The libaes_siv dependency

[Eric S. Raymond]

Jason Azze via devel <devel at ntpsec.org>:
> I predict that ESR will ask me for an alternative approach. He won't
> like my recommendation. It's to use a feature or development branch
> for a change as big as the introduction of NTS.

Even if I liked using feature branches (and I don't; see

https://blog.ntpsec.org/2017/04/09/single-head-provable-steps.html

for explanation) they wouldn't solve our actual problem.  NTS isn't an
experiment that may or may not be merged at some point in the
indefinite future, it is our main line of development because Cisco
has promised us money to have it ready to ship in our next point
release. Thus, putting the work on a branch would be pointless.

This is also the reason I haven't made an NTS config option. The only
theoretical justification would be giving people who build from source
the option of avoiding the increase in code footprint from NTS support.
But when you start asking who actually might *want* that...

It's no help to distro packagers, who are going to want to compile in
maximum features rather than a minimal build.  Actually, the additional
code volume for support of NTS is so low that it's difficult to see it
being a problenm for *anyone*, especially when bear in mind that NTP
Classic users put up with code 4:1 larger.

A branch or config conditional would also have costs in test complexity
that I don't care to incur.

In every scenario I've gamed out in my head, the right answer comes out
the same: Keep It Simple, Stupid! One line of development, no new code
conditional, power through the integration problem somehow.

You've changed my agenda for the day. I think now I have to revisit an
approach I was avoiding: including libaes_siv as a git submodule. I can
male it work, but submodules are fluky and irritating.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

My work is funded by the Internet Civil Engineering Institute: https://icei.org
Please visit their site and donate: the civilization you save might be your own.