Scott & Mark Learn To... Are Apps Dead? - podcast episode cover

Scott & Mark Learn To... Are Apps Dead?

Feb 04, 202613 minSeason 1Ep. 32
--:--
--:--
Download Metacast podcast app
Listen to this episode in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more

Summary

This episode dives into the debate about whether traditional apps are being replaced by AI agents and chat interfaces. Scott Hanselman and Mark Russinovich argue that while text-based UIs are resurging and dynamic interfaces blur boundaries, reliable, secure, and repeatable workflows still demand well-built applications. They emphasize that AI is best suited for reasoning and ambiguity, not as a substitute for fixing poor UX or replacing robust SaaS platforms.

Episode description

In this episode, Scott Hanselman and Mark Russinovich dive into a wide-ranging conversation about the future of software, debating whether apps are dead in an era of AI agents, chat interfaces, and automation. They explore the resurgence of text-based and terminal user interfaces, the limits of using large language models as stand-ins for deterministic workflows, and why reliability, security, and repeatability still demand traditional applications and SaaS platforms. Along the way, they unpack common misconceptions about AI replacing apps, argue for better UX and APIs instead of throwing AI at broken systems, and emphasize that AI is best used for reasoning and ambiguity, not as a replacement for well-designed software.  

 

Takeaways:    

  • AI tools don’t eliminate the need for well-built apps 
  • Chat and terminal interfaces expand, not replace, software 
  • Dynamic interfaces blur boundaries, but durable apps still anchor workflows 

 

Who are they?     

View Scott Hanselman on LinkedIn  

View Mark Russinovich on LinkedIn   

 

Watch Scott and Mark Learn on YouTube 

       

Listen to other episodes at scottandmarklearn.to  

         

Discover and follow other Microsoft podcasts at microsoft.com/podcasts   

Hosted on Acast. See acast.com/privacy for more information.

Transcript

Are Apps Dead? The TUI Resurgence

Mark, are we getting to the point where it's gonna be there's no apps. It's just talk to and disembodied agent and it's gonna fire up a loop. I know you're poking me because we've talked about this My head explodes when people say apps are dead or SAS is dead. Okay, Mark and Scott learned that apps are not dead. I'm interested in your your

One of the things that I think we need to think about we need to work on is that Mark and I feel like once we've done a topic, we've done a topic and the topic is now done. Yeah. But there's the agentic fun, explorations, and people really like these shows. Because it's real world and it's factual and it's it's got clarity. So that means that it's, you know, Mark and Scott Vibecode for the sixth time. So it makes it a little hard for naming, but it does not a bad show. It's just not a a topic.

I do think that we should start looking at hacker news and tech meme and finding things that are somewhat topical, but not necessarily something that's gonna get us to spin a news cycle. Mm-hmm. So that we can talk about that Frank will get mad about. Yep. Because um like like like my like two eys, like I'm really excited about two's right now.

Um, textuise and I don't I mean this yeah, and you know what T Mux you don't agree. Tmux is becoming like the the everybody's awesome tool of choice now. Have you seen that? Yeah, dude. But that there's also arguments so that's a show, like that when you talk about twoies.

Because Jay has asked me to do a video on TUIs and I did a whole deep research thing and I spun up a bunch of agents to experiment on different TUI styles and then then MCP UI, Goose just released MCP UI, but they're doing it with Chrome and Canvas and another guy did Claude Canvas. So basically you're asking it to prototype stuff.

Should it prototype and React? Should it automate Chrome? Should it make a new TUI? Should it, you know, how do you manage agent swarms? What's the cube cuddle for agents? Like there's a lot of fun conversations happening there. People are using TUEs because they're tired of the weight. Yeah. Um Web View two.

Defining Apps and AI Limits

Huh. I don't necessarily think that that means that the next action is terminals again. Well um that c ties into another a debate that I've even just had with some people yesterday of are apps dead? Somebody claimed. Well, but see, Sam Salachi said apps are dead. Yeah, I am. And I'm like, no, they're not. But see, this is the thing, right? What is an app is a whole philosophical conversation that we can have. But we said our apps dead when DCOM came out.

