Skip to content
510: Technical Debt is Technically Risk episode artwork

Oct 13, 2026 36 min

510: Technical Debt is Technically Risk

from The Bike Shed Hosted by thoughtbot

Debt metaphors abound! In this week’s episode of The Bike Shed, Aji and Joel ponder the question, what are risks in a code base? From looking at technical risk as a different lens on the idea of technical debt, to the perils inherent with introducing debt into a system, this is certainly the episode to listen to if you don’t want your own projects to blow up in your face. — Your hosts for this episode have been thoughtbot’s own Joël Quenneville and Aji Slater. Don't forget you can now also watch the show over on our YouTube channel! If you would like to support the show, head over to our GitHub page, or check out our website. Got a question or comment about the show? Write to our hosts: [email protected] This has been a thoughtbot podcast. Stay up to date by following us on social media - YouTube - LinkedIn - Mastodon - BlueSky © 2026 thoughtbot, inc.

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

510: Technical Debt is Technically Risk

0:00 36:53
of matches

Transcript Auto-generated, may contain errors

Hello and welcome to another episode of the Bike Shed, a podcast from your friends at Thoughtbot about developing great software. I'm Joel Kinville. And I'm Adi Slater, and together we're here to share a bit of what we've learned along the way. So, Adi, what's new in your world? Well, I have recently joined a new project, so I'm moving on to a team of a client that has been a Thoughtbot client for a really long time, so there are a lot of very thought-by practices, and onboarding to a new team and project is always drinking from the fire hose.

There's so much to learn, so much context to take in, new team members to meet all of these things. But one of the things that has been set up by Thoughtbotters of the past is an onboarding, it's labeled as a questionnaire, but it's basically a little take-home test about the code base and the architecture of the system and things like that. So, asking questions like, what is the slowest request coming into these couple of services, or how would you find out this information and provide a link and that sort of thing in it? It gives you kind of a guided way into the tools that are most commonly used and the places that are most commonly referenced, and just sort of things that are top of mind for somebody who's working in this project team and to learn about the code bases a whole as well. That's a really interesting idea. Did you find that was helpful for

you? It was super helpful. There's so much that is usually kind of picked up as you need it in the first era of joining a new client or a new project team that having just this little guided tour that was sort of learning your own pace and didn't just hand the resources over in a document. Like, a lot of the answers were in a lot of the documentation that's been building up over time, so far I think pretty well maintained. I haven't run into anything that is out of date yet and caused me consternation, but I'm sure I will. Any longstanding projects can have that. But, yeah,

it was incredibly helpful just to have a little bit of a point in a direction and say, answer this question. This is a kind of question that we might have to look into often, and just being pointed towards resources without just being spoon fed them, because those are facts that are so much easier just to lose, especially picking up and reading and gaining so much information all at once. The little extra step of having to do some of the work yourself and having to do a little bit of digging and a little bit of learning and uncovering makes that stick and makes it just such much more usable. And it's right at the tip of your fingers for recall. And there's just something much more palpable about the skills that you pick up in that way. At least that's been my experience

and has been, I think, for a lot of people joining this project, because it's a practice that's stayed in place and been kind of updated as new people join the team, or as the priorities and focuses shift. You can easily swap out a question if you've moved on to a new tool, or if there's something that's maybe more relevant to the more recent work. I can imagine a world where maybe this idea was taken and it wasn't executed well. It might feel a little bit patronizing. Yeah, yeah, for sure. There were definitely some questions that

I think had a little bit of, they were maybe more relevant or more in the zeitgeist of the team when they were written. So there was a little bit of back and forth with one of the questions that was like, what are you asking here? What do you want me to find out with this? It was a little vague, but getting a little bit of the backstory, you find out that, oh, that would have made sense because people were talking about it in sinks and all of this knowledge that came along with it. So we're able to refine that. I think also the amount that teammates would kind of check in and ask about

how it's going and stuff like that made it feel very collaborative in a one-step removed sort of way that because of the way that people were taking your working through these questions seriously, as actual tickets on the board, then it felt like you were working with them of the past who kind of set these things up. And also people were willing to answer questions and things like that. So it wasn't sort of treated as like, go over there, do this, and then you can come join us and work on the project once you've answered this or cleared this hurdle. It was a little bit more, yeah, so it was less kind of patronizing in that way. And it was really set up as like, this is useful, this is helpful work, and you're onboarding quickly through this is part of the work of the team.

