Some blogging myths | Hacker News Reader

Some blogging myths

jvns.ca

245 pointsGavinAnderegg62 comments

My three tips for blogging more:

1. Write TILs. The only justification needed to publish one is "did I just learn something that was new to me?" - having such a low bar makes it much easier to publish frequently.

2. Write about projects you have shipped, no matter how small. I like to think of this as the price I pay for distracting myself with yet another little side-project.

3. Learn to hit "publish" when you still aren't completely happy with what you have written. The alternative is never publishing anything at all!

> 1. Write TILs. The only justification needed to publish one is "did I just learn something that was new to me?" - having such a low bar makes it much easier to publish frequently.

I would add "today I just explained" (whether an internet forum question, a new colleague, or my nana ;)

> 3. Learn to hit "publish" when you still aren't completely happy with what you have written. The alternative is never publishing anything at all!

Perfection is the enemy of progress! (And there's always an edit button ;)

My blogging flow doesn’t have an edit button. I email a blog post which gets published for the world to see.

So, there is not always an edit button.

The limit to the way to edit a blog post is your imagination

1. Delete it and write a new one

2. Close your account and send DMCA takedown notices to the server, then replace with a clone of your old posts plus an updated one.

There's /always/ an edit button, just, it might be shaped a bit differently than you have previously imagined ;)

Somewhat in the spirit of this article, your #3 point:

> 3. Learn to hit "publish" when you still aren't completely happy with what you have written. The alternative is never publishing anything at all!

...reminds me of the very well known quote that most will know but maybe some won't:

"If you are not embarrassed by the first version of your product, you’ve launched too late." - The LinkedIn Dude

Yes, and I'd like to add a #4.

4. Publishing is a side effect of writing. So write as much as possible... Keep spooling the brain to disk, noodle on many bits and bobs, email friends those half-baked thoughts, let them accrete into more baked thoughts, GOTO 3.

#1 through 4 is how I re-started blogging: https://www.evalapply.org/posts/hello-world/

And perhaps a #0 too... giving myself the okay to blog. In a "permissionless" medium like the WWW, all that is required of a person is to make up their mind and slam that publish button.

> 1. Write TILs. The only justification needed to publish one is "did I just learn something that was new to me?" - having such a low bar makes it much easier to publish frequently.

Exactly this! I started publishing TILs (just very short articles, most of the time I use them as resources for myself) [0]

Sometimes I keep an one-liner and develop it into a full-length article. (e.g. This one [1] was like, today I learned not all tool extract same colors form the same image)

[0] https://kiru.io/til/ [1] https://kiru.io/blog/posts/2023/extracting-colors-from-image...

Great post. I'll propose one more "myth": writing a blog doesn't benefit you personally, except by drawing attention. I've found that writing about a topic can also be a great way to solidify your own thinking on that topic. There ideas that I've understood much better after blogging about them. Basically, you're using the entire Internet as your "rubber duck".

“How do I know what I think until I see what I say?” -- E. M. Forster

“How do I know what I think until I see what I wrote?” -- ChatGPT

Apt for ChatGPT which literally has no capacity to think outside of the process of writing.

Hey Steve! One more thing re: drawing attention: you're drawing the attention of people interested in the same topic, and often thinking along similar lines as you. And that's a fun community to connect with, even if you're not saying anything new or benefitting from your writing in other ways.

Arguably the main reason of writing a blog is to collect and organize thoughts in one's mind. Especially when writing about technical topics/reviewing scientific literature.

True, I've added blog writing to my learning process. I'll sit down to write about a topic and find out that I don't really understand it.

It helps me to put stuff in a structured manner for myself and other to review later.

Often I'll ask questions to myself that I didn't during the learning process. Like "Do you really understand what a controller manager is? What is the difference between a controller manager and custom controller in Operators?"

Julia Evans is a better blogger than I'll ever be, but I've been trying to get back into blogging. These tips are great, and all things I need to keep in mind. "You need to be original" and "writing boring posts is bad" are two thoughts that often keep me from writing on a subject.

