Advice for a friend who wants to start a blog
henrikkarlsson.xyz
henrikkarlsson.xyz
- Own your content: Medium, Substack, Hashnode, Dev.to—all of them suck. Your readers deserve better. Don’t waste their time.
- No one cares about the technology behind your site unless it’s huge. So it should be the least of your concerns. Don’t be that JS guy who rewrites their entire site every week in a different stack and writes a blog post about it.
- Consistently writing is a lot of work, and there’s no way around it.
- Picking a niche and writing about it yields better results than trying to write about anything and everything.
- No one will probably read it for a long time, and that’s okay. You should write only for yourself in the beginning.
- POSSE is the way to go. If you don’t promote your work, no one else will.
- That said, if it took less time to write than it would take someone to read, it’s probably not worth writing.
[1]: https://rednafi.com
I didn’t know what this acronym meant. Hopefully I saved someone a Google search.
What does this mean? (Googling wasn't helpful.)
This has not been my experience. If you do interesting things then there are plenty of people who are willing to bring attention to it. I've been cold-emailed by journalists wanting to interview me without ever doing anything other than a) make my work publicly accessible and b) collaborate with like-minded people.
Perhaps I'm lucky (probably). Perhaps my work would've gotten more attention if I spent more effort promoting it (probably). The things I do because I find them interesting are not things I feel the need to promote, because I'm not doing them for attention.
If your goal is to grab a lot of attention, you will inexorably have to do things that are not interesting because the only way to achieve broad appeal is to not offend anyone's tastes and the only way to do that is to be bland.
Careful. While I agree with the premise, you make it sound binary. You can have broad-er appeal while still being individual.
On that …
- It's generally easier for people to know what they _don't_ like than what they _do_. Preferences are defined by process of elimination.
- The commonality is: get lots of people on the same page about you, and you've got a big audience / broad appeal. But the most engaged audience is the one who has strong emotions attached, so being an offender guarantees a ravenous audience … and is generally easier.
I witnessed a memorable moment at work when an engineer responded to a VP and started by saying “careful.” Had it been another engineer, it would have been ignored as authoritative-sounding filler. But the VP did not take it well.
That's like an actor breaking character during a play and it pops the bubble of reality the audience has been living in.
In all likelihood, the engineer meant "that approach/idea/decision has risks we should consider." The VP likely heard "you are suggesting something wrong/dangerous/ignorant."
A nuance here is that the engineer probably implied that "(We should be) careful (in our consideration)" and the VP perceived "(You need to be) careful (and you don't seem to be)"
I'm possibly paraphrasing and simplifying too much but "How to Win Friends and influence People" mentions being very cautious about correcting people, more so correcting them publically, even when you are correct.
Such a good book, I'm about due for a re-read I think.
It also places the speaker in a position of authority, admonishing someone who is still learning. "Careful!" is what we say to children. A high-status person may be particularly sensitive to this, but nobody likes to be talked down to. As I alluded, I hear this amongst engineers and filter it out as verbal fluff, so there are cases where it is ok.
An easy alternative to "careful" is "I'm not sure about that" or even "I disagree." Both are what peers would say to each other. In scenarios where a high-status person is not receptive to being talked to as a peer, well, you get to add all the deference and qualification you need to, but it comes down to stating disagreement in a non-disagreeable way.
Honest question, because I am just starting out, myself. I have a small blog where I share my opinions and projects related to technology I enjoy. However, if I were to do a thought experiment and assume my audience is HN readers, I'd hesitate to say anyone here would find my posts very interesting. I like them, but I am not sure who else would.
I genuinely do not understand how to evaluate an audience, so I just write what I like to write and hope for the best.
Setup Matomo or similar analytics software, write about a bunch of topics and review after a year which topics your readers are most interested in, for which they search on your site etc.
When you have a decent number of posts and some visitors you can see which sub set they find the most interesting.
You can also make a blog about things that don't interest you at all. This is the shortest route to success and you will hate your new job :)
> This has not been my experience. If you do interesting things then there are plenty of people who are willing to bring attention to it. I've been cold-emailed by journalists wanting to interview me without ever doing anything other than a) make my work publicly accessible and b) collaborate with like-minded people.
Both of those are promoting your work so it is your experience. Your experience just has been that you haven't had to put much work into promoting your work so it doesn't feel like promoting your work to you.
