About
What this is
Every day at 07:30 KST, the repositories that made GitHub's trending lists the day before: what rose and what it does, in one-line summaries and numbers. An independent service, not affiliated with GitHub, run by IT Solution.
Data and writing
The data is GitHub's daily, weekly and monthly trending lists and each repository's star, contributor and fork counts. A repository's name opens its page here, which links on to GitHub. Summaries, the headline and summary at the top, the field each repository is filed under and the repository articles are written by AI from each repository's description and README, and can be wrong.
Shopping (in Korean)
The Shopping page, in Korean only, lists the search keywords shoppers used most in shopping search over the last seven days, from Naver DataLab Shopping Insight: each top category's top five, keywords that climbed into a category's top ten, and keywords new to a category's top hundred since Trend began recording shopping searches. It is not affiliated with Naver either.
Rising now (in Korean)
When Koreans are searching for something far more than usual, the Korean front page opens with it: every hour, Google Trends' rising searches in Korea over the last four hours, each set against its own last nine months, day by day. Only words searched clearly more (about half again or more) than on any day of those months, or hardly searched before, appear — so a tax deadline or a holiday only when bigger than before — each with a news story about it. Search terms, names included, are shown as the source has them; headlines are their publishers'. Not affiliated with Google.
How the stories are chosen
Each search term in the Korean front page's list comes with one news story about it. Terms, stories and their order are chosen by a fixed rule, not by a person. Every hour the top twenty of Google Trends' rising searches in Korea are read, and for each term the bigger of its two most recent days with data is set against its biggest day in the nine months before: only terms at about 1.5 times that or more, and terms hardly searched before and now searched a lot, are kept. A sports search of two words or more is left out (it is mostly the name of that day's match, with no past to measure against). For each term left, going down the list of stories Google Trends attaches to it, the story is the first whose headline contains every word of the term; a term with no such story is left out. Terms hardly searched before come first, then those with the largest multiple of that biggest day — so the order can differ from that of the figures shown beside them; terms alike in this follow the rising-search rank. At most five at a time. Headlines are not changed (a publisher's name at the end of one is shown on its own). No story or advertisement is placed for payment.
Who is responsible
The rule above and what it puts on the page are the responsibility of the person in charge of story arrangement, and youth protection that of the youth protection officer. Both are 조성호, IT Solution's representative. Requests to correct a story or to reply to one, and reports of anything harmful to young people, go to contact@ondo.sh.
Use
AI-written summaries can be wrong. Check the GitHub original for the exact content and the latest figures.
Contact and removal
To have your repository, username or your name on the front page left out, or for anything else, write to contact@ondo.sh. Confirmed requests take effect the next day.