Someone Made a Fake Wikipedia for AI Hallucinations, and It’s Basically the Internet Eating Itself - VICESomeone Made a Fake Wikipedia for AI Hallucinations, and It’s Basically the Internet Eating Itself VICE
Show HN: Snitchmd – Cloudflare-protected URLs into clean Markdown via DockerShmauthor here. Built this for myself, putting it out in case it's useful.Needed any URL as clean Markdown for LLM context — including Cloudflare/anti-bot sites. curl gets HTTP 403 on those, raw HTML is 80%+ nav noise eating context, paid SaaS (Firecrawl, Jina) wasn't an option for me.It's a Docker wrapper around two existing OSS tools — CloakBrowser (stealth Chromium that passes Cloudflare) and rs-trafilatura (HTML → Markdown). No new scraper, just glue. Runs locally, my URLs stay on my boxToken reduction (raw curl HTML vs snitchmd, tiktoken cl100k_base):- cloudflare.com/learning/bots — curl: HTTP 403 → snitchmd: 0.8k- docs.docker.com/engine/install — 187k → 0.9k- en.wikipedia.org/wiki/LLM — 222.7k → 29.7kHeads up: passes Cloudflare, can't solve "click traffic lights" captchas (reCAPTCHA v2, hCaptcha)MIT. Happy to answer questions
Show HN: Vectree – Learn complex concepts through AI-generated interactive SVGsHi HN,I launched Vectree about 3 weeks ago as an after work side project. It's basically a "visual Wikipedia" where you explore concepts through interactive, zoomable SVGs.Link: https://vectree.ioWhy I built it: Whenever I encountered a new complex concept, I just wanted a quick visual way to understand the basics. I started manually prompting LLMs to explain things to me by generating SVG schematics. It became so useful for me that I decided to automate the process and turn it into a web app.How it works: You can browse the public graph of concepts totally for free. You click on different parts of an SVG diagram (nodes) to drill down into sub-concepts.Bring Your Own Key (Private Lab): While browsing is free, I highly encourage creating an account and plugging in your own paid Gemini API key. This unlocks a "Private Lab" where you can: - Architect your own private concepts from scratch - Regenerate existing concepts (most concepts I generated with "flash" Gemini model) - Publish your p
Show HN: I tried beating the stock market with Gemini 2.5 for 2025Hi,I ran a fun experiment trying to beat the danish C25 index with just Gemini 2.5 in simulation.I fed Gemini 2.5 with stock data, danish news articles, Wikipedia & Google trends top lists and much more like Eurostat data to see if I could get Gemini to beat the danish C25 stock index. (Novo Nordisk, Carlsberg, Mærsk etc.)It could only see the current days data and backwards and was prompted to not take in forward data from its training data. It was prompted to follow these strategies among others::- Take a longer term perspective - Be conservative with position sizes - Only trade when seeing a clear opportunityIt did pretty well, but ended up doing a couple bad moves selling at bad times - it almost feels like it panicked.Data is from (01-01-2025 to 11-12-2025).-Any ideas on improving it? I think next step is using more a Chain of Thought process and multiple personas to evaluate each investment.
