Three hundred videos.
When this channel hit 100, I wrote about wanting to build a quieter, more useful corner of the internet. When it hit 200, the post was about how the second hundred was the one that defined what Dargslan is actually about. This post is harder to write.
The third hundred turned into something I didn't plan for. Somewhere between video 250 and video 300, the channel stopped being a YouTube channel about IT and started being something else: a curriculum.
I want to talk about how that happened, because I don't think it was intentional, and I don't think it would have happened without the audience pushing harder than I would have on my own. If you're building anything in tech education — a course, a blog, a YouTube channel, a paid newsletter — there's something useful in here.
The moment I noticed
It was a comment, not a metric.
A viewer wrote something like: "I started with your Python series in March, finished it last week, watched the JavaScript foundations playlist this weekend, and now I'm three videos into the AI for Developers series. I think this is the only resource I've used in a year."
I read that and realized something obvious in retrospect: people weren't watching individual videos. They were following a path. And the path existed not because I'd designed it, but because the videos had accreted in a shape that made one lesson lead to the next.
A curriculum is a path. By video 250, that's what the channel had become — and it took a viewer's offhand comment for me to see it.
Five tracks, all in parallel
By video 300, the channel runs five tracks, side by side:
- Linux Foundations — terminal to sysadmin, the work that pays the bills for hundreds of thousands of people
- DevOps — pipelines, infrastructure, the layer above Linux that turns a Linux operator into a Linux engineer
- Python — zero to senior-shape, structured as a 13-week series because the structure is what most beginners are missing
- JavaScript / TypeScript — modern web, modern tooling, the fastest path from "I want to ship something" to "I shipped it"
- AI for Developers — Cursor, Copilot, Claude, ChatGPT, but framed as workflows, not gadgets
I didn't plan five tracks. The fifth one — AI for Developers — didn't even exist as an idea when video #100 went up. The track emerged because every week, more questions in the comments were about "how do I use AI in my actual job", and the answers warranted their own dedicated thread.
The shape of the curriculum was decided by what people kept asking.
What the third hundred taught me
Three things, mostly:
Length is a feature, not a cost
I used to apologize for twelve-minute videos. By the third hundred, the videos that performed best — and the ones I'm proudest of — were the ones I refused to cut for length.
People don't want short videos. They want videos that respect their time. Those are not the same thing.
The shortest path from confused to competent is sometimes a four-minute Short. More often it's a twenty-minute walkthrough that doesn't skip the part where you have to wait for the build to finish, or doesn't fake-confidence past the moment where the real engineer would actually go check the docs. Length isn't bloat when it's the actual work.
A curriculum is a community decision
I cannot, sitting alone in front of a microphone, design what people need to learn. I can only respond to what people are asking.
The longer this channel runs, the more obvious it is that the audience is doing most of the architectural work, and I'm doing the implementation. My job is to be an honest builder. Their job is to point at the gaps. That division of labor is what makes the result a curriculum and not just a content library.
The unfashionable topics are the most valuable ones
Half of the most-watched videos on the channel are about things nobody talks about on Twitter — basic Linux administration, fundamentals of HTTP, cron jobs, log rotation, the unglamorous middle layer that every working engineer relies on every day.
That's where the gap is. That's where the channel earns its keep. Every time I've been tempted to chase a trending framework, the data — and the comments — have been screaming the same thing: cover the boring fundamentals first, deeper than anyone else is willing to.
What I'd do differently
If I started this channel over today, I'd think in series from video #1.
Most of the early videos were one-offs. Standalone lessons, no series context, no structural arc. They worked individually but didn't connect. The series-format I introduced around video 80 — and solidified by video 200 — is what made the curriculum legible. Every new viewer who finds the channel today enters through a series, not through a random video, and that changes the entire experience.
If you're building a channel, a blog, or a paid course: think in series first. Standalone videos and posts are fine, but a series is what holds a curriculum together. It also dramatically reduces the cognitive load on the creator — you stop deciding "what should I make next?" every week, and start deciding "what's the next lesson in this thread?"
What Chapter 3 looks like
Three things I'm building toward in the next hundred videos:
Cross-track projects. Right now the five tracks are parallel. What's missing is the project that uses all of them at once — the "build a real product end-to-end" video that crosses Linux, DevOps, Python, JavaScript, and AI tools in a single narrative. That's the next big thing.
Curriculum landing pages. Each track gets its own dedicated page on the website, with the videos in order, the prerequisites flagged, and the recommended sequence. The YouTube channel is the content; the website is the curriculum.
Quarterly community votes. Every quarter, the next series gets chosen by the audience. I publish three options, the community votes, the winner gets made. The channel is built by the people who watch it. The voting just makes that explicit.
If you're building something in tech education
Three things I'd tell you, three hundred videos in:
The audience is smarter than the algorithm. Optimize for the people who already trust you, not for the discovery mechanic. The algorithm rewards good answers to specific questions; it does not reward generic content optimized for it.
Your most boring topic is your most valuable one. Whatever feels too obvious to make a video about is probably the video that gets watched most. Test it.
Series compound. One-offs don't. The single biggest change I made between video 100 and video 200 was committing to series. Three years in, that's the lever I'd pull on day one if I started over.
Thank you
Three hundred videos. The math is unromantic — that's a video roughly every three days, every week, for nearly three years. But that's not what this milestone is.
What it is: three hundred conversations with the people in the comments. Three hundred questions answered. Three hundred small bets that someone, somewhere, was looking for exactly this lesson today.
The channel exists because those bets keep landing. If you've been here since the early ones, thank you. If you just found this — welcome. Chapter 3 starts this week, and the only thing I promise is that the next hundred will be made the same way the first three hundred were: slowly, honestly, and shaped by what you ask for next.
🎥 The milestone video: youtube.com/@Dargslan
🌐 The five tracks (free): dargslan.com
What's the one lesson you wish existed in your field but doesn't yet? Drop it in the comments. The most-asked ones get made first.
Top comments (0)