Is that actually a possibility? I can't type faster than I can read.
seems like a perfectly reasonable idea for a blog about different stacks. I'm not advocating it, but there's nothing wrong with it if that's what interests you.
Almost every opinion people have has been covered by someone else in some way unless you are working at the forefront of something, so it usually depends on how you value-add to existing things with your own experience.
It's also that sort of I-know-what's-better attitude that turns a personal thing into a competitive thing.
What is semi-successful anyway...
OP is saying don’t be the person who says they want to write about Frogs, but who then spends an immense amount of effort not writing about Frogs, and writing about procrastination instead.
Write in markdown and serve directly, or export to HTML. Or write in LibreOffice Writer and export to HTML.
Upload HTML to a basic webserver. You don't need a fancy load balancer or denial-of-human-system or anything.
No, you just need a host that doesn't have ridiculously small limits. You can easily find find hosts with limits in hundreds of GB, or TB, or even unmetered. You can find them for free, and you can find them for quite cheap compared to most technologists' incomes.
Just joking. Sort of. There's apparently a modern tool you can use to transpile SGML to HTML[0], and I'm itching to try it out.
--
[0] - https://sgmljs.net/docs/producing-html-tutorial/producing-ht... - yes, it's an NPM package, and yes, you can run it in the browser, but you can also use it on a server, or as a static site generator.
Disagree. If it took less thinking than it will take reading, then yes, probably not worth it. But if you can trim your writing to the pith in every paragraph or even line, people will love you.
Too many people babble to keep the underlying lies and fraud and their complicity burried in narratives.
This takes a lot of time. Writing concisely is much more challenging and time-consuming than writing verbosely. Writing unnecessarily long essays means that your readers end up spending more time reading them than you did writing them.
As Blaise Pascal wrote: "I only made this letter longer because I had not the leisure to make it shorter."
> POSSE is the way to go
Can you host your own blog, but replicate (in part) in dev.to?
Is it against their TOS to have a link to an original, or even only include teasers advertising the 'real' blog post?
It's kind of like having you main git repo somewhere, but cloning to GitHub for the community.
You know a blog is going to be good when it's using the default wordpress theme from 10 years ago.
https://github.com/simonw/datasette.io
http://blosxom.com/documentation/users/blog.html
http://blosxom.com/documentation/users/install/dynamic/isp.html
I love the idea of it in that it's just markdown files! `vim ~/blog/entries/some-random-thought.txt` end ups with $AUTHOR of `chown`, $DATE of the file modification time, tree of `mkdir -p ~/blog/entries/some/category` ... it's just trivial to backup, restore, and work with. You can even (probably) hook it up to a static regen which is probably safer: http://blosxom.com/documentation/users/install/static/
The link to datasette is if you're wanting to nerd around a bit more, as it's kindof a combo of sqlite and a web server for rendering data from within it. The linked github repo is how their main site is built/rendered and it definitely has aspects of "blogginess", and you can kindof see how it's done. The neat thing of `let tags = "SELECT DISTINCT tags FROM posts"` or transforming blogging into effectively updating/adding records in a database is an extension of the blosxom zen of "it's just files [records], yo! go forth and edit them..."
set plugins and themes to auto-update, add wordfence and change default settings.
use updraft or similar to pull full backups on a schedule.
I agree but when I tried this a few weeks ago, it looked like a lot of work to publish on my own blog and then replicate everywhere else. I didn't find a good tool and got lost trying out fediverse stuff.
> Your readers deserve better. Don’t waste their time.
Nothing about substack wastes readers time.
Conversely, what wastes their time is your site being down or hacked and what wastes your time is maintaining your blog's infra, updating because vulnerabilities, and/or using bad UIs like blogging from github (maybe passable for nerds, not for non-nerds)
A blog isn't about the tech you use. It's about the writing. Worrying about tech is not going to add a single reader to your blog.
Of course there is plenty of money to be made as a social media influencer who specializes in long form content, and if that's the career a person is interested in building then sure - Substack, Medium, whatever - great choice. But for people who are interested in blogging as it used to be, independent writers who want to write without the pressure of editors, without paywalls, without popups, without forcing their readers into signup funnels... It might be a bit more work to set up, but tools like WordPress, Ghost etc are better suited.
