Habits vs. Motivation, Claude Co-Work, and the Apple Saga Continues

Tim Williams: Hey there and welcome to episode 7 of the Rubber Duck Radio, with me again is Paul Mason. At this point we might as well just call you the Co-host, what do you think? Paul Mason: Hey, I'm fine with that. When we started this project I wasn't sold on having the extra time to sit and talk, but I find it cathartic. Tim Williams: I'm glad you're getting some enjoyment from it. How was your weekend, getting out with the family lately? Paul Mason: It's been cold up here, so the outdoor activities are far and few between at the moment. I spent the weekend playing around with Claude Co-work while I watched the kids to let my wife have a girl's weekend out. Tim Williams: Nice. Is it your first time using it? How did you like it? Paul Mason: I feel like it's a neat platform to build out the kind of micro-projects that we, developers, are already building by just creating a new folder, loading in a bunch of files for context then firing up Cursor or Claude to build the project out. Paul Mason: This sort of creates a structured approach around that by tying it together in a UI and giving some basic tools to give people inspiration on what they can do with it. Paul Mason: What I did was use it to help me put together a slide deck on a tool we're presenting to a client soon. I loaded the full email context, wireframes of the tool, then started to slowly chip away at the presentation. Paul Mason: Here's where it helped in huge ways. I am not particularly good at putting together presentations, so I used it mostly to help with the marketing speak aspects, and translating the complex ideas into benefits. Paul Mason: I just loaded in all of the technical details of the project and gave Co-Work the objectives, then steered my way through building out a presentation plan. Paul Mason: All in all, it saved a ton of time, but wasn't radically different than how I would have done it in Cursor. I see the market they're reaching with this, and I think it's a fantastic play by Claude. Tim Williams: Yeah, I think it's a tool that reaches deeper into the technically competent, but not a coder type of user. This is a bridge somewhere between ChatGPT and Cursor. Tim Williams: Where Claude Code is even more primitive in its TUI interface, this hits a market of users that aren't going to be familiar with using the terminal to do anything. Opening a terminal window to them feels like hacking the matrix. Paul Mason: For that market, this might be a smash hit. I think Anthropic needs the network of developers they've built up to push this to their networks of technical managers, pseudo-coders and the like. Tim Williams: You're right. I see a ton of Anthropic and Claude marketing, and I think their developer marketing is OK. I'm not sure they're going to reach this new market with the same type of messaging. I'm no marketing pro though. Paul Mason: I'm sure they'll figure it out. Tim Williams: Yeah. Anyway, the saga with Apple continues. Paul Mason: They rejected Falconry Journal Pro again? Tim Williams: Two more times in fact. At this point I'm extremely frustrated. Their policies and procedures are very well designed to push small developers like me to monetize on platform. There's basically no way around it without a legal team and a few million dollars to burn. Paul Mason: So did you end up integrating Apple's In App Purchases like they wanted to with the last set of rejections? Tim Williams: I did. That wasn't enough though. They wanted me to hide all mention of the fact that you can subscribe to the app through the web portal for any user that's outside of the United States. Tim Williams: It's a one two punch. In the United States you have to provide both options. You have to support Apple In App Purchases if you have any unlock able paid content, no exceptions. Except of course if you have that legal team. Tim Williams: To make it even more complex, Apple argued that the content in my application needs to be accessible without creating an account. Meaning users can access the pro features, which all require cloud compute and storage completely anonymously. Tim Williams: I can't build a mental model around what they expect, it just doesn't make sense to me. Almost every application I have on my phone that does anything useful requires an account login. What are they smoking? Paul Mason: Let me get this straight, not only are they forcing you into on-platform payment management, from which they are going to take their 30% tax. They're also forcing you to completely re-architect how your application manages data? Tim Williams: Yup. I'm beyond frustrated at this point. Paul Mason: Is there no other appeals process you can follow? Tim