PROPOSED, change of stance, release metronome

[Eric S. Raymond]

Hal Murray <hmurray at megapathdsl.net>:
> What's the criteria for 1.0?
> 
> Do we have a list of things that need to be fixed or implemented?

We do.  It's in devel/TODO.  It's mostly build-system stuff:

   == Checklist for a quality 1.0 release ==

   === Build system ===

   * Document build files.

   * Cleanup and separate features in pylib/*

   * Add 'snapshot' feature to dump config status into a JSON file for collecting
     build + platform information

   * The install production needs to warn of stale ntpdc and ntpdate binaries.

   * Add support for enabling all conditional checks to ensure none are broken.

   * Add support for disabling specific conditional tests to ensure nothing is
     compiled into the base in a surprising fashion, e.g. --undefine=FOO should
     suppress generation of any FOO #define into config.h.

   * Add support for enabling hidden features like autokey and/or
     internal symbols for testing, e.g --define FOO=BAR generates #define
     FOO BAR into config.h.

   * Add monkeypatch for rmdir, rmtree under Windows.

   * ntpdsim does not quite build.  There is something wrong with the way waf is
     setting up the linkage.  To reproduce, remove the "if 0" near the end of
     ntpd/wscript.

   === Code ===

   * We need a build time option to disable all dangerous operations
     that ntpq can produce.  First we need to enumerate them.

   * systime.c needs patching to put ntpdsim's hook back in place. Deferred
     until the ntpdsim build is fixed.

   * Hal has seen a core dump in ntpdig.  It has not reproduced, so we don't
     know what's going on there.

   === Testing ===

   * Unit testing.  "waf check" needs to run all the unit tests, not
     just the small fraction it does.

   * We need to live-test various refclocks.  It would be nice
     to test the full matrix of refclock x platform, but that's too
     much work.  We should probably test as many refclocks as we can
     on at least one platform and test the NMEA, Atom, and SHM drivers
     on most platforms.

   * The script tests in scripts/t should be investigated, perhaps
     they are salvageable.

   === Documentation ===

   * Short doc on how to tell your ntpd is running correctly.
     Perhaps lifted from GPSD Time Service HOWTO?

   * The format and organization of the NEWS file needs a rethink.

We've actually done the last one, I think.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>