> writing boring posts is bad

This is highly subjective, what you think is boring might be super interesting to another person.

The mistake that too many people have, is thinking that your blog has to appeal to the masses. Unless it goes 'viral' and attracts millions of views they think it is a failure.

Instead, focus on topics that might only be 'not boring' to a small subset of the population. HN tends to attract people with very particular interests. I often browse through a hundred or more posts before finding one that truly interests me.

My own blog is dedicated to those who are passionate about data management. This is a small group indeed, but who cares if less than 1000 people read my posts. All it takes is a few to learn something meaningful from it to make it worthwhile.

I am facing a similar dilemma. Lately, I have been writing about data structures and the "original" aspect has really taken over my writing. Breaking down the content into a readable format and complementing it with intuitive figures can become a challenge. I have two almost-ready drafts sitting on my computer, but I cannot gather the courage to post them as I feel like they might be missing more analysis. I am not sure if this is necessarily good or bad. I have been trying to treat my blog as something I enjoy, with content that will still be relevant in a decade or so. Perhaps, I need to take Julia's advice and write more boring posts, like some of my favorite things in my bashrc.

Maybe you ought to adjust the approach. Rather than have a complete and finished reference now, post what you have so far to show what you have been trying. Someone may reach out with additional insight. Though whether or not you get any feedback at this stage, it can lead up to a post that covers everything that you want once it's ready

You can always edit old posts or just create follow-ups and link them together. Or just keep it as an open topic that you periodically post smaller updates on

You might check out what Steve Randy Waldman is doing with his "drafts" blog. He has found it freeing: https://www.interfluidity.com/v2/9807.html

Another myth that dovetails with the author's mention that pageviews don't matter: The only way to get traffic is to write about something a lot of people search for, thus the entire industry around Ahrefs and similar Google Search tools.

And yet. The blog posts that have been my breakout successes, when measured by pageviews or cold outreach or opportunities, are always about incredibly niche things. How to fix one specific thing on an obscure printer. My rambling ideas about a broader topic. If anything, the more niche, eccentric stuff that only matters to you helps you find your people, and that's where the magical connections happen. And the traffic, weirdly, as no one else is catering to your people as well as you can.

My biggest problem after blogging regularly for 15 years is that I continue to have more ideas and interesting things to write about than I have time to write.

And the more I write the more ideas I get.

edit: This has especially been a problem for the past year because I’m also under a book contract. Most days writing the book takes priority, but then my mind is exploding with blog ideas that don’t fit in book. But I don’t have energy to do both.

I think the deeper issue is that there is no "one right way".

I subscribe to the "you need to be original myth" for example. If I don't feel I can contribute anything to the existing corpus of text then I won't write about it. This does not mean that there can't be pre-existing content; I did an explainer on bitmasks for example – hardly a new concept – but I felt I could add something to the existing corpus, so it was "original" in that sense. I also tend to only write on things I'm fairly knowledgable on (with one or two exceptions) and while no one is ever "100% correct" I spend quite a bit of effort ensuring it's as close as I can be. All of this is what works well for me, so that's fine.

Fundamentally: it's your site, and you can do whatever the hell you want with it. Purple text on a pink background? It's your site! Lost of swearing? It's your site! Furry drawings? It's your site! Gay leather kink pictures in the sidebar? It's your site!

The idea of editing already-published blog posts is unappealing to me, in case someone bookmarks or saves a link to my article before it was edited. Unless I make it clear what content was edited and when (more work!), the reader might be coming back to something different the second time they visit my page.

It's this fact of the web that forces me to go to The Internet Archive's "Save Page Now" whenever I come across something remotely interesting, and I don't want to be part of the problem!

Before someone mentions that most likely nobody will care about my edited page - it's a matter of principle! For example, I'm sure many of us here take the time to license our toy projects, even though the vast majority of them are effectively unknown to the rest of the world.

These are all great tips, and I need to keep them in mind. I have a blog post I've been trying to finish for months, and pretty much all of these things have been holding me back.

