The history of this blog goes back to 2011. Its main goal was to share my findings and solutions to specific problems related to infrastructure engineering, Linux and coding.

Back in the days, it was IMO one of the best ways to learn, but also to share knowledge with the community. Looking at referrer statistics from those times, Google Search was responsible for something like 90% of the traffic. People were actually finding solutions here, commenting, contacting me directly via e-mail - man, there was so much good stuff happening!

And 15 years later, it all kind of broke apart.

Today, 99.9% of the traffic I see seems to be AI agents scraping the content. People changed their habits too (can't blame them!) and it looks like the first obvious place to ask for a solution these days is an LLM.

My blog looks mostly dead in terms of actual, human readers. Yes. It is sad.

So - does it still make any sense to keep posting?

I went for a longer hike. Actually, three sunset hikes. And spent plenty of time thinking about the whole tech-related blogging thing in 2026. Does it still make any sense?

Before I share my thoughts on this, allow me to share some of the photos I took during those hikes. The Chamonix area is unique!

The motivation

One of my original motivations for posting - and the one that looks mostly killed by LLMs - was helping others solve their problems.

It was really a special feeling to know that after spending hours, or sometimes days, tinkering with a problem and then posting the solution, someone found it, applied it and said:

"Hey Maciek, big thanks, you saved me tons of time!"

I don't think there's a way back to this. At least not at the scale it used to happen.

But there was another motivation: learning.

It follows the well-known idea associated with Richard Feynman: "if you truly want to understand something, try explaining it to others". Not his exact words, but you get the point.

Posting about technical subjects was always challenging. Many times I found myself asking additional questions and challenging my own understanding while writing a blog post. And how many times did I actually discover a problem with my understanding that I then had to fix?

So this motivation stays.

It overlaps a lot with conference and meetup speaking, where preparing to explain something to others forces you to understand the subject much more deeply yourself.

There's something more, however. Not really a motivation, but rather a kind of self-retrospective.

Looking back at the blog and scrolling through some of the posts makes me really happy after all those years. It brings back good memories, but it also lets me see how my interests, knowledge and opinions evolved.

So maybe it's not a strict motivation for writing a blog, but it's definitely a huge pro of having one.

So what now?

Over the last couple of years, Google has been steadily transforming Search around AI. AI Overviews became widespread, and at last Google I/O the company presented AI Mode as an end-to-end AI search experience. By 2026, asking a question and getting an AI-generated answer directly inside Search is simply becoming a normal part of using Google.

That's a huge change for websites like this one.

At the same time, while reviewing this blog's statistics, I learned something interesting. After filtering out most of the AI-related crawlers, Google actually dropped to second place in my referrer stats - just behind DuckDuckGo, and very close to GitHub and Bing.

And it got me thinking. Why should I share my findings with AI bots when there's no feedback and very little return? Why should I pay the electricity bill and spend my CPU cycles - remember, I self-host all my blogs! - so AI crawlers can scrape this place multiple times a day? Maybe I should simply block all AI traffic?

Google makes this particularly interesting. There is Google-Extended, which lets publishers restrict some uses of crawled content by Gemini without removing the site from regular Google Search. But the underlying crawling is still done using Google's existing crawler infrastructure. And I completely don't trust them. They turned evil during the last decade.

Maybe I should go even further and block Googlebot as well?

DuckDuckGo and some of the other search engines don't bother me nearly as much. And people who actually want to follow the blog can still use RSS or the newsletter.

I'm really thinking about doing it right now. Blocking AI bots isn't that hard, and there are plenty of interesting ways to approach it. Who knows - it might even generate some more blogging content 😄

I'd like to finish this post with something optimistic. It would be something like:

"I won't stop posting. Never. It keeps me sane."

And also:

LLMs may have killed some of the audience for technical blogging, but they haven't killed the value of producing the original material.