34 Domains Behind Popular Hacker News Stories Are Now Completely Unregistered

·1 views

A developer audited approximately 5,000 archived Hacker News stories using the platform's official APIs and found significant link decay, particularly in older posts. Among 2,127 unique registrable domains identified, 133 no longer resolve and 34 are entirely unregistered. Stories from 2007–2011 with over 300 points showed the highest dead-link rate at 24.5%, while stories from 2012–2014 clocked in at 16.5%. Several high-profile threads — including ones about a PayPal CEO response and a viral cardboard arcade story — now point to fully expired domains. The audit also flagged a silent failure mode where sites redirect old article URLs to their homepages, returning a healthy 200 status code while the original content is gone.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Four Design Lessons from Building a Block-Editing Language for AI Agents

Developers building IWE, an AI agent memory system, found that requiring agents to re-emit entire documents for small edits caused a 9% structural malformation rate in their first benchmark round. To fix this, they created a block-level editing language that lets agents target specific document nodes—headers, paragraphs, tables—without rewriting whole files. The system introduces 'expect guards' that force operations to fail loudly if the number of matched targets differs from what the agent anticipated, reducing silent errors. Document keys must be derived deterministically from metadata, making duplicate entries a structural impossibility rather than a prompt-discipline problem. These design choices reflect a broader principle: for AI agents, safety constraints should be enforced by the engine, not left to instruction-following.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Tokenization Explained: How AI Models Convert Text Into Numbers

Tokenization is the process by which AI models break input text into smaller units called tokens before processing it, since computers work with numbers rather than raw language. Each token is assigned a numerical identifier called a token ID, allowing the model to perform mathematical operations on the text. Tokens do not always correspond to whole words; uncommon or lengthy words are often split into smaller subword units, which helps models handle rare terms, names, and technical vocabulary. This subword approach means a model does not need to store every possible word in its vocabulary, instead combining reusable pieces to represent new or unseen words. Tokenization is equally applicable to programming languages, enabling AI coding assistants to process and generate code across dozens of languages such as Python, JavaScript, and SQL.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Why JavaScript Evaluates '2' > '10' as True: String vs Number Comparison

In JavaScript, comparing '2' and '10' as strings yields a surprising result of true, unlike the numeric comparison which correctly returns false. This happens because JavaScript uses lexicographical ordering when comparing string values, evaluating characters one by one as in a dictionary. Since the character '2' ranks higher than '1' in character ordering, '2' is considered greater than '10' as strings. When one value is a string and the other a number, JavaScript automatically converts the string to a number before comparing, restoring expected mathematical behavior. The key takeaway for developers is that data types fundamentally affect how comparisons are evaluated in JavaScript.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer spends 10 hours tracing AWS timeout to a self-inflicted credentials bug

Developer Themis Lex deployed an AI-powered PDF assessment tool called Themis Lex to AWS Amplify for the Women in AI Accelerator Spring 2026 Build Challenge, only to encounter a 28-second timeout error immediately after launch. The app, which uses Claude via AWS Bedrock to generate role-specific AI guidance for court clerks, appeared to work perfectly in local development. The root cause turned out to be a hardcoded credentials object in the AWS SDK client that passed empty strings instead of omitting the field, effectively blocking the SDK's automatic credential provider chain. Because the failure occurred before any application logic ran, the try/catch block logged nothing, making the bug resemble a platform timeout rather than a code error. After ten and a half hours of debugging and a series of fixes including switching to response streaming and a lighter AI model, the app was successfully shipped.

0 comments Read more at DEV Community