Hi, Maven Shared IO was last released as 3.0.0 in December 2015. Since then it has had no user traffic at all: of the 45 issues the repository has ever held, 29 are the historical JIRA import, 15 come from a code-reading sweep across a dozen Maven repositories in July 2026, and one is mine. None is a user reporting that the component misbehaved for them.
It also has no consumers left. No Apache Maven component imports, the last one that did, dropped the dependency in 2020 by copying the classes it needed. The two references that remain are test-scope only: an integration-test fixture in maven-assembly-plugin pinned to 1.1, and a test-classpath entry in a Jackrabbit plugin that imports nothing from it. Open Source Insights reports three direct dependents on 3.0.0; it gives counts and not names, so I cannot say who they are. The git history looks more active than that, and I want to be straightforward about it rather than have someone find it in the first reply: there are 98 commits since the 3.0.0 tag, 21 of them this year. Nine of this year's are Dependabot and most of the rest are fixes for the July sweep. That is a component being read, not a component being used. Finally, the download package has no Maven 4 story. DownloadManager exposes Wagon in its published API, and DefaultDownloadManager obtains the Wagon, the credentials and the proxy from maven-compat's WagonManager, which Maven 4 does not have. Fixing that means breaking the published API of a component nobody calls. I therefore propose that we retire maven-shared-io. If this vote is successful I will make one final release, which also ships the fixes that have accumulated since 3.0.0, making it clear on the component site that it has been retired. After that the source code repository will get archived, meaning it's read only in the future. The process for retiring a component is described here: https://maven.apache.org/developers/retirement-plan-plugins.html The vote is open for 72 hours. [ ] +1 Yes, it's about time [ ] -1 No, because... Sylwester
