Hi! I'm Eduardo Lávaque, and this is my blog.
Writing software for 11+ years. Eschew hype; focus on performance.
Living in Switzerland since 2017.
Hi! I'm Eduardo Lávaque, and this is my blog.
Writing software for 11+ years. Eschew hype; focus on performance.
Living in Switzerland since 2017.
I had the pleasure of discovering https://lilura1.blogspot.com via a post on X/Twitter.
This woman has been blogging for 12 years. Might even call it journalism, even if it isn't strictly "news". And her blog is very detailed, one can tell lots of love has been put into it.
It's hosted on Blogger. I'm only concerned that Google is very well known for shutting down services with no warning, even if they are in wide use, so I hope a backup of her whole website has been made somewhere.
I read several articles, here are a few I'd like to highlight:
I realized multiple things while reading through her posts.
The vocabulary on the internet has certainly been dumbed down, for lack of a better word. Reading her blog I read multiple words that I don't recall reading in a while, or have only now heard for the first time. Not regular English vocabulary, but rather specialized vocabulary. Like "save-scumming", I never heard that term, but as a kid playing console emulators, that's how I played games by default, and I never had a word for it as an adult.
In any case her writing is, on average, of a higher caliber than most things I read on the internet. Curiously, I never had to wonder whether it was written by an LLM or not, her writing style never suggested as much. But when reading books and other materials pre-2019 sometimes they still sound LLM-ish, and now I wonder if it's just because modern writing styles are rather dumbed down, compared to older writing styles. When I read materials from the 70s and earlier, it sounds much different and I, almost, never get a hint that it sounds like an LLM.
Her writing style shines through wonderfully. And I think that's one of the best compliments a reader can give about writing style.
Perhaps the current writing style is more homogenized simply becase people read less now than before.
But let's move onto the content itself, which also made me reflect.
They made me reflect a lot, on my own taste, and what should go into me producing content for others to consume.
I find her to be very much a purist. She doesn't appreciate things getting watered down. She is a real Geek, as far as I can tell.
She is also a gatekeeper. She keeps the gate. She only allows the real ones in, so to speak.
And I used to be much more in that class of people, generally speaking, although perhaps never on her level. At some point in the last 10 years I've mellowed out, as I was distracted by other things that weren't my craft. I've been a casual geek, let's face it.
But recently I've been thinking about how it would be more positive to be more of a geek, and more of a gatekeeper.
I have interacted the most with the web development subculture. I interacted shortly, around 2013-2015, with the hacker subculture through the Arch Linux forums and the nixers.net community IRC.
Until just now, as I'm writing this even, I was going to talk about how the programming subculture, particularly web development subculture, is not gatekeep-y at all, and how that's a problem. But while trying to find a particular post on X/Twitter, which I couldn't find, which speaks poorly of their search, I realized lots of people are complaining that there is a lot of gatekeeping going on. About music, about art, about programming.
And I had the realization, wow, with AI coming into the scene now, the MOPs and the sociopaths have really gotten in here and are complaining that people are holding some sort of standard. They would like AI slop to be an acceptable standard.
The web development culture considers itself incredibly open, and not gatekeep-y at all. If you make something cool (note I said cool, not good) you will get lots of attention. The general programming subculture, somewhat unlike the hacker subculture, is primarily a fad-driven subculture, what is "in" now is what gets attention and everybody does it, not what is good or robust or rigorously tested, or even logical for the given problem. Those things are "boring", and they're a bit too engineer-y for the culture, in factual reality.
Just yesterday I read about how the async/await fails concurrency, and yet it won the culture war, probably in no small part thanks to ES2017 introducing it, making Promises the defacto standard for async APIs. In addition that guy produced an HTTP 1.1 framework that can be tested with deterministic simulation testing, why is that not the standard of software engineering?
