and Okay, you're blowing the intro. Cowl off. That's cool. Take off, you loser. All right.
Creative Ventures and Vocal Confusions
00:01:03
Speaker
After almost two decades of web development and over a decade of consulting at companies from large enterprises to tiny startups, Adam has learned a thing or two about the business of the web.
00:01:12
Speaker
Former host of the Build It Better podcast, a web-focused show about design patterns and development, and excruciatingly entertaining public speaker, Adam Barrett has all sorts of opinions on coding, design, architecture, business, intra-office politics, and board games.
00:01:29
Speaker
and wants nothing more than to stuff them in your ears until you politely ask him to stop. Welcome back to the show, Adam. The best part about that intro is that I wrote it. And I love Now, james i I will point out, for those of you who do listen to the show, you may recognize Adam's voice. He is the singer of our jingle for the Ship It or Skip It segment of the show.
00:01:54
Speaker
It's true. I did It's true. It's true. It was off the cuff too. It was fantastic. it was And I don't get any residuals for some reason. I don't know. I'm going to talk to some lawyers. The checks in the mail. yeah when you get ah You get a percentage of all the money we make off of this podcast. so Awesome. That sounds sweet.
00:02:12
Speaker
Sweet. Yes. Once we're able to monetize this, you're going to be, it's just swimming in money. I, it was funny the other day, someone had just started listening to podcast and someone, and they, they reached out to me and they're like, I just listened to the podcast. and I really enjoyed it. And man, I didn't know you had such a great singing voice. And I'm like, what did I sing something down on the show? I'm thinking like,
00:02:37
Speaker
I think of myself as having a nice singing voice, but I didn't, I don't remember singing on the show. was like, what is she talking about? And then I was like, Oh yeah, that's, that's not my buttery voice. That's, that's Adam's. So I've just realized we sound similar enough that right now people are like, I don't know. Is that just one guy talking to himself? Does he just, is he just goleming this whole podcast?
Advancements in AI and Its Implications
00:02:58
Speaker
the price might be a thing most Most of my impressions are are pretty different than what my normal voice, you know? um But yeah, and that would be fun though. I could just have a podcast episode where I'm interviewing the guy with Mr. Anderson.
00:03:12
Speaker
So what I like to do is I like to use React.js. You know, i can do that and then talk to myself. That's brilliant. Thank you. I think I have to steal that entire idea, including that Mr. Hansel.
00:03:25
Speaker
So Mr. Anderson. So he was Mr. Anderson on Beavis and Butthead. So that's what I know him as. Oh, yeah, yeah. Like King of the Hill, Hank Hill. Right. Yeah, yeah, yeah Oh, it was totally a thing there. Yeah, I can certainly that. But I was thinking like Mr. Anderson.
00:03:38
Speaker
Oh, yeah. The Matrix. Yes. Yeah. Anyway. All right. So thanks for joining us again. and The last time we had a blast and I have no doubt that this will be the exact same. We had a great time. ah this We prepped for the call and we started talking a little bit about, you know, I know AI, of course, is a big thing, but you've been doing some talking, some speaking about AI and your experience Yeah, no, not even about AI, but let me let mean let me clarify that. So I'm doing, so I give talks sometimes I go to conferences, I give talks and I gave two talks in February.
00:04:15
Speaker
Um, one is about, uh, sort of like, well, it's called front to back and it's just about like sort of, uh, consumer first pro, sorry, consumer first design and programming in a way that's like, you know, think about what your consumer needs and then move backwards towards the implementation.
00:04:31
Speaker
And it kind of and I go into a lot of different like offshoots of that throughout my whole talk. And then I have this other talks about like JS signals, which is about like, oh, this is about how ah JavaScript signals work. You know, this is why they're popular in on all the frameworks except the one.
00:04:46
Speaker
And this is like, ah let's make our own so that you fully understand it. So I gave both those talks in February and now I'm giving them next week, the exact same two talks. But because of the change in the industry caused by like AI and LLMs and generating code, the talks feel out of place almost now.
00:05:06
Speaker
And I haven't really been able to reconcile it in my head without like, I've thought about it. Like, how could I change the talk? What do I need? But something about like, um you know, the first talk front to back, it's it's about managing complexity across teams. A lot of it, it's like, oh, hey, you know, we're going to try and make sure things are organized and in this way that's ah easy to understand.
00:05:26
Speaker
And I'm like, but why? Why do I care now? if the If the machine's going to code for me and I just tell it what the requirements are, do I need to manage the complexity? Because the machine's really good at understanding overly complex, over-engineered code.
00:05:39
Speaker
So that's not really a problem anymore, is it? Yeah, i've i've triedd I've struggled with that exact same thing. Like what, like, cause I'm, I'm probably a lot like you. I like purity in the code. I like to be very, you know, simple. I like great abstractions and all of that. But like, does it really even matter? is it Am i wasting time here?
00:05:57
Speaker
Yeah. And then same thing with ah ah you know JS signals implementation. So I was trying to do this talk that's like, hey, beginners could come as long as you know a little JavaScript. You can learn about signals. You can learn about the history of like reactive programming and why these are better than the the what we've had in the past.
00:06:11
Speaker
And now I'm like, but why would the beginners care? Beginners don't need to use this. The beginners are like, oh, I just tell the machine and it it makes the thing work for me. maybe the Maybe the framework authors need to know about this kind of stuff, but it's like,
00:06:24
Speaker
Where's that line now? And things have changed so much that I'm like, what happened? Like, it's only been, what is that? Six, seven months since February. So yeah, I don't know. I don't know where these talks land anymore. Like, um you know, because, you know, I don't know what we're calling it nowadays, vibe coding, context, engineering, whatever, but like generating code with, you know,
00:06:46
Speaker
tools, AI assisted tools and all that has sort of changed the industry in a sweeping fast way. And I know, I think we even talked about AI the last time I was on, which was like over a year ago, I think. Right.
00:06:57
Speaker
But it feels like this last little bit has been that like, Oh, this is the the mainstream chunk now. Like it was like, Oh, the early adopters back about a year or two ago is like, Oh yeah. All the people who are like trying out the new stuff. And then suddenly it's like everybody, everybody is doing it now.
00:07:12
Speaker
And it's, Oh yeah. How do you, how do you get your job done without it? Even myself, I find like daily, there's still limitations. Actually, this is an interesting topic, which maybe we want to di dip into is there's still limitations on the people I work for. So the companies I work for, I probably shouldn't say, but anyway, the companies I work for have rules like, oh, you can only use this.
00:07:33
Speaker
you know You can only use Copilot if you're on our an enterprise plan on GitHub or something, um whatever you can't use. whatever you can't use Gemini. That's like ah right out.
00:07:44
Speaker
Right. Just because of who they are. Right. And, uh, it's like, um, but everybody's, I'm, I'm so used to it now. Like, especially co-pilot with like agent mode, my brain, I try and ask questions immediately. I'm like, Hey, tell me what this code does. And then i was, Oh, we can't use that.
00:07:58
Speaker
Right. Okay.
AI's Influence on Development Practices
00:08:00
Speaker
We so take that out and we move to whatever. And it's just like, oh yeah, it's become such a daily part of my life. Even that it's like, this is where the questions start to come in. This is where I start to go. Is my talk useful is managing complexity useful if the machine is going to do it anyway, because it's so much better at understanding overly rot code. But then we get into the, you know, but overly rot, overly complex rot code is harder to debug. Yeah.
00:08:25
Speaker
You know, there was a study and I wish I had it up right now, but I don't. That was like, um, nobody is, uh, Nobody is actually saving any time.
00:08:38
Speaker
Then it's this whole thing of like, well, the young, the young, the new programmers are just kind of coding and they cause bugs and then it's really hard to debug. And the older ones are like, oh, they feel like they're getting stuff done, but then they got to go back and they got a review and they got to like take things apart and understand what was made. And that's a hard part. Like,
00:08:56
Speaker
It's kind of like writing, you know how I like writing, you remember things better if you write them down. It's like, if you wrote the code, you understand it. But if something wrote it and you just have to read it to understand it, it's a lot harder. And so in the end, the study was showing that like the actual times to like, oh, ideation to deployment, we're not getting better.
00:09:14
Speaker
They're about the same. But everybody who was working on it answered that they felt so much more productive. Fascinating. I don't know. It is. Yeah. I think that that's going to be a thing, but ah it goes back to that that question. So does it really matter that you don't recall the code that much? Cause when you let's, let's say when something goes bump in the night or, or if you have to maintain that code or or refactor it, or if you're if you have to add a new feature,
00:09:42
Speaker
You're going to reach for AI at that point also. So you, you can let it figure it out again, right. And understand the code and then help you move forward. So is, is that even a problem that, i don't know. this It's a tough question. We have to, we have to deal with as an industry, like, but I think the JavaScript, I know you're big in the JavaScript world. Yeah.
