1. Prologue
2. Parts
3. Frame Prep
1. Bottom bracket
2. Head tube and headset
3. Cutting the steerer tube
4. Wheels
5. Rotors, calipers, and handlebars

Phew. Slow to pick this series back up this year for obvious and non-obvious reasons.

When I started this build, the wheels were the most intimidating part of the work. That’s not to say that the rest is trivial, but with the biggest unknown behind me, it’s slightly downhill from here.

In my next working session I managed to install the brake calipers, brake rotors, and the handlebar.

Brakes

At this point the rim vs disc brake situation is a choice already made by your frame for you. I’m not aware of any bike frames that have mounts for both disc and rim brakes, so if you have your frame, you already know what kind of brakes you need.

For years now disc brakes have been considered the obvious choice for better braking and the default for an average new bike. New bikes with rim brakes absolutely do ship, and rim brakes have a significant loyal fan base, but it’s kind of a vinyl vs streaming type deal. The argument is that disc brakes, especially hydraulic disc brakes, offer much stronger braking power and better braking modulation than rim brakes, especially in wet conditions (not uncommon in Seattle).

I am here to tell you that in my personal experience, that is far from a given.

The first disc brakes I ever used were the SRAM Force 11 hydros, and they are absolute garbage. They are the blemish on an otherwise amazing bike. I had nothing but trouble with those brakes from day one; loud squealing, weak and mushy braking action, more squealing. Stay away. The Grand Cru Long Reach rim brakeset I have on my Surly Steamroller, coupled with the Paul Component canti brake levers make me feel so much more confident in stopping than the SRAM ones.

I’m not claiming the general rule that disc brakes > rim brakes is complete fallacy. It’s not. I’m just saying that mediocre disc brakes can be shit, and really good rim brakes are way better than mediocre disc brakes.

For this build, not repeating the SRAM Force experience was critical to me. Two things allowed me to avoid this: the Crust Evasion frame makes it easy to use much bigger rotors than the ones on the bike with the SRAM brakes, and I was given the recommendation to consider the very expensive, but very excellent Growtac Equal mechanical brakes.

But first, the rotors.

With rotors you generally have the choice between 6-bolt or center lock mounting. Except this is yet another choice that you will have already locked in with your hubs. In my case, I chose hubs with bolt mounts for the rotor. I chose 180mm rotors for the front and 160mm for the back.

Next, the calipers.

There is, uh, a big complicated mess when it comes to mounting standards for disc calipers. Flat mount, post mount, IS (International Standard). Which one’s legacy, which one’s good, which one has more than one name, which one needs an adapter, what kind of adapter is it and what size rotor does it let you use, it’s all bad and unfortunate. I spent a lot of time checking and re-checking the mounts on the frame and the adapters I needed, and thankfully I got it right.

That’s it for mounting braking hardware! The cables and levers will come on a different day.

I managed one more easy thing in this session.

Handlebars

There are many handlebars out there. Generally you are choosing between drop bars and flat bars, and I went with the Salsa Cowchipper, which are drop bars that have a significant flare and not-so-significant drop.

Checkpoint

Finally, my bike starts to look like a bike.

Next, drivetrain.

This was a year of substantial subterranean change with minor surface-level manifestations.¹

Trends

I listened to a lot more music in 2025 than I did any year since 2015 (with one exception: 2021).

I listened to fewer podcasts.

The Storygraph tells me that I read the same number of books in 2025 as I did in 2024 (I abandoned a couple in 2024 though, and none in 2025) and the number of pages is about 200 fewer in 2025.

All of this was pretty much as intended, so I call it all a success.

I like my job

It feels so weird to say this. In the world that we live in, in the year 2025 anno Domini, I feel the kind of self-consciousness I’d feel if I was announcing to the world that I won a billion dollars in the lottery. It feels like a huge brag with a not-insignificant amount of guilt for having something so good when so many other people don’t. All of us should be doing jobs we like. All of us should be working for employers who don’t treat us, or make us feel, like disposable wipes. All of us should be working in enterprises whose incentives are not so brutally misaligned with what’s good for human beings. The current situation is so deeply inhumane.

That said — I do like my job, a lot. In 2023 I joined the Allen Institute for Brain Science, and I will likely stay at the institute for as long as there is a job to stay for. The Allen Institute sometimes seems too good to be true: a well-funded non-profit whose mandate is to do complicated, expensive bioscience research and then share it publicly for free. No profit margins, no ARPU, no ad tech. Just people showing up every day to help the world understand cell science, immunology, and the brain, and cure neurodegenerative diseases without charging anyone for it. I mean damn, I did win the lottery.

Seattle

I still live here, and I still feel like I don’t belong.

Sunny Side Up

I guess I should come clean and admit that as I write this, Sunny Side Up is fully built and operational. The blog posts are lagging behind because they are a lot of work to make. The bike itself rode like a dream for all of 40 km before I tried something stupid and broke my foot. In my defense, I lacked sufficient information to determine that what I was about to attempt was in fact stupid, but in any case 100% my fault 0% Sunny’s fault. It’s already been two months and I need to wait one more before it’s not risky to ride my bike outdoors again.

Bicycles remain the most effective grass-touching tonic for the dark days and weeks of the soul I sometimes go through.

I officiated a wedding

If you think you are immune to the emotional impact of the words “by the power vested in me, I now pronounce you…”, think again.