And everything was gonna be a decom you know but I I've an app for me is something that has a user interface that lets something that is Uh durable, a piece of code that's durable and has a user interface. That's what I would call an So David Fowler vibe coded Tally, his money system. Yeah. It has no user interface. That's it. And it transforms it into a money potential source. Let me say user or a API interface.

Okay. An API interface. So the API interface in the case of Tally is it's a CLI. Yeah. But it is icon you know, when people say an app is dead, I think what they're saying is, you know what, we're just gonna you you want to do something, you're just gonna ask the chat bot to do it. Well yeah, you're just gonna say, Oh, you know what? Uh go f go here. Uh here's my expense reports, go OCR them and then file it.

And you know what? I didn't use an app. I didn't use an expense report app. I didn't use the chat box. And the chat box had to f and this is where I'm like, no, that makes no sense at all. No su like Why? Okay, so here's why. This is the whole Scotty and Star Trek computer. Yeah. Like that's what they want us to do.

Filing expenses. Yeah. Everybody that's got to file expenses going to their chat bot and saying, figure out how to OCR this and figure out my expenses. It's just stupid. You're going to have an app. That is like the the deterministic here's You know, upload your expenses. Mm-hmm. I'm gonna call OCR on them. I'm not asking AI to figure out how to do this. I just know like that's a flow workflow that is very deterministic.

Very it's it's gonna requ it requires execution over and over again. I need to make sure it's reliable. I need to make sure it's secure. And you're saying that that should be an app because that's a problem that we're throwing an AI at when we just should fix the expense reporting. That's right. And there's so many cases like that where you don't wanna leave yourself to the whims of an AI that might screw things up and it's

Not just a one like if this was a one off case of like, you know, nobody else in the world is ever gonna want to do this followed by that, followed by that. Right. Sure. Have AI do it. But if it's something that is repetitive And something that requires high security and reliability, you're gonna wanna codify that in a deterministic manner. And you know, maybe there's AI sprinkled in there something like, you know, AI is gonna OCR it.

Yeah. But the rest of the workflow should just be deterministic. Totally agree. This is one of my you got me on a rant here because they have a rant after'cause that that's also I hear Saz is dead. I'm like, no, it's not for the same reasons. Saz is not dead. And UX is dead. Chat's the new interface. No, that's not true either. There's so many cases where humans need to review something or they need to input something.

that chat and text is just incredibly inefficient means of doing it. Yeah. And in those cases you're gonna have a UX like a drop down that's yes or no. You're gonna say, you know, in the chat bot's gonna say yes or no. Um, you know a form? You're gonna have a chat chat with a form? Yeah. So the thing is though, th th this there's there's pushback because the the terminal folks are using ask the ask user question tool. Have you seen this? No.

Okay. So the way that people are getting around this is this tool for Claude called Ask User Question. And you tell it about the tool and then it makes dropdowns and yes nos and interviews and it's basically dynamically generating a SAT multiple choice questions and stuff like that.

AI for Reasoning, Not UX Fixes

So rather than interviewing you, it's like, Okay, what behavior do you want? Retry, fail, cue, and then you get to do that. So their dynamic user interface creation is gonna blur the lines. where you ask for a thing, you're like, I wanna go and do this problem, I wanna resize these videos. It could do that and script it. It could generate an entire app, and that would be an app by your definition, or somewhere in the middle, it dynamically creates a Okay, well let me ask you this.

Lo so let's let's just test this. Zoom it and video clipping. Mm-hmm. So often do I do it? Let's say AI is so good now that it it could do that in one prompt. Create that clip. So what's better? What I just did, which is create, you know, have it create it so now everybody can use it. Or yeah, every time somebody wants to trim something, they go, Hey, co pilot, create it. I want to trim this video. And then it's like, okay, one second, let me go create a UX for that for you.

Mm-hmm. And you're totally right. Yeah. We don't need just in time UIs for stuff that is like the if something's done twice. Then make it deterministic. Totally agree. Uh when people are using LLMs as a poor man's that that's a a phrase that people use. So as a a simplistic scripting language, you're wasting power, you're wasting the power of this thing.