Williams: I spoke to the head of the App Review department last week, and all he did was reiterate the points and double down on the policy. There wasn't a modicum of compromise in anything he proposed. Tim Williams: He actually criticized features, like using a code to unlock and register a phone as one of your active devices. I pointed out that this is how Netflix, Hulu and other providers manage activating and deactivating devices and he acted like he's never heard of it. Paul Mason: That seems mean, you're at the mercy of Apple because of their market dominance. I'm surprised there are no new class action lawsuits like the Epic Games vs. Apple. Tim Williams: There are actually. The issue is they haven't been as high profile as the Epic Games one. You'd have to know to look for them, but there have been a few pretty big ones. Tim Williams: Cameron vs. Apple was settled in 2021, that was a coalition of iOS developers led by John Cameron that alleged that Apple's 30% commission and anti-steering rules violate the Sherman Act, Clayton Act and California's Unfair Competition Law. Tim Williams: There was Miller vs. Apple which was actually dismissed by the court in 2022. This was also a coalition of indie iOS developers claiming the 30% cut is unlawfully a monopolistic tax. Tim Williams: Then there was Sullivan vs Apple whose claim was that Apple's mandatory in-app purchase rule violates California's UCL. Tim Williams: This one was actually Partially granted, a preliminary injunction forced Apple to allow metadata buttons linking to external payment sites, but the case was later merged into the broader Cameron settlement. Tim Williams: Now I'm no lawyer, but this all seems to be related to the Epic vs. Apple lawsuit. That lawsuit and their small success provided a legal foothold for the rest of the lawsuits to move forward. Essentially it set a precedent. Paul Mason: Now it makes sense why Epic vs. Apple was actually covered so widely. It was indeed epic. If you will. Tim Williams: Oof, that was a bad one. But I'll let it pass. Paul Mason: You know I can't pass a good pun. Tim Williams: I know. And this is why you're the co-host and not the host. Paul Mason: Alright, taking it too far. Moving on. Tim Williams: Joking of course. Anyway, all of this work and research to release an app that's positioned to just pay for itself. I am beginning to explore moving this completely over to a progressive web app. Paul Mason: Are there any device APIs that you couldn't implement if you went that direction? Tim Williams: That's what I'm not sure about. I haven't closely tracked the progressive web app space lately, but I suspect there might be issues with the geolocation functionality, possibly some restrictions with the file management functionality and that sort of thing. Tim Williams: Part of what makes the app sticky is you can do a heck of a lot of your data entry just by snapping a picture and uploading it to the app. Tim Williams: This iteration of the app is designed to be local first. This means that everything is designed around optimistic local updates. Tim Williams: It works completely offline using SQLite to store the relational records, then uses a delta sync method to make all records eventually consistent. This single constraint I forced on myself caused the app's complexity to explode over the first iteration. Paul Mason: I can imagine. I've played with CRDT systems a bit but haven't dived into the minutiae. That stuff flies over my head. Tim Williams: This isn't the first system I've built with this, but it is the most complex one. The CRDTs have to operate to sync SQLite with DynamoDB, but also have eventual consistency for all files and images that users upload to the app. Tim Williams: On top of that, the system has to be completely secure ensuring that the only people who have access to their data are the users who created it, not even me the developer can access data that isn't explicitly shared with me. Paul Mason: Ah, no super admin trap door in this one, huh? Tim Williams: No. To make sure my audience fully trusts the app, safety and security were some of my primary concerns. What I learned from the previous app is that this type of security sometimes makes providing support very difficult. Paul Mason: Right because you can't just copy their records and replicate a bug, you have to use their anecdotal evidence and try to rebuild any bugs from scratch. That's my biggest gripe with this type of locked down system. Paul Mason: It makes debugging ten times more complicated than it might be otherwise. Tim Williams: Yeah. Though I built my previous app this same way, I didn't address those concerns with it. This time I purposely built hatches to allow people to explicitly