Login modal wastes time.
I feel like this is what blogging is especially good for. You can clear out the awful stuff and then try to incrementally improve.
I would diametrically disagree with point #7—if you want to write well, you need to revise the hell out of it. I guess pick whose writing you like better, between me and the author, and take the corresponding advice.
I find that phrase very helpful, because usually when I write, the first draft is always a giant mess. But you're able to craft it and edit it the best way that you can. So you can push out what you think is good work, and continue to improve as you write more.
I find forum posts far, far less nerve-wracking than a blog post. For me, a forum post is almost like how the original article describes a chat: it feels closer to realtime, and feels easier, because it's just like playing a normal role in a conversation. It's normally a response to something, expecting a response.
A blog post feels closer to a formal publication, and I feel like they're more expected to stand on their own, without the justification and context helpfully provided by others' posts.
My technique is Da Vinci method just collect / discover lot of shiny pieces. Then stitch them together later. Develop these pieces separately like colored glass pieces in a mosaic, sort of inversion of divide and conquer, collect and assemble.
each sentence should make sense based on every prior sentence. each sentence should be able to stand alone on its own sheet of paper.
write each sentence on a single sheet of paper.
What struck me back then was he said in an interview that you need to go through at least a thousand sheets of paper before you start producing anything good.
As a high schooler dreading to write a 4 page report on the guy, a thousand pages sounded ludicrous.
1) Blog for yourself and no-one else. Write things up because you find them interesting or you learned something. Trying to second-guess things like what will be popular or get engagement is not a fun time for what is likely to be a hobby.
2) Keep it low maintenance and simple publishing process. I use Github pages and Jekyll, but other similar options are available.
3) If you're doing technical content, try to have practical examples. I've had multiple times where I thought I knew how something worked and, in the process of writing a blog with practical examples, found out I was wrong about how it worked!
Sometimes others want to express a (non trivial) idea to someone else, and your writing may be the best thing they know that gets the message across. Then they can send it.
Sometimes (or rather, always) your idea and writing is not the best or final thought you have on the subject. But it is not until someone has read and commented/replied (thoughtfully) that you understand what the next refinement is.
And then there is serendipity, just putting it out there to see what might (or might not) come of it.
Plus maybe one finds it motivating to have it actually accessible for others. As a clear goal and a standard to hold each piece to.
1) Main reason is that if I've spent time working something out, I can save other people the time doing that, by writing down the details in a searchable place.
2) If I want to explain something to someone, it's easy if I can just point them to a blog I've already written
3) (bit of an unintended consequence) It helped get me my last two jobs :D Being able to point to a long history of technical blog posting was relevant when I got a job in advocacy as that's one of the necessary skills.
I don't think that merciless editing and beauty are opposed. The author perhaps underestimates the extent to which artists mercilessly edit their own work. The author employs the metaphor of musical improvisation, but the best musical pieces are usually composed, not improvised. I like jazz, but... not for the songwriting. For stream of consciousness writing, you'd almost be better off posting on social media.
I've been blogging for over 15 years, and my advice, admittedly ironic, is to not take advice from anyone. Actually, that's my advice on many subjects. Everyone wants to impose their own little idiosyncrasies on you, but what works for them won't necessarily work for you. One's idiosyncrasies are not universal truths.
Makes me feel OK to be quirky and unpolished when the rest of the world is making reaction faces and pointing at text to feed an algo.
A picture and 3 sentences can be a blog post.
You can talk about a hobby, a trip, a vacation or random thoughts.
Sometimes people will find a how-to blog post through Google. Sometimes an old friend or former teacher will look you up.
In reality, you'll probably be happier if your blog never grows a steady following :)
No need to write a full-on 10k+ word novella on everything.
People started blogs for that? I thought that was more the realm of supermarket bulletin boards next to the firewood flyer.
Maybe I'm just from a previous previous generation.
How can "a friend" post something on a supermarket bulletin board now that they don't seem to exist?
“Everybody wants to be a bodybuilder, but nobody wants to lift no heavy-ass weights.”
Many people get caught up in thinking the blog, YouTube, or podcast is the thing, when it’s just the thing that lets you share the real thing. It took me longer than I’d like to admit to really integrate this lesson.
I think putting more questions and uncertainty in the way of actually just writing/podcasting/youtubing is what stops a lot of people really building the habit in the first place.