Humanties Last War# Humanties last warFor many generations, knowledge has been a tool to ensure that humankind is able to produce a value of its own. This power is on the edge of shifting towards a few companies that rely on its value closed-source. Leaving the integrity of knowledge solely in their hands. Scientia potentia est has become more distinct and clearer, being centric around a few.> Phase 1 - Mining The real value is not here yet. The costs are immersive, and cloning every source of data is pain. How good that there are open source projects, that provide the data for free (https://index.commoncrawl.org/) a few Petabytes should be a good starting point.> Phase 2 - Creating dependency, for less than the value Let’s start with a business 101 here, Predator pricing (https://en.wikipedia.org/wiki/Predatory_pricing) should be a good starting point. We are lazy, as we just collected 6.4 PB of data we just leave the rest of the collection to the users that pay that we can use their data (https://help
Show HN: Resrap – A Parser but in ReverseI built Resrap, a Go package that takes a grammar in ABNF format and generates infinitely long sequences of syntactically correct code...either completely randomly or with seeds for a deterministic generation.ABNF is a modified version of EBNF(https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_f...) I made for this project, basically means you can specify when generating code 20% of lines will be if{} blocks and 50% will be while{} blocks which allows for more natural code generation, plus support for infinite generation of code.It’s very fast...it generated ~40 million tokens of C syntax in about 26 seconds on my laptop and supports multithreading which actually saw boosts in performance since its very easy to parallelize.I originally made this for a typing-test project (I didn’t want to store code snippets manually), but it turned out to be useful in other contexts too, like: - Stress-testing parsers and linters - Creating non-copyrighted “lorem ipsum” code for tech demos - G
Wikipedia Faces Flood of AI Bots That Are Eating Bandwidth, Raising Costs - PCMagWikipedia Faces Flood of AI Bots That Are Eating Bandwidth, Raising Costs PCMag
Show HN: Mutable.ai Codebase chat that uses a Wiki for RAGHi HN!We wanted to introduce our codebase chat that uses a wiki of your repo to do retrieval instead of traditional vector or keyword RAG.We have been pleasantly surprised by the quality of responses. We think creating a language knowledge base for LLMs to do RAG is the future. The way this works is we first write a Wikipedia style article on your codebase complete with diagrams and citations to your code.We’ve run it on some of the most popular repos. If you want us to run in an open source repo please comment below. You can also run it on private repos and the permissions will follow those of the repository on GitHub. The wikis auto update via a GitHub bot and even show you previews of the new wikis in a diff view.Please check it out and let us know your thoughts.Thank you! OmarPS Here are some examples:https://wiki.mutable.ai/hashicorp/terraformhttps://wiki.mutable.ai/ggerganov/llama.cpphttps://wiki.mutable.ai/ethereum/go-ethereumhttps://wiki.mutable.ai/NVIDIA/TensorRThttps://wiki.m
Show HN: Wikipedia Golf – find the fewest clicks between two random wiki articleHi, I'm Rizqi, and I just made a web-based game, called Wikipedia Golf!What is the game, and how does it work? Simple. First, the game find two random Wikipedia articles--the first article and the destination article--and then you click links from the first article to another... until you find the destination. If you made it to the destination wiki article, then cool! If not, you can also give up :(.You can play this game alone, or with friends, by creating a room. After creating a room, the game will prompt you with a random six-characters code and you can make your friends enter the code to join the game. Mind you, the multiplayer feature is a bit rushed so forgive me if you find some bugs...This game is actually my first finished side-project (for real), so I hope you don't mind some bugs here and there. This game is also hosted on free-tier Fly.io and, additionally, it's my first experience hosting it to public with so many people so forgive me if you had a hard time accessing it.S
Show HN: flashcasts: flashcards in a private podcast feedFrom the /about page:I just wanted to study biochem and spanish while digging holes in my garden.Computers recently learned how to talk like humans. Actually, they've been talking for a while[1], but it only recently became affordable[2].[1] https://en.wikipedia.org/wiki/Speech_synthesis#Electronic_de...[2] https://openai.com/pricingAnd by some kind of nerd magic, the cost is low enough for me to share it freely on the internet! If you create an account, you can create mix-and-match flashcast decks and tweak all the knobs.Knowledge probably shouldn't be a class issue, so this site is committed to a frugly[3] pricing model. And yes, there's a lifetime license for all y'all with subscription fatigue.[3] https://taylor.town/fruglyIf you're curious about the Craigslist vibes, I went all-in on a cheap web[4] stack. I want this site to run on a 2010 potato phone clinging to one measly bar of cell service. Of course it's ugly, but at least it's somewhat accessible and not eating our sunshine[