share data with me when they're having issues. Tim Williams: But to your point, this type of data access pattern made the app so much more complex than having a super admin back door where I have access to everything. I had to be thoughtful about the patterns I put in place to make it very apparent to users when and what they are sharing with me. Tim Williams: This was probably the most exciting challenge of this app. Beyond trying to build something better than the previous iteration, which was a bigger challenge than I was expecting, the challenge to make it a better experience long term for me was fulfilling. Paul Mason: You're speaking as if this thing is to market already. Tim Williams: It's not that, it's that I feel confident this time I can provide a better experience all around. Paul Mason: Makes sense. Tim Williams: The first iteration was truly just a tool that I needed, so I built it. I released it to the community because I thought other people might find it useful. I didn't know if there was actually a market for it. Tim Williams: Now I know there is a market for it, so I'm building for the future. My aim still isn't to make it a big moneymaker, but make it what I had initially wanted it to be once it began to catch on. Paul Mason: What did you want it to be? Tim Williams: More of a platform. In iteration one, I build only the native interfaces. There were some social features like the ability to share some data in read-only mode, but nothing beyond that. Tim Williams: This time I want it to be more social. Allow people to collaborate, share and build around it. And of course, this time I added AI features. A knowledge-base of raptor biomedical information, falconry and hunting regulations and MCPs to let people ask about their own birds habits and health. Paul Mason: You just couldn't resist the AI features could you? Tim Williams: Nope. Too useful in my opinion. Paul Mason: Anyway, maybe we should move on so we don't alienate the parts of our audience that don't care about birds of prey and your weird sport. Tim Williams: Not a bad idea. One thing occurred to me to talk about beyond the realm of code, and I think I want to make this sort of thing a regular segment. Just musings on life and productivity. Paul Mason: I like it, ok shoot. Tim Williams: I've been musing over this idea that most people operate off of motivation. I used to be this way myself as well. Do you know what I mean? Tim Williams: Things like going to the gym or cleaning the house or spending time on that hobby you want to get better at. People tend to only do these things when they feel motivated to do it. Paul Mason: How would you do it otherwise? Tim Williams: Habits. I realized that people are defined by their habits, not by their motivation. How often do you actually do the thing when you're just motivated to do the thing? Paul Mason: True. It's a low success rate. Tim Williams: So we are basically defined by our habits if you think about it. We aren't defined by what we want to do, but by what we actually do. Paul Mason: An interesting way to put it. Tim Williams: And who among us actually follow all of our motivations? Like in December you make a resolution to go to the gym the next year, that's your motivation. It's aspirational thinking until you follow through with it. Paul Mason: Yeah, that's one of the reasons I don't make resolutions. Tim Williams: This got me thinking, if I am defined by my habits, do my habits reflect who I want to be? So the first step is to take inventory. Good habits vs. bad habits. Paul Mason: Uh oh, are you going to air your dirty laundry? Tim Williams: I'll keep it somewhat general, but ok here we go. Bad habits I'd like to cut. One, too much time getting distracted by social media. I don't use social media much for the actual social aspects outside of building and maintaining communities around things I'm working on. The problem is I get distracted by it. Paul Mason: That's what it's designed to do. Tim Williams: Yeah, I'll get sucked into a political debate or some other useless argument, then lose an hour or two. I hate it. I need to be more conscious and intentional about it. I don't yet have a plan to tackle this one, but being aware of it is a first step. Paul Mason: Any others? Tim Williams: I'm an introvert, so forgetting to reach out to friends and connections to just strike up a conversation to let them know I'm thinking about them. It feels so unnatural to me so I avoid doing it even though I do regularly think about people and have a small amount of motivation to reach out. Paul Mason: Makes sense. On that one I'm the opposite. Extrovert here. Tim Williams: Then the last one is that I have too many hobbies and have a tendency to start projects and