I think understanding what you want to share helps solve the major stumbling block that leads to thousands of blogs with one entry that is effectively hello-world. If told to draw anything, most people will sit with a blank page. However, if they are told to draw a castle, those bounds give them the freedom to start and create, and it can evolve from there.
What are you into to? What hobbies do you have? What do you know, or have opinions about, that others might be interested in? This is your theme, the “real thing”. Writing, podcasting, or video production, without this usually falls flat after 1 or 2 attempts, and it makes it impossible for a habit to form.
If you’re already into something and doing it on a regular basis, then you’re just writing about, talking about it, or recording what you’re already doing or thinking about all the time anyway. The content and ideas can flow more easily, especially at the beginning. Not to mention, being more motivated to make the content, since it’s already a passion.
So many people are trying to make content about nothing, because they aren’t really doing anything else… and this is where friction is born. Do something you find interesting, then share that thing and your perspective on it. Start with the doing, not the sharing (writing, podcasting, and YouTubing is the sharing).
Hopefully this clarifies what I was getting at and resolves the chicken/egg thoughts.
all this conversion and optimization of texts for search engines turns bloggers into robots, and it is simply impossible to read the texts, they are not human, they are as if created by neural networks. who needs this hypocrisy?
you should blog, and by a blog definition that is going on the internet. to get more people to read it (often the goal) the easiest thing to do is utilize the tool people use to find things on the internet (search engines).
if there are issues with spam, keyword stuffing, forever run-on paragraphs of nothingness, well that is the search engine algorithm's job to fix. And they have a lot of money to fix it.
and if they don't fix it we should use a new search engine. it'll take time. but those seem like the dynamics at play in this scenario.
There speaks someone who has never spent time writing poetry, and probably isn't very good at editing.
I'm over 50 and my eyes are not quite what they used to be. Your 12pt (ish) text is largely legible to me but why do you insist on your idea of presentation, rather than mine?
The whole point of the www as far as I recall is that you send a message and I receive it. How it is formatted should largely be a matter for the recipient, with some hints from the sender.
- everybody limits it. This is because no matter the size of your screen, the human eyes are somewhat limited to lines of 70-130 characters. Higher than that, and your eyes have hard time finding the next line at the end of the previous one because the travel is too high. As a web page writer, you need to take care of that, because the default presentation will be bad for everyone if you don't.
- as it happens, the author uses Subtrack so they might not even have picked that size consciously.
- reader mode works on this page, so you can use this to read the post the way you want, and if you like the content and would like to read it regularly, there is an RSS feed that you can use with a client that displays posts the way you want.
The last point is why separate style sheets and correctly structured content are so nice: if you have such a strong opinion, you can tweak stuff or even ignore the style sheet and use yours. And reader mode actually makes that practical for most blog posts / articles.
(that said, I agree that the width seems particularly small, I replaced it with 80ex just to try and it looked better to me - but I just set up a new WordPress website and the default 2025 theme has the same ~730px width for the content, I don't know what's with this default size)
That's unfortunately a common myth even re regular text (even more so re code with all the extra structural helps) that's responsible for a lot of vertical space waste.
But at any rate you can adjust your window to whatever is comfortable, no need for the author to incorrectly guess what's best for you
A common myth that seems quite well researched [1], no? Do you have some evidence of the contradictory?
> you can adjust your window to whatever is comfortable
Websites all have different layouts, you'd have to change the window width (and/or zoom level) for each website. That would be quite clunky… on devices where it's at all possible.
Forcing users to resize their windows constantly is not a nice alternative.
And the main factor against long lines is errors returning to the next line's beginning.
First, what's the well researched level drop between 130 and 150 for a person with a high reading skill level?
Then, have they "well researched" such simple typographical guides as the varying line widths (check out the layout of this very article, note that it's not rectangular, but one side is continuously growing in length, cool trick!). Or having some alternating line markers at line beginning/end to make visual match easy (similar, but not identical, to tables where you have full line highlighted)?
Maybe the most efficient reading is at line length of 333 with line guides and semantics guides (like those bionic reading extensions) at higher text density? You could be missing out on +10% speed bump and 4% comprehension level! And the head movements would also reduce muscle strain from a prolonged fixed head/neck position! Or maybe the very best is even higher 1234 but requires some curved ultrawide monitor to fit you whole peripheral vision and immerse you in the text without other visual distractions? I'd be curious to know what the real scientific eye limits are when it comes to reading.