In September I traveled to Kansas City, MO to officiate the wedding of two close friends. They were smart enough to ask me back in March, which was far enough from September that my anxiety was not greater than my desire to do something meaningful for my friends and I said I would do it. What did I know about officiating a western wedding? Basically nothing. I got ordained sometime in July and started preparing in earnest a month later. Like some things in life, this is the kind of experience that is as rewarding as the amount of forethought and preparation you put into it. In the end I was very happy with what I said to the couple and how the whole ceremony went. Most importantly, I think my friends were very happy too.

Year of Linux on the desktop

If that joke got any more tired it would quite positively be dead.

But. 2025 was the year of Linux on the desktop for me. I haven’t owned a personal laptop in years. Earlier this year I was visiting a professor at my grad school alma mater and saw a strange, tiny laptop connected to a display and peripherals on his desk and decided to get one to try it out for myself. Frankly its origin makes it somewhat suspect and I use it with a certain degree of paranoia, but I did what I could and then installed Arch Linux + Hyprland. Whereas my many prior experiments with Linux over the years were breathtakingly brief, this one stuck. I’ve been using it to do Real Things for a few months now.

Linux always set the gold standard of the “batteries not included” experience in computing, and while more things work out of the box now than they used to, the spirit definitely lives on. Basic things like audio output and brightness control, screen locking, timer-based sleep and suspension, even setting a desktop wallpaper, all need libraries to be installed and configuration to be set.

A couple of weeks ago a story was making the rounds about a person whose Apple account was deactivated after they tried to redeem an Apple gift card that was tampered with (?).

The question mark in the parentheses is the reason why this story is a scary story instead of a nothing story. The scary thing about something like this happening with any of the providers of walled digital gardens is there’s a very good chance that you will be unable to speak to anyone who can do anything about the fact that you can no longer access your passwords and/or cloud storage and/or email and/or messages. That’s if you can speak to anyone at all. The only reason this got resolved five days later is the coverage the story got. If it were you or me, who knows. There is no shortage of stories on the web of people getting locked out of critical accounts from major tech companies and finding nothing but chat bots and dead ends to get help from.

So, in a I-never-thought-I-would-consider-this moment, I am rethinking my enmeshment in the remotely lockable fiefdoms of companies like Apple. Really it’s only Apple at this point. It’s really hard to give up the platform benefits: iCloud sync, passwords everywhere, Apple Pay everywhere, Messages everywhere, especially when I still have some trust that the company’s incentives are aligned with mine (mostly… for now). The hoops I jump through to get a file from my photo library to this Linux machine are legion, especially if I’m trying to be careful about what I install and enable. But it might be a situation where my initial investments accrue compounding interest over time. And more importantly — it’s fun!

Next year

Writing non-technical posts is working well for me, and I hope I do more of that next year.

This might change, but one way I will sense how well the year is going by how little screen time I have in my average day or average week. This will of course be difficult, and some activities like writing shouldn’t count towards that timer.

There was a lot of travel-for-others in 2025. My plan is to almost exclusively travel for me in 2026. My hope is all travel involves some kind of bike touring, domestic or international.

Life is short

Months ago I read a quote by Albert Camus in Emergent Strategy and it’s been the best kind of recurring intrusive thought ever since:

The only way to deal with an unfree world is to become so absolutely free that your very existence is an act of rebellion.

So really my main wish for you and for me is to stop waiting and let go of some shit in 2026. I wish for us to be free.

1. Prologue
2. Parts
3. Frame Prep
1. Bottom bracket
2. Head tube and headset
3. Cutting the steerer tube
4. Wheels
5. Rotors, calipers, and handlebars

Ahhh. It’s been a shake or two of a lamb’s tail. Life. You know? Sometimes you love something and then it breaks your fifth metatarsal. ANYWAY. Wheels!

Okay so you need wheels. You can buy wheels. Many do. I’ve done it. There’s no shame in that. But sometimes you need to put the packet of Nescafé back down and go make yourself a good cup of pour over coffee.

So here I will talk about building your own wheel set. I had never built a wheel set before, but I wanted to try, and it was a lot of fun! Building wheels can get kind of addictive to the right personality type. And, I’m writing a blog post about building a bike and you’re here reading it. Sooo…

Choices

Ack! Sorry, there’s a bunch of choices you will need to make, and they are all kind of interdependent. Like many other things when it comes to building a bike, you can go the easy way and make standard (and good!) choices, and everything will be pretty easy. Or you can start to get fancy, and your options get restricted accordingly.

Here are the parts involved in building a wheel set:

• Rims.
• Front hub and rear hub (they are not interchangeable with rare exceptions).
• Spokes.
• Spoke nipples.
• Rim tape.
• Inner tube (maybe?).

Here are the kinds of choices you can make about all these components. This may seem overwhelming. Just remember that standard choices are always available and you will likely have an easy time of it if that is what you want.

• Rims: size: 700c, 650b, 26”. carbon or aluminum, symmetric or asymmetric, disc brake or rim brake, number of holes, tubeless ready or not.
• Spokes: number, butted/blade/other type. You will need to determine the length of the spokes, which is determined by the lace pattern you want to go for, your specific rim measurements, and your specific hub measurements.
• Hubs: number of holes, thru axle or quick release (this is decided for you by your frame really).
• Spoke nipples: brass vs a worse material, color, whether you will get really fancy and add washers that allow your spoke nipples to adjust position so as not to bend your spokes.
• Rim tape: you need a specific kind of rim tape if you’re going tubeless, or have more choice if you’ll use tubes.
• Inner tube: see rim tape.