Not really a "myth", but I'd also say: Don't box yourself into writing about one specific topic.

I find blogging to be a really rewarding thing to do - but I gotta remember that I do it for myself and not others. When I start to drift away from this then I start to suffer. It's an interesting thing to balance, since you're publishing things online.

One strategy that works well for me is to first do all my writing and note-taking in an actual notebook. This lets me write freely and explore ideas. Then I can make a decision about whether or not to polish and publish.

About the "posts need to be 100% correct", I sometimes read through my old blog posts and find a bunch of errors. If it was a post on HN you can no longer edit it, but if it's your own blog you can just edit and make corrections! Sometimes I add a note like "Me from the future..." it feels like time travel.

I'm a blogger too, not as famous as Julia, but I'm too famous for my tastes. [1] [2] [3] [4]

I agree with Julia's points perfectly! In fact, some of those four posts were "boring" (I thought), but got a lot of traffic.

I just want to add two more myths that are unfortunately pushed in today's clickbait culture.

Myth: You need to make your posts clickbait.

One of the posts linked about used clickbait-ish titles ("Considered Harmful"), but they weren't very clickbait-ish in general. Yes, my thesis may have been opposite of mainstream, but I don't think the titles were clickbait.

Myth: You need to make your posts appeal to people's emotions, i.e., make them ragebait or similar.

Those four posts are my highest traffic posts on HN, and they were all level-headed posts.

I had an earlier version of my "Considered Harmful" post, and it was ragebait, but it didn't get much traffic (thank goodness!).

Be constructive.

I think if there's a theme to Julia's points, it's that: be constructive and useful. Sometimes, that means being boring. Sometimes it means having a short post, sometimes a long one. Sometimes, it means repeating something. Sometimes, it means skipping concepts.

My $0.02.

[1]: https://news.ycombinator.com/item?id=34662666

[2]: https://news.ycombinator.com/item?id=30965805

[3]: https://news.ycombinator.com/item?id=28736238

[4]: https://news.ycombinator.com/item?id=29671325

> > Myth: You need to make your posts appeal to people's emotions, i.e., make them ragebait or similar.

> Those four posts are my highest traffic posts on HN, and they were all level-headed posts.

I think the examples you've linked, it's the topic rather than the reasoning that's going to incite emotions; especially if your take is contrarian, not mainstream, or discussing unresolved problems.

Perhaps, but I argue that they didn't raise emotions much because people were awesomely calm in the comments on HN.

> I still don’t know if all of those statements are true (is it true that PHP programs can’t have long-lived persistent TCP connections? maybe not!)

They can, depending on how the PHP program is coded. PHP even supports non-blocking I/O.

One thing that comes to mind about comments.

I think writing blog posts may be hard, but writing a constructive comment is hard also. People are not good at it, and I often find myself writing something and then just give up in the middle and deleting it again. Not just because finding the right words is hard (especially if you're not a native speaker), but some of the same questions go through my head as well (Who is the intended reader? Hasn't someone commented on this before? ...).

Thus I would suggest assuming the commenter has spent more time thinking about what to write than you might expect. And even if not, someone else may get some value out of their comment, so even if it isn't you, give them the benefit of the doubt. The point about assuming they're confused, not angry, is spot on I think.

There is probably a connection to the point about writing becoming harder with more experience. Each sentence you write will trigger more warnings in your head, or it might become harder to put yourself in someone else's shoes. To keep going, you probably have to learn to just accept and expect imperfection and just keep going, cut the editorial phase short at some point and just publish or you'll never get done.

The next presumably angry/ignorant/dumb commenter you encounter may have just done that, and by necessity. You never know.

Also, if you can write a really long reddit/HN comment about it, you can probably write a blog post about it. A lot of my blog posts come from either that or ranting to a friend.