not finish them. Too many dangling projects out there with no time to finish them up. Paul Mason: Yup, I share that one. Tim Williams: I need to be more intentional about the things that I spend my time on. I need to accept that some projects I don't intend to finish, I'm just doing them to learn something or explore something. Tim Williams: So to close the loop I need to be intentional about setting them aside as explored with no intention of finishing, so they don't reoccur in my mind giving me shame for not finishing them. Paul Mason: I like that. It still gives you the freedom to explore, but with constraints. This way you can safely set things aside with intentionality. Maybe you can close the loop by inventorying what benefits that exploration gave you? Tim Williams: Brilliant idea. I think that would solve some of the shame I feel for not finishing a project. Paul Mason: See I don't have the shame complex around the unfinished projects, I feel bad about the messes I leave everywhere. Tim Williams: Making your wife clean up after you? Paul Mason: No, making my wife yell at me for making messes everywhere. Tim Williams: I'd be pissed if I was her. Just kidding. That does lead to a transition on what I think my good habits are though. One is cleanliness. I think I have a solid set of good habits around keeping my house, car and anything I come in contact with on a regular basis clean and if not clean, then organized. Paul Mason: Ok, this is one I can improve on. Spill the beans, how do you manage it? Tim Williams: I make a point of cleaning the whole house every week. Dedicate a whole hour or two to it. This serves as a checkpoint to keep the place from falling into complete disarray then costing a whole day of cleanup. Tim Williams: Then to maintain, I make a point of always cleaning up dishes immediately after I use them. No waiting. When dishes pile up over several days in a pile in the sink that one crosses into the disorganized and dirty category. It makes me feel physically ill. Tim Williams: I think it's this way because when I'm depressed, this is one that tends to slip. It's a sign to me that depression is pending or already present when I'm not generally aware of it otherwise. Paul Mason: I'm onboard with this one. I feel the same. Though I do occasionally let a nights worth of dishes soak, I'll feel bad if I start work the next day without dealing with them. Paul Mason: But I guess that's one of those things I allow myself more leash on than you do, and I'm gonna blame it on having kids. Tim Williams: I might be the same in your situation, but I have no excuse, therefore I don't allow myself that grace. Tim Williams: Ok, my next habit I think is positive and possibly defining is that I make it a point to get up everyday and workout. Either a strength building exercise with weights and pushups, or at least a 5k run. Paul Mason: That's impressive. Everyday? Tim Williams: Except for weekends a lot of the time, but on the weekend I try to get outside for some sort of physical activity. That helps round out the week and not become incredibly monotonous. Paul Mason: I could stand to learn from this one. I am not a morning person. The thought of getting up early to spend time doing anything before work for me sounds like hell. Tim Williams: The misconception people have is that I wake up motivated to do this. I don't. Tim Williams: I wake up dreading it exactly as you'd expect. The difference is holding myself accountable for doing it means everyday that I do it becomes easier and easier to the point where I don't think about it much any more. Tim Williams: The problem comes when I let the habit slip. Every day that I let it slip it becomes harder and harder to slip back into it again. Paul Mason: I get that. Tim Williams: So to sum up, here's my philosophical thought. You are defined by your habits, not your motivation. Inventory your habits if you want to know how people think of you, you'll get a better 3rd person view of yourself. Paul Mason: I think what makes this interesting is that all of those habits you just listed have really obvious parallels in how developers work. Tim Williams: Exactly. That's where my head went next. If habits define who we are as people, they absolutely define what kind of developer we are. Paul Mason: Ok, let's hear it. What are the bad developer habits first? Tim Williams: Oh, there are a few that I see over and over again. One big one is relying on motivation to clean things up later. Paul Mason: Ah yes, future-me will handle it. Tim Williams: Right, uh. Future-me is very optimistic and wildly irresponsible. Paul Mason: Same guy who's gonna write tests after the feature ships. Tim Williams: Exactly. Things like documentation, refactors, naming, deleting