As you can probably tell, all those decisions are entangled. If you decide on a 32-hole rim, you have to go with a 32-hole hub. Rim or disc brake configuration is decided by your frame, and so is the type of axle that goes through your hubs. If you want to go tubeless, you will need rims and tires that are tubeless ready. If you want wide tires, you need to make sure both the frame and the rims support that width.

Take your time researching your options.

How to build a wheel

Look, let’s cut to the chase. Watch these Park Tool videos, and you will learn almost everything you need to know:

• How To Build A Bicycle Wheel - YouTube - once I collected all my components, I basically followed this video step by step and ended up with a built wheel. Remember the part about making standard choices? I decided to go with a 32-hole wheel with 3-cross lace pattern, and those are the same specs in the Park Tool video.

• Which Spoke Pattern Should I Choose? | Between Two Wheels Episode #2 - YouTube - “lacing patterns” are how you entangle the spokes to make your wheel round and strong. There are standard patterns: 3-cross, 2-cross, radial (basically no cross, looks good, makes the weakest wheels, do NOT do this with disc brakes).

See these two incredibly detailed write-ups from the Park Tool website:

• Determining Spoke Length for Wheel Building | Park Tool

• How to Build a Wheel | Park Tool

Spokes

You need your spokes to be cut to a specific length, and that length depends on the dimensions of your rims and your hubs. You can use this great spoke length calculator with an impressive pre-built list of rims and hubs and their dimensions. Once you have your spoke lengths, you can either order them to that length, or stick with the DIY spirit and cut them yourself.

Spoke cutting and threading machines are not easy to come by, but if you have access to one, it’s likely to be some kind of niche or custom-made affair. This is the one I used:

Lacing the wheel

I’ll keep this very brief: watch the first Park Tool video I shared above and your journey might look something like this.

Truing

Truing a wheel is using a spoke wrench to tighten or loosen the spoke nipples, which in turn tensions or detensions the spokes themselves in order to make sure the wheel is as round as possible. You are shaping the rim by pulling it towards and around the hub. The process involves using something, ranging from a proper truing stand to a zip tie on your frame to detect where your rim is wonky (the technical term is “out of true”) and adjust the spoke tension accordingly.

Dishing

Dishing the wheel also involves tensioning and detensioning of spokes, but the goal is to make sure the rim is centered in the hub, specifically centered between the two ends of the hub that contact the frame. For this you use a specific wheel dishing gauge that helps you measure the offset and confirm improvements after adjustment.

Whereas truing involved tensioning and detensioning neighboring spokes on the circumference of the wheel, dishing involves tightening the spokes pulling to one side of the hub, and potentially loosening the spokes pulling the other way. I say “possibly” because at this stage it’s likely that your spokes are overall not tight enough, so pulling the rim to one side by tensioning the spokes pulling to that side without loosening the other side might be the appropriate thing to do to increase the overall tension. How do you know the spokes are the right amount of tight?

Tensioning

This comes last after truing and dishing, and it involves using a tension gauge to measure the tension of every spoke to make sure that no single spoke is an outlier in terms of tension, and that all spoke tension measurements are somewhat close. This can be… a little tedious if you’re a beginner, but you start to get a feel for it after a while.

The tires

Okay the skeleton of your wheel looks good! Now it’s time to add the cushion air pillow bits. Some call them “tires”.

I like to wait until a trend is nice and tired and almost out of fashion before I finally decide to try it. People have been raving about tubeless setups for so long, and we’re just at the point where they’re moving on to the new hotness (TPU tubes apparently?), which means my moment has come. I’m finally trying tubeless.

Tubeless setups have a reputation of potentially getting really messy and really frustrating. Instead of an inflatable tube inside your tire, you seat the tire directly in the rim. The tire bead should seat well enough that it’s essentially air tight (or almost so, a bit of air leaks over time). You then pour sealant fluid inside the tire which helps seal small punctures you might accrue on your adventures.

This was a pretty easy setup for me. The tires slipped onto the rim just fine, and I didn’t even have to use an air compressor to inflate the tire fast enough to seat it, a floor pump worked just fine. After the tire was seated, I pulled the valve core out (releasing the air but not unseating the tire), poured the sealant in, and inflated it back up again.

Nice, we built a wheel! Honestly these 27.5” × 2.6” beasts look awesome. Next, I’ll install the brake rotors and calipers, put on the cassette, and slap on some neat handlebars.

1. Prologue
2. Parts
3. Frame Prep
1. Bottom bracket
2. Head tube and headset
3. Cutting the steerer tube
4. Wheels
5. Rotors, calipers, and handlebars

Tube, Steerer

The steerer tube is the tube that connects your handlebars to the two bottom blades that hold your wheel.

New forks will come with a steerer tube that is longer than almost anyone will need. Part of prepping your frame for your body and your components is to cut the steerer tube down to a proper size.

You will only need to do this for a new fork. Although I suppose you may want to do this if it’s been cut for someone else who needed more steerer tube up top than you.

