Experimental implementation of liquid-hg

[Pierre-Yves David]

On Wed, Jan 19, 2011 at 10:56:32AM -0500, Eric Eisner wrote:
> Could frozen and liquid changesets be implemented entirely as published and
> unpublished? Every push or pull would transparently mark copied changesets
> forever into the frozen category (on server and client), without the need
> for a pushkey protocol. All local changes are by definition unpublished and
> are free to be rebased, mq'd, stripped etc.
> 
> I'll admit to not following all of the paths of this discussion, so which
> proposed use cases require an explicit marking of liquid history?

Counter argument are in this email.

http://selenic.com/pipermail/mercurial-devel/2011-January/027591.html

-- 
Pierre-Yves