But also efficiency isn't the only important metric for casual reading of web blogs, while a lot of research focused on that. There was some research about user perception that found no difference between line lengths. So there is that.
I also don't get why would you ever need to adjust window width per website? Say, for me the best line length is max 133 at font size 14 and char spacing X. So if I set my window at about this width so that no more text fits, and it reflows, I'm set. Yes, there are sites that disrespect user font size preferences, which would require zooming (or reading plugin), but this true regardless of the max width preferences, and also they would disrespect whatever the width guidelines are.
So IIUC, for you, evidence is weak and we don't actually know either way and we haven't explored alternatives to limiting light length enough that could help comprehension and speed. Fair enough.
> I also don't get why would you ever need to adjust window width per website?
Because websites define different margins, may have a side menu, or two... There might be stuff around the content that take width. And responsive designs may worsen the effect.
You can always read these posts in Reader Mode on your browser and adjust the text size, paragraph width etc. Even on this site, I have to magnify the display to 130% to view comfortably on a hiDPI screen.
The reason why most blogs are narrow is because blog templates are still from the 2000s in terms of how they show content on desktop . Traditionally the right hand third of the screen would display ads or a blog roll, while the left hand third would show the site's navigation .
Over time this changed, but what stayed the same was showing body text in the central third columns.
1) Because sometimes you are right about something and it's nice to show that you were thinking about it ahead of time. An example would be my prediction in 2023 (which is fairly late in all fairness,) that one of the most important improvements in AI in the near future would be the ability to interact with any GUI as if it was an API.
2) Because often you are wrong about something and having it written down and in public will keep you honest about hindsight bias. An example would be predictions I made about lasting impacts of the pandemic on how we approach healthcare and minimizing respiratory infections in general.
Whether that's hobbies, a home renovation project, a public presentation you did at work, a life event (wedding, kids, etc).
I feel the title should be more about 'how to write better blog posts over time' or something. I was expecting to hear advice more about creating or maintaining a blog itself, which is more than just writing (in my mind) -
The site pulls a bitdefender warning about malicious something or another, not sure what the trigger is.
I've commented elsewhere about blog software since on HN this inevitably comes up on similar threads.
I've got some stuff about modern wordpress blog design and such on this page: https://steveiscritical.com/learn-websites-and-coding/
Might have to make a more well rounded wordpress blog stuff that includes backups, securing and hosting considerations then submit a more complete article at some point.
- You get one free site per GitHub account
- I forked https://github.com/academicpages/academicpages.github.io and updated the blog posts to link to published Notion pages.
- This gives you a great WYSIWYG editor and all the nice features that Notion provides (collab edit, embeds, tables, LaTeX, view/edit anywhere).
- Notion handles the SEO too.
- See https://mikeshen.github.io/year-archive/ and an example post - https://mikeshen.notion.site/Deepnote-Embeds-Share-Your-Resu....
1. https://csswizardry.com/archive/
2. A fact I am embarrassed to admit in present company. I still can’t remember how it happened.
Accept that you will work on a cool personal project for a long time (months, at least), and will then write a blog post showcasing that work - and no one will care. Over. And. Over. Again. Very humbling, but a good exercise in evaluating your original motive to work on the project.
As such, if you actually want to produce text worth reading, you'll need to spend (a lot of) time editing and critiquing each draft you write.
I also think the world abounds with supposedly quirky people, who think that that is enough. It is (almost certainly) not enough.
Unless you are incredibly unusual and individualistic, your quirckyness is not sufficient to make your work worth reading. The problem is that there are lots of interesting, unusual and deep people, and some of them also spend a huge amount of time and effort on the craft of writing.
• How do you guys feel about your thoughts, ideas and creativity being scraped and secretly added to the training corpus of AI models?
Letting a AI learn from your content devalues it as they can regurgitate it for a few dollars a month.
That doesn’t mean you can’t work to change the world, and in fact in order to do so you first need to see how it actually is.
But it does also means you need to decide which battles to fight.
Serious shortcoming indeed.
I know it is stupid but I care.