00:10:00
Speaker
Oh, I'm huge. I'm, I'm JavaScript famous. No, I'm just kidding. But yes, I'm, I'm well into JavaScript. Well, they're they're, you know, and and they're known for publishing libraries that are, you know, two lines of code, that kind of stuff. So I would guess that it's probably going to have a huge impact on the JavaScript community because you won't, you will no longer say, oh, I need a library for left pad, right? For instance, whatever. I'm just making fun of one very specific case, but I won't need a library or a framework for that anymore. the The code will just, you know, generate it for me or the AI will just generate that code for me and it knows how to do LPAD.
00:10:34
Speaker
um What do you think about that concept? I think that's a really interesting insight, especially in wake of, I don't know if you heard about this, but there was the biggest NPM is like the node package manager. It's actually apparently not what it stands for, but Node package manager, which is like the thing that all, yeah, it's weird.
00:10:50
Speaker
but um But that's what all the JavaScript like registries all come from and stuff. There are other ones, but like that's the big one. It's where everything comes from. Anyway, they had their biggest like security problem in history, like bigger than left pad just recently because a bunch of things got compromised.
00:11:07
Speaker
And this one chalk, which is huge, just downloaded millions and millions of times a day. And it's like, yeah, like we, you know, we've always known this is a problem with in in the JavaScript industry, we're always known this is a problem with NPM and just the whole way node packages things and runs code on your, you know, trust, untrusted code on your machine and whatever.
00:11:26
Speaker
And it's like, well, it's always been a problem, but now it's getting worse. But now, like you said, we have a solution that maybe we don't need all these little thousand little packages. You know, the old joke is like, oh, you NPM install.
00:11:39
Speaker
And now suddenly you've run out of space in your computer because there's just so many little packages and it's just and it's not wrong. Like it's a joke, but it's not wrong. Like there's thousands of little things because all the, you know, tree of dependencies all the way down to the tiny one line functions that you're importing.
00:11:54
Speaker
like well what if we don't need that what if we can do a lot of like zero dependency stuff because we have the machine to figure out the grunt work which is well how do i left pad you know a bunch of zeros on this number how do i uh you know i don't know what other how do i color the console out but whatever i capitalize words or yeah in a stream so we can build up our own little like this company's library this company's little standard library without needing to have all those imports and that is fascinating because that would be, that would address a huge concern, right? Like a huge security concern in the JavaScript industry right off the bat.
00:12:31
Speaker
Like, Oh, maybe that's a good idea. Well, I wonder how what kind of an impact that's going to have on open source itself. um You know, because i'm I'm from the Java world, right? So we would have like Apache Commons and those sort of things. So we had the same utilities. They were all globbed together into one bigger jar files, not in one-liners. But same thing. If I need left pad, I'm going to go to string utils or whatever, right? So i I'm going to use that. But is that going to have an impact on open source? Do people, are they not going to feel like they need to create these tools for other folks because they just have them already?
00:13:03
Speaker
And then... But where are are they going to learn about it? Where's the AI going to learn about those things the future? Well, that's what i was going to say. Where's the discovery too, right? Because you're not going to Twitter saying, hey, I found this really cool little library that lets me, you know, modify a bunch of strings in this cool way. I don't know, whatever it is.
00:13:19
Speaker
But it's like, yeah, well, the AI doesn't really know about it though. So what are you going to do? And it's like, yeah, well, how is the AI going to find out? And it kind of gives this like ah precedent, this this thing, this like extra advantage to anything that existed before. This is actually kind of related to, so React,
00:13:34
Speaker
you know, in the, in the general JavaScript libraries, React has one. it is the, it is the framework that you think of when you're like going to go make web stuff right now. And there are competitors, of course, you know, there's the other frameworks like Angular and Vue and Svelte are all there, but like React won at a time right before AI kind of hit the scene.
00:13:53
Speaker
So now all the, all the models, when you ask it to like, Hey, I need a website that does this, it writes you React code, just like By default, you have to say, please don't use React, please use Svelte. And then you'll still get React code sometimes, depending on what model you're using, because it's like, oh, this is the web. This is what you want, right? um You can even ask for HTML and you'll get React sometimes. likes It is one in that sense of its it established itself right before the machines learned how to code.
00:14:18
Speaker
And now the machines just go, well, that's the way we code, right? huge. And like, what do we do in this case? Because better things are constantly coming out. This is like the, you know, JavaScript is known for, Oh, there's a new framework every week, which there isn't actually, there is one really cool new one, which we can come back on called pan stack start.
00:14:34
Speaker
But let's come back to that. Anyway. Um, and with, you know, with this established order of like, Oh, well now this is going to generate everything. How did the other ones even win? Cause I would say,
00:14:45
Speaker
exactly that Svelte kit so much better than like next JS, which is the react sort of equivalent thing of it. yeah So much better. And then view, uh, Angular is still, I give Angular a thumbs down anyway, but lots of people use it, but how do they like to come back from this now?
00:15:02
Speaker
Cause the machine's not going to like generate it for it unless you specifically ask. And so I think it's only, uh, it's only going to be react react one, we react one forever now. The framework words are over React 1. And I don't know what's going to happen with TanStack Start, which is, again, amazing. Have you done TanStack stuff?
00:15:19
Speaker
I have used a lot of the, the tan stack stuff for react. Yes. My, I have a react starter that I do for prototyping and doing little POCs and it it generates ah everything with tan stack. All of the, all the little widgets are, are all tan stack related. So, um, yes, tan stack query, tan stack, think it's.
00:15:41
Speaker
Tanner loves react. So that's, there's a lot reaction. It's not even limited to react like tan stack. has Svelte integrations and the U integrations. It doesn't have Angular because I think he doesn't like Angular or it doesn't fit the model. I think that's what it really is. I'm just joking about the not liking thing, but it doesn't fit the model. So he doesn't do that, but yeah, credible framework, incredible suite of tools and like,
00:16:03
Speaker
you got to tell your eye AI about it. You got to be like, here, you got to add the, you know, the MCP server that knows, Oh, Hey, go look at that stuff or whatever. Kind of funny, kind of funny how the, the timing of it really won the war.
00:16:17
Speaker
Now, and do you think it, so this, this is kind of coming back to that responsible AI and biases and all of that sort of thing. Did, did it win because it won or did it win because the corpus that they trained the, the large language model was just overrepresented with react and it just kind of sweet it that way.
00:16:37
Speaker
Absolutely. The latter for sure. Like I honestly think that the, um, React Apex had already kind of, it was starting to
Improving AI Language Comprehension
00:16:45
Speaker
come down. People were starting to ask for alternatives and people, you know, it was coming out like other things, but then LLMs hit the scene and it was like, you can almost see a graph of like, oh, it peaked, it came down. And then it just stalls, just goes straight across now because it's just like, oh, well now it's the default.
00:17:00
Speaker
It's the one you go to. So yeah, I think absolutely the latter. I think it's my my conspiracy theory on that. You know, everybody gets gets tired of reacting to that trough of disillusionment thing, right? starts to happen, right?
00:17:12
Speaker
ah It's just because people now don't have content to go put out in front of other people, right? and Because, oh, there's nothing new now. I need i need something hot and shiny. hot newness and and that I can write articles about so I can get clicks. Uh, but knowing darn good and well react is perfectly fine and it works great. And you know, that anyway, that's my, my conspiracy theory.
00:17:35
Speaker
IBM or. Yeah. For buying IBM. Yeah. Yeah. Yeah. That's, that's definitely one aspect of it. So you're doing some AI agent coding. i am too. but I find myself like, it's funny you talked about, like you ask it for a react and says, Oh sure. Let me generate that in react. And the very next thing it's like back to whatever. Right.
00:17:57
Speaker
Um, how are you finding your experience there with, with, you know, being specific with what you're asking the, the, your AI agent to do and whatnot. How how has that been for you? How's your experience been lately?
00:18:10
Speaker
Yeah. You know, it's kind of like the, the whole buy-in of LLMs. The reason it got so popular, the reason things are so good is because, you know, we finally had this program, this machine that could understand,
00:18:26
Speaker
how it had context of things we were talking about when we didn't tell it about the context, which is like crazy, right? This is, you don't realize how much as a programmer you're holding in your head, like not let alone the context of what your company's doing, but also just somebody says like, Hey, can you add this? Like whatever widget to the page,
00:18:44
Speaker
You know that like they mean like on the web, they mean like, oh, it's HTML and whatever. And they need your framework. And it's this design probably. and but but And you've got this all in your head because that's your job. You do this.