I feel like there's some of that that's sort of trying to navigate the boundary between sort of tacit knowledge that lives in the head of the current team versus things that are written down. Yeah, yeah. And it might be an interesting way to try and answer some of those onboarding things or those knowledge transfer things that technically there already is documentation because of like, I could have just read the AWS CloudWatch or you know, there is documentation for these tools. So the team writing out these different things is maybe not as like just putting down some documentation for this is not going to be all that effective for communicating some of that. But it really was like, here's an investigation that's coming to us and so like, go do that investigation in a low stakes sort of way. And I see a little bit of that like kind of shared understanding of like how we

go about answering these questions. And how about yourself? What's new in your world? So I was recently at Rails world in Austin, Texas. I gave a talk there really well. While there,

I got to do something for the first time. I wrote a mechanical bolt. It was hoping you were going to talk about this. There were pictures in our slack. How was it?

Kind of intense. I would imagine. It's a little bit scary. Yeah, it hurt a little bit too. Yeah, that all scans that makes sense. I'm not really surprised by any of those. How

well did you do? Not necessarily time wise. How do you feel about your performance? I think I did pretty well. People seem to think I sort of stay longer on than they expected me to.

So it exceeds expectations. Great. Love it. Love to see it. So the mechanical bolt was actually a thought-bought production. This was at some point Amanda for

the Rails Foundation had put out a request and said, Hey, I've got this really cool idea. It's kind of weird. I'd love to have somebody sponsor it. I can't tell you what it is. Are you going to trust me? Wow. All right. And we were like, we're in. And yeah, so it turns out

this was a, Hey, do you want to sponsor a mechanical bowl for the Austin, Texas, Rails world? There was a lot of fun. Did many other people try it or was this a, yeah, not just a thought-bought experiment? Yeah, no, it was a lot of people at the conference where there was like sort of right outside the convention center where they had a bunch of picnic tables. People would hang out there or that's where lunch was happening. So a lot of people got a

chance to try. That's awesome. Yeah, I've never done a mechanical bowl. So I'm a little bit envious that you got to try out this new experience. So, yeah, I think there's a lot of people who sort of maybe saw it and were like, it's fun to watch other people do it. I don't know that I'd want to do

it. Sure. Any of those person, you might have a lot of reasons why maybe that's not the right choice for you. 100%. Yeah. Yeah. I can't find a waiver and everything about, you know, hey, I assume all the

risks associated with this and all that. So, yeah. And I think both ways of experiencing mechanical bowl are equally valid and fun and enjoyable is to watch other people fall on their face or to fall on your face yourself. Exactly. They're both enjoyable but for different reasons. Speaking

of risks though, I know those have been something that's been on your mind lately. They're sort of thinking about what are risks in a codebase and particularly maybe correlating the idea of technical risk almost as a different lens on the idea of technical debt. Yeah. I have been thinking recently about just this idea of technical debt and why is there a bit of a disconnect between the kind of gravity that developers talk about technical debt with and sort of the rest of different teams, maybe not kind of feeling the same peril about the risk inherent with introducing debt into a system.

I'm sure you've experienced it. Maybe they're like flippancy that other people with different roles on a team might be like, oh, yeah, sure, we'll clean that up later. And devs are like, no, we need to clean that up later versus the thing is down we can move on that it might feel we get reflected back from our teammates sometimes. And so I started to think like maybe if someone hasn't had a similar experience of debt as other individuals, there's not the same emotional stakes tied with debt as there might be with other terms that we kind of have a shared understanding with as we're working on these kind of projects. And I thought, well, risk does kind of carry a shared understanding in that way. And what is technical debt except something that like may or may not blow up in your

face later, right? Right. We're using it as a proxy term for what actually is, which is risk. Yeah. And there are a lot of things about the metaphor of debt that like work out really well, but there are a couple of places where it falls down. And I thought maybe that risk

picks up a lot of those same ways that works well, but also maybe forms a little better with some of the ways that the debt metaphor doesn't. What are your kind of initials on up there? The first thing that jumps out at me is that risk is a scale, not just a magnitude. So you can say we have a lot of debt, we have a little bit of debt, but not all technical debt is equal. Some of it is significantly riskier than others. And sure, if you're going deep in the financial