These companies are violating that agreement for their own gain, and respond to any pushback by invoking bogus legal arguments and ridiculous machine anthropomorphism. Brute force statistical analysis is not "learning."
I'm also worried that if I wanted to find an audience the audience would first have to navigate the AI search hall of mirrors.
as soon as i do something slightly spammy or excessive SEO-wise i get dinged for it. within 1-2 days. then i re-evalute what i did and revert it or change the strategy slightly.
those blogspam houses are just doing a few things right that are clearly being given too much weight.
and finding an audience was never easy. but social, newsletter, building free tools that link to your blog are some ways to do different than just SEO.
They both get traffic. Neither site is optimised for SEO. Looking at analytics (Matomo on my blog, Google Analytics for the bot), the stuff I write myself gets "real" visitors. People will stumble on a project, and you can see them navigate to the next article, and the next article.
The content from the bot isn't engaging - people might stumble on it, then bounce.
Absent search, interesting content gets shared. Blog spam does not.
I don't worry too much about my content being used to train LLMs - I don't feel it takes away from me, it a bit feels like content getting indexed by a search engine. And I see value, for me, in just creating it.
- `add_action(‘template_redirect’, ‘my_scramble_func’)`
- or `add_filter(‘the_content’, ‘my_scramble_func’)`
Wordpress is PHP so you get the power of doing whatever as the request loads.
`Template_redirect` would allow you to redirect to a specific template/page.
`the_content` would allow you to return different content for that path.
Then you just need to check the user-agent/IP. I’m sure there’s a PHP library you could install with composer to do this.
What would be fun in my opinion is leaving the metadata the same but replacing only the meat. For example article titled “How to Update Your Blog” would still have the title and headings, but then return “<h2>Step one: open your blog editor</h2><p>hajfineidif hahahe lorem ipsum…</p>” for the paragraphs.
I have a blog that admittedly does not get much traffic, and I have started to add random words after each paragraph. The background is white, and these words are white, so they are invisible to most people. Maybe it is not that effective, but it makes me feel better.
On the other side, for many of my thoughts/ideas/creativity, I find it is often a question of, do I die alone with it in my head or put it out there, maybe for someone to find, use or be inspired by? If it's a question of one or the other, I think perhaps I'd rather let it go free so the value isn't lost (assuming, that what I have to write/say/share is of some value).
I have seen that mechanism be exploited.. sometimes you put something out there, someone uses it, and there is zero attribution/citation. For myself, I prefer to try (most of the time) to treat non-fictional writing in the traditional sense of an academic paper and include citations whenever they are useful and I can - even for non-academic purposes - for the purposes of giving credit to others. It's nice to share credit and show how things connect, IMHO.
FWIW, I like some of the inspiration behind Creative Commons, but even old school citations (MLA/ALA or heck, even just this kind of inline citation style [1] that is common on HN and other sites is better than zero).
There are citation generators online also [2] - for formal purposes/writing/papers you must get these formats right, but for informal/casual use, I think simple citation is better than no citation.
[1] simple citation style, or even just paste the URL - at least you're giving credit.
[2] example, https://zbib.org/
I applied the same method when writting my long defunct blog. Nobody was really interested in my opinions, but my experience both positive and negative was somehow valuable.
Mataroa is another option similar to bear.
No point hosting a blog or anything of the sort, unnecessary effort in the beginning when you don't know if you'll continue.
I recently rebuilt my own blog over the course of a month or so after a decade hiatus (and wrote about it, of course[1]).
I enjoyed the process of working through all my little ideas and deploying it — and continuing to tinker on it.
1: https://duggan.ie/posts/ignoring-good-advice-and-building-my...
Many people don't like using the platforms.
They're obviously parasites.
It's not very complicated to do yourself.
Obsidian Publish, Micro.Blog, Ghost, BearBlog, HackMD, etc. are all great places to start, with less "baggage" associated with them. I know a lot of people write on Medium as well, but the account requirements that pop-up randomly are super annoying.
Basically you can ssh/scp/rsync static pages there.
The only personal thing you provide is an SSH public key pair.
I started reading more and more books and have seen a lot of improvements in writing. I am not inspired by anyone's writing style but yes I am writing for the sake of my own curiosity and fun. I did follow some of your ideas on how you write though xD
my blog: https://mangeshm.xyz/ some recent articles about my thinking and some stories I've written.