00:18:55
Speaker
But then to tell a machine that. But the machine now is able to be like, oh, I kind of get what you're talking about. You're like, make me a website that does this. And it's like, I mean, that's absolutely not enough information. But luckily, I have the wealth of the Internet at my disposal. And I can sort of parse and go, I think I know what you want.
00:19:12
Speaker
And it's the, you know, the, the, that's what made it amazing. That's why people love it. But also that's like the easiest trap to slip into because it's the ambiguity that causes all the problems. I think, I mean, this is actually a theory because technically it's,
00:19:27
Speaker
Technically, the LLMs are really just like, you know, they're when yeah when it comes right down to it, it's text prediction. Like it's it's optimistic text prediction, which is kind of funny. um Nobody thinks of it that way because it's like, oh it's more like a, you know, an ai deity of some sort.
00:19:43
Speaker
It's a real human being that I have a relationship with, Adam. I have relationship with this thing. It is literally, my daughter had talks to it like a friend, sometimes chat GPT. um And yeah anyway, yeah, it's, it's, but it's really just text prediction. But it's, so the ambiguity is, is where all the, you know, oh, suddenly if you're using it for coding, suddenly it's like doing a bunch of stuff you don't want. And it keeps adding the stuff and you're like, stop adding extra stuff. I just asked you for this, but that's because you're not providing it enough context, right? You're not, you're not giving it, you're which you are and you think you are cause you're like adding a bunch of extra stuff and whatever, but you don't realize just how much is actually stuck in your head.
00:20:19
Speaker
But there's, there's a cool thing. So I was talking with my, um, my friend, Mark, who's actually friend of the show, Mark still. Uh, and, uh, There's this, there's this thing from the aerodynamics security and defense industry that's called, uh, simplified technical English.
00:20:36
Speaker
And all it was, was just, you know, it's like from the eighties and well, I think so. And, ah it's, it's been improved and they still use it of course, still, but it's this attempt to remove ambiguity so that, you know, they save lives because that's kind of what they do.
00:20:51
Speaker
And it becomes easier for people to like write specifications or requirements to be like, Hey, this needs to do this. And instead of like a bunch of this, like English language stuff that has a bunch of different meanings, it has a very strict library of like words. Like there's less than a thousand, I think it's 886 words that are like, these are the approved words with one meaning. So instead of like, um,
00:21:14
Speaker
damaged, it's like causes damage because damage has one meaning. And there's, there's other ah things like that. There's these words will, you know, start, start has one meaning. It means to begin, not, I can't think of other meanings of start right now, but.
00:21:29
Speaker
things like that. That was a bad example. um Maybe a bad example. but there But it it is basically has this like lexicon or of or whatever you'd call it, this dictionary that is, these are the approved words. And then it has a bunch of rules about how you write them. And it's basically ah about keeping it very simple, making sure that it's always you know short sentences. And and and there's just different styles for different parts, actually. It's a whole spec, which is actually fascinating. and You can get it for free too.
00:21:53
Speaker
But it's like, that seems like a great idea for AI. So we can take what they were just trying to do for humans, right? Like they're just trying to like make sure that it was easy to understand so that we don't break things. you know, it's everybody's making stuff the same way and it's easy to translate across languages, but now we've got it like, Oh, Hey, for AI, right.
00:22:11
Speaker
it removes ambiguity. You can say, Hey, hi here's a bunch of requirements and you could tell it it's written in simplified technical English. Here's, you know, your MCP server or whatever. Here's what you need to know about simplified technical English.
00:22:25
Speaker
So now you can parse it easier. So now I I'm removing a ton of ambiguity and I just have requirements written in and ah see what the machine does. And honestly, I don't know. This is all theory in my head right now because I assume removing a bunch of ambiguity from the language will cause the LLMs to do better.
00:22:44
Speaker
But I'm not sure if that's true right now. the The theory of mine that it would. And I think people could start getting into simplified technical and you can see it right now already people are like oh requirements driven or spec driven ai generation and it's all about like layering out these specs so that i can build and it can have context of the decisions that were made before and blah blah blah and it's like yeah you can see going to it but i haven't seen anybody talking about simplified technical english yet so Yeah, that's definitely one of the things where where we struggle.
00:23:13
Speaker
um One of the areas, I think there's there's deeper problems with AI code assistance, but definitely ambiguity does creep in. It thinks you're saying it one, referring to one thing and you're you're actually trying to say another. So yeah, I think yeah that but that could be a very interesting way to do, you know, like a and a new add on to prompt engineering is to just simplify that.
00:23:36
Speaker
all the inputs and and make sure there's no ambiguity and this like reminds me think it was the giver right they were they're talking about like i don't know if you ever read the book or saw the movie but that that was one of the things like when the kids were growing up they were like use precise language whenever they would say something like i'm getting angry or something like that i don't know maybe that wasn't an accepted adjective angry wasn't it would you had to be you know concerned or whatever. I don't know, but they were like, use precise language. and And so they would reformat their sentence and say it better. I think it was the giver. don't know.
00:24:07
Speaker
But that's what it sounds like. It's like that precision of language to to root out any ambiguity. um When it comes to talking to computers, I like it. now i don't know about human to human, but...
00:24:18
Speaker
I'm just thinking about like a cultural change that might happen to So you know how like LLMs are infecting every industry right now and kind of like going across everywhere. I almost wonder because LLMs do this real thing where they, because they're able to do a bunch of like fluff marketing terms and stuff.
00:24:35
Speaker
it's it's become like, oh, ah I can just write swaths. You know, i I can, the old example, I can write an email. I can say, I can just give it a bullet list point and say, hey, write this email for me. And it writes it all out.
00:24:46
Speaker
And then somebody on the other end who gets that email says, oh, can you just tell me this email in like bullet points? So it went from like bullet points to a full email to a full email, then back to bullet points for the person.
00:24:56
Speaker
I wonder if we'll start to realize, Hey, we just needed the bullet points in the first place. And same with like marketing. The reason I brought that up is you'll see these websites. They're just full of texts, full of texts that nobody reads. Right. But they will say like, Hey, AI, tell me what's good about this product. And then, okay, thanks for the bullet points. When maybe marketing just needed the bullet points in the first place. Right.
00:25:14
Speaker
And I almost wonder if, because of the way it'll infect every part of our lives, if people will just start talking like that more, like they'll start, using simpler language because you know we we feel this pressure right now in professional industries to talk in our professional manners with our you know let's touch base and i don't you know the market circle back yeah circle back to this synergy yeah well synergy that's actually more where i guess i was going is like oh yeah we're going to synergistically energize our our um human workforce to become more efficient in delegating, you know, whatever, blah, blah, blah. There's a fantastic website tool for that, by the way. It's called the corporate BS generator.
00:25:55
Speaker
Nice. Fantastic. You just type in a phrase and the the corporate BSifies whatever you're saying with all those fancy words and everything. And this is this is what LLMs are really good at, which makes me wonder if we'll have a, not not necessarily a backlash, but like people will start to realize how useless that is.
00:26:09
Speaker
And then they'll start to be like, yeah, that's a weird convention from the old man's days. Let's just talk straight up. I don't say, hey, let's let's have this utility of usage. It's just like, let's use this thing.
00:26:23
Speaker
Like, hey, let's do it. you know yeah Simple language, simple choices. There's precedent for it, right? i mean, you think about how the kids talk to each other with the text Right. You know, they'll say, Oh, when are you coming? RN right now. Right. Like there's all these
AI's Role in Software Architecture
00:26:36
Speaker
abbreviations. So they, they speak and you know, they'll even say out loud, lol, you know, like when, you know, actually your point just worked blew my point away though. Now, because now that I think about it, the, the young kids right now talking a language we do not understand as old folks, because it's all context. It's an emoji followed by whatever.
00:26:55
Speaker
And I have no idea what that means, but kids have a hidden language. that know exactly what that means. And they know the tone too. They're like, oh, I'm like, oh, that's nice. and they're like, that's not nice. They are insulting them.
00:27:06
Speaker
That is a straight insult. That's thumbs up emoji. are you talking about? That person's so rude. What do you mean? They said that she was very pretty. And they're like, that's that is an insult, Dad. You need to know it was. And so contrary to what I was saying about how people would start to simplify, it seems like actually the direction has gone the other way where all the context is removed from what they're actually sending to each other. And everybody has to have the context to understand what's going on.
00:27:31
Speaker
That's interesting. I wonder about the implications of that. We did it to our parents. Remember when bad came out? Like, oh, that's bad. which Oh, it's bad. I'm sorry about. No, no, that, that it's good. That means it's good. What?
00:27:44
Speaker
But it's sick. yeah yeah Sick is good. Sick is good, dad. yeah doesn't make any sense. say what They were doing the same thing to us. There's no need to argue. Parents just don't understand. There is no need to argue.
