For a 0.X.X branch, which is experimental anyway, I'm not too concerned about strict semantics.
What about Go ? On Wed, Feb 11, 2026 at 5:19 PM tison <[email protected]> wrote: > It looks OK to apply to datasketches-rust. > > I'd add some points that, since datasketches-rust is still under 0.x, it > would look like: > > main > | - 0.3.x > | - 0.4.x > > ... and based on Rust's Cargo implementation [1], 0.4 may introduce > breaking changes compared to 0.3. > > [1] > https://doc.rust-lang.org/cargo/reference/semver.html#change-categories > > So in datasketches-rust, I may prefer: > > 1. Keep the main branch only for now, as the release goes straightforward. > 2. If we make the first backport release, we check out 0.N.x branch from > latest 0.N.M. > 3. Once any backport release happens, or after datasketches-rust releases > 1.0, follow the strategy above. > > Best, > tison. > > > Lee Rhodes <[email protected]> 于2026年2月12日周四 07:38写道： > >> So far in our DataSketches project, we have adopted a common strategy for >> numbering and organizing branches for releases for the languages Java, C++, >> and Python. It looks something like the attached. >> >> I think that all of our languages should follow the same strategy. If so >> I'll add it to our website. >> >> Thoughts? >> >> Lee. >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] > >
