Blog

Barcode formats: Code 128, EAN-13 and UPC compared

What separates the common one-dimensional barcode symbologies, why a retail barcode needs a GS1 prefix, and how to print one that actually scans.

Character limits that actually matter in 2026

The real limits for title tags, meta descriptions, social posts and SMS - plus why Google measures pixels, not characters, and why emoji count as more than one.

CSV to JSON: converting without corrupting data

Why CSV parsing is harder than splitting on commas, what nesting does to a conversion, and the type coercion that quietly destroys zip codes and IDs.

HEX, RGB and HSL: one colour, three notations

How hex codes map to RGB channels, why HSL is the right notation for building a palette, and how contrast ratios are actually calculated.

Metric and imperial: which conversions are exact

Which unit conversions are defined exactly, where US and imperial measures diverge, and the volume and temperature traps that produce wrong answers.

Morse code: how it works and how to learn it

The structure of International Morse Code, the timing ratios that make it readable, and why learning it by lookup table caps you at a few words a minute.

Password strength: what entropy actually measures

What actually makes a password hard to guess, why forced rotation and complexity rules backfired, and how many bits of entropy you should be aiming for.

Percentage calculations: the four formulas

Percentage of, percentage change, reverse percentages and percentage points - the four calculations people mix up, and the arithmetic behind each.

Images to PDF and back: what conversion does

How an image is embedded in a PDF, what page size and compression to expect, and why browser-based conversion is the safer default for private files.

Roman numerals: the rules and the 3999 limit

How subtractive notation works, which pairs are legal, why 3999 is the ceiling, and what to do about numbers the system cannot express.

Subnetting explained: CIDR, masks and host counts

What a /24 actually means, how to split a network without a calculator, why two addresses per subnet are unusable, and how IPv6 changes the exercise.

Unix timestamps: seconds, timezones and 2038

What a Unix timestamp counts, how to tell seconds from milliseconds, why leap seconds are ignored, and how to store dates without a timezone bug.

301 vs 302 redirects: which to use and what chains cost you

The difference between permanent and temporary redirects, when 307 and 308 apply, and why every extra hop in a chain is worth removing.

Base64 explained: what it is, and when to use it

How base64 encoding works, why it makes data 33% larger, the difference between base64 and base64url, and the one case where inlining is worth it.

Broken links: how much they matter, how to find them

What broken links actually cost you, how Google treats them, and the fastest way to find every one on a site without paying for a crawler.

Building in public without it becoming the product

What building in public actually gets you, what it costs, which numbers are safe to share, and the trap of optimising for an audience that will never buy.

Churn rate: three versions, and which one to watch

Customer churn, revenue churn and net churn - how each is calculated, why they diverge, and what counts as normal at small scale.

SaaS conversion rate benchmarks, and why yours will look bad

Realistic conversion rates by traffic source and funnel stage for small SaaS, how to calculate yours honestly, and what actually moves it.

Core Web Vitals: what to fix first and what to ignore

LCP, INP and CLS explained with the thresholds that matter, the fixes that actually move them, and an honest account of how much they affect rankings.

Cron expression cheatsheet with the mistakes that bite

Every cron field explained with worked examples, plus the off-by-one and timezone mistakes that quietly break scheduled jobs.