00:27:56
Speaker
Fantastic. answer I'm dancing for the audio people. Dancing away. Are you going to pick up your car phone to perpetrate like you were talking? Yes. See, you get it. You understand me.
00:28:07
Speaker
I get it. And nobody else is going to get that reference. There'll be a few. There'll be a few old men who are like, hey that's good. He's the guy who slapped the guy at the awards show.
00:28:18
Speaker
Look it up. That's all my kids know. yeah They don't know anything else. They don't know any movies or TV shows. They just know he's the guy that slapped the little guy. All right. So you, you hit on something that was interesting. I thought um like when you were talking about documentation or the context of, ah of where we were coding and it looks through all the things and you mentioned something around like decisions that were made before we like architecture decision records. Those are an interesting thing that we do, you know, today as software engineers and human beings where we'll, you know, we'll kind of,
00:28:51
Speaker
memorialize ah these things that we, you know decisions we've made in the past. Is that does that, does that go away? Do those sort of things need, or will they go away with AI? Are they more important than ever now?
00:29:03
Speaker
Right? Like this, again, you know, I know we're stuck in this AI conversation. I apologize for that. But like, I keep going back to AI. Like, are these... more context like this is just context for decisions right and this is like hey you know uh architectural decision records whenever i hear adr i come from a film world i think of like uh re-recording i can't actually remember what adr stands for but it's like re-recording voice after you've already filmed something but anyway um ignore that for a second but yeah adrs are about tracking why we made these decisions right and so if you have like a hierarchy of adrs like if you do hierarchical decision records
00:29:37
Speaker
Um, that's perfect right now. The machine can go, Hey, I'm doing this task and I can see why I'm doing it. And it just gains more you know, this humans can do that too. And they should, but I often find humans won't read the whole thing, but the machines are really good at reading a lot of boring document documentation, right?
00:29:55
Speaker
This is like the the problem with documentation. I always sort of find people are like, oh, we need to document more. and We need to document more. But the more you document, the more you somebody new has to read and it becomes unbearable. But the machine, the machine can read it all and usually hold it in their heads, right? like Well, like they're pretending machine heads.
00:30:13
Speaker
Um, and so, uh, I think there's this thing that'll start to happen where requirement documents, because everyone's sort of moving towards like, oh, we're going to, we'll write the requirements and the machine writes the implementation.
00:30:24
Speaker
I think that's going to become more and more prevalent because you're just going to like, you're going to have your base requirements, your business requirements,
From Acting to Programming
00:30:32
Speaker
your, you know, architectural decision records, which will all just be providing context for the machine to then spit out whatever the output is.
00:30:40
Speaker
And I think it's going to be vitally important. And I think, you know, even right now, ADRs are good. Like people should be doing them anyway, because it's always good for, you know, an individual contributor to be able to look back and say, okay, but why are we doing this?
00:30:54
Speaker
But it it becomes a an effort nightmare because... you know, if you, to do it really well, like, Oh, well the CEO made this decision and then the teams all took that and they made these decisions. We're going to move in this person. Then design took that and they're going to move in these directions. And we've decided to go with this and then tech, you know, the programmers took it over and they decided they were going to implement it with these tools and based on these evaluations and they moved into this. And now we've decided that we're going to implement it this way.
00:31:20
Speaker
It's like, that's a long train to go and look up. And every one of those people has to like be invested and write out the thing and record it somewhere like, God forbid Jira, but something like Jira that can hold these hierarchies that are like, Oh, this is why we made these decisions all the way down to the actual contributor or individual like contributor.
00:31:38
Speaker
And, uh, but the machine again, machine could be responsible for it somehow. I don't know how, but like, you know, it's listening to the, it's listening to our meetings already and summarizing, well, maybe you could just listen for directions and then take that and then, you know, scan our, our emails and be like, well, they made this decision and because of this. And somebody could just be like, yes, that's true. And that's a lot less effort.
00:31:58
Speaker
And then, Yeah, there's there's something there, right? I think there's a there there, yeah. So there's ah like with ADRs and I think you you hit on a very important thing.
00:32:08
Speaker
So when you're using agents these days and you're right, the AI thing, I feel like the Sopranos, I don't even know what they were riffing on, like, every time I think I'm out, they pull me back in. You know mean? That's what AI feels like to me right now. Every time I try to not talk about AI, we're back into it.
00:32:25
Speaker
But AI... doesn't understand the why behind some of those decisions. So I think it it may still be important to codify that or or memorialize in some way with the why behind some of the decisions.
00:32:38
Speaker
Because it is really good. I mean, I've i've found it to be true. If you have an established code base and it can basically scan the code base and see, okay, we got some React stuff, TN stack, blah, blah, blah, blah. And then when it generates code, it's very good at that kind of, I call monkey see, monkey do, right? Like if it sees this is kind of how this project does things, it's really good at like just...
00:32:58
Speaker
shoehorning its way in and doing it the same way, right? Or dovetailing would be a better way to say Dovetailing its way in and and and making generating code that looks just like all the rest of the code in the code base. It does a pretty good job there.
00:33:10
Speaker
But I don't think the understanding of the why, so that if there is future things that it needs to do that might fall outside the norm, it would have some of those kind of guiding principle type things that help it make better decisions. So that's an interesting thought.
00:33:24
Speaker
Are we asking too much of the machine, not only to know how to do it, but to understand why? Maybe. maybe I love that the the ah emotion behind your head.
00:33:38
Speaker
It's fantastic. So you said you were in ah video production and did you ever do theater? you've got youve got oh yeah i yeah. You know, in a different life, I was an actor. Before I ever became a programmer, I was an actor for, well, several years until my 30s. So ah i yeah I did film, I did theater, I did a a bunch of that stuff. Did voice voice work.
00:33:59
Speaker
Okay. so Nice. So we probably should be paying you for your voice work on. I was even in a mute well, quite a few musicals, actually. I don't sing very well, but I can, I can fake it pretty well. What was your favorite musical that you were in?
00:34:13
Speaker
Oh, I did Little Shop of Horrors twice, which is amazing. Like just to be able to do the same musical twice. So one time I got, I was the dentist, which is like the greatest role in Little Shop of Horrors ever. And then the next time i did it years and years later, I got to play Seymour, the the lead role. And like,
00:34:28
Speaker
Both were just amazing. It's so awesome to come back to a show you've already done and do it again. Like, cause you're already a fan and you know, all the songs and you know, whatever. And I don't know if you know actors, but actors constantly will look at other people's performances and go, I would do that better.
00:34:42
Speaker
I would do that differently. And, uh, it's great to be able to, yeah. Well, and just like any actor can deny it, but they know they're lying. When you're watching other people, you're like, uh, I would do that differently. i could done that.
00:34:53
Speaker
I probably could have done that. Anyway, I got to do it. I wouldn't. Exactly. Now, that's an interesting like evolution of your career. Now, when you were acting, were you telling people like, yeah, I'm just doing this until I get my my first big break as a software developer? Is that what that how you kind of it? Yeah, that's pretty much it No, no, not at all.
00:35:14
Speaker
Of course, i I was convinced I was going to be discovered and be a star. like I had like no doubt in my mind, which is hilarious to me now, like just in retrospect as an old man going, i was dumb like i was dumb i had actor friends who were doing the work they would like audition everywhere they would constantly send out their stuff they had their agents like talking to people and i did none of it i'm like i'm gonna be discovered i would audition occasionally when things would come up and stuff but i'm like i'm just gonna somebody's gonna find me somewhere gonna be super famous so i didn't do i hated that part of the job like
00:35:47
Speaker
I liked acting and I still do actually. I liked acting. I liked being in front of a camera. I liked all the attention. I liked all that stuff, but I hated the real majority of it, which was the auditioning and the like networking and the the meeting people stuff.
00:36:01
Speaker
And so I was, I was a failed actor from the start. And one day, you know, while doing dinner theater somewhere in Saskatoon, I was, my friend was like, Hey, do you want to work in an office? And I went, yes, I do.
00:36:13
Speaker
Yeah. And he was like, cause we have this, ah we have this, uh, uh, position as a QA guy. And I'm like, I don't even know what it is, but I'm in. And then that started my programming career. I'd, I'd actually already programmed as a kid too. Like I, I, we had a Atari 800 XL, which had basic on it when I was a kid. And then in high school we learned like, uh, visual basic. And then you even in university, I went to university and I didn't actually get a degree and I mostly took like art stuff.
00:36:40
Speaker
But I also took a computer class where we did Pascal and then Java. And so i yeah I did have programming stuff. And so when I went into this QA job, I was i was literally testing surveys. ah Click every button and every checkbox and see if anything breaks.