Show HN: Hovercard.js, simple hovercard utility for web pagesOne of my favourite features of Wikipedia is the hovercard. You can hover over a link and see a preview of the content at the link destination.I made a small JavaScript utility that accepts a JSON object with link information and enables hovercards on a personal website. I am "eating what I cook" per se by running this on my website. You can hover over any link in blog posts on my personal website and see hovercard.js in action.Here's an example: https://jamesg.blog/2023/05/19/airport-pianos/ I also maintain a Python package [1] that makes it easy to retrieve meta information about sites for use with hovercards [1] (The format is title, image, and description, so you can use any tool you want, though!).[1] https://jamesg.blog/2022/10/12/hovercards/
Ask HN: What comes after the “pale blue dot” enlightenment?We are all but smaller-than-insect organisms in a pale blue dot [0].Our entire lifetimes are but small flashes, mere nanoseconds in the scale of the universe.Realizing all this through Science and Philosophy- I am in despair. I don't see the point of any long-term goals anymore. I don't see the point of anything other than immediate fruit-bearing activities like spending time with friends and eating.What to do? How to get past this?Many here has gotten the pale blue dot enlightenment, I assume.What still drives you? How and why do you still have long term goals?Can you share with me your perspective and/or any reading material on Philosophy/something else?(Note: I am in situational depression connected with my current job + AI scare. I hope to get out of the bad job hole in about a year. Does it get better after the situation changes? What to do in the meantime?)[0]: https://en.wikipedia.org/wiki/Pale_Blue_Dot
Tell HN: Vitamin A Linked to Energy LevelsPlease share tips or experiences related to vitamins/supplementsVitamin A is pretty scarce in many diets, yet a 2010 paper shows that sufficient vitamin A produces 30% increase in ATP production. [0] ATP facilitates energy transfer between cells [5]Lots of sources including wikipedia vitamin A page [6] don't seem to recognize the role of Vitamin A in energy transfer between cells. [2] [4] Wikipedia page for ATP does not mention the link with Vitamin A either. [7]There are reports of Vitamin A supplements producing psychological productivity energy boost akin to caffeine. [1]One cup of boiled carrots is sufficient daily intake of vitamin A. [2] Boiling the carrots increases vitamin A absorption due to breaking down the cellulose in carrots. [2]Other sources include cooked tuna, cooked squash or pumpkin, cooked sweet potato, cooked spinach/kale/collard/chard/bok choy, or beef/fish liver. [3] If you aren't eating a good amount of these foods every day you are likely deficient.VitaminA Vit
Ask HN: Is there a website which has currated truths, but not WikipediaI'm looking for a resource which I could consult for various things throughout life - something akin to, but not limited to -1) how much vitamin d should I take and are there useful combinations, i.e. D3+K2?2) which limb streches should I do to get most benefit as a beginner; what about as an intermediate?3) which screws should I use when drilling a specific wall type?4) recommended daily dosages of protein, carb, fat based on activity level?5) common active ingredients one should look for when treating common ailments?and so on. These and similar questions are not rocket science. Guidelines, while may be shifting with advancements, have been established for decades, yet anything outside of a simple query usually directs me to some content farm, ad farm or similar. Even most books published within this decade are guilty of this, inflating the word count for seemingly no reason.Wikipedia is often the wrong resource for the job, because it's aims are academic, not practical - yes you may
Ask HN: Models for Collaborative Websites?I am playing with the idea of creating a niche information website, where I want users to be able to contribute and improve the content.I am looking at the Wikipedia and StackOverflow models to get inspiration on how to foster quality contributions. Is there any other good example of a website where good quality content emerges from user contributions?Using a pure Wiki or Stack-exchange model is not an option, because:- I want to be able to impose a schema for the different items that can be documented, to improve searchability, so a wiki is a bit too unstructured, but- the general structure should be closer to an encyclopedia than a Q&AThe aim is to provide a collection of quality information, which is missing in that niche, not to make money, in case that matters.
How Google Can Open Up Image Labeler To Entrepreneurs And Accelerate A.I. ProgressGoogle image labeler http://images.google.com/imagelabeler/ allows people to play a game labelling images, thus give data towards their machine learning research on image recognition. Why don't they open this up to entrepreneurs ? (or amazon, yahoo, or some startups take up the opportunity to do the same)What I'm suggesting is that a "gold rush" type situation form around the building of classifiers (a type of machine learning algorithm), such as that occurred with website creation. There are millions trying to get an income onlne creating content for adsense and affilate programs (and for free with wikipedia and open source). Most will fail at generating reasonable income but in the process a large amount of wealth has been created.The same could be done for classifiers/artifical intelligence. For each classifier it would require collecting a thousand or two pieces of data (image/sudio/text) to create a classifier, an effort similar to creating a small website. For each use of the cla