Make the non deterministic thing generated script. Yeah. It's also like we blew up like we're not a scripting language. It's not a scripting and AI should be used for cases where you need reasoning. Ambiguity. It's for great places where a little bit of ambiguity happens. The expense reporting thing pisses me off. I've seen people try, oh yeah, look, Claude can automate my browser because but what you're saying is

Your UI for your expense reporting system sucks so bad. Yeah. That you need the power of the world burning computer to fix their UI. Versus the Expense Report company throwing AI at putting an API on it. And they love to go and do the whole like look at me OCRing my receipts rather than the business process problem, which is why are receipts in that format format? Can I get the receipts from American Express?

Why am I physically taking pictures of receipts? Why are the receipts entering my brain and then exiting my brain, entering my phone, going to iCloud, coming back down? Why do I have expense reports at all? It's kind of like filing taxes in the United States.

That's stupid. The government should just file the taxes for me. The company should just do the expenses for me. So it's very much high on your own supply to think that that's the way to solve the problem. It's like saying, uh, the root cause is I can't touch that. Yeah. That's just too hard to fix. So what I'll do is throw a whole bunch of crap and, you know, baling wire and yeah. I've seen another another great example is um using an AI to book travel.

Yeah, that's that's dumb. What it is that's dumb. You're gonna end up going to the wrong place on the wrong dates, probably. Yeah. Well, where is my travel API? Yeah. This is the thing. We had decom, we had Corva, we had RMI, we had REST, we had SOAP, we have we have it we have M M C P it's just like we're inventing remote method invocation over and over again and then putting an ambiguity loop around it.

And then we're surprised when things don't work out. Are we agree on that? Well, so going back, is SaaS dead? No. And I'll tell you another thing. Like if you take a look at dynamics, which is where people you would file expense reports. That's a service. It's a cloud service. It has hundreds of features, hundreds of workflows.

built into it and it's dynamically you know extensible as well. But it's running in cloud servers all over the world. So it'cause it's gotta honor regionality, you know, data sovereignty and latency. And it's got security controls all over it because this is sensitive data. And, you know, the people are like, oh, you know what? SAS like that is dead. People, you know, a startup is just gonna be able to go create a dynamics killer based on AI. Like, what are you

That doesn't make any sense. I love it when people are gonna vibe code SAP. Yeah. Like I'm looking forward to that. And I'm like, what about scale? What about regional deployments? What about security? What about compliance? What about all the features that it has? that are built on real world workflows and experience with co real customers over

at this point decades. You're just gonna go vibecode that in a in a few months or in a year or two years. Like it just doesn't make any sense to me, uh, that people extrapolate this. And not just that, but we talk about when do you want to use AI? Again, a whole lot of what these SaaS applications do is very deterministic. You're not gonna go Yeah, anyway. Let me ask you this. That's cool. This is a whole other show.

The Future of Expertise and Software

When people talk about like, oh, knowledge extraction, like that's not it's very anti human. Like people know stuff. Wha what is the I don't want to extract knowledge from humans. I wanna unlock it. I want people to be enabled by these things, not replaced by these things.

If you were gonna take something that an expert knew and then people seem to think that they can turn that into a markdown file and then have like Racinovich dot Md and it's like, okay, I I know everything, Mark knows it's in Markdown now, everything's cool. Let's go. Right. Yeah. So then what is the role of expertise in this context? Well, how are you using this to get smarter? How can I put my expertise in a markdown file? I don't It'd be a very big markdown file. It'd be monstrous.

Yeah. Yeah, your ego could fit in there as well, but we need another power. You know, uh I think it was Quentin Tarantino who said he who is most likely to make declarative statements is most likely to be called a fool in retrospect. That was so let's go and see three, five, ten years later, return to this episode of Mark and Scott and see if Mark is uh wrong or not. Did apps get replaced? It's twenty thirty. Where are we? Are we at the uh the Hillside Mall with Marty and Doc? Okay.

Like and subscribe, make sure that you leave a comment. I cannot say enough that we would rather have a excited group of people who watch the show and comment and let us reply to you than a whole bunch of views. It's much more fun to be engaged. So if you are a viewer, uh I'd love to hear people who are

first time callers, longtime listeners uh that have something to say, may make sure to offer suggestions on future episodes of Mark and Scott because uh we are reading the comments and we will put episodes together if we think that we can uh answer your questions in the show. With somebody else. So we got some more beers. Yeah. We would appreciate that.

This transcript was generated by Metacast using AI and may contain inaccuracies. Learn more about transcripts.
For the best experience, listen in Metacast app for iOS or Android