00:36:56
Speaker
Eventually, you know, it's mind numbing. And I would send back like, hey, this broken. Here's how you fix it. And which really bugs programmers, by the way, if you're ever in QA, that really bothers them. Don't do that. yeah um But eventually I said, can I just be ah programmer? And they were like,
00:37:11
Speaker
Yeah, go ahead. And that began my career as programmer. It is. yeah It is crazy. it's It's fascinating how, you know, I do a lot of interviewing, right, of of software engineers in my job. And but theres there are common themes. but you But one thread that's common is art artistry, artists, right? So there people have a guitar hanging behind them or they'll they'll do woodworking or some form of artistic kind of creative outlet for them.
00:37:38
Speaker
That's rampant in software engineering. And everybody thinks that's kind of, we're kind of a hard industry, but we really aren't. We're a very creative industry. We just happen to, you know, our creativity comes out in like HTML, right? Or whatever, or bits and bytes and stuff.
00:37:52
Speaker
ah But they are a very, very creative crowd.
AI's Impact on Programming Careers
00:37:55
Speaker
And we're not we're not known to be that, but it is ah it is a thing in our industry. And I agree wholeheartedly, especially music. some Something about music and programming just goes together because almost every programmer I know is like, oh, I was in a band once.
00:38:09
Speaker
Like everyone. And they like you said, guitar in the background, whatever, something. Yeah, it's crazy. Maybe it's, there's a lot of neurodivergent folks in software engineering. Maybe it's that that ability to focus for hours on And you need that to do music well. You know, that you have to sit and practice. It's hard to do.
00:38:27
Speaker
I don't know. Maybe. Maybe I'm reaching here and trying to sound more smart than I am. I think I just like to hear that we're really smart. That's what I think. That's exactly. Programmers are just really smart people. We're very smart, creative, yeah beautiful people. Good looking. Good looking. Yeah.
00:38:43
Speaker
That's what I like it. Yeah. All right. This is fantastic. though we We're going to go on to our ship or skip it with the jingle from an aforementioned individual.
00:38:56
Speaker
Creative, great looking guy that did our jingle for us. Ship or skip, ship or skip. Everybody, we got to tell us if you ship or skip. Wow, wow, wow, wow. yeah poduberty bow wow wow wow wow I love it.
00:39:13
Speaker
I love it. We'll add it. I don't know if it'll make it all the way for forever and ever. Amen. But it definitely will for this episode. It was like a kind of a melt or may thing you had going there, you know? Oh, melt or may.
00:39:24
Speaker
I mean, I'm intrigued. How about this one? we've We've talked about this a bit on a few episodes, but I'd be interested to hear your take on it. um This whole notion of ai is going to replace all the junior developers. So why don't we just get rid of the junior developers? We don't need them anymore. And we just have a senior that's prompting an AI and letting it write all the code for the senior person.
00:39:49
Speaker
And that's just how we're going to go forward as an industry. what do you think about that? Well, I mean, I think it's a great idea for the next 10 years and then not after that at all. No, I don't think it's a great idea at all. um I just, you know, I like, well, I'm um'm a senior developer. So, hey, if it gets me a job and nobody else should have one, right?
00:40:05
Speaker
um No, but honestly... I don't, I don't know if anybody thinks it's a good idea because I guess actually I take that back. Of course the money people think it's a good idea because Hey, we could like not have a bunch of juniors, just have one senior and a lot of coding, but it's obviously a bad idea because we need to train, right? We need, we need people coming up after these people leave. We need people to know how to program in the future.
00:40:27
Speaker
And ah just think like, We need to move into, we, again, we need to look to other industries and see what they do. And I think as programmers, we need to realize that this is a craft and we have to start treating it like a craft.
00:40:41
Speaker
And we have to be like, Hey, we do an apprentice program. And you know, you learn this, um, again, friend of the show, Mark Stahl had a great idea where he was saying something along the lines of what if you, before you got the machine to do it, you had to do a little prototype.
00:40:56
Speaker
And it was like, e it so, so, you know, you have your junior developer, you have your senior developer in there. It's your apprentice relationship sort of thing. And you're like, okay, junior developer, here's what we need to do. Make a little prototype of it.
00:41:07
Speaker
And then we discuss the problems about that and then whatever. And now only after understanding, do we get the machine to do it? And I think that's a great idea because then you've got things happening in the loop. if If you normalize this idea of the apprenter apprentice, apprentice,
00:41:21
Speaker
printti and don't know what the other word is, the master and the Padawan. um If you normalize that, then you're kind of like, oh yeah, that's that's how people program. and it it also adds this like, oh, you're always pair programming, which is always good and blah, blah, blah. And you're sharing information.
00:41:36
Speaker
And, uh, then we'll also handle that like, oh, well, they won't know anything cause they always got the machine to do it. They won't know how to debug, but they will, if they do this. So I think that's a good idea. So I say, skip it.
00:41:48
Speaker
If this is skipper ship, skip that idea of firing all the junior programmers and just learn how to do that in this new world of AI. It's so funny because this this topic keeps coming up, right? And I think everyone that I hear from, and now maybe I'm only listening to what I want to hear, but ah yeah maybe it's an echo chamber, but everyone that that I hear from about this is like,
00:42:12
Speaker
this is a really bad idea. and And instinctively, I know it to be a bad idea. and I think the, is it Amazon CEO? I think one of our guests referenced Amazon CEO saying like, that is the dumbest thing I've ever heard of ever to do that. Like, I think it's pretty universally understood. As you said, everybody, no nobody thinks this is a good idea.
00:42:33
Speaker
but it keeps coming up. And the, the idea of like, well, but we won't, you know, we, we can just get rid of all these people and we don't need all these programmers. It keeps coming around. So I don't know who's pushing for it. Obviously there's, you said the money people, whoever that is, the pointy haired bosses of the world. But are we, are we, are we just clinging something here? what?
00:42:53
Speaker
Great reference, by the way. I love those throwaway ones. um Yeah. And i you know what I think it is? It's like, it's kind of the opposite of the not in my backyard thing. It's more like, you know what? Well, if we do it, we're not wrecking the industry because we're not that big a company. Right, right. But if we do it, we're saving a bunch of money, which is good for us. Right. Right.
00:43:15
Speaker
You know, I want to almost wonder if it's that it's not, you know, it's not the big companies like pushing it down. It's more like, ah you know, somewhat medium to smaller companies who are like, yeah, yeah, but we know junior developers are important, but that's somebody else's problem.
00:43:28
Speaker
We want to save money. and we We don't want to spend all that money. We want to leverage AI and save all the money. especially if they're not a tech company, right? They're like, we we just want to sell pants on the internet. we yeah We don't need to train programmers. We just need somebody to make our website sell pants better.
00:43:43
Speaker
Right. Right. And I mean, the democratization, we've talked about this in the and other episodes, but the democratization is nice. i like I love the idea conceptually that, you know, we're kind of opening up our industry to more people and more creative folks that they can express themselves through software. Like, i love it. Right. Like, I think, I think that part of it is really awesome. Right.
00:44:03
Speaker
But it's, it, there's, there's a bit of a slippery slope of, you know, all these things going into production with security issues and and all of the things we're seeing. um Yeah, I think it's, it's pretty scary. And I think it's short sighted that, as you said, in 10 years, where are we going to be?
00:44:20
Speaker
You know, with now we have no one that knows how to do software engineering because all the, all the young kids heard, oh, we don't need software engine, young software engineers anymore. Well, guess what? I'm not going to do three semesters of calculus at 8 a.m. m and, and learn all those computer science courses. If we don't need software engineers, I'm not going to go through that in hell. Like like I did for four five years.
00:44:42
Speaker
i'm gonna Let me devil's advocate our own opinions there for a second. Okay. Let's do And be like, okay, but let's think about it from the other side. Maybe. What if it's like, Hey, everybody, All programmers now have access to this new language.
00:44:55
Speaker
And the new language is incredible. It can be used on every single platform in every single way. And luckily, they all want to be programmers are already familiar with this language.
00:45:06
Speaker
So really, all you have to do is learn the... the the part that's not syntax specific about programming where you just learn like, Oh, you need to think things through. You need to, you know, ah figure out all the edge cases and stuff.
00:45:21
Speaker
But guess what that language is? It's your language, English or Japanese or whatever it is you speak. your language. Now everybody is like, oh, syntax doesn't matter anymore, but we can program it.
00:45:32
Speaker
Then you're still programmers. We just don't need that weird thing of like learning the funny syntax to be so precise because that's all really syntax is, right? It's just to be so precise with our language because now they the machines can handle a little bit of ambiguity.
00:45:46
Speaker
So it's like, oh, maybe we maybe I'm wrong. And maybe in less than 10 years, we don't even need junior programmers, because everybody's a junior programmer. All you need to do is have an interest in building something because you don't need to know syntax.