metaphor, the finance team always has risk assessments tied to various forms of debt. But we usually don't do that kind of analysis when we're talking about technical debt. Whereas if we're going directly to the risk mental model here, I think that maybe allows us to talk a little bit more clearly about what the cost of this particular sort of shortcut has. Right. Something that I think the debt metaphor sort of misses a little bit is if you are adding some technical debt to a very stable part of the codebase, that is not going to have to change very often. And so if you're adding some debt that increases the cost of change, but you might not

have to change that for a very long time, then you might never have to pay that debt. You're not a crewing interest or anything on this in the same way that you might be with financial debt. So risk, it's very common to speak about this is a low risk that we're introducing here. Whereas debt is maybe less like it obviates the the gradation between something that is harmful or maybe just kind of an irk, right? Yep. And I feel like a lot of devs are not great at navigating the conversation

around technical debt. And we all love a refactor, we love a rewrite. The temptation to just label any part of the codebase that's janky debt, because we didn't get to have an extra week to perfectly polish it up. So at what point is it actual technical debt or technical risk versus perfect being the enemy of the good? Right. Yeah, that's a really good aspect of that that I don't think I'd consider

the accumulation of would be nicest that can get into the conversation of debt. And maybe because of some of that kind of falling into that category, it's easier to brush off all debt as the would be nicest that the dev wants to do, right? Whereas if it can pass that threshold of like, do we as the good calling at risk? Oh, it would be nice refactor that I would love to get to one of these days or like make my soul feel good, but I know it's not gonna deliver real value. I don't think crosses that threshold where I would say like, Hey, there's a risk here, right? Right. There's a risk to my soul.

But yeah, in this case, we're like, we're building production systems that have to like do things in the real world. We're not doing an art piece where we get to like polish the final little piece as much as we want before we release it out for public consumption. Yeah. And I think even like art there in some ways, there's the great quote that art has never finished, it's just abandoned, right? Or that's maybe paraphrasing it, but that's exactly right. Like even you could always be

polishing and you could refactor even further and even further. And so like it shrinks even further and further away from that threshold of risk, right? But could still be labeled debt. And so maybe, yeah, that's why that metaphor doesn't doesn't land emotionally across the other side of the conversation that we might want to have happen sometimes when we're talking about adding that or accumulating debt into our code base. I feel like there's something kind of at that threshold or beyond where we might not like it for a variety of reasons, but not only are we going to punt on it for now, if nothing else changes, it is never going to be worth our time to come back and fix that.

Right. Right. And that maybe hurts our souls a little bit. Yeah. But that's the reality of things.

Yeah. Yeah. It might hurt a little bit, but honestly, with move on to the next project and in the same way that I can look at code that I wrote a month ago and say, who wrote this? I will have moved on. It will hurt for a moment, but it's not something that's going to stay with you. On the other hand,

you don't want to be constantly cutting corners and giving sort of a degraded experience to clients and just be like, well, you know what, clients will have what they want, or they'll sort of get what they paid for. Yeah. I think that's also sort of an interesting phrase there that you say, though, is like constantly cutting corners because in much the same way that you could always polish and you could always perfect like there will always be a corner that's caught at some point. It's just you have to decide advantageously to delivering value for the product and the users, like what corners are important to keep sharp versus cutting them, right? Or even almost like the level of you might even call like resolution. You want to have on this corner? Yeah. Yeah. Absolutely.

How like perfect when you keep zooming in is it have to remain perfectly sharp versus like like beyond this zoom level doesn't matter if it's sharp here, it's good enough. Right. Yeah. This part of the code base doesn't require scanning electron microscope precision, right? This is only ever going to be at a much lower resolution. So this is naked eye and moving on here.

