GitHub trending archive
GitHub shows only the current trending list. This archive keeps the daily list of every day since our collector started, each one with the time it was read.
Why keep the old lists
The trending page on GitHub has no history: once a day has passed, its list is gone. Keeping each day makes it possible to answer questions the live page cannot — when a project first appeared, how many days in a row it stayed, what else was popular the week a tool you use was launched.
Each stored day is the main daily list (all languages) as it stood at the last collection of that day, in UTC. Nothing is edited afterwards: repositories that were later renamed or deleted remain as they were listed. The archive starts on the first day our collector ran and grows by one page per day; there is no data for earlier dates, and none is reconstructed from other sources.
To see the current list, go to today. How the collector works is described on the method page.
More lists to explore
The same data cut by period, programming language and subject.
By period
By programming language, today
Questions and answers
More detail on the method page.
What is stored for each day?
The main daily list as it was at the last collection of that day in UTC: position, name, description, language, total stars, forks and stars gained that day.
How far back does the archive go?
To the first day our collector ran. GitHub does not offer past trending pages, so days before that cannot be reconstructed, and we do not fill them in from other sources.
Why does a day show the list from late in the day?
The collector runs every three hours and each run replaces the entry of the current UTC day, so the stored list is the last one read before midnight UTC.