This one is demonstrably false. Your personal written style is what's important. Also, you have hands-on experience, which is also demonstrably more than any "AI" has. I urge you to ignore this kind of doubt or consideration.

- What kinds of posts actually worked (for learning, career, network, opportunities)?

For learning, "book report" type posts, just to solidify what I've read in my mind, maybe drive a little experimentation to ensure I've concluded correctly. I've decided not to collect any metrics so that I don't follow from behind, so that I don't end up doing clickbait. Career and network opportunities have not arisen from my blog.

My "public notebook" posts get more traffic, and I've referred back to them, but for me, these are mostly Linux sysadmin topics. I'd wager these are most valuable to people that find them for very specific problems, like seeing LLDP info from inside a WiFi access point or fixing GRUB problems on particular hardware.

- Any practical format that lowers the bar (length, cadence, themes)?

I have not discovered anything for this, alas. I use Hugo, I have a couple of little shell scripts to do monthly counts of finished vs draft articles. I try to stay at or above 5 posts a month. I'm not sure that helps lower the bar, which I interpret as "provide motivation to post".

What would I do differently? Start a blog years before I actually did so.

I'm happy to correspond, my email is in my HN profile.

Re blockers:

- Novelty: I routinely search for very niche, "boring" information, and am disappointed by how few in-depth blogposts I find.

- “AI can explain most topics better than I can”. I doubt it! I rarely find current AI as valuable as a good blog post. It tends to be shallow and regress to the mean, and b/c of hallucinations it's untrustworthy, so a lot of time is wasted fact-checking.

- Fear of shipping: if it isn't relevant, nobody will read it (unless you're already famous)

Re questions:

- What made it worth it for you?

Clarifying my thoughts, connecting with strangers who think about the same things, the leverage "having a platform" produces (it opens a lot of doors), and gaining prestige in certain niches.

- What kinds of posts actually worked (for learning, career, network, opportunities)?

I don't think this is simple to answer until the heat death of the universe. Traffic stats is a very poor estimator of value delivered. Which posts I am most proud of, and how much traffic they got, are weakly correlated.

- Any practical format that lowers the bar (length, cadence, themes)?

Things that you are obsessed with. It's a tonne of work writing a good post, and sometimes you publish it and nobody cares, so it has to be intrinsically rewarding.

- If you were starting today, what would you do differently?

I don't know! Probably put less effort into trying to appear intelligent/impressive, which rarely works anyway.

These are my off-the-top thoughts based on over a decade of blogging.

Writing is cathartic, but you already know this.

I have a small personal blog myself.

There is no money in blogs if there ever was. The money moved away to social media a long time ago. Leave the blogs to human people talking to each other and showing off their gardens and pets and hobbies.

Opportunities. You don't need many readers, you just need the right readers. I'm a big believer in making your own luck - putting things in place that make luck more likely to strike. Having a collection of writing online that people might stumble onto is very effective way of doing that.

> What kinds of posts actually worked (for learning, career, network, opportunities)?

I've written a bunch about this in the past. TLDR version:

- Stuff I've learned: TIL style posts that describe something I've learned recently

- Stuff I've found: links to things that are useful, with an explanation of why they are useful

- Stuff I've built: descriptions of projects I've completed

What to blog about: https://simonwillison.net/2022/Nov/6/what-to-blog-about/

My approach to running a link blog: https://simonwillison.net/2024/Dec/22/link-blog/

> Any practical format that lowers the bar (length, cadence, themes)?

TILs are an incredibly liberating format. You don't need to be describing something that's never been written about before - just something that's new to you today.

> If you were starting today, what would you do differently

I'd use static publishing on GitHub Pages on myname.github.io so I don't even need to run any web hosting or buy a domain name.

it's very silly to analyse trivial hobbies this much; if you want to write a thing, do so.

For me, the main motivation is that I enjoy reading other people's blogs, and hopefully my posts give someone ekse a similar enjoyment

I had a few attempts to lower the bar (tags for low effort, short and shitpost so far), but it feels like a crutch and hasn't worked long term for me.

I don’t do it often anymore (lack of time) but used to be a somewhat active blogger. It helped with my own understanding of the topics I wrote about.

Learning to hit publish even when you're full of doubt is the cure for self-doubt. Stop letting doubt rule your life and do the things you want to do!

> Any practical format that lowers the bar (length, cadence, themes)?

My recommendation, short posts at least once a week revolving around a single topic

> If you were starting today, what would you do differently?

I would not have built my own blog from scratch, I would just use one of the many fine options out there. Be realistic, you likely will not get many readers, at least not for a while. The value of blogging is what you learn about writing and the topic you write about it.

Today's AI is built on human-made content, and if we want "more" AI then we will need more human-made stuff. So it's a moot point. Unless you are OK with AI causing a plateau in human progress, don't let it get in the way of you (a human) from making progress.

That said, I cannot really comment on your first or third blockers. I have the exact same problems.

LLMs are likely more attentive readers than most human beings and in a way a blog might achieve even greater reach by virtue of being read by an LLM and incorporated into its "understanding of the world." (Or whatever is the right metaphor.)

> “Nobody needs this” / “It’s not original”

We need it more than ever. Who cares if it's not original, AI slop isn't original either.

> “AI can explain most topics better than I can”

Don't write tutorials.

> A bit of fear: shipping something that feels naive or low-signal

Life is about overcoming your fears.

If you want to have your content discovered online, I'd say you might be in for some trouble, although I don't think AI is the cause, only an accelerator on that. Blogs for readers learning are probably in decline, you're unlikely to get any outreach based on your posts for networking.

However if, like me, the writing process is the point – you're trying to clarify your thoughts, learn something new yourself, or have a document you can share with colleagues when they ask you to explain your opinions, I think blogging is valuable. While you won't get direct outreach, you can share it on your CV or send it to recruiters and you might get noticed when applying for jobs.

1) AI absolutely makes new blogs hard to get any traffic, unless you are already famous somewhere else

2) That said it’s still worth writing even if it’s just for yourself