dead code. All of that gets punted down the road because you're motivated to ship, not to maintain. Paul Mason: And the problem is you never actually feel motivated to clean up tech debt. Tim Williams: Never. That motivation does not arrive. You have to build habits around it or it simply doesn't happen. Paul Mason: What does that habit look like in practice though? Tim Williams: Small, boring, repeatable things. Like never leaving a function worse than you found it. Or every PR includes some cleanup, even if it's tiny. Paul Mason: That's very leave the campsite cleaner than you found it. Tim Williams: Exactly. And it compounds. A year later you don't feel like a hero, but you also don't hate your codebase. Paul Mason: That's a win. Tim Williams: Another bad habit is treating AI like a vending machine instead of a collaborator. Paul Mason: Meaning? Tim Williams: Meaning devs copy-paste without reading, without understanding, without integrating it into their mental model. Paul Mason: Yeah, that's becoming... let's say, noticeable. Tim Williams: AI accelerates whatever habits you already have. If you're sloppy, it makes you sloppier faster. If you're thoughtful, it makes you dangerous in a good way. Paul Mason: So the habit isn't use AI, it's how you use it. Tim Williams: Exactly. Asking why something works. Asking what tradeoffs were made. Asking what happens when it breaks. Paul Mason: Otherwise you're just accumulating mystery meat. Tim Williams: And mystery meat code always expires at the worst possible time. Paul Mason: Usually Friday afternoon. Tim Williams: Always Friday afternoon. Paul Mason: What about good habits? Besides the obvious write tests, which everyone says and half the people ignore. Tim Williams: One good habit that took me years to build is slowing down before making something permanent. Paul Mason: Like what? Tim Williams: Database migrations. Public APIs. Auth models. Anything that's hard to undo. Paul Mason: Oh yeah. The this seemed fine at the time category. Tim Williams: Exactly. The habit is asking: Am I locking future-me into something here? Paul Mason: That's an architectural habit, not a technical one. Tim Williams: Right. And those matter more as you get senior. Another good habit is finishing the loop. Paul Mason: Meaning? Tim Williams: Deploying something and actually watching it. Logs, metrics, error rates. Not just shipping and moving on. Paul Mason: Oh, that's a big one. A lot of people mentally check out at merge. Tim Williams: Shipping is not the end. Shipping is when the rubber hits the road. Paul Mason: Sometimes loudly. Tim Williams: Very loudly. And the habit of listening to that feedback instead of getting defensive is huge. Paul Mason: That's not even just code, that's ego management. Tim Williams: Exactly. Which brings it full circle. Developer habits aren't separate from life habits. They're the same muscles. Paul Mason: Discipline over motivation. Tim Williams: Consistency over intensity. Paul Mason: And probably fewer half-finished side projects sitting in GitHub. Tim Williams: Or at least being honest about which ones are experiments and which ones matter. Paul Mason: I like that framing. It removes guilt. Tim Williams: Yeah. Habits shouldn't be about shame. They should be about alignment. Paul Mason: Alignment with who you want to be as a developer. Tim Williams: Exactly. And if you don't like what your habits say about you — that's actually good news. Paul Mason: Because habits are changeable. Tim Williams: Slowly. Boringly. Repetitively. Paul Mason: The worst kind of work. Tim Williams: The kind that actually sticks. Paul Mason: Alright, so let's zoom out a bit. Those habits make sense at the individual level, but things change when you're leading a team. Tim Williams: They do, and this is where habits either scale beautifully or blow everything up quietly. Paul Mason: Quietly is the scary part. Tim Williams: Always. The first habit that really matters when you lead is consistency. Paul Mason: In what sense? Tim Williams: In expectations, feedback, standards. If your reactions are inconsistent, the team spends all its energy trying to predict you instead of solving problems. Paul Mason: That's such an underrated point. People start optimizing for mood management. Tim Williams: Exactly. And that kills trust. A good habit is being boringly predictable as a leader. Paul Mason: Which sounds counterintuitive, but it's comforting. Tim Williams: Right. The team should know how you'll respond to a failed deploy, a missed deadline, or a bad idea. Paul Mason: And if they don't, they start hiding things. Tim Williams: Which leads to the next habit: creating psychological safety through repetition, not speeches. Paul Mason: Meaning? Tim Williams: You don't say it's safe to fail once in a meeting. You show it every time something goes wrong by not assigning blame. Paul Mason: Yeah, people watch what you do under stress. Tim Williams: Always. That's the real interview. Another habit that scales is writing things down. Paul Mason: Ah yes, the unsexy superpower. Tim Williams: Decisions, rationale, tradeoffs. If knowledge only lives in your head, you've created a bottleneck. Paul Mason: Or worse, a bus factor of one. Tim Williams: Exactly. Leaders who don't document become accidental single points of failure. Paul Mason: What about delegation habits? Tim Williams: Huge. A bad habit is delegating tasks. A good habit is delegating outcomes. Paul Mason: That's a subtle but important distinction. Tim Williams: If you hand someone a checklist, you've taught them to wait. If you give them a goal and constraints, you teach them to think. Paul Mason: And that's how teams scale without you being in every PR. Tim Williams: Exactly. Which brings us nicely into the dark side of this conversation. Paul Mason: Uh oh. Tim Williams: Bad habits that AI makes worse. Paul Mason: There are so many. Tim Williams: There really are. The biggest one is outsourcing thinking instead of effort. Paul Mason: Oof. That one hits. Tim Williams: AI is incredible at accelerating execution. It is terrible as a substitute for judgment. Paul Mason: But it feels so confident. Tim Williams: That's the trap. A bad habit is trusting fluent output without interrogating it. Paul Mason: Especially for junior devs. Tim Williams: Especially for juniors, but honestly seniors aren't immune either. Another habit AI amplifies is cargo-cult architecture. Paul Mason: Copying patterns without understanding why they exist. Tim Williams: Exactly. You get a perfectly reasonable, looking system that fails spectacularly under real constraints. Paul Mason: Because no one understood the tradeoffs. Tim Williams: Right. AI doesn't feel pain. It doesn't feel pager fatigue or production anxiety. Humans do. Paul Mason: That's such a good way to put it. Tim Williams: Another bad habit AI makes worse is avoiding uncomfortable conversations. Paul Mason: Oh interesting, how so? Tim Williams: People use AI to write feedback, performance notes, even tough Slack messages, and never build the muscle themselves. Paul Mason: So they sound polished, but hollow. Tim Williams: Exactly. Leadership is about clarity, not eloquence. AI gives you eloquence on tap. Paul Mason: But clarity still requires ownership. Tim Williams: Yes. And the last one I'll mention is speed addiction. Paul Mason: Oh yeah. Tim Williams: AI makes it easy to move fast, which can trick teams into thinking fast is the same as effective. Paul Mason: When really you're just generating more surface area for bugs. Tim Williams: Exactly. The habit leaders need is deliberately slowing certain decisions down, even when tools make speed tempting. Paul Mason: That's discipline again. Tim Williams: Always comes back to discipline. AI doesn't remove the need for good habits, it punishes bad ones faster. Paul Mason: That might be the real takeaway. Tim Williams: Yeah. AI is a force multiplier. It multiplies who you already are. Paul Mason: Which is both exciting and mildly terrifying. Tim Williams: Very much both. Paul Mason: I feel like this episode kind of snuck up on us. We started talking about tools and app store pain, and somehow ended up doing philosophy. Tim Williams: Yeah, but I think that's kind of the point. The tools change, the platforms change, the rules change. The habits are what stick around. Paul Mason: And whether you're writing code, leading a team, or just trying not to lose your mind, habits are doing most of the work behind the scenes. Tim Williams: Exactly. Motivation is loud, but temporary. Habits are quiet, but persistent. Paul Mason: And AI doesn't save you from bad habits, it just puts them on fast-forward. Tim Williams: Right. If there's one takeaway here, it's that none of these tools absolve us from thinking, caring, or being intentional. Paul Mason: They just raise the stakes. Tim Williams: So if you don't like where things feel chaotic — in your codebase, your team, or your own workflow — the fix probably isn't a new tool. Paul Mason: It's changing what you do every day. Tim Williams: Slowly. Boringly. On purpose. Paul Mason: On that uplifting note. Close us down Tim. Tim Williams: We'll wrap it up there. Thanks for listening to episode seven of Rubber Duck Radio. Paul Mason: As always, if this sparked a thought, an argument, or mild existential dread, mission accomplished. Tim Williams: We'll be back next time. Until then, take care of your habits, because they're definitely taking care of you.