I would’ve said that cutting your steerer tube is the ultimate one-way door in your frame prep, but guess what, there exist steerer tube extenders that will let you add on some height if you cut too much. That said… don’t put yourself in the position where you need to use one. There’s a lot of debate on their durability, and this is a part whose failure can be quite catastrophic for your corporeal being. If you will buy one, buy a good one.

Also note that a common recommendation is to not remove every millimeter of steerer tube you think you don’t need. If you think you might sell your bike one day, then the less steerer tube left on your fork, the fewer the number of bodies your frame will fit. Also, your body can change. If you want to keep your bike forever, your body might¹ get less flexible with age or lifestyle changes, and you might want to raise your handlebars to a more comfortable position. Or, you could injure your back and might need a more relaxed posture on the bike for a few weeks or months.

On the opposing side of these very reasonable recommendations is an aesthetic argument (and isn’t it always). Having a bunch of leftover steerer tube over your stem is considered not a nice clean look by many. Unfortunately I am one of those many, so I tend to set my handlebars at a reasonably comfortable and conservative height, then lop off the remaining tube. No steerer over the stem.

The Measuring

This… is tricky. How do you know how much steerer tube you need? I think if you are very informed about your measurements on a bike, and the geometry of the frame and fork you have, you might be able to estimate this. Doubtful though, because each frame has slightly different geometry. In my experience, if you take your new frame to a bike shop and ask them to cut your steerer tube for you, they maybe eyeball your size and the size of the frame, throw a bunch of spacers on there, measure that and call it a day.

Since I did neither, what I did do is err on the side of caution and leave a very generous amount of steerer tube on the bike. As I mentioned, once you start riding your bike, you can always move your stem down one spacer at a time until you find your right fit.

The steps are simple but require precision:

step	photo
Your headset has to already be installed. Once that’s done, install your fork in the frame - this is temporary.
Install all the spacers you intend to use. This is the part where you factor in a margin of comfort or lack thereof. The spacers are what is doing the actual measuring for you. Also install the exact stem you are planning on using (this is important because stems vary in their stack height, don’t use a dummy/stand-in unless you know it’s the same height as your final stem).
Use a tiny pick or knife to make a mark on the steerer tube at the exact edge where your stem/spacer ends.
Remove the stem and spacers, remove your fork from the frame, and make a second mark about 2mm below the original mark. This is important because eventually the goal is to preload the bearings and slightly compress this entire stack when you install your stem cap and bolt and tighten them. The 2mm allows room for that compression, otherwise you perfectly match the length of steerer tube to the height of the un-compressed stack, and as you tighten you will run into the steerer tube and your headset and bearings will not perform well. Make sure the 2mm is not 5mm or 6mm because you also want to ensure that when you tighten the bolts on your stem, you are tightening around actual steerer tube material, not empty space. If you tighten around nothing, you could crack your stem and, once again, sadness ensues.

The Cutting

You will need some tools. Traditionally, you will need a vice, a saw guide, and a hacksaw with a steel blade. I think you can also use a steel pipe cutter, but I went with the traditional approach.

Take your time. This is the one-and-three-quarters way door. You’re about to cut some hecking steel here.

Make sure the saw guide is good and tight in the vice. Take your time aligning the second mark you made on the steerer tube with the guide, and tighten the guide on that tube very well. Place your saw down the guide a few times to make sure you land on the mark you made.

Once you’re set, start sawing. Consider good sawing technique, don’t try to push down on the saw, your goal is to push forward, then pull back. You’ll slowly make your way down the tube.

If you really go for it without many pauses, the metal could get pretty hot by the time you’re done! So once the excess piece falls off, handle with care.

The final cutting step is to use a file or a tube deburring tool to clean the outer and inner edges of the tube. You want it to feel smooth enough to the touch that it wouldn’t cut anyone or scratch other metal easily.

The Star Nut

Hang in there! We’re almost done.

The star nut is, you will be absolutely astonished to hear, a nut shaped like a star. A weird star. The star nut is hammered down the steerer tube and holds its place by the friction of its star edges wedged against the inside of the steerer tube. The nut is what you screw your stem bolt into, and that bolt compresses the entire stack (stack = stem cap + stem + spacers + head set + headtube + fork crown) as you tighten it.

A star nut is usually included with every headset, but you can’t install it until after your steerer tube is the correct length for the build. You will be further astonished to learn that yes there is a dedicated threadless nut setter tool into which you screw the star nut and then hammer into the steerer tube. Now that we’ve cut our steerer tube down to size, time to hammer our star in.

The Final Installation

Are you still there? Have you been hydrating? Because this is the moment! Time to put it all together.

Time to put your fork through your frame for the last time. For now. Coat the steerer tube with a layer of grease (it never hurts, almost always helps). After installing the fork, install your spacers and your stem. This time, the star nut is in there, so you can also install your stem cap and bolt.

Remember the 2mm adjustment we made? This is where it matters. You want there to be a few millimeters between the edge of your steerer tube and the top of your stem (or spacer), but you want that edge to be above the bolt that you tighten on your stem.

The Second Optional Cutting

A final note.

Remember when I said that I erred on the side of caution and left what I thought was a preposterous amount of steerer tube on the fork? Turns out I kinda like it that way, and I don’t think I’m going to move the stem down any time soon. I like the more upright posture this puts me in, and I think it matches the non-performance focused party pace riding I have in mind for this bike. It does move a bit of weight and pressure from my hands to my sitz bones but I think my butt and my bones will adapt.

