Patrick McLaren patrickmclaren

I'm hunting for the best solution on how to handle keeping large sets of DB records "sorted" in a performant manner.

Problem Description

Most of us have work on projects at some point where we have needed to have ordered lists of objects. Whether it be a to-do list sorted by priority, or a list of documents that a user can sort in whatever order they want.

A traditional approach for this on a Rails project is to use something like the acts_as_list gem, or something similar. These systems typically add some sort of "postion" or "sort order" column to each record, which is then used when querying out the records in a traditional order by position SQL query.

This approach seems to work fine for smaller datasets, but can be hard to manage on large data sets with hundreds (or thousands) of records needing to be sorted. Changing the sort position of even a single object will require updating every single record in the database that is in the same sort group. This requires potentially thousands of wri

The introduction to Reactive Programming you've been missing

(by @andrestaltz)

This tutorial as a series of videos

If you prefer to watch video tutorials with live-coding, then check out this series I recorded with the same contents as in this article: Egghead.io - Introduction to Reactive Programming.

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

	class Foo
	attr_reader :bar
	def initialize
	@bar = 123
	ObjectSpace.define_finalizer( self, self.class.finalize(bar) )
	end

	def self.finalize(bar)
	proc { puts "DESTROY OBJECT #{bar}" }
	end

	// ES6

	class AngularPromise extends Promise {
	constructor(executor) {
	super((resolve, reject) => {
	// before
	return executor(resolve, reject);
	});
	// after
	}