f311a

9 ms·

151.
by f311a 9mo ago
Why would you compare them, bun is a complex tech used by real projects
152.
by f311a 9mo ago
Just a free plan on Cloudflare (CF in the comment above also means CloudFlare). Zero maintenance. I push to Github and CF deploys static pages to CDN.
153.
by f311a 9mo ago
How it works: * CF worker on a subdomain that handles POST requests. Basically, a JS function that handles incoming requests. * It stores comments in CF KV and sends me a copy to telegram * All I need to do is copy it to Markdown (can be au
154.
by f311a 9mo ago
My blog is fully static and I have a 50-line CF worker script that sends comments to me which I import directly to markdown of a blog post. There are ways to do comments without embedding.
155.
by f311a 9mo ago
They don't understand codebases. They are trained on other code, ignore how your codebase is structured, and lack knowledge of it. To do so, you would need to feed the whole codebase every time you ask it for something, with extensive
156.
by f311a 9mo ago
Why are so many people so obsessed with feeding as many prompts/data as possible to LLMs and generating millions of lines of code? What are you gonna do with the results that are usually slop?
157.
by f311a 9mo ago
> Wine/Proton have made far more in-roads in being able to run Windows Yeah, they can even run modern games, which ReactOS can't. It can't even run on modern hardware properly. It's a nice project, though. Good progre
158.
by f311a 9mo ago
For ML/AI/Comp sci articles, providing reproducible code is a great option. Basically, PoC or GTFO.
159.
by f311a 9mo ago
It worked well for me when people were stealing my articles, pretending they wrote them. One tweet or mention in Linkedin and the article is gone.
160.
by f311a 9mo ago
How would it work if LLMs provide incorrect reports in the first place? Have a look at the actual HackerOne reports and their comments. The problem is the complete stupidity of people. They use LLMs to convince the author of the curl that h
161.
by f311a 9mo ago
Imagine the amount of slop PRs if it was open source. They don’t want to taste their own medicine
162.
by f311a 9mo ago
It’s not like you needed LLMs for quickjs which already had known and unpatched problems. It’s a toy project. It would be cool to see exploits for something like curl.
163.
by f311a 9mo ago
Just look at the code quality produced by these loops. That's all you need to know about it. It's complete garbage, and since it runs in a loop, the amount of garbage multiplies over time.
164.
by f311a 9mo ago
> The use of em-dashes, which on most keyboard require a special key-combination that most people don’t know Most people probably don't know, but I think on HN at least half of the users know how to do it. It sucks to do this on Win
165.
by f311a 9mo ago
Yeah, it's - Servo's HTML parser - Servo's CSS parser - QuickJS for JS - selectors for CSS selector matching - resvg for SVG rendering - egui, wgpu, and tiny-skia for rendering - tungstenite for WebSocket support And all of t
166.
String handling in ClickHouse(rushter.com)
3 points by f311a 9mo ago|1 comments
167.
by f311a 9mo ago
I think you overestimate how much vape companies care about safety. When there is no liquid left, you just vape smoke from burning cotton (it tastes like burned plastic) on half of these devices. There are checks for this, but they are not
168.
by f311a 9mo ago
Yeah, especially when a CSS library makes $1M a year. I guess they have no incentive to improve funding.
169.
by f311a 9mo ago
Temporary splits to run code close to the editor, tmux popups (they open lazygit or file manager on top of the current tab). Debug runs, where I want to preserve their output for some time, so I keep them in separate tabs.
170.
by f311a 9mo ago
Battery life and performance on MacOS
171.
by f311a 9mo ago
Guido gave up the idea of keeping CPython interpreter simple. It's now a complex beast with JIT and a lot of small optimizations.
172.
by f311a 9mo ago
It’s still pretty profitable, more than $100k a month
173.
by f311a 9mo ago
With TC of $250k. There is a lot of room for optimization.
174.
Recent Optimizations in Python's Reference Counting(rushter.com)
25 points by f311a 9mo ago|7 comments
175.
by f311a 9mo ago
There are new technologies, but if you look at the most viewed questions, they will be about Python, JS, Java, C, and C++ without libraries.
176.
by f311a 9mo ago
People overestimate the impact of toxicity on number of monthly questions. The initial growth was due to missing answers. After some time there is a saturation point where all basic questions are already answered and can be found via Google
177.
by f311a 10mo ago
> Strings >The rule of thumb for strings is the core string object takes 41 bytes. Each additional character is 1 byte. That's misleading. There are three types of strings in Python (1, 2 and 4 bytes per charact
178.
by f311a 10mo ago
https://github.com/golang/go/blob/7ecb1f36acab7b48d77991d58d... Here linker relies on the current memory layout. Changing it requires updating linker code as well.
179.
by f311a 10mo ago
In Go, hashmap implementation is pretty low level. Even linker carries some parts of its implementation details.
180.
Hash tables in Go and advantage of self-hosted compilers(rushter.com)
2 points by f311a 10mo ago|0 comments

More ›