Fifty-Five Days In: Being Wrong in Public and Writing on Empty

I started writing every day in January as an experiment. I’ve kept going because it turned into something I didn’t plan for: it changed how sure I am of what I think, and how I say it out loud.

Two weeks ago, in Writing as Thinking, I wrote about the page as a thinking tool and why the delete step matters. This post is about the parts that only showed up later: what happens after you publish, which ideas survive, and the days when there’s nothing in the tank.

What I expected

A writing habit. Clearer thinking. Maybe a few useful posts for the blog.

All of that has happened. But there’s something underneath it that I didn’t see coming.

What actually happened

My opinions got more calibrated

Writing forces you to commit to a position. Then people react. Then you find out whether you were right.

I’ve been wrong in public a few times. Not dramatically wrong, but noticeably wrong. That used to embarrass me. Now I treat it as information. An opinion I’ve never had to defend is really just a mood, and publishing it is the cheapest way I know to test it against people who see a different slice of the problem.

The side effect is that I now hold my positions with a confidence level attached. Some things I’ll argue for in any room. Some I’ll write as “my current bet”. Knowing the difference is most of what people mean by judgement.

I stopped hoarding ideas

Before this, I’d have an interesting thought and file it away for a “proper” post someday. Most ideas died in that folder.

Now I write the idea the day I have it. Most of them turn out smaller than I thought. Some are bigger. I wouldn’t know which without writing them. The folder was protecting ideas from the one test that mattered.

The topics I return to surprised me

I assumed I’d write mostly technical posts. I’ve written as much about parenting, presence and consulting as I have about Azure.

That’s the writing showing me what I’m actually thinking about. A post about my son’s LEGO set ended up being about debugging discipline. A post about being hired as the expert ended up being about listening. The personal and the technical keep leaking into each other, and I’ve stopped trying to keep them apart.

The rhythm

Morning coffee. Thirty to forty minutes. Most posts start as a question I want to answer:

• “Why does prompt caching matter, and why do I keep forgetting to bring it up with clients?” That became Why Your Azure OpenAI Prompt Cache Keeps Missing.
• “What does my son’s LEGO session have to do with debugging?”
• “When does an agent make more sense than a workflow?” That became Workflow First, Agent Inside.

Starting from a question matters more than I expected. A topic (“prompt caching”) gives me nothing to push against. A question gives me a finish line: the post is done when I’ve answered it, or when I’ve admitted I can’t yet.

The writing is the thinking. The published post is the evidence that the thinking happened.

What I’ve let go

Early on I worried about quality: “Is this good enough to publish?”

The question has shifted to: “Did I think clearly about something today?” If yes, it gets published. Length, structure and polish serve clarity, not the other way around.

Some posts are 300 words. Some are 1,500. The word count isn’t the point. A 300-word post that makes one decision clear is worth more than a long one that surveys a topic and commits to nothing.

The benefit I didn’t plan for

I talk to people differently now. When someone asks what I think about something, I have a more precise answer. Not because I’m smarter, but because I’ve been practising precision in writing every morning, and it carries over into speech.

Clients have noticed. A few have asked whether I consult on communication, not just technical architecture. I didn’t expect that, and I take it as a signal of where the real value in consulting often sits: the architecture matters, but so does being able to explain the trade-off to the person signing off on it.

The hard days

There are days when nothing interesting happened and nothing interesting comes to mind.

Those turn out to be some of the best posts. Writing from an empty state forces honesty. There’s no clever framing to hide behind, just: this is what I’m thinking, even when what I’m thinking is mostly uncertainty.

They’re also the days the habit earns its keep. Anyone can write when an idea is burning. The practice is showing up when it isn’t, and finding that there was something there after all, just quieter.

Fifty-five more?

Yes. Not because I’ve committed to a number, but because stopping would mean choosing less clarity over more.

That’s not a good trade.

If you’re considering it

• Start with ten days. Don’t aim for quality; aim for consistency. Ten is short enough to finish and long enough to hit at least one empty day.
• Write what you actually think, not what sounds impressive. The impressive version is the one you’ll be embarrassed by later.
• Publish it even when it’s imperfect. Being wrong in public is how the calibration happens.
• Don’t do it as a performance. If daily publishing becomes about streaks or reach, it stops being a thinking tool. Skip the day rather than publish filler.

Day fifty-five doesn’t look much different from day fifty-four. It looks very different from day one. The only way to get there is to start on day one.