Hi!
I would say that GuixSD is more alphaish than Guix, but I’m fine calling
it “beta” as well (in fact, I took a step in that direction when I
updated the web site a couple of days ago for the fundraiser after we
had had a similar discussion on IRC; look at this blue note at the top
;-)).
The main blockers to 1.0 IMO are:
• full-disk encryption in GuixSD (see in particular
<http://bugs.gnu.org/21843>);
• ‘guix pull’ made faster & more secure (how do we authenticate code?
how do we know it’s the latest code?);
• stable build farm infrastructure;
• clean upgrade path, revocation, and delegation for the keys used to
sign substitutes (currently we have no easy way to provide an
updated key for hydra.gnu.org, or to revoke the existing one, for
instance.)
Then of course there are many other improvements that can be made,
notably usability improvements, of which having a fully working GNOME
desktop or a nicer installer are one aspect.
I think these are highly desirable, but not blockers for 1.0, in the
sense that 1.0 should provide a stable implementation of the core
features, but not necessarily every feature one would hope for.
WDYT?
Ludo’.
