Welcome to [cybernetic] prison – part dukes
Monopolising the Web by corroding the language used to create it to the point of meaninglessness.
In my first post on cybernetic prison I talked a lot about how people lay themselves victim to what is essentially a mind prison. For all its horror, in that framing, people are as free to exit the prison as they let themselves believe. It becomes a self-resolving value prop type of thing.
But let’s say you know better about all this. You know that Twitter isn’t showing you the posts in the order they were made, and you’re gonna sit down and do something for yourself about it. No more complaining, no whining and begging to unaccountable billionaires to do the right thing. How hard could it be, after all? The content is already public.
Well, that’s inconvenient. How do you fake having JavaScript, then? Honestly, you can’t. Sure, you could cobble something together that’s Turing complete and has virtually zero language features, but that’s a losing game of cat-and-mouse with websites that don’t want you scraping their data.
Part one explained how many people in what we call ‘tech culture’ don’t even realise how siloed they truly are. Now we learn that the deck isn’t just stacked against us softly, but also stacked hard against us. All of the code has been written from day one to pay lip service to the open web. Tech has taken to pissing on Aaron’s grave.
The best outcome you can get in this environment is something like Brave – which, as much as I must count our blessings for having, is little more than a stasis in technology, as opposed to an active degeneration into ad-tech hell that would be the norm now if Google et al had their way. Brave still makes the same abortive gambles the rest of tech does: cryptocurrency-backed optimism, labour-intensive development practises, you know the works. The very things that I would argue led us into this mess in the first place.
The fundamental issue with how we make software is the hard culprit, while the soft siren is all of the baseless optimism that profit dotcom-era hucksters (you know the ones). They blow smoke up everyone’s asses, and the end result is software that is so ridiculously and warrantlessly complex that nobody understands it, requiring a revolving team of brainiacs on call 24/7 to deal with it. It’s pretty convenient if you’re a billionaire hedging on the zero sum game of Billy G’s so-called “IQ wars” where you’re ‘collecting’ and owning geniuses like fucking Yu-Gi-Oh! cards.
Said geniuses also did this to themselves, and I’m categorically tired of it. Every time you have a technical discussion about some software at hand, and you cannot make a sound argument for why you included some code or why you avoided the more obvious solution, you are wrong.
I never want to hear an argument in favour of some added complexity that does not have an immediate, obvious benefit to the architecture of the project at hand. Everything that doesn’t plug into an active cog needs to be justified by its design. Everything else is hogwash.
This does not mean I treat such a maxim absolutely, either. It just means that you cannot treat shit like ‘big endianness’ like it is of paramount importance, or really even worth having a meeting about. It’s not.
It’s time to stop being insane and having insane conclusions like “we can’t get rid of that because somebody somewhere 3,000 years from now might use it”, or “we won’t add it until the last second because we can’t name somebody who is going to use it today.” Jesus fucking Christ, people. Get a grip!
Look. Hinterlib 2 has some code for big endianness. It’s there because it’s not a bother, although I will admit I don’t foresee a big utility in big endian machines going forward. I’m not going to pretend to you this is objectively wrong or right, because my whole point about this is, that doesn’t fucking matter. To be completely honest, I put it there because I don’t feel confident enough about eschewing big endian machines writ large with this software that is highly heterogeneous in its targeting. I’ll tell you what I didn’t do: I didn’t overthink it.
Quit being so autistic that you have to neurotically insist on some rational quorum about everything, because that is precisely how you become unhinged and get lost inside the rectums of people like Marc Andreessen and Elon Musk, which is hurting you way more than any potentiality about endianness ever will.
And hey, if all of this seems like splitting hairs to you, because the kind of work you’re doing is mundane and you take support libraries and computer architecture for granted, you’re not an engineer and I’m not talking to you. Since there are tons of code monkeys out there who think they’re architects or engineers because they wrote some Rust code or C++ for a stint, I have to put that out there.
There are a lot of other specific angles I’ve talked about the ruinous state of complexity on this weblog. I haven’t spoken nearly as much about the social rot powering it, although I have. Here’s some further reading to round this post out:
Anyways, thanks for reading.