Hello, friend. I'm in the very same boat :)

My thus-far latest post is exactly that, a cross-post of a long-ass wall of text I ended up making in one of the Slacks I frequent. https://www.evalapply.org/posts/cold-restart-total-outage/

I blog regularly, and given many other bloggers are commenting, let me ask: how do you do analytics? I am currently using Google Analytics, but they're changing it and a nagging me to change the configuration, but I wanted to move away from them for a long time for something more privacy-oriented (I just want to know how many people are reading my posts, maybe from which countries, and that's all - no ads). Would appreciate honest suggestions (no ads).

I'm using Mataroa[0] as my blogging platform and it comes with its own analytics. It's very simple, and gives me the following:

- Read counts per post, for the last 25 days.

- How many RSS pulls, for the last 25 days.

- How many site visits, for the last 25 days.

Nothing more, nothing less. I'm pretty happy with them, and have no intention to change it.

[0]: https://mataroa.blog

For the longest time, I did not want analytics. I don't write for traffic or eyeballs, and I abhor being profiled for ads. My purpose is to think via writing, and as a side-benefit feed conversations about the stuff I am interested in.

Recently though, I added Cloudflare analytics (no cookies, no deep profiling) out of sheer curiosity, in the spirit of the innocent surprise and joy of seeing "hits" from across the world on my very first website in the mid 2000-s. Sooner or later I'll get rid of it.

As Julia writes in her post, page views aren't important to me. The occasional email conversations that fall out of my posts are!

(edit: better words)

I use GoatCounter (https://www.goatcounter.com/). It's very simple to set-up and privacy-oriented (no cookies).

I am now using this. Thanks for the hint... it works perfectly, unfortunately the ad blockers block it, but that's what they're for I guess...

I don't know why, but I hate having to use anything with a DB. So most of the options (although really good) were a no for me.

I ended up using https://goaccess.io/ because it doesn't require a DB, _and_ because it was in the package repo of the distro I use on my server, so it requires basically no setup. Just editing the config file.

This is nice advice. I ended up with choice paralysis even deciding how to set up and publish blog posts. Static site generators with deployment pipelines? Something more traditional like Wordpress? Other?

Gosh this is a horrible spiral I've been going through as well. My personal experience is:

1) WP: You start with promising yourself to never touch any PHP and will still end up in this mess. I always regretted setting up WP few days/weeks after.

2) Gatsby/Next: I'm saying this as one of those frontend hipsters: For a personal blog, just don't. The npm ecosystem imho got so messy that it always added more pain to actually start writing.

3) Hugo: just do it. It's small, fast, nice and once you wrapped your head around their taxonomy and theme model, it's really nice and easy to extend and fun to actually blog.

My 2 cents. Your mileage may vary.

jeez why not just use one of the million services.. php... npm... do you also start your own fire, like with a tinder box, to cook some food?

I recommend https://bearblog.dev (free) or https://ghost.org (paid). Both let you use your own domain.

It's fun to set up static site generators but I think it's better to pick something you can start writing in immediately instead of spending time thinking about what to use.

If you end up enjoying blogging you can switch to different software later if you like. But don't let that stop you from starting to write now.

Being fluent in another language than English, I’ve always wanted to take my blog multilingual.

It’s been years now. And Ghost’s internationalisation support is still a joke.

Give https://blogs.hyvor.com a try. It supports multi-languages by default. (I'm the founder)

Oh! It could be interesting for some.

Personally, wanting to build more things, I’m moving away from blog management systems, towards content management.

If I had discovered bearblog I probably would have never written my current static site generator. Bearblog is exactly what I was looking for because you don’t have to make any choices. Just write!

This is really great advice and I appreciate it. You're right about having something to start writing in immediately. It's funny getting tied up in the process when I haven't really even discovered if I have much love or knack for writing yet!

All the replies to my comment have been super helpful but the mindset to write immediately is important. I'm grateful for the suggestions and one of the other replies also sound positive for bearblog - it could work out well and I'm going to check it out :-)

I use github.io with Jekyll.

IMO, it's perfect:

- it's the default GitHub blogging platform and well documented, the static generation is handled by GitHub.