I thought we were engineers, and yet continuously our solutions are subpar when subjected to actual engineering. It's because there is in fact no engineering discipline, when it comes to programming. And I know Robert C. Martin likes to use that word, but he is very poorly disciplined himself even on fundamental factors like performance. (For those that are missing the point about Martin, having a framerate of "high 20s" with "hundreds of objects on the screen" is an abysmally poor showing, even Doom could maintain that in 1993, with hardware that was thousands of times weaker than modern hardware. The fact that Martin thinks that's acceptable or at all impressive tells us that he cares little about performance, which is far from a real craftsman.)
All this lack of proper engineering is leading to the Node.js ecosystem, and other ecosystems, being pwn'd on a weekly basis at this point, affecting millions of websites and servers. Some of them are technical in nature, but some are of a social engineering nature, where the developer is pressured into downloading an unknown binary. Are modern developers so AI-slop-rot-brained that they fall for the oldest tricks in the book? Is modern software so volatile that it asking you to download suspicious update doesn't ring massive bells in your head? The answer to both of those is yes.
But perhaps more likely, the reason those very smart people fell for social engineering, a product that copied Microsoft Teams so well that they thought it was the real Microsoft Teams, and downloaded and executed a binary from it, the most likely primary reason that happened is because Microsoft's products are so shit, that a knockoff can resemble it. Can you imagine the quality of your software being so shit that somebody can make a knockoff that is believable? It lags, it loads slowly, it constantly asks for you to download and install things (that you probably don't want to)...
I never install Teams on my personal machine, for the record, I always use the web version.
Anyway, I think I should be more gatekeep-y. I should, for the better good of my craft, focus on becoming really good, an actual geek, and invite others to do the same, but not allow them to introduce bad things (the opposite of good things) as the standard, and stake my career on that. I've already done that at my last job on some level, didn't end well but that wasn't the only reason why.
I mean that with no respect (lol).
One can read in the 1950s people complaining about how newspapers are full of, for lack of a better word, bullshit.
Yet, people always fall for the Gell-Mann Amnesia Effect.
Briefly stated, the Gell-Mann Amnesia effect is as follows. You open the newspaper to an article on some subject you know well. In Murray's case, physics. In mine, show business. You read the article and see the journalist has absolutely no understanding of either the facts or the issues. Often, the article is so wrong it actually presents the story backward—reversing cause and effect. I call these the "wet streets cause rain" stories. Paper's full of them.
In any case, you read with exasperation or amusement the multiple errors in a story, and then turn the page to national or international affairs, and read as if the rest of the newspaper was somehow more accurate about Palestine than the baloney you just read. You turn the page, and forget what you know. ― Michael Crichton
I also find myself looking at newspapers hoping to get accurate information. But within paragraphs one can spot logical mistakes.
Perhaps I should simply stop reading the news entirely...
In any case, Lilura1 complains about the journalism from last century about computer games. How unserious they were, in fact, they were not computer game geeks.
In recent times we had the GamerGate fiasco. Which, to my best knowledge, was a big drama, which originated by a popular new game which was "hard", to be rated extremely low by a games magazine. Somehow some gameplay of the journalist who published the review was leaked, where they demonstrated they not only were not a gamer, they were nothing close to being adept or experienced at games, struggling even with the concept of a double-jump.
I don't care about the drama itself, it just taught me that game journalism is simply untrustworthy going forward, when it was happening.
In the web develpoment subculture, we have as an example, and I don't mean to pick on him but it's just one of the few I'm aware of, Theo - t3.gg. If 536k subscribers doesn't speak for itself, and given his videos get 100k+ views on average I don't think that sub count is fake, this guy is popular in the web development space.
In a recent video titled "We all fell for it..." which is about vibe coding, where he's reading an article that is critical of vibe coding, he admits within the first couple of minutes that "he barely looks at code anymore".
Now, he's a Sociopath as far as I'm concerned in this context. And he is a thought leader in the space.
I was never fooled, and all of these negative side-effects of vibe coding I have predicted, and I have documented. I stopped using AI in my primary workflows the moment I felt like I was getting dumber as a programmer. Now I only use it when I explicitly do not care about the code quality, and I'm not hoping to learn anything from the project I'm working on, or when the next priority after "it's working", is that it is done fast (note this is not for long term projects, obviously). Not to mention costs, ethics, morals, all of those good topics involved in using LLMs for anything.