Yeah. Like when I was in the circus and like the third show of the day our makeup would be getting a little bit like messed up, we would stand like 10 yards apart because that was the very closest any audience member was going to be and we'd be like, is this is this readable? Is this good enough? Do I need to go fix this? And like sometimes that's the actual question. It might feel sloppy

because you're so close to it. But when you kind of step back to look at the project and the business and everything at a whole, it's like that is great. It's actually great. And it's not just like, it's workable and we're going to ignore it. It's like, no, that was the correct level of resolution to dial into here. And I think that that is sometimes uncomfortable reflex for us. And because

that's hard for a lot of developers to navigate that correctly, sometimes maybe then gives teammates and other disciplines, maybe a bad sense of what techno that means or a sort of like, oh, does are always crying wolf about this stuff and they're always looking to spend time on refactoring and it doesn't, what do we get from it? And I think that's, yeah, that the risk is maybe a better communicator of how it will impact people outside of development, right? Because I think you just said something like really key right there. Because now you're measuring in terms of impact outside development, as opposed to this code, if ends my aesthetic sensibilities. Right, right.

Yeah, when you're saying like, Oh, there's technical debt here or something, you have to go through another couple layers of that abstraction of that metaphor to get to the point of like, well, this could come back and haunt us later that if we need to change this feature, which it looks like down the road, we might be doing, it's going to take longer because we didn't clean up this last step. Whereas if you're going to the risk, you can say, well, what is the risk? It's like, well, right now the risk is that any change here, which is likely to need to happen soon is going to take longer to do unless we do this first. And sometimes that's just sort of spreading out that work. Sometimes it actually is potentiates and doing a small thing now will make many harder changes later, easier versus trying to do them at the same time. Right. And so when you are specifically

saying like, this is the impact, this is the risk, you're less obfuscated, why it's important to you to change now. I think one of the things that was interesting about the idea of technical debt was this sort of big and if you lean more heavily into the metaphor, accumulating interest. And I think that sort of risk paradigm, you can get from that as well, because you're saying, Hey, we made some changes now, we got the feature out live, it's working, there's some cleanup that probably needs to be done. And that would be that would make the cost of implementing a feature next month more expensive. But if the cost of fixing it now, as we finish feature one versus the cost of doing that as pre work on feature two next month, if the cost there is the same, why wouldn't we want to defer it?

Because we can get more value now by doing something else. So that's almost like a, they get going back to the debt metaphor, a zero interest loan. Right. Right. Now oftentimes say, well, context is fresh in my mind, I have all these ideas, I can do this now. And so the cost

is lower. But being able to differentiate between those two things, just because the cost of change later is expensive. If it's by the same constant value that you would have to change it now, it's really then just becomes a scheduling question. Great. And if in the future, we're not 100% sure we're going to do that story. Well, now there's a chance we might never need that

refactor. And so punting on it can make sense as much as it sort of breaks our heart. Yeah. Yeah. Does then the idea maybe of interest or compounding interest work well to communicate those situations where the longer debt or a certain decision might exist in the codebase, more things will start to depend on that decision. Is that kind of like, incruing, a current interest or maybe like a compounding

interest sort of thing, like the longer we wait, it's actually going to stack higher and higher and higher that this is going to cost more to change. Is that then somewhere where that interest and demand for is like perfectly apt? I like that. But I think that it's maybe still a proxy for the thing that you're actually trying to say, which is fixing this now versus later is or punting on this as a higher risk choice. And here are the reasons why it's a higher risk choice. As opposed to a situation

we say, look, nobody touches this, we're only going to change it if we wanted to feature why. And we're not even sure we want to do that. So now that's a very low risk thing to leave in there. Of course, all of these risks have to be kind of multiplied by their probability of having. And I think that's where talking terms of risk is the thing that more of a business-minded people are constantly having to make decisions based off of risk and sort of how big the risk is, what the impact is going to be and the likelihood of that risk coming to be.

Yeah. And especially when you're trying to communicate the necessity of an amount of work, when you're talking about people whose jobs it is to look into the future and make plans about what might be coming down the road, it's all probabilistic for them. This is most likely that this is going to be necessary. This is less likely. If you're already talking in this kind of language of like a probability and like potential futures, then it's going to actually going to have less levels of translation to need to get to talking about the same units of value with other teams that are already in that language. Yeah. And sort of cost of change is only one potential risk that

you might have. You might have just sort of general carrying cost as your code gets more and more gnarly. It's hard to even do things like find root causes for bugs. As you get more gnarly, it may be also the risk of introducing about gets a little bit higher because it's all these like jinky edge cases that are not sort of coalesced nicely. If you're cutting corners around things like performance. Again, performance is never like binary performance, non-performance is scale of

where is good enough. But if you're saying, you know what, we decided that this is a new feature, not that many people are using it, we're going to ship it with no performance optimizations. And we'll come back to it if that ever becomes a problem. That could be a very valid decision, but that is inherently a risk balancing conversation. And one that should be done not just with the dev team, but in concert with the product team, with anybody that's doing sort of strategy at the company as people are sort of thinking out, where exactly we want to balance that risk versus knowing, hey, we've got 10,000 dealers users of the site, and we expect that this percentage of them are going to use this feature day one, that might require a very different sort of profile.

