Re: Another look?

From: Junio C Hamano <gitster@pobox.com>
Date: Jan 4, 2026, 04:17 UTC
Message-ID: <xmqqo6nanh8o.fsf@gitster.g>
In-Reply-To: <aVnTc8GmqH+5Ui3M@ubby>

Nico Williams <nico@cryptonector.com> writes:

> Sun Microsystems, Inc. used a rebase workflow from 1992 to its end
> ...
> It worked like a charm.

That's good for you. But I do not necessarily think merge workflows are bad, and my advice was certainly *not* about avoiding merges.

A linear logical progression of commits that is about a single theme is much nicer than drunken-man's walk that is also a linear sequence of commits. The distinction between them has nothing to do with merges.

And a history that bundles together a collection of linear logical progressions with merges of these topics one by one is much easier than rebasing these unrelated topics into a strictly linear single strand of pearls, by making it much clear where each of these topics concludes.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.