There is a spectrum of programmers, from "I love programming" to "it's my job" to "I hate programming, it's a means to an end". I'm on the left side, I guess Theo is middle to right side of that, because those were the only people that jumped on the vibe coding train.
But that's just to say, he's a journalist on YouTube, adding one data point to the point I'm trying to make. He's definitely not a programmer on YouTube, because I never see him programming.
That being said, of course there are good journalists out there. For some good journalism on AI, there is the Wading Through AI podcast. or Ed Zitron's work.
Now, I knew that. While the scientific method itself is great, which science gets funded is always biased, even when it is publicly funded.
But you know, I thought scientists at least made sure their alibi was airtight you know?
arXiv, in a recent tweet, made it clear that vibe-written research papers are against their policy. A college professor chipped in asking how can you expect people to read every citation. Maybe it's just me, but that was my genuine expectation. How can you be willing to put your name on a paper that you don't understand fully? Maybe that's just me though, 'cause the bugs come to hunt me later while I sleep anyway, so I better be damn sure I know what it is.
This is a side-point, but with the above points in mind about journalism, I actually find it funny that Wikipedia's policy only accepts information from certain "notable" or "verifiable" sources, like peer-reviewed (😉) scientific papers or serious journalistic (😉) sources.
This came to my attention because the Odin programming language, which I find very enjoyable to write in, and I think is a cool language, has a history with Wikipedia where Wikipedia mods don't want it to have a page.
2026-05-20 Edit: I read this very thorough review of that situation between Odin/GingerBill and Wikipedia, I share it because I think it's worth your time to get further context on this matter: Odin, Wikipedia and Engagement Farming
I think there is definitely an under-current in our culture, that being average is OK. Nowadays perhaps it's somewhat of an over-current.
Somehow the underperformers, the ones that have some kind of anxiety about their performance, or the highly empathetic kind of people, got a hold of the zeitgeist of our modern culture.
With a bit of thought, I can only come to a couple conclusions about that.
The first is that, well, the inadequate got a hold of the zeigeist, so they try to make their inadequacy normal and acceptable.
The second might be that, with full good intentions, the high empathy types are doing a similar thing, for the inadequate.
The third might just be that because there is no "higher" level to aspire to, as previous cultures might've had by trying to make things "worthy of the gods", or "for God", so to speak, the standards dropped, until it meets the average.
A fourth interpretation might be that due to the rise of Bullshit Jobs, as proposed by David Graeber, people have to make their bullshit jobs more acceptable to themselves, and to society, lest the whole thing falls apart.
But why do I think about it like this? I mean culture still appreciates excellence, right? I don't know. The mainstream music is excellently produced, but has little to offer for the soul. Modern programming, with AI and all that, has little to offer for the soul or for the stakeholders (no matter how much the bubble likes to claim it has much to offer, it is simply more expensive than the value it hopes to produce, even if it does produce some value). The modern internet is is, as I spoke previously on this post, very homogenized, everything is on a limited amount of platforms, few people have their own blog where they frequently write their thoughts down.
Excellence requires deep thought, deep work, deep human soul invested. But modern culture wants all of that to happen without any of that.
If I want excellence, I need to express high standards and maintain them.
But how do I have high standards to express them, in the first place? Well first I need to be dedicated. And I need to genuinely spend time doing hard things, for fun! I need to stop spending time on things that aren't valuable, like scrolling social media (I genuinely don't do it much, but I do it more than zero, which is not ideal, zero is ideal).
And I need to somehow make an island for myself, where what I've made, speaks for itself.
But things speaking for themselves is not enough. Perhaps I need to also promote my ideas more widely. Perhaps I should open up other channels or avenues of expression, to promote my ideas. Otherwise even if I try to raise the standard, there is nobody there to listen.
