Ask HN: What are some technical topics I can blog about?

When you start something new or run in to an issue doing something create an article.

Detail what you are doing and what documentation your are following. Outline the steps, pitfalls/potential pitfalls and how to overcome them.

Almost everything we do whether something new or doing something new detailing the steps and providing tips to complete it or improve it is going to help someone coming down the same path after you.

So take inspiration from your own work.

If you use a framework or language that's a good topic to dump all the knowledge you come across for others following in your footsteps.

Every topic has been written about, but there is always new versions of everything so detailing the steps again and including anything missing from the docs will be helpful to people.

For those interested, i wrote a blog post on this exact topic[1] about a year ago.

1: https://blog.benroux.me/untold-benefits-of-a-software-blog/

If you want to write about what you know, do it. Even if someone else has written about it, you may write about it differently. Maybe your "voice" will reach people that didn't like another blog post about the same subject.

If you decide, as saluki suggested to write about something you don't know; do as he suggest and provide details and references. This could be more interactive, if people are trying to learn along with you.

The key is to write -- both on schedule and as well as you can. Adhering to a schedule means people can easily follow along and know when you are posting. The more you write, the better you get.

It'll cement the knowledge in your mind, and perhaps you'll have insights that others haven't had. If you're aiming to write about stuff no-one's ever written before, you'll never get around to actually writing anything.

Why?

> but almost every topic I think of has been written upon by someone, probably better that what/how I would have written it.

So?

If you want to write the most authoritative blog post on some technical topic, why? Why be the best? Why that topic?

Write what you want to write about. In a year or two you can look back and revise/delete your old posts if you want. Do you think your favorite blogger's first post was amazing? Probably not.

At this point, you'll be better served by building a publishing habit than picking the mythical perfect niche. But, if you're concerned, in my experience, all the good niches either have huge amounts of competition or they're so incredibly bleeding edge that they're risky to learn about.

Learn to write for yourself first. If you set up Raspian on a Raspberry Pi over the weekend, write about it! If you popped a CentOS install disk and set up a server, write about that. It doesn't matter if there's 10000 blog posts about the same exact thing, you can turn even mundane topics interesting as long as you have a good writing style- and that's what you're going to be exercising by blogging.

Find a topic you want to learn (or understand at a much deeper level) and blog about it.

Agree with the write for yourself mindset.

I am kind of doing the same thing on my blog https://flaviocopes.com where I write every single day (personal challenge) about some particular thing related to Web programming, and I mainly write for me, because I don't have a good long term memory and every time I need to use X I reference what I learned n months ago about it. That's much better than searching again and again "that post" you remember that clarified the thing for you.

Turns out this is useful for other people too, as I can see in the positive feedback I get.

After some time I started rewriting old posts to improve them / add more things I learned, as your writing style evolves.

So, my advice is, just start and stick to it.

The people that wrote the better version of your article probably weren't great writers from the start. My advice would be to just start writing. You will get better and you will find along the way what you want to write about or maybe you will find you don't want to write at all.

You could write about new programming languages or new algorithms that just came out. Anything with nice animation is a plus. (I.e. to illustrate the algorithm)

Another thing that I find interesting is interviewing great people in the field. Often, there are questions I wish I could ask them that most other interviewers don't because they're not technical enough.

You can write about weird bugs that you encountered, how you fixed them and what lessons you learned.

Any pointers on how to start blogging about a side-project? I've been working on a prototype, but it isn't quite operational yet. I'd like to start laying the groundwork, but I'm having trouble distilling my thoughts into actual blog posts.

Write for yourself to help learn a skill and improve your ability to write future articles.

Worry about being the best written explanation.

For instance patio11 has gathered a huge blog audience, and he didn't do that by writing about things no one else had written about, or knowing more then everyone else.

But his writing is fucking amazing, and that's why I can read 20 pages of his writing while being barely able to make it through 3 paragraphs of other authors.

Quantum computing understandable for simple tech people (not quantum physicists). There is so much noise (as in SNR) on the internet about this topic, and I think I have a reasonable grasp of it, but I don't have a big audience nor the time to write it all down. A good source on that would be great.

Machine learning: where are we and does it hold any promise for future work, or is it a dead end beyond optimizing what we're already doing with it? Neural networks sounded like "oh, with enough compute power, we can simulate a brain right?" but so far the models show no level of true understanding whatsoever (to the best of my knowledge).

------

While writing answers on the IT Security StackExchange website, I often find that answers of mine which I found dull and uninteresting (they logically follow from basic knowledge) are popular in terms of views and votes. Usually it's somewhat predictable, but I'm regularly surprised by what people find interesting and useful. One example was about receiving email without a "to" field ('how did it end up in my inbox?'), which I thought was a niche question (only people who run their own mail server would have that, I thought, and they know SMTP right?), but 80 upvotes tell a different story.

And like other commenters said: it's not so much about what topic you choose, but about how well you can explain it. Of course pageviews is your quality of writing multiplied by the general interest level (bad quality but huge interest might still get you a lot of views), but most of the time there turn out to be a lot of people interested if you only explain it well, so the quality matters more than the topic.

That's a defeatist attitude that will sap your energy and make sure your blog is short lived.

Now, I definitely understand the sentiment, because there are a lot of smart folks out there writing on the internet. However, it simply isn't true that you don't have anything to say, even on a topic that has been written about over and over again.

What I'd suggest is writing about something that you are passionate about and willing to investigate. And I'd suggest primarily writing for yourself and your understanding rather than to be a famous blogger or find a job (those might be byproducts, but they're unlikely, whereas gaining knowledge is pretty much guaranteed).

The topic could be a problem domain (real estate tech, fintech) or a specific technology (AWS RDS, Google cloud spanner) or even development techniques (TDD, mob programming).

Pointing to other interesting posts and riffing on them is a great way to get started, but do be prepared to spend some time writing original content. Don't feel you have to find your perfect topic immediately, you can wander a bit.

Source: I've been blogging since 2003.

Do you use any proprietary or closed source products? Many (not all) of those have poor quality, perfunctory documentation, and if you work with them a long time you'll development knowledge by means of experimentation, reverse engineering and support tickets.

Try blogging some of that, or develop it into some alternative documentation, book, or presentation. Become a well-known independent expert.