Frame Prep is Complete

Next stop: wheels. You can build them yourself!

1. Prologue
2. Parts
3. Frame Prep
1. Bottom bracket
2. Head tube and headset
3. Cutting the steerer tube
4. Wheels
5. Rotors, calipers, and handlebars

The background

The head tube is the part of your frame where you install your headset. Just like a bottom bracket holds or contains your cranksets allowing them to rotate smoothly using bearings and a good fit, a head tube holds your steerer tube (the tube connecting your handlebars to your fork) in place while allowing you to turn the handlebars smoothly and without play.

And just like with bottom brackets, there are many headset standards. At a most basic level you need to know if your frame takes a threaded or threadless headset; the modern standard is mostly threadless, and that’s what my frame takes. Besides that, you need to know:

• The diameter of the steerer tube and whether it tapers. If it does, the diameter of the top headset cup will be different from the diameter of the bottom headset cup.
• The inner diameter of the head tube at the top. This determines the outer diameter of your top headset cup.
• The inner diameter of the head tube at the bottom. This determines the outer diameter of your bottom headset cup.

In my case, yet another reason to love the Crust Evasion frame is its use of a steel fork with a standard steerer tube diameter (1 1/8”) and therefore a very standard headset size: EC34/EC34 in SHIS terminology. I don’t want to mint bike laws that have too many exceptions to be directionally useful but: I think in general you have a higher chance needing components of common standard dimensions if you go with a steel road or gravel frame, and things start to get more rare and proprietary as you move to carbon bikes and mountain bike geometry.

Okay now we know about the head tube. What are we prepping again? You already know about facing. Reaming is the equivalent of chasing from the bottom bracket prep step. Since this is a head tube for a threadless headset, we don’t chase threads, we ream the inside of the tube to make sure it is round enough and has the correct diameter to match the headset meant for it.

Unlike with chasing the threads on a bottom bracket shell, there’s a good chance you don’t need to ream the head tube on a new frame, but you won’t know for sure until you try to install the headset and run into trouble. And unlike the low risk of gently threading and unthreading the bottom bracket, trying to press fit the headset into a head tube that is somewhat un-round could damage your headset cups. Sooo… I recommend you ream the head tube anyway.

The prep

Of course there is a tool from Park Tool company made for purpose, the HTR-1B: Head tube reaming and facing set.¹

One thing I didn’t mention in the previous step (but you would know if you watched the Park Tool video on chasing bottom bracket threads) is that you have to use a good cutting oil whenever you are chasing, reaming, or facing. I regret that none of my photos show this, but cutting oil was used in all the steps that should’ve included it.

On the reaming end nothing much happened. I suppose it means the head tube’s internal shape and diameter were more or less spot on. I shaved a bit of paint and metal off the edges, but not as much as I’ve seen in some videos. So that was kind of easy and uneventful. Fret not though, events are on the horizon…

Installing the headset

Boy. We arrive at the first “this did not go as smoothly as I hoped” step of this build. The challenges I faced here were not fundamental, they were aesthetic. Warning to the reader: if you are as obsessive about centering logos as I turned out to be, and you are choosing between a few headsets, maybe pick a headset without any logos.

Before we continue, and as is custom, I introduce the Park Tool HHP-2: Bearing cup press. I’ll say here that there are people who will say that you don’t need a fancy schmancy dedicated bearing cup press, that you can do this with a long bolt and a couple of washers and nuts, that you can even just use a piece of wood and a mallet or something. I suppose the fact that people get this done with those tools is proof it’s possible. Might you scratch or dent or bend your headset cup in the process? Maybe! You do you boo. Me, I’m using the fancy tool.

How this works is that the headset cups are pressed into the head tube one cup at a time. First, remember to grease the inside of the head tube. You then place the cup on the matching head tube end, then pass the long pole of the bearing cup press through the head tube and connect the bottom locking mechanism to it, then slowly, mindfully, rotate the threaded handle to press the cup into the head tube.

So far so good, right? RIGHT? Yes, right, because the top cup has no logo. The bottom cup, however:

This post is getting long and we need to keep moving, so in summary: I don’t know if I was doing something wrong, but what drove me crazy about installing that bottom cup is that as I rotate the handle and press the cup into the head tube, the cup rotates ever so slightly. This means that no matter how good I was at centering that “W” before pressing, the “W” moved. After the first try, I tried to move on but my eyes just could not look away from a “W” that was a few millimeters off center. So I used the RT-1 headset cup remover to push the cup out and reinstall it three times before I got the right offset on the “W” pre-pressing so that it turned into perfect center by the time pressing was complete. Just… know that this is a thing that might happen.

Finally, we install the remaining components of the headset.

The fork crown race may be of the split or non-split variety. If it does not have a split, it behooves you to use a proper tool to hammer it down evenly and without damage. However, in this case it did have a split, so it fell into place with pretty low finger pressure.

We are finally done prepping the head tube and installing the headset. I admit the frustrations I had with aligning that stupid logo put me in a bit of a bad mood by the time I was done. But also a lot of good lessons were learned that day.

You’ll notice that not all headset components are installed. The top cap cannot be installed until we cut our steerer tube. If you are not going to do that right away, be aware that your fork can slide right out of that head tube unless you secure it with something, so use a strap or a placeholder stem to keep that fork from falling off.

