Technical state of the project

[Eric S. Raymond]

With 0.9.7 just shipped and our usual bravura performance of
CVE-dodging accomplished, this seems a good time to start thinking
about our objectives for the next round of development.

The background fact is that the great decrufting/hardening is
effectively done (with one exception I'll get to) and the codebase has
come through it as stable as a rock.  Months of burn-in on the test
farm RPis has produced not even one single crash or unexplained
anomaly.  We're down to 60KLOC, which is 26% of the codebase size
at fork time.

If we had to ship something labeled 1.0 tomorrow for political or
funding reasons I'd feel comfortable doing that. We're down to mostly
polishing and testing now.

There are a couple of places we can go from here.

PACKAGING: The most direct path to a 1.0 release would be to work up packaging
metadata for major distributions, and recruit package maintainers.

SOCKETFU: There's one promising C cleanup still pending - simplifying the
socket handling so instead of scanning all system interfaces ntpd just
listens on one wildcard socket.  This would allow us to drop a couple
KLOC of code and eliminare a dependency on the (non-standardized) code
for interface scanning.

We've also got a bunch of minor issues on the tracker, mostly related
to the waf recipe, that ideally ought to be fixed up before 1.0.

I think it would be a good thing to set achieving some subset of these
as our condition for shipping 1.0, and then move on that.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

"To disarm the people... was the best and most effectual way to enslave them."
        -- George Mason, speech of June 14, 1788