# A Distribution Metric's Silence Reflects the Channel Not the Usage ## Claim A distribution metric reading zero or low means the project isn't using that specific channel, not that the software has no active users. <!-- atomicity: 1 idea ✓ --> ## Scope & conditions - Applies to any software distributed through multiple possible channels (WordPress.org directory, GitHub Releases, direct git deploy, package managers) where each channel produces its own, non-overlapping metric. - Doesn't establish that any single channel's metric is reliable even when present — GitHub asset-download counts are still noisy (CI jobs, mirrors, repeated downloads inflate them), a separate limitation from the channel-dependency claim. - Most consequential when someone compares two projects' health by a metric that only one of them actually produces (e.g., judging a git-deploy project as "unused" because it has zero GitHub release downloads). ## Evidence anchor.host's piece on self-publishing WordPress plugins outside WordPress.org observes that GitHub exposes per-asset download counts and repository stars, but "the metric only exists if you actually use the release channel" — a project that deploys straight from git, with no tagged Releases, produces zero download-count signal regardless of how many people run it. The author's fix is comparing several recent releases (not one lifetime total) normalized for how long each stayed current, rather than trusting a single raw number. ## Links - [[3. Resources/Literature Notes/What Visibility Do You Have When Releasing a Plugin Directly on GitHub]] — because evidence-for; the source article this claim generalizes from. - [[3. Resources/Permanent Notes/Absence Of Detection Is Not Absence Of Occurrence]] — because extends; both claims warn against reading an instrument's silence as the absence of the underlying event, in different domains (security incidents vs. distribution metrics). - [[3. Resources/MOCs/Open Source Governance MOC]] — because applies-in; project-health reporting depends on knowing which metrics a given distribution architecture can even produce. - [[3. Resources/Permanent Notes/Usage Is Not Adoption When a Connector Fails in Silence]] — contrasts: this note reads a zero as an unused channel, and the connector case reads a nonzero session count as use when the developer may have stopped after a failure. ## Tests / falsifiers - Find two functionally similar plugins, one distributed via WordPress.org (active-install counts) and one via GitHub Releases (download counts); confirm whether observers correctly attribute the metric gap to channel choice rather than inferring lower adoption for the GitHub-only project. - If practitioners already reliably account for channel-dependency when reading these metrics, the claim is more a reminder than a genuine correction of a live misreading. ## Maintenance - Last tended: 2026-08-16 - Next review: 2026-11-16 ## Source - [[3. Resources/Literature Notes/What Visibility Do You Have When Releasing a Plugin Directly on GitHub]] — reports the release-channel-dependency observation and the exposure-time-normalization fix this claim generalizes from.