How LinkedIn Uses Event-Driven Architectures to Scale (6 minute read)
LinkedIn leverages event-driven architectures and the actor model to scale its infrastructure. The actor model is a concurrent computation model where independent actors communicate and coordinate solely through asynchronous message passing. The company uses this approach in several systems, including one for collecting and processing server metrics in near real-time and another for monitoring server consoles.
|
Supporting Offline Mode in TanStack Query (11 minute read)
Effect Schema, a validation library, can be used for achieving full offline support with TanStack Query. It allows for bi-directional parsing, eliminating the need for separate schemas for encoding and decoding data. This article goes through how to store data locally (using IndexedDB or LocalStorage), implement cache removal, set cached data, and handle mutations.
|
Reactive HTML notebooks (12 minute read)
Celine, a reactive HTML notebook library, can be used to improve scientific publishing workflows by consolidating data exploration, analysis, and typesetting within a single HTML file. The library uses existing technologies like Observable, Plot, and various WASM runtimes (Python, R, and SQLite) to enable interactive elements and diverse data visualizations directly within the HTML document. This article goes through various implementation examples, such as interactive code cells, plots, and even a slideshow presentation mode.
|
|
Against best practices (5 minute read)
Blindly following “best practices” without considering context leads to poor decision-making. Instead, each "best practice” should be evaluated critically in the context of the application itself.
|
The Tragedy of Running an Old Node Project (3 minute read)
This developer attempted to update their four-year-old Gatsby website but ran into difficulties running the old Node project. The `npm install` command failed repeatedly due to missing Python 2 and incompatible C++ compiler versions. Troubleshooting involved installing Python 2, discovering a required older Node.js version (v12.2), and ultimately resolving the build errors after two hours.
|
Socratic Questioning - "Meetings are a waste of time" (5 minute read)
This article uses the statement "Meetings are a waste of time" to illustrate Socratic questioning: clarification, challenging assumptions, seeking evidence, changing perspectives, exploring implications, and questioning the question itself. Each step is demonstrated through examples related to the initial statement to show how Socratic questioning can be used as a tool to think critically and challenge prior assumptions.
|
|
Kokonut (Website)
KokonutUI is a collection of 50+ modern, customizable UI components built with Next.js 15 and Tailwind CSS. These components are designed for quick and easy copy-pasting into applications.
|
AI SDK 4.0 (7 minute read)
Vercel's AI SDK 4.0 is an open-source JavaScript/TypeScript toolkit for building AI applications. This release adds PDF support for document analysis across multiple providers, computer use support with Anthropic's Claude for app interaction, and continuation support for generating long texts.
|
Tips.io (Website)
Tips.io is an AI-powered website builder that uses Tailwind CSS. It has an HTML editor with autocomplete, AI-assisted content generation, and pre-built themes and components.
|
|
20 things you didn't know about Google Scholar (5 minute read)
Google Scholar, launched in 2004 by two researchers, aimed to improve access to academic information by creating a comprehensive online research library. Early challenges included slow internet speeds, leading to the creative "Sneakernet" method of delivering research data on hard drives. Beyond its core function, Scholar has revealed unexpected uses, from family history research to uncovering unusual and funny aspects within academic publications and legal documents.
|
An Uncanny Moat (7 minute read)
The "uncanny valley” is a phenomenon where almost realistic depictions of humans appear unsettling - this applies to AI. Increasingly realistic AI chatbots, while convenient, erode our ability to distinguish between human and artificial interactions and potentially reduce empathy. Deliberately stylizing AI interfaces, making them easily distinguishable from humans, could mitigate these risks.
|
Traits are a Local Maxima (22 minute read)
Rust's traits, while popular, suffer from the "orphan instance problem," where implementing a trait for a type from a different crate is restricted. This leads to potential coherence issues. A possible solution is local coherence using implicits. While implicits address the orphan instance problem, they introduce new challenges, including increased code repetition, potential loss of correctness guarantees, and reduced code readability compared to traits. Both traits and implicits represent local maxima in language design, but each have their own trade-offs.
|
|
Fast GraphRAG (GitHub Repo)
Fast-Graphrag is an open-source framework for Retrieval Augmented Generation (RAG) that uses graphs to streamline agent-driven workflows.
|
MailCatcher (Website)
MailCatcher is a simple SMTP server and web interface that allows developers to view emails sent during development by configuring their applications to send emails to it.
|
|
|
Love TLDR? Tell your friends and get rewards!
|
|
Share your referral link below with friends to get free TLDR swag!
|
|
|
|
|
Track your referrals here.
|
|
Want to advertise in TLDR? 📰
|
If your company is interested in reaching an audience of web developers and engineering decision makers, you may want to advertise with us.
If you have any comments or feedback, just respond to this email!
Thanks for reading,
Priyam Mohanty, Jenny Xu & Ceora Ford
|
|
|
|