- I can run a local generator on my laptop and see how the content will look like before pushing content to the site.

- free

- no hassle with domain names. The thing is automatically linked to your github.com account.

- git revision control out of the box. I have 80+ branches with blog posts in various state of completion.

- markdown to write the content

- people can clone my blog if they want to, for their own use, or to fix things and submit PRs. (It has happened a few times.)

- longevity: I think github.com to exist longer than most other blogging platforms, and that it might very well outlast me.

- Instead of https://tomverbeure.github.io, I could change the URL to my own domain, one doesn't have a github.io suffix, but why would I?

Good tip (I am inspired to migrate to this actually) and a very nice blog.

Either you want to just get it done or you enjoy the process of tinkering on the infrastructure.

If you want it done: pay somebody a small amount to do the things. Sometimes it's even free.

If you like tinkering: pick a static site generator. Whichever one you pick will be wrong, but the process of setting it up and getting the first entries through will be a learning experience. In a bit you will have an opinion on what went well and what didn't, and be able to learn from others.

(I use Pelican with a bunch of plugins and a repeatedly abused CSS file. It's not right for everyone, but I like it.)

Just use whatever is easy to get started with for you and go from there. My first version was something like:

  #!/bin/sh
  for p in *.html; do
    cat .head    >out/$p
    txt2tags $p >>out/$p
    cat .footer >>out/$p
  done

And that worked for me. It was simple to set up and it allowed me to focus on the writing itself, rather than all the plumbing surrounding it. You can always switch tools later if need be (as I have done).

Sometimes programmers can get all hung up on the tooling: "I want to do X", "Oh, I need to find or write software to do X", spends more time tinkering with or writing software for X instead of doing X.

If your barrier of entry is setting something up, I'd definitely go with wordpress as it makes it quite seamless when writing new posts.

As for me I use 11ty [1] for my blog [2]. I quite like 11ty as I just write markdown files and it gets converted to static content. I'm also testing the grounds on using it for an online magazine, basically want to do some css garden tricks for different types of content. Seems to work quite well for this as I want as little JS being sent as possible.

[1] https://www.11ty.dev/

[2] https://azemetre.com/

I started writing plain .md files in GitHub and published using their default jekyll template for about a couple of years while. In the meantime, I messed around with several static generators before eventually just writing my own. Gives you something to blog about too :)

i wrote blogs simply becauase i enjoy writing and having fun, and i would have thought that is the primary motivator for most bloggers. and i really enjoyed writing training courses. given up now because of health and other problems.

a couple of mine:

https://latedev.wordpress.com/

https://punchlet.wordpress.com/2009/12/01/hello-world/

My primary motivators for starting a blog are trying to improve my writing and having an outlet for all of my thoughts and opinions that I don't want to yell at strangers on the street.

i disagree with a lot of this. there's enough incorrect and duplicate rubbish on the internet without encouraging more of it.

In the old days - before narcissism became an acceptable personality trait - people were encouraged to think, learn, research and be original before subjecting the world to their writings.

I'm not that old but i miss that

I imagine that my generation is far from the one you're talking about. Could you elaborate more on what it was like "in the old days," what has changed, and what you miss? Or is it hard to put your finger on?

Agreed. And i'll say something even more downvotable - i don't think Julia Evans's blog is very good.

I’m curious.. what is it about her blog that makes you feel that way?

Meh - its all contained under the jvns.ca domain though - you can skip that easily.

To paraphrase Rich Hickey, The only people entitled to say how blogs 'ought' to work are people who run blogs, and the scope of their entitlement extends only to their own projects. Just because someone blogs something does not imply they owe the world accuracy, originality, interest or education.

As a reader of a blog you are not thereby entitled to anything at all. You are not entitled to contribute. You are not entitled to depth. You are not entitled to be the intended audience. You are not entitled to gain value from the blog.

----

And your complaint can be easily dismissed by noting that the article author blogs about things she's learned, and that reading all existing content to see what is original is impossible.

Reply on news.ycombinator.com