Right. Right. And the sort of, you know that you're sort of building on shared assumptions at that point too, right? Because you're saying we expect 1% or whatever will be using this day one, unless you're specifically gating that, right? Like you could be off by an order of magnitude on your assumptions, right? And so, it's having those conversations of like, if these things hold

true, then this will be right. Otherwise, this risk will, you know, will be much more important, but like cost of that will come do much sooner. Something that's really interesting was real scale. I was talking with some of the folks from Aura Frames. I think that is a very small, like private social media network thing for your family that is photos only, or for your family, you invite your group. So if you have a small group and you share photos and that's kind of it,

but it is tied to one or more digital frames. And so the photos can sort of then just like rotate through the digital frame. And a common thing that we'll do is people get that for their parents or grandparents, set up in their home and then various members of the family can just upload photos to this group and then they scroll through in their, in somebody's living room. These devices are very frequently bought for Christmas. And so there's a giant spike of traffic. And

Christmas morning, as everybody opens their frames, plugs them in and uploads photos. What's interesting is there's a little bit of a signal about what the likely spike is going to be because you're buying physical frames. Usually in the lead up to a big holiday. So you can a little bit guess based on sales volume, what your volume on Christmas day is going to be. All the lots of assets there. And so they do all sorts of really interesting things with load testing and things

like that. But I can imagine that some of these sort of risk calculations that you and I are sort of hypothesizing here are things where they have to sit down with some pretty big, scary numbers around like, okay, well, what if we've seen growth year over year, what if our spike is even bigger this year than we expected? Can we handle that? And what is the cost of falling over on Christmas day? Yeah, that's probably not a product that gets sold again to pass on to more relatives to join the group then if you fall over on that very important day. Right. So yeah.

So that's a very high risk. But also, you wouldn't want to be like, okay, we're going to ship the naive version and not bother on performance optimizations. But also at some point, you have to cap your optimizations at a certain level. They have a very sort of data driven approach. And so they're sort of balancing that risk. But that is the kind of conversation where you need to sit sit in with

the stakeholders and say, look, let's get these numbers. Here's the traffic we expect. Here's the code change we're trying to make. Here's the business cost if we go down. Here's the cost of spending more time on optimizations or buying more hardware. And then how do we balance these so that

we're comfortable with the amount of risks that we're taking? Yeah. And being in those conversations, I think is really great to be able to speak in the terms that are sort of shared terms across the entire business. And so maybe something that developers, some of us out there might have less comfortability, speaking in business terms or things like that. But this is just a little shift from debt to risk that can help spread that gap a little bit and help everybody in that room where it happens to all be speaking the same language. It's a much more holistic term. And it has to

sort of take into, or maybe the risk is not taking into consideration all the other numbers, but it sort of has to be shown next to the other numbers. Because an event that's very risky has a large downside, but is highly unlikely is not something that you need to worry about quite as much. So you can't just say this is high risk. You have to say, okay, but what is the likelihood of that risk? So now you have to pull in a few different numbers. And also keeping it like we've been saying,

keeping it connected to the impact, right? Like it could be very high risk. But if the impact is really low, then that's not going to be as drastic a concern, right? I think this is something that I really like navigating as a consultant. Because as a consultant, oftentimes my role is to do this with a hard work of the analysis of what is the cost of doing these different features or doing this cleanup or picking a technology or even like a few different architecture paths. And then

bringing that back to the team and say, Hey, here are the cost benefits on each, but specifically, what are our risks associated? As a consultant, you don't always get the opportunity to decide, but you're there to provide expert advice so that the actual decision maker can make an informed choice. And that's where I bring a lot of value to a client. Sometimes I might be technical leadership, sometimes it's even like leadership outside of the engineering organization. And say, look, we're looking to make some big decisions, maybe make a big bet with a lot of money. And I don't want

to be going to this blind or just going off of vibes. And so I can say, look, here are five choices we can make with the impact to the risk, the implications, sort of give that holistic picture so that then my client can make an informed decision, one where they're taking on risk in a level where they're comfortable in a way that sort of fits their broader strategy for their business. Right. And to help them understand that risk, so that they know if it is something where they're comfortable taking on. Yeah, exactly.