Next stop: we shall cut through steel by hand.

1. Prologue
2. Parts
3. Frame Prep
1. Bottom bracket
2. Head tube and headset
3. Cutting the steerer tube
4. Wheels
5. Rotors, calipers, and handlebars

1. Chasing and facing the bottom bracket shell.
2. Reaming and facing the headtube.
3. Cutting your steerer tube to size.

Not all of these are required, depending on the condition of your frame and the specific part being prepped. We’ll address each step in a dedicated post, so here I cover step 1: the bottom bracket.

Beware: There are things that you can try and mess up and try again when it comes to building your bike, the frame prep steps outlined here are not in that category. You can permanently damage your frame or fork doing these things. Do your research well, and ask for help from someone who knows better (like I did with step 1) to avoid sadness.¹

Chasing and facing the bottom bracket shell

“Chasing” is short for chasing the threads. This means using a tool to retrace and clean up the threads that are already cut into the shell.

There are many bottom bracket standards and not all of them use threads. If your frame has a press fit bottom bracket, you got nothing to chase.

You might need to do this if your frame is:

1. Brand new, or
2. Very used

First try to gently thread your bottom bracket into the shell. Please note the emphasis. If the threads of your bottom bracket are not meeting your bottom bracket shell with light/medium-light finger pressure, or if it threads in for a bit but then seems to get stuck, you probably need to chase the threads.

Why might this be needed on a new frame? I think it’s because the ED coating applied to the frame in the finishing stages goes in and over the threads, gunking them up just enough for them to not mate properly with the bottom bracket threads.²

“Facing” is shaving the edge, or face, of the bottom bracket shell to make it as straight and even as possible. This shaving usually removes some paint and a tiny amount of metal in order to achieve this evenness. You want that edge to be as straight and clean as possible for your bottom bracket cup to sit evenly against it, and I suppose if that doesn’t happen your bearings and maybe your spindle might wear out prematurely.

There is an excellent Park Tool video on chasing bottom bracket threads. You should just watch it.

This went fairly smoothly. I had the sense to ask a bike mechanic friend to provide some adult supervision because I really didn’t want to start this journey by screwing up my frame and having to worry about the repairs.

Of course, when you’re done, it makes sense to install your bottom bracket in the same session. Grease your threads well, and use a torque wrench to tighten properly

Don’t skip the grease, and don’t skip the torque wrench. This admonishment will be repeated throughout this series. Grease any threads that do not have something else like threadlocker applied to them. For example, if you have disc brakes and your hubs use screws instead of a lockring to secure your rotor, the screws come with threadlocker pre-applied. Do NOT grease them.

In part 3b we will cover headtube reaming and facing, and headset installation.

Resources

• Sheldon Brown page on bottom bracket standards.
• Park Tool page on bottom bracket standards.
• Park Tool Tech Tuesday video on chasing bottom bracket threads.
• Park Tool article: Bottom Bracket Tapping, Threading, Chasing, and Facing.
• Park Tool BTS-1 tool.
• Internal Rust Proofing for steel frames - Education / Finishing - Custom Framebuilding Forum

To build your bike, you need the frame, and you need components. There are many components to consider. Here’s a simple list in general order of front to back:

• Handlebars.
• Brake levers.
• Shifters: they might be integrated with the brake levers.
• Cables: brake, shift.
• Cable housing: brake, shift.
• Bar tape, or grips.
• Stem.
• Headset: this will also likely give you the stem cap.
• Spacers (if your headset is threadless).
• Front wheel, and all that makes that up: hub, spokes, rim, rim tape, tube, tire.
• Front brakes (if you’re not doing brake-less fixed gear).
• Bottom bracket.
• Crankset.
• Chainring(s).
• Front derailleur (if not a 1x setup).
• Pedals.
• Chain.
• Seatpost.
• Saddle.
• Rear wheel.
• Rear brakes (if you’re not doing brake-less fixed gear).
• Cassette (or fixed cog/single speed freewheel).
• Rear derailleur (if not a single speed setup).

There are more optional components that are not required for the bike to be ride-able, but you likely want for function or safety:

• Lights, front and back.
• Fenders.
• Racks.
• Bottle cage(s), bottles.
• A good bell (almost want to move this to the list of required parts).
• Bags: handlebar, frame, saddle, panniers, etc.

So how do you start? Do you have to have all your components chosen or in hand before you start, or can you acquire them as you build? The answers depend on a few factors that can make this easier or more difficult.

What kind of build?

This is a different question from “what kind of bike?” For example, you might want to do a parts bin build. If you want to use used or salvaged components then the acquisition can be unpredictable. Of course, in a post-2020 world the availability of new parts for a brand new build or a swanky build can also be unpredictable due to the concussion that 2020 was and the additional bruising caused by the aftereffects on the bike industry, geopolitics, tariffs, and basic common sense.

How standard are the components you have in mind?

Which is partially determined by how standard your frame is. The Evasion I chose is an example of a relatively standard frame: 73mm BSA bottom bracket, 1 1/8” steerer tube and a non-tapered head tube, pretty common hub spacing in the front (12 x 100mm) and rear (12 x 142mm), regular round 27.2mm seat post. All this means that you have a lot of options for components like headsets, handlebars, hubs, bottom brackets, and seat posts that are compatible with those specs.