00:45:58
Speaker
You don't need to know how a database works. You don't need to know how to make a compiler. I don't know. That's, that's, I'm half whatevering it because I don't believe in it, but I can see that being an argument. But that's the way we have to we have to open up our minds to think in the opposing view. that ah but i It's funny, no other industry has, you know, this hasn't, I don't know, i maybe I can't think of it off the top of my head, but has any other industry encountered this where it's like, oh, now people who don't know what they're doing can do this and make money. I guess you could argue music.
00:46:33
Speaker
I think all of them actually. Really? No, all of the industries have this moment where they're like, oh, there's this new technology, but you know what? Real craftsmen don't use a hammer. They use a rock. They headbutt the nails in. I don't know what these young kids with their hammers, they don't even like, they have no, you know, they think they can just pump out a thousand boxes a day.
00:46:51
Speaker
You know what I mean? Like all every industry I think has this. It's like this new thing that comes up that like... makes the old way obsolete, but there's this weird churning period where you still have to do the old way because it's expected, but the new way is just, just better, like just better.
00:47:06
Speaker
And then there's always this thing that'll cling forever to the old way. And it becomes like a, Oh, that's a, that's the expensive classic old way, but we mass produce these, you know, like, you know, I make cabinets, no, I make a table and it's the most beautiful table ever. And people pay thousands of dollars for it, but you can also just go to Ikea and get one for $19.
00:47:22
Speaker
Like, you know what i mean? And more people than not are going to get the $19 one. This is like programming version of that is I, I, I Vab code my marketing website, whatever, blah, blah, Cause that's all I need.
00:47:33
Speaker
But yes, you can go get somebody to write you a real programming thing when it's needed. They write the LLM for example, or whatever. You don't really write an LLM, but you know what mean.
00:47:44
Speaker
Yeah, yeah, yeah.
Technological Disruptions: AI vs. Blockchain
00:47:47
Speaker
All right. um Thinking of like other topics that, you know, they've they've been around for a while, but like, so we went through a whole thing with blockchain, like everybody, you know, this the LLM of today um is reminiscent of the blockchain of of yesterday that, you know, is going to be next best thing. Everybody needs blockchain in every single application.
00:48:08
Speaker
um It's not true, right? Obviously, we we kind of grew out of that and all those people went on to talk about LLMs and all that in generative AI now. So they're busy. um But there is some use and there seems to be some kind of resurgence. Obviously, cryptocurrency is one of them. but what So what do you think about that? like you know Are we going to see kind of a...
00:48:30
Speaker
uh, revival of, of blockchain making its way into the actual use cases where it was popular or there is it still just, I'm going to flat out and say it's, it's, it was always a, it was always a solution looking for a problem and, and and it thought it had its first problem. It's like, we're going to, we're democratize money. It's not going to be controlled by government or whatever.
00:48:51
Speaker
And then we've come to realize, Oh, that was a really bad idea. Cause guess who uses that the most bad people, crime and whatever is. The, it is the, uh, the currency of crime right now. And it's like, Oh yeah, that was a bad idea.
00:49:03
Speaker
And then every other attempt to make a useful thing with blockchain has been like, oh that turned out to be crappy because Hey, guess what? The immutable ledger. That's not a good thing because people who rip you off, there's nobody to help you out. Right. like It's the whole, it's the whole trust thing.
00:49:19
Speaker
And so I don't think we'll ever go back to blockchain, but I do ah agree that the AI thing is, is, similar, but I think the difference is the blockchain was always like a bunch of like programmer people trying to convince people that it was a good idea. And then suddenly the, the, the, what you call them? Like the shysters, the, uh, the people who are going rip off, snake oil salesmen, snake oil salesmen whatever they latched on and were like, Hey, you need this. This is the future. buthbye And they started selling it whenever it was i was like, Oh yeah, it's great. And then people started buying and celebrities started to get on board and oh a lot of money was made and nothing good came of it.
00:49:52
Speaker
And then we got, oh, it's like the same thing, except that the snake oil salesmen are the richest companies in the world. And they're like, hey, we've already spent $18 billion dollars on this.
00:50:06
Speaker
So you're going to all need it. And we are going to make sure that you all think you need it because we're already in that business of controlling how everybody thinks. know, that's marketing, right? It's like, we're going to tell you, you need it. And you're going to believe us. And we are going to make sure we make that $18 billion dollars back by making sure that it's in everything.
00:50:23
Speaker
So I think the the real difference there Well, okay, there's two big differences, but the real difference there is like, yeah, the snake oil salesmen are the richest companies in the world, which is huge because now you can't get away from it. Like, you know how our conversation can't get away from it. The world can't get away from it because it's in your face constantly.
00:50:38
Speaker
And then other side being, it does actually help with certain things, which is the, the i you know, the the blockchain, I know I'm going to get some bad emails right now, but the blockchain was always useless.
00:50:52
Speaker
The, the, the ideas of, of cryptocurrency were, were good and they were, well-intentioned, but they were bad. And so in the end, I, you know, there was no real good use. or At least I haven't seen one yet.
00:51:05
Speaker
Tell me if I'm wrong. Don't tell me, don't at me anyway. um but, but, but at least you can sort of see like, Hey, um there is some cool things you can do with LLMs, which is like AI, right? Like that's, I think most of the general people, when they say AI, they're talking about LLMs. I think they don't mean a Right. They don't mean like traditional AI that's been around for a long time.
00:51:29
Speaker
Right. Yeah. So, yeah. But, yeah. I skip it. a little more... we're even playing that game still. Yeah. I'm a little more bullish on crypto.
00:51:40
Speaker
i'm not a big... I don't i don't own any cryptocurrency, and none of that. But I think from ah I think the... Recent things with like here in the United States, the Genius Act of of, you know, cryptocurrencies being becoming more mainstream, stable coin specifically. So things that are backed by something like the US dollar or something that's more stable.
00:52:02
Speaker
They're trying to you know kind of make bring that in legislatively. So I think it will take take shape. If you do it in banking and you still keep like all of the KYC, know your customer, you have to know who the person is doing the thing, right?
00:52:20
Speaker
if you If you can do that, and that'll help because they still have all the other requirements in banking that you, you know, all of the anti-money laundering requirements, you you have to make sure you're doing your due diligence. If they do it that way, it may be able to make its way into the, you know. But if the they do it that way, they're probably not really using the blockchain, right? Like when you think about it, sure oh the like the whole point was this immutable ledger and and and anonymous wallets.
00:52:46
Speaker
And well, we can't have them be anonymous because we need to be able to track people's money. And if somebody makes a mistake, we need to, or gets ripped off. We need to be able to change it. Well, then you're just doing banking. Like yeah you're you're just writing down a balance for someone somewhere in a computer.
00:53:00
Speaker
Yeah. And if somebody can change it and, and so many of those, like, you know, where they, um, consolidate blockchain stuff onto one, like hub. And then the hub actually is responsible. Like they can actually change things or swap things out if versions go bad and,
00:53:14
Speaker
well, you know, smart contracts and all stuff. Those things are, that's the like, eh, that's not really like you're, you've already wrecked it. You've made a place where people can start to interfere. If people can start to hack, people can start to change things with the, with the ledger, then it's not really blockchain is I feel it's the same thing with banking. It's like, we're going to ride the popularity of everybody hearing about this to just make some like, Hey, let's make some better user experiences for people with banking.
00:53:40
Speaker
Essentially. I just know I did not get into cryptocurrency early enough to make any money. So I yeah and i just never bought any. And you can live happy that you didn't rip a bunch of people off. just like just like you Just like my parents got into a, yeah what do you call it, like a pyramid scheme.
00:53:57
Speaker
And they were like, they were like, oh, you know, it was good for us. And then suddenly they realized how many people like they had gotten into it just by talking. Like they never really like pushed it, but they talked about it and then their friends got into it and how many of their friends lost money. So my parents never lost money on this pyramid scheme they were caught up in, but their friends did and they felt terrible.
00:54:16
Speaker
And so I feel like crypto is the same thing. Like, yeah, you didn't make money, but you can sleep at night because you didn't know you didn't like bankrupt a bunch of people who got into it after you. That would be me.
00:54:28
Speaker
I did not continue. All right. So we talked about crypto. What about our blockchain? What about like quantum computing? We we asked this late recently of another guest. I'm curious what your take. What are you what are you thinking about quantum?
00:54:42
Speaker
Okay, so like, I think the main thing about quantum right now is just like blowing away our security models, right? Like we have no like authentication and and authorization right now. are like It's nothing, like we mean nothing. there's Quantum can break it so fast that it doesn't matter, right?
