Text Studio

Questions people actually ask

Which title-case rules does the case converter follow?

AP style, applied as a fixed list rather than a length rule. These 23 short words stay lowercase inside a sentence: a, an, and, as, at, but, by, for, if, in, nor, of, off, on, or, per, so, the, to, up, via, vs, yet. Everything else is capitalised, and the first and last word of every sentence is capitalised regardless — so "The Art of War" keeps its lowercase "of", while "What Are You Waiting For" capitalises "For" because it ends the line. Hyphenated compounds get each part capitalised unless the part is itself on the list. Chicago style differs (it capitalises longer prepositions such as "between" and "through"), so if your house style is Chicago, expect to fix a word or two by hand.

Which Morse standard does this use?

International Morse as defined in ITU-R M.1677-1 (2009) — the 26 Latin letters, the ten digits, and the standard punctuation set (. , ? ’ ! / ( ) & : ; = + - _ " $ @). Letters are separated by a single space and words by a slash, which is the printed convention in the recommendation itself. Decoding is more forgiving than encoding: a slash or any run of two or more spaces is read as a word gap. Morse has no upper and lower case, so anything you decode comes back in capitals, and any character with no assigned code is listed rather than silently dropped.

What do the readability grades actually mean?

Five of the six are US school grade levels: a Flesch–Kincaid of 8.0 means a typical eighth-grader — around 13 years old — can read the passage without re-reading. Flesch Reading Ease is the odd one out, a 0–100 scale where higher is easier and 60–70 is ordinary plain English. They disagree on purpose, because they measure different proxies: Gunning Fog and SMOG weigh words of three or more syllables, Coleman–Liau and ARI count characters per word and skip syllable guessing entirely. The consensus figure is the average of the five grade scores. Treat all of them as rough: they measure sentence and word length, not whether your argument makes sense.

Why is SMOG flagged as unreliable on short text?

McLaughlin calibrated SMOG in 1969 on samples of 30 sentences, and the formula assumes that much material. Below it the complex-word density swings wildly on a single long word, so the studio labels the score "rough" rather than hiding it or quietly refusing to show it. The other five formulas degrade more gracefully but are still noisy on a paragraph or two — a few hundred words is the point where the numbers start meaning something.

Is fancy text safe to use in a bio or a heading?

Not really, and the tool says so where you generate it. Those characters are not bold or italic formatting — they are separate Unicode letters from the Mathematical Alphanumeric Symbols block, borrowed for decoration. A screen reader either announces each one by its formal name ("mathematical bold small a") or skips it, so a name written in fancy text can be unreadable to someone using assistive technology. Search engines do not match them to the plain word either, and some apps and older fonts render them as empty boxes. Use them for a one-off flourish; never for a heading, a link, or anything that has to be read aloud or found.

Is any of my text uploaded?

No. Every one of the fourteen tools is JavaScript running in this tab — the case converter, the diff, the readability scores, all of it. Nothing is sent to a server, nothing is logged, and nothing is stored between visits. That matters more here than on most pages, because the text people paste into a diff or a readability checker is usually an unpublished draft, a contract, or someone else’s confidential document. Close the tab and it is gone.

What does regex mode in find & replace support?

Your browser’s own JavaScript regular-expression engine — character classes, quantifiers, anchors, alternation, backreferences, lookahead and lookbehind. Capture groups work in the replacement box: search for (\d+)px, replace with $1rem. The pattern runs with the global flag, plus the ignore-case flag when "Match case" is off, and "Whole words only" wraps your pattern in word-boundary guards rather than making you write them. If the pattern does not compile, the exact engine error is shown and your text is left untouched rather than half-rewritten. One honest limit: regex mode compiles without the Unicode flag, so \p{…} property escapes are not available there — plain-text mode is the Unicode-aware one.

Why does one emoji turn into four groups of binary?

Because the binary tool encodes UTF-8 bytes, not code points. ASCII characters are one byte each — eight bits — but accented letters take two, most non-Latin scripts take three, and emoji take four. Encoding bytes rather than code points is what lets any character you can type survive a round trip unchanged. Decoding ignores how the groups are spaced, so "01001000 01101001" and "0100100001101001" give the same answer, and a leftover partial byte is reported instead of guessed at.