A frame can be more challenging, and your preferences can add to that challenge. You might have a tapered head tube, or a frame that has a suspension fork or more suspension features. You might need a more-difficult-to-work-with bottom bracket or a proprietary seat post or weird seat post shape. On the preference side of things, you might really have your heart set on Campagnolo parts or certain vintage components that are a world unto themselves.

It helps to understand where your frame and preferred components sit on the standard:niche axis.

How particular are you about this build?

This interacts with the previous two questions. Within the options enabled by your frame and filtered by your preferences, how particular are you about details like color, finish, brand, condition? The more particular you are, the more patient you will have to be.

Inter-component compatibility

This is not a question, it’s a gentle warning. In addition to figuring out what you want out of each component, you will have to research compatibility between/among components, especially if you’re going for a parts bin build or mixing and matching between brands and types.

You want to use a specific crankset? What is the BCD? Is it compatible with the chainring you have in mind? What is the Q factor, is it compatible with the frame, the frame’s chainline, and your body? Is the chainring compatible with the chain that you have to choose to work with your cassette? Is the cassette compatible with your rear hub? Is the derailleur compatible with the cassette? Are your integrated levers compatible with the derailleur and cassette you want? Oh also are your levers compatible with the calipers you have in mind? And are the latter compatible with the frame mounts?

I’m not trying to scare you. If I figured this out, you can definitely figure this out. I’m just saying that you will want to do research as you collect your parts and get a sense for what your options are.

In fact, inter-component compatibility can be turned to your advantage! One of the things that can really help you with indecision is limiting your degrees of freedom. Narrowing down options makes choice easier (most of the time?) As you start connecting the dots between your components and eliminating options that just won’t work, your degrees of freedom will narrow and consequent choices will be easier.

Resources

Good news, the web is still alive as I write this, and bike enthusiasm and maintenance are topics that are not yet overrun with generative model slop. There are a lot of great resources out there.

Top recommendation has to be Park Tool Company YouTube channel. The videos are insanely excellent and almost every topic is either covered in a dedicated video or a Tech Tuesday video. I learned so much from that channel.

If you learn better from reading than watching, Sheldon Brown’s Home Page is the de facto bible of bike maintenance. It still looks like it’s from the 90s, and that’s a good thing.

Finally, I learned a lot from the component maker websites. If you’re working with relatively modern parts, chances are the specifications and compatibilities are documented somewhere, and those numbers are important to know and match between your parts.

Try to have more fun than tedium sourcing your parts, and find comfort in knowing that if you do this step well, this will be the most annoying part of the entire process. Next stop: frame prep.

Achtung! Achtung! This internet site will be about bikes for the next while.

I haven’t felt like writing about much lately.¹ I used to be excited by tech and technical projects, but that hit a wall around late 2022 / early 2023. Politics and civil society are… yeah. Yeah. A handful of my other interests were hit by the same wall that flattened my interest in tech. Much of it is coming back, slowly, but it’s been a struggle to connect with a lot of things that used to make me happy.

You know what never stopped making me happy though? Bikes. The simple bicycle. You go outside, you spin the legs, turn left, turn right, feel the wind on your face, stop for a coffee and a danish, and it feels like things just might be okay. At least they are okay right now.

n+1

An overdone joke in cycling circles is that the ideal number of bikes to have is n+1 where n is the number of bikes you own right now. Ha ha. It’s overdone, but it’s not a bad joke, because like a lot of good humor it’s based in truth.

Long story short: the other two built and operational bikes I have are both comfortable on paved road and light trails, but not on anything rowdier than that. With a maximum tire clearance of 35mm, the majority of bikepacking routes around where I live are simply not an option. Even when I rode the Olympic Discovery Trail the somewhat rocky segment near Lake Crescent (which I think belongs to the Olympic Adventure Trail) left me white-knuckling the steering of my loaded bike for what felt like hours but was probably just 20 minutes or something.² Not fun.

Given that bikepacking and touring are what I want most of my cycling in the foreseeable future to be about, the tire problem needed a solution. I explored the option of modifying one of my current bikes to take wider tires, but when the framemaker I consulted said that 40mm would be optimistic and 38mm likely, I didn’t even bother asking for a quote. The simplest mathematical solution then becomes: n+1.

Sunny Side Up

So I’m building a new bike around the Crust Evasion frame. It’s steel, which introduces the second law of bikes in today’s post after n+1: steel is real. It has the mother of all tire clearances, which I will exploit with a monstrous 27.5”/650B 2.6” wheel set, still leaving .2” of clearance I could’ve used. It has braze-ons everywhere you can imagine, so it will accommodate all the cargo and accouterments a cyclotouriste would want. And it’s set up for disc brakes with rocker dropouts that support geared or single-speed drivetrains. It’s perfect.

I like naming things. My other two bikes are Bowie and Dozer, the former because it’s a raw titanium finish with purple color streaks: strong Aladdin Sane vibes, and the latter because it’s a black Surly Steamroller and the thing does feel like a bulldozer in all the right ways. The Evasion is meant to be something different, something playful, colorful, dare I say something happy. And so it’s Sunny Side Up :)

Finally, I’m building the bike myself. It’s not the first time I’ve assembled a bike, but there are some things that I’ve had a bike shop do for me before that I’m learning how to do this time around.

And since a blog is obviously the first place someone would check when they want to research building a bike in 2025, the posts in this series will document the process in detail likely considered excruciating by the average reader. Have fun! I know I will.

