Text guides for people who work with words and data
Six guides, each written to answer one question properly: how counting works across writing systems, which case style belongs where, how to clean a pasted list without losing rows, how JSON errors are structured, why encoding is not encryption, and what a slug does to a search result.
Word Count: Why Chinese Counts Characters and English Counts Words
Why the units differ, how 400 characters per minute and 200 words per minute are applied, and where counters disagree.
Read the guide →The Seven Case Styles and When to Use Each
What each case style signals, the small-word and acronym rules, and the three conversions that quietly break links and code.
Read the guide →Deduplicating and Sorting Lists: The Right Order for Pasted Data
A repeatable six-step cleanup order, why natural sort beats plain alphabetical, and the traps in spreadsheet paste.
Read the guide →JSON Formatting: A Self-Check Checklist for Common Syntax Errors
The seven failures behind nearly every parse error, how to read line and column numbers, and the valid-JSON traps that bite later.
Read the guide →URL Encoding and Base64 Are Not Encryption
What percent-encoding and Base64 really do, the btoa Unicode bug, and why a JWT payload is readable.
Read the guide →Slug Generation Rules and What They Do to SEO
How search engines tokenise URLs, eight generation rules, transliteration choices, and the redirect discipline for changing a slug.
Read the guide →How these guides are written
Each guide comes from a specific problem: a style guide that contradicts itself, a list that would not deduplicate, a payload that failed at 2 a.m. We explain the mechanism instead of listing commands, because the mechanism is what transfers to the next tool. Where a rule is a convention rather than a requirement — Title Case small words, for example — the guide says so and names the style manuals that disagree.
Nothing here is generated filler to fill a template: no invented statistics, no "10x your content" promises, no comparison tables against products we do not use. If a claim cannot be checked against a documented standard, it is presented as a recommendation, not a fact.
Written by the TextBench team. Last updated November 19, 2026.