oss-launch

Start with name and license clearance, then run a readiness gate where a cold reader follows only the README to first success. After that, write a category-first one-liner, pick one anchor channel, sequence amplifiers around it, and capture a metrics baseline the evening before launch.

GitHub Trending ranks stars gained inside a time window, and the selection formula is unpublished, so no placement can be promised. The realistic target is the language-filtered page, and concentrating effort into one day produces the slope these surfaces reward; never solicit or coordinate stars.

For developer-first projects, a tech aggregator like Hacker News is the anchor and Product Hunt is a secondary channel, since its audience skews product and founder rather than deeply technical. Product Hunt's durable value is SEO; Show HN requires something people can actually try.

No. Soliciting upvotes, comments, or stars gets the submission, the account, and sometimes the whole domain penalized, and the penalties are invisible from outside. Pointing people at the project is fine; asking them to act on ranking signals is not, on both Hacker News and Product Hunt.

Localize the failure by funnel stage: clicks with few stars or clones means the README lost the ten-second read, while clones without downloads means install or first-run friction. Only a wrong-channel diagnosis justifies relaunching; the rest require fixing the artifact first.

A launch caravan only pays off with at least three genuinely announceable items, a conference or fixed date on the calendar, and staff to hold a named owner per day. Below that threshold a single spike delivers more outcome per maintainer-hour, and a caravan dilutes into several ignored posts.