I am haunted by a recurring hypothetical. The hypothetical comes out of how I feel about music, and what music means to me.

Some people don’t care about music at all. Some people use music as a utility: help me focus, hype me up for my workout, set a romantic mood for date night. And some have music TIG welded into their identity, and that’s the group I belong to. Growing up a skeptical queer-in-my-own-way teenager in a religious, conservative, bigoted community, I was particularly angsty. I remember how I felt discovering music whose anger spoke to mine. I remember buying a clean copy of Eminem’s The Marshal Mathers LP on cassette¹, and listening to Meteora and Hybrid Theory on my Walkman late at night. Chester Bennington didn’t know that I, personally, existed, but he did know that I, silhouette and template of a teenage boy struggling with being a tetrahedron in a world obsessed with grinding out every edge and corner, existed.

Music was never just a tool for me. Music is asynchronous emotional communication and energy exchange. Like other modes of art, it touches me not because the tunes are pretty, but because another person felt something, probably strongly, creating it. What did they feel? Do they feel the same thing I feel? Did they think of what I and others would feel when we heard it? Music was how I knew I was not alone.

So, the hypothetical. The hypothetical is: what happens when a year, two years, a month(?) from now, I listen to a song that touches me the way “Numb” by Linkin Park, or “Comalies” by Lacuna Coil, or “Postcard Blues” by Verbena touch me, and then I discover that it was made by a generative model? Not partially, not in an advisory capacity, but totally, completely. There was no person screaming into the microphone, and there was no drummer tearing out the callouses on their hands whaling on the drums.

In that moment, I fear an irreparable internal fracture, perhaps a shattering, where for the very first time I will lose all trust in the emotions that naturally arise in me when I listen to a song that moves me. It won’t even be music anymore, not to me. The reason we feel something when Chris Cornell sings “I am not your rolling wheels, I am your highway. I am not your carpet ride, I am the sky.” is because Cornell existed outside the song with all his motivations and struggle and pain, and the pain of all the other writers credited on Audioslave’s “I Am the Highway”. In my hypothetical there will have been no Chris Cornell, no Chester Bennington, no Cristina Scabbia.² There was no consciousness thrashing.³ No one felt nothing.

There’s a camp out there that believes that my hypothetical cannot happen. That bytes created by generative models will always remain perceptively soulless, and that a relatively low ceiling exists on the creativity of whatever non-sentient entities produce. This post cannot hold the full extent of that debate, but in short I think I agree when it comes to deep knowledge creation and moral reasoning, but I’m not convinced the stochastic mimicry is not or will not be good enough to satisfy the hypothetical.

To be clear, I dread this not just because of the possibility or inevitability of it, but because of the people out there eagerly trying to make it happen. The people excited to figure out how to make it happen. Late-stage capitalism was a prescient name because there are apparently worse stages still to come: Nihilist capitalism? Solipsistic capitalism? What else can I call the kind of capitalism where people don’t just want all the money, but want to commoditize into the dirt the value of everything you could spend it on?

1. Last.fm users
2. who still collect and play CDs
3. and who know their way around a command line

So, yeah, it’s niche, but I use it every day!

I just added a couple of small features worth mentioning for posterity.

Barcode scanning

The first version of scrobble required you to enter the CD barcode yourself. This was fine. I mean, I lived with it. But obviously it’s a pain in the ass to type out that long number, especially on a smartphone screen.¹ I tried to hack my way around that by using the iOS Shortcuts OCR action to process an image and detect the numbers, but I always had to fix the result before submitting. Obviously the tool needed a proper barcode scanner.

This turned out to be very easy to do, I just had to use the barcode.BarcodeDetector class from opencv-python.

So now you can take a photo of the back of your CD case and pass the path to that image. e.g.:

scrobble cd --verbose ~/IMG_1234.jpg

If you’re thinking it seems like a pain to take a photo, save that somewhere, and pass the path to that to a command, consider that iOS Shortcuts is still the primary way at least I use this tool.

Choosing which tracks to scrobble

I didn’t expect this to be something I would need, but I realized it was when I listened to my copy of MF DOOM’s MM..FOOD.²

My copy of the album is a double disc release with one audio disc holding all the tracks and the other an extras DVD-Video. The MusicBrainz release represents the DVD with a single 1:04:19 track called “MM..FOOD Drive Tour”. Well, that’s nice, but I don’t want an hour long track I never listened to being scrobbled.

So scrobble has a new --track-choice flag that lets you choose which tracks from the release to scrobble. Right now the feature requires charmbracelet/gum to be installed, and the experience of using it is very cool.

This feature is also useful when you can’t listen to an entire album in one sitting. I try not to let that happen, but it happens, and when it does it feels “wrong” to have the whole album scrobbled at once without gaps.

Up Next

scrobble is pretty simple and it’s built on the shoulders of people who did real work, like the authors of musicbrainzngs , pyLast, and the other dependencies the tool has.

Here’s another problem to solve. Believe it or not, some CDs don’t have barcodes. Sorry to tear the fabric of your reality like that, but it’s true. What do you do if you own a copy of the US release of Vol. 9 & 10 by The Desert Sessions and you look at the back and find no barcode?

It would be nice to be able to pass a url to a Discogs or a MusicBrainz page and scrobble the tracks from there. Maybe Last.fm album pages too? We’ll see.