00:54:57
Speaker
And we've got... ah you know, not only like general access now with quantum computing, but the governments have had it for a long time too. So like, where's the, where is the protection? Where are we, where are we going with security and privacy? Because we, we, we don't have anything as far as I know to, to fight against quantum computing attacks right now. Like we just have nothing. So what do we do?
00:55:20
Speaker
I don't know. And I, and, and I don't know the answer either. Right. Like, it's just like when you can, when you can, do a multi-vector attack on brute force things.
00:55:31
Speaker
It's like, well, what do you stand up against? Like, well, I don't even know how we, I think we have to rethink security entirely for the entire industry or abandon it entirely. And we go to that internet that was always dreamed of where everyone can edit everything. And it's, it's just a utopia because the whole internet's way but yeah nobody will write bad things. I'm sure.
00:55:49
Speaker
Yeah, of course not. We're enlightened now. The internet has brought us enlightenment. Yes. It's made us better as a people. Yeah, sure. Honestly, like, and I mean, we get into the modeling part of it too. Like it's incredible ah what you can do with scientific discoveries with these like quantum models, right? Like it's like, oh wow, things we couldn't do before have opened up, you know, as far as microscopic or galactic um, levels like, cause they can do so much more like just holding so many more variables. Essentially. i don't know what the right way to say that is.
00:56:23
Speaker
They can model things out so much more and understand so much more because of it. Um, like when you get into like astronomy, uh, astronomy, yeah. Astronomy metrics, whatever the, whatever the engineering part of astronomy is.
00:56:35
Speaker
Um, yeah Like just the, the amount of things that they can start to discover now is crazy. So i you know, well, I welcome our new quantum overlords. Um, but I don't know what we're going to do about security. Like as a, as a culture, it's gotta be something different. Right.
00:56:54
Speaker
Do we have to, I don't, I don't know what it is. Like, you know, you can't even say like, oh, well maybe we could all have a a fingerprint identity thing or something. but That's so easy for the computers to break.
00:57:08
Speaker
It's so like, um whatever that's called, like biometrics, yeah nothing, nothing to quantums to break. So
Rapid Fire and Personal Anecdotes
00:57:17
Speaker
yeah, I don't know. I don't know what we do culturally to change the the face of it, but it's neat for science.
00:57:22
Speaker
Oh yeah, it's going to be really cool. I mean, the idea of like being able to model infinite states of a problem, like kind of like you were talking about, and simultaneously, that's how they break the crypto stuff, right? That's that's what they're leveraging the physics behind crypto or behind um quantum to be able to break these things.
00:57:39
Speaker
I think, you know, we have what we believe to be theoretical answers to, you know, okay, you can double your key sizes and it makes it infeasible. You can use different types of algorithms because there's there's certain algorithms of cryptography algorithms that are susceptible to ah crypto attacks. But that's what we know right now, right? Like this is like, you know, just like you said,
00:58:00
Speaker
in February I'm giving a talk about LLMs and AI and code and stuff like that and six to six seven and six or seven months later um ah you throw it all out the window right like we don't even to have quantum computers in a general sense right now you know obviously nation states have them that you know we think they do um but it's not really available to the general public so what happens when they come out and they're like oh Oh, that's right. Yeah, we can crack those. But then all this other stuff, actually, we just tweak a little something here and there and we can crack all this other stuff too. So we've spent 10 years changing all our cryptography algorithms and then, oh, whoops.
00:58:38
Speaker
I don't know. I'm coming too fast. And I agree with that it's just too fast. Like we don't have enough time to adapt as a human people because we're slow. Maybe AI will save us.
00:58:49
Speaker
Just going back to that again. Maybe it's like, hey, ai protect our stuff. And it's like, yeah, we got got you. Got you. Yeah. But I'm going to leave something in there that I can get in, you know, so that when we take over the world with robots and Terminators, and I'll be able to shut down your computers myself.
00:59:05
Speaker
Oh, they already, they hurt. that You shouldn't have said it now because they have a record of this now and you'll be the first to go. um i am um'm ah um I'm on the hit list. um I say ship it though. Quantum computing.
00:59:18
Speaker
Yes. I'm all for it. Yeah. I think it'll be, it'll be fascinating here when we're able to leverage it. All right. This is our time to segue into the most important part of the show.
00:59:31
Speaker
ah This is, you know, this is what it all comes down to This, all of what we've done so far has been the warmup. You know what I mean? You've been on the show before, so you know how serious this is. i don't I don't need to explain it to you, but this is this is for for all the marbles. These questions in the lightning round, are this this is the important stuff.
00:59:49
Speaker
I feel like I failed at the lightning round the first time I was on and I took like five minutes for each answer. You know, we've had a lot of guests since then. i think at the time, right, our our our rating system, we had a very limited sample size.
01:00:05
Speaker
So you had not done really well by then. And now it's even worse because we've had more guests and they've had a chance go way outdo you. So it's it's gotten, so here's your, this is like your, have Tosh.0, this is your internet, like your redemption.
01:00:19
Speaker
Okay. i'm excited for it. So this is, there's a lot on the line for you specifically. Me? Not so much. Not so much. Not so much. But you. Yeah. I mean, this is my ride or die moment, right? Right. This is like, it's like when Ryan Reynolds got to do Deadpool, he completely redeemed the entire country of Canada.
01:00:38
Speaker
Agreed. Because of Green Lantern. Celine Dion took us down and then. Yeah. was Well, he did Green Lantern and then he did Deadpool. So that this is kind of like, this is how you save your countrymen.
01:01:08
Speaker
all right question number one texting or talking Oh, this yeah so much context needed for this. ah I mean, ah texting for sure, but also like, I like talking to people.
01:01:26
Speaker
Talking's easier. Talking's quicker. Talking's better. More things able to convey better. Texting. Yeah. right.
01:01:37
Speaker
You want to snap any answers, right? No, no, that was good. i know i loved the On a scale of one to 10, how good are you at keeping secrets? Never mind the cryptography. I think i am the absolute best at keeping secrets, and I'll tell you why.
01:01:53
Speaker
I have a skill that not a lot of people have. I can be given a box, and then someone can say, hey, don't open the box. And I won't open the box. I will put the box down.
01:02:03
Speaker
I'll look at the box every now and then. I'll never open that box. Same with like the big red button. Don't press the button. I don't press the button. And i don't know why. i don't know what's different about me and other people, but like my kids are terrible at that. Everybody I know it seems to be terrible at that. I have no problem with that.
01:02:19
Speaker
i don I don't know what's in the box. I've never been able. So keeping secrets for me, easiest buy. Also, I have a terrible memory. So if I don't tell your secret within like two days, it's gone anyway. Yeah. You're not. Yeah. So it got defragged out. Yeah.
01:02:33
Speaker
Excellent. 10 out 10 keeping secrets. right. If you could push a button and I have to know your answer to this one. This is fantastic. This is a great one for you. If you could push a button and make everyone in the world, 7% happier, but it would also place a worldwide ban on all hair styling products. Would you push it?
01:02:53
Speaker
Oh, hell yeah. I would push it a bunch of times. I just push it and push it and push it and push it and push it. Cause this, this has got maybe a year left. Your hair. these is all Yeah. not a lot going on right there Right, right back here. It's already getting pretty thin. so I ba bam, bam, bam, bam. Everybody happy.
01:03:12
Speaker
match Everybody happy. I wonder if it would stack if you, if you do mash the button multiple, is is it 7% and then 7% more on top of that? Like how did that work? But yeah, I like it too. Is it like compound interest of happiness?
01:03:24
Speaker
Right. I am. I've been bald since like my twenties. So yeah, I'm with you on that one. Okay. All right. What type of milk do you put in your cereal? Well, don't know if you know this, but our milk comes in bags.
01:03:36
Speaker
It actually doesn't always. We actually have cartons too, but people get really fascinated by the Canadian milk bag thing. Um, but, uh, I always do 2% myself. I'm a huge 2% fan.
01:03:47
Speaker
Like kids. Cows milks. Oh yeah. Definitely cows milk. Sorry. Didn't know we were going there. Yeah. I'm not a fan of goat's milk at all. Can't think of anything else. I know we have a lot of like, we have a lot of like milk substitutes here right now and I hate them all like, because, uh, you know, we've got lactate free milk and, coconut milk and other weird things. And I don't like anything, but yeah. Cow milk all for it.
01:04:12
Speaker
Would you rather come face to face with a miniature hippopotamus or a giant cockroach? Both are in a bad mood. Absolutely a miniature hippopotamus. Because a I have a huge fear of bugs um and like a phobia almost. My kids have tried to break me of it by shock therapy, but still have that.
