← The Daily Witlet

Credits and licenses

Our games use original game code and vector artwork, along with open language resources. We gratefully acknowledge the contributors.

Swapple

Word lists are adapted from the English Speller Database (ESDB/SCOWL), copyright Kevin Atkinson and other contributors. See the complete copyright and permission notices. Word frequency filtering uses Robyn Speer's wordfreq.

Linklets

Word relationships are adapted from Princeton University WordNet 3.0 (license; full deployed notice) and ConceptNet 5.7, compiled by the Commonsense Computing Initiative (full contributor credits, CC BY-SA 4.0). Relationships are filtered and transformed into puzzles.

Radar

Radar uses Open English WordNet 2025 (Open English WordNet contributors, CC BY 4.0), Princeton WordNet, ConceptNet 5.7 and ConceptNet Numberbatch 19.08 (ConceptNet and Luminoso Technologies, Inc., CC BY-SA 4.0). Reference: Robyn Speer, Joshua Chin and Catherine Havasi, ConceptNet 5.5 (AAAI 2017). See the Numberbatch credits. We filter and adapt lexical relationships and convert vector similarities to game-specific scores. Thanks also to Robyn Speer (wordfreq, copyright 2022). Word frequency data include the freely available SUBTLEX resources by Marc Brysbaert and colleagues, and data from OpenSubtitles, Google Books Ngrams, Wikipedia, Leeds Internet Corpus and ParaCrawl. See the full wordfreq notices and credits.

Open data and software

ConceptNet-derived relationship data and Numberbatch-derived score data, to the extent they are adaptations of those resources, are offered under CC BY-SA 4.0. This does not apply to independent game code or artwork. Our use of wordfreq's data follows the applicable upstream data licenses. Development tools include Python, NumPy, NLTK and wn; the website uses browser-native JavaScript, system fonts and inline SVG artwork.

For licensing questions or requests concerning source attribution, contact the operator via Feedback.