And I feel like that conversation is way more valuable than just showing up and saying, hey, there's a lot of technical debt. I think this is maybe my jaded anti-capitalist state coming in a little bit, but there might be, as you get into conversations with teams that are very different from just your lived experience of yours, there might be people that are like, yeah, debt, whatever, who cares? If they haven't had crushing amounts of student debt or something like that, that can be a dangerous word for some people, but for other people who have maybe not had to struggle with it in the same way. Like, okay, yeah, we put that on the books. And then we took on that debt to do a thing and we'll pay off that later. And it's like, it's never been a problem.

Right. So like, that might not be as pointed a word as you need it to be sometimes when there's something you're trying to get across. Like, no, this is a risk. This is something that we need to pay attention to labeling a debt might not get the same impact that you're looking for. Because people are thinking about it through sort of their personal lived experience rather than through the lens of like, this is a tool that a business has. And here's how we service this debt. And here's

how if you don't have a finance background or haven't worked a lot with the finance department, maybe you're not quite as familiar with the impacts and the trade-offs and the risk management that goes around with debt. Right. Right. Yeah. It's again, it's there are more layers out there where this might not be a shared term. Even if you have this one aspect of it that might be shared,

there are other aspects coming in that like, we might like to think that we are all completely just like product and metric and all logical people making those decisions, but like, communication and decision making and everything like emotional states always come into that, right? And if you're trying to communicate like, no, I am afraid of moving forward with the code base as it is using a word that doesn't strike terror in someone else's heart. Like, you think it might it's not going to get your point across in the same way. Right. And that's always a challenge with metaphors, right? They rely on sort of a shared emotional

connection to that metaphor. Yeah. And that's I feel like risk is much less metaphor, or at least it feels less abstracted away from the thing than I said this term multiple times. I think what we're dealing with is a proxy measurement. We're using a metaphor to try to emotionally communicate risk.

We're actually trying to communicate risk, but we use the debt metaphor as a sort of proxy way of doing that. And as you mentioned, different individuals, especially if they're thinking in their personal capacity, rather than their background and sort of business finance, might have like very different emotional reactions to debt. One person is like, Oh, debt is the worst thing ever. And you like, you want to have zero at all times. And like, you never take that on. If you have even the slightest

little bit, it's, you know, a big alarm. And other person's like, no, that's the thing that you cycle through it, you service it, you all the stuff. And it's the thing to be managed. And so one person's trying to like raise the alarm by using the word debt. And then another person is like, yeah, it's a risk management thing. Right. This is a tool we use that is, you know, a constant

feature of our work. So yeah, maybe the by using the sort of the real thing that we're trying to talk about, which is risk. Maybe that helps to clear up some of those conversations. Yeah. And it's always clarity, understanding, and shared assumptions and mitigating those differences of assumptions and communication, because that's where all the problems are concerned from.

So I'm curious that in your own work, is that a maybe a shift in vocabulary that you've started doing for yourself talking more in terms of risk than debt? I think so. Yeah. I think focusing more on like, what is the impact of the outcome that I'm afraid of, or just concerned with or thinking about, right, like whatever level of alarm that I'm trying to raise, right, is has been much more effective in communicating why I am bringing up something for consideration. Whereas like saying this is this big piece of debt that we're taking on, it's yeah, it's it's fuzzy. It's not precise in communicating. And so switching my terminology and kind of the

framing that I'm using has been pretty effective in making sure that we're on the same page and like getting my idea across and also helping the team, you know, communicate about something more effectively. So it's something that I found useful and I want to get out there and see if other people can take advantage of that and have more productive discussions because of changing that framing device a little bit. Yeah, let's get our industry talking more about risk and sort of impact this visible across the team. 100%. On that note, shall we wrap up? Let's do it.

Frequently asked questions

How long is this episode of The Bike Shed?

This episode is 36 minutes long.

When was this The Bike Shed episode published?

This episode was published on October 13, 2026.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this The Bike Shed episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.