10
points
jbreckmckye
Joined 3,237 karma
- > David was successfully treated in this way: his motivation levels went back up after taking a drug that stimulates dopamine receptors in the brain. Because of this, he was able to get a new job, become independent and even find a partner
I've always wanted to hear the stories of patients like David. What happens when you can tell your family and friends - I'm not a screwup, this has a physical cause? How do people interact with you who've previously only seen you in your decline?
A long time ago I read an account by a financial journalist who'd had some kind of brain injury and temporary lost his faculties. It damaged his professional relationships and the implication was, most of his former colleagues remained unduly sceptical - even after years of normal functioning.
- You are right. An observation, even before the rise of LLMs, was that computers were becoming more human-like and humans were acting more like machines
The sculpting force of algorithms is bite sized zingers, hot takes, ragebait, and playing to the analytics
- As explained in other comments, I only used Gist because publishing to my own site failed. So GH is my redundancy :-)
- I published it as a Gist because my own blog deployment pipeline was in a non-functioning state.
- Eh, I'm not convinced.
1. Culturally, using `unwrap` is an omerta to Rust developers in the same way `panic` is an omerta to Go devs;
2. In the Rust projects I've seen there is usually a linter rule forbidding `unwrap` so you can't use it in production
- But I could screw it up in Go, if I made the same assumptions
Ultimately I don't think language design can be the sole line of defence against system failures; it can only guide developers to think about error casesfvs, err := features.AppendWithNames(..) if err != nil { // this will NEVER break panic(err) } - > If you figure out how to do this completely, please contact me—I must know!
I think you want to use a TypeScript compiler extension / ts-patch
This is a bit difficult as it's not very well documented, but take a look at the examples in https://github.com/nonara/ts-patch
Essentially, you add a preprocessing stage to the compiler that can either enforce rules or alter the code
It could quietly transform all object like types into having read-only semantics. This would then make any mutation error out, with a message like you were attempting to violate field properties.
You would need to decide what to do about Proxies though. Maybe you just tolerate that as an escape hatch (like eval or calling plain JS)
Could be a fun project!
- Why not?
It's better to have diverse, imperfect infrastructure, than one form of infra that goes down with devastating results.
I'm being semi-flippant but people do need to cope with an internet that is less than 100% reliable. As the youth like to say, you need to touch grass
Being less flippant: an economy that is completely reliant on the internet is one vulnerable to cyberattacks, malware, catastrophic hardware loss
It also protects us from the malfeasance or incompetence of actors like Google (who are great stewards of internet infrastructure... until it's no longer in their interests)
- I learned Go this year, and this assertion just... isn't true? There are a bunch of subtleties and footguns, especially with concurrency.
C++ is a basket case, it's not really a fair comparison.
- For a long time, yt-dlp worked completely with Python. They implemented a lightweight JavaScript interpreter that could run basic scripts. But as the runtime requirements became more sophisticated it struggled to scale
- I'm nowhere near as senior as you (12 years). But I've reflected on this a lot recently. I love the technology and feeling like I'm building things, but staying hands on will always limit your scope. Taking the management path improves your scope but the work is - well - just less fun than programming.
My answer right now is to try and build more things myself. Small apps, CLIs, retro games. Not libraries or much OSS stuff so much as actual "products" that give me creative control and concrete outputs. It's hard to make the time though.
Outside of my career I'm also trying to cultivate other works, like my YouTube channel and my writing. Creating a video that 250k people enjoy is at least as meaningful to me as crushing my OKRs
We are all mortal beings: there will inevitably be more things to do than time to do it, and it's easier to ruminate on options than commit to something that feels "suboptimal".
To give an example of that. I spent a lot of time wondering if I should have gone into academia / literary criticism rather than tech, because of a vague sense that because I was very good at something, that's where I should put my efforts. Is that sound reasoning though? I probably achieved more "value" for society working as a programmer, than writing about Chaucer.
So to summarise it may be a choice of making peace with the lower scope / autonomy of hands on work, and finding that satisfaction outside work. If that suggestion makes your soul revolt, though, it may be you have to compromise and take the managerial path
- I have observed it too, it is heavily inspired by economics and mathematics.
Saying "it's better to complete something imperfect than spend forever polishing" - dull, trite, anyone knows that. Saying "effort is a utility curve function that must be clamped to achieve meta-optimisation" - now that sounds clever
If I was going to be uncharitable, I think there is are corners of the internet where people write straightforward things dressed it up in technical language to launder it as somehow academic and data driven.
And you're right, it does show up in the worse parts of the EA / rationalist community.
(This writing style, at its worst, allows people to say things like "I don't want my tax spent on teaching poor kids to read" but without looking like complete psychopaths - "aggregate outcomes in standardised literacy programmes lag behind individualised tutorials")
That's not what the blog post here is doing, but it is definitely bad language use that is doing more work to obscure ideas than illuminate them
I'm also sketching out a concept for a YouTube video explaining how retro game upscaling actually works on a technical level.