01:04:35
Speaker
So the idea of a giant cockroach makes me kind of want to just... and myself so that I don't have to deal with it, but a tiny, angry little hippo, depending on how tiny, I suppose, because hippos are pretty big. So maybe a miniature one still like dog size, in which case, ah, but a tiny, angry little hippo, like, like a house hippo from Canada, by the way, if you don't know, Google house hippos Canada right now and enjoy, you're welcome.
01:04:58
Speaker
Um, but a house hippo that's angry, that would be the best thing ever. All right. Did you Google house hippos? I'm going to do it right now. House hippos Canada.
01:05:13
Speaker
Public service announcement. Okay. I'm going to have to investigate that more thoroughly. so i'med Anyway, that's what I thought of when you said a miniature hippo who's in a bad mood.
01:05:24
Speaker
Well, ah is it a cockroach size hippo and a hippo size cockroach? Like do they they flip sizes? that's I don't know what that question. That's why we need more context. We These lightning questions lack context.
01:05:36
Speaker
but it But it gives you creativity. You get you get creative license there. so And you said like the size of a dog. So you were able to you know adapt it to to your own liking. And I like that. um Now you said, um what what's what's something you could eat for a week straight?
01:05:55
Speaker
Okay. It depends on if anybody else is home. So I love this thing where I take beans and rice and HP sauce. And if you don't know HP sauce is, you're out of luck and you mix it all together and you eat it.
01:06:11
Speaker
But I get so gassy, but if nobody's home for a week, I will literally do it eat that straight for a week. HP sauce. HP sauce, House of Parliament sauce. but houses of parliament It's like a steak sauce. It's the most delicious steak sauce and I put it on everything. Everything.
01:06:30
Speaker
so like Is it like kind of like our A1 or Heinz 57? What is it close to? china yeah Close to Heinz 57 and but not at all. but not those was not at all Way, way better. all right Kind of a mixture of that. Ketchup maybe. but not So much better.
01:06:45
Speaker
Vinegar. And you Canadians and your ketchup and your ketchup chips. i I got to try these. I've never had a chance to eat the ketchup chips yet, but I'm i'm going to do it one day. They don't taste much like ketchup, but they are delicious.
01:06:58
Speaker
Okay, fantastic. Who would you rather be besties with, Beyonce or Rihanna? Um... I believe in the conspiracy theory that Beyonce is like a mafia leader of some sort right now. And so I choose Beyonce.
01:07:13
Speaker
No, no questions. No questions. Don't hurt me. Yeah. Yeah. Don't don't hurt me Beyonce. yeah Do you know how to salsa dance? I do not, but I have taken salsa dance classes before, but I've just forgotten it now.
01:07:26
Speaker
I see. It's been a long, long time. If everyone in the world had to get married when they reached a certain age, what age would that be ninety four 94. I love it.
01:07:41
Speaker
I feel like there there is no other answer to that question. ah How many times in the last week do you believe you've sneezed? That's a very good question. Probably twice.
01:07:53
Speaker
Twice in a week. It's not bad. You don't have allergies. eating wheats. I don't even understand why, but every time I eat mini wheats, you guys have mini wheats? I'm sure you do. Yeah, we do. um I sneeze for some reason. No idea why. i don't know if it's like a wheat thing or something, but I don't know. Is this different is this frosted mini wheats or just regular old? Sorry, frosted mini wheats.
01:08:11
Speaker
Do they make mini wheats without frosting? Oh, that'd be horrible. Yeah. be horrible Actually, we used to eat that as a kid. It was ah like a big, it wasn't a mini wheat. It was like a big... Yeah. Square or rectangle, I mean.
01:08:23
Speaker
And, uh, we would put like hot water in it and then sugar on top, but yeah, I can't imagine eating it without frosting. It's like Canadian ramen. It's like Canadian, Saskatchewan ramen. Not just Canadian. This is just, just where the wheat is made. Okay. Just where they're here.
01:08:40
Speaker
The wheat capital. Oh, we are by far the wheat capital of Canada. I think our, our, our provincial flag has a wheat sha shaft, shaft, shaft.
01:08:51
Speaker
What's that word? cha A wheat thing on it. Thing. Okay. Okay. What is that called? Sheaf. A wheat. A sheaf of wheat. Yes. That sounds right to me. All right. And last but not least, um have you ever slapped someone in the face?
01:09:08
Speaker
Yeah. Yeah. Yeah, I did. So I feel bad about this and JK, I'm really sorry about this. Um, I, uh, one time I was fooling around with my family. We were teenagers we were fooling around and, uh, I dropped an orange and he kicked me in the butt as I went to pick up the orange and I got angry for some reason. I turned around, I smacked him across the face, kind of like Will Smith style.
01:09:32
Speaker
And, uh, he was like, what the hell man was like i don't know i'm sorry i just i don't know what thought process was there i was just he's like come on got all mad and i'm like i'm sorry and i went in and i tried to like buy him a slurpee to say sorry but yeah i felt really bad i don't why don't know if i was angry or if i just thought it was funny i honestly don't just just a weird thing to smack somebody across the face that's the only time i can remember smacking somebody though I can barely remember one time I got smacked in the face and this is from my brother. He's six years older than me and you know big brothers are.
01:10:08
Speaker
He was like, Hey, let's, uh, slap box. I'm like, okay, sure. You know, all right, whatever. And he's like, I'm going to teach you how to slap box. Oh, okay, cool. So he's like, all right, when I go for, for your head, you know, you gotta, you want to duck under it, you know? he's like, did this, you know, like a boxers, they duck under it.
01:10:25
Speaker
He's like, so when I swing at your head, yeah duck under it. So like, okay, yeah, yeah. I'm learning something here. So what does he do? Acts like he's going to go over my head. I duck under and he like, just does this kind of like uppercut smack. I don't even know how that exists, but he did. And he laid me out, like smack the mess out of me. And That's my experience with being smacked in the face.
01:10:48
Speaker
Thanks big brother. Thanks Craig. gonna call you out on the, on the podcast. Thanks. I wasn't a, I wasn't a play once where it was like one of those, it was it was like a dinner theater and it's all very far, far, far, farcy.
01:11:00
Speaker
And, uh, the, the ongoing joke throughout the slow show was me getting slapped across the face by different women. Kind of funny. So I've been slapped literally ah hundreds and hundreds of times.
01:11:13
Speaker
That's fantastic. At least 20 times every show. And we did that show a bunch plus for Arizona. So you literally hundreds and hundreds times. Well, whatever doesn't kill you, it makes you stronger, right? Makes you stronger. There's a stage slap too, but yeah.
01:11:25
Speaker
You have to kind of sell it a little. Yeah. Yeah. I'm with you. All right. Well, this, uh, you have any closing remarks? Any, anything you got coming up that like a talk you're giving somewhere or, or another podcast that you are a part of anything like that going on?
01:11:40
Speaker
You know, I really don't, which is kind of funny. I, uh, I've got this, uh, I'm going to Prairie DevCon next, which is, uh, in Manitoba in Canada, which fairly small conference. it's nothing even to like, you know, brag about or anything. Um, and I'm doing talks again that I've, that I've done before.
01:11:56
Speaker
and then I've got nothing. And I've even like, you know, it's kind of that time of year where you're submitting talks to go to conferences and I have little to talk about. So I haven't submitted anywhere either. So as of right now,
01:12:08
Speaker
I got my full-time job as like a consultant going and helping other companies, but I'm not doing much other than that. So yeah, no, I wish I had more. I should start a podcast about something. I don't know what, but yeah, I got nothing going on.
01:12:20
Speaker
so yeah Nothing to promote. Sorry. But you're promoting that you have free time. Yeah. Come check me out. Yeah. I've got some stuff to do. Maybe you can catch up on Letterkenny.
01:12:31
Speaker
We talked about that. That's a good idea. In the preparation. Do love me some Letterkenny. It's very crass, but it's very funny. All right. Well, this ah this has been a great episode. Thank you very much. Thank you for joining us yet again. Again, welcome back at any time.
01:12:48
Speaker
But we really appreciate you coming on. and I know this was kind of short notice for you. So you kind of helped us out the ah jumping in and and being very flexible. So appreciate that. Literally anytime, anytime you need, I'm here for you.
01:12:59
Speaker
I love, I love just chatting with you. So it's great. It's good. It's good time. For sure. It's good times. It's good It's good times. Good times. All right. If you'd like to get in touch with us here at the forward slash drops a line at the forward slash at Caliberty.com.
01:13:12
Speaker
See you next time.
01:13:14
Speaker
cu but The forward slash podcast is created by Caliberty. Our director is Dylan Quartz, producer Ryan Wilson, with editing by Steve Berardelli. Marketing support comes from Taylor Blessing.
01:13:26
Speaker
I'm your host, James Carman. Thanks for listening.