Word Counter

Beyond raw length, this counter reports unique words, the number of distinct lowercase word forms in the text. The ratio of unique words to total words is the type-token ratio, a basic index of lexical variety used in stylometry and language assessment. That ratio is strongly length-dependent and falls as a passage grows, because common function words recur while new content words appear more slowly, a consequence of the Zipfian frequency distribution of natural language. Comparing the ratio across texts of different lengths is therefore invalid, and researchers use length-controlled measures instead. Reading time here uses a deliberately conservative 200 words per minute rather than a research mean, since informational and technical material is read more slowly than narrative prose. Character counts matter because most publishing constraints are stated in characters rather than words. Google truncates meta descriptions at roughly 155 to 160 characters and title tags at around 55 to 60, although the real limit is rendered pixel width, so wide capital letters truncate sooner than narrow lowercase ones. A single SMS holds 160 characters in the 7-bit GSM alphabet but only 70 once any character forces UCS-2 encoding, which is what happens the moment a curly quote or emoji is inserted. Posts on X have allowed 280 characters since 2017, doubled from the original 140. One structural caveat: whitespace segmentation is meaningless for Chinese, Japanese, and Thai, which do not delimit words with spaces, so word counts for those scripts will be badly wrong.

Counting text predates computers by centuries, and the original unit was the character rather than the word. Compositors setting metal type were paid by the thousand ens, a measure of the horizontal space occupied by set type, and publishers estimated book length by character count long before word count was practical to obtain by hand. Telegraphy made words billable: operators charged per word with minimum message lengths, which is precisely why telegraphic style dropped articles and pronouns. Twentieth-century publishing settled on the standard manuscript page of about 250 words, a convention derived from double-spaced typewritten pica text at roughly 60 characters across and 25 lines down, and freelance writing rates are still quoted per word. Automated counting arrived with early Unix. The wc utility, which reports lines, words, and characters, was present in the First Edition Unix distribution of 1971 and defined the whitespace-delimited notion of a word that virtually every counter since has inherited, including this one. The analytical measures came from linguistics: George Kingsley Zipf described the frequency-rank law underlying vocabulary growth in work published through the 1930s and 1940s, and Wendell Johnson introduced the type-token ratio in 1944 as a measure of language behavior. The 160-character SMS limit was fixed by Friedhelm Hillebrand in 1985, who judged from sample sentences and postcards that 160 characters were sufficient for most messages.