[bind10-dev] WIN32 port

[Francis Dupont]

> I think it all depends on when we want to prioritize the Windows
> support.  Unfortunately I don't think we can afford time for it at
> least until the September release.

=> as usual the real choice is between a little time now or a bigger time later.
BTW if the test machine is ready we can talk about to add a WIN32 build & check
on it. I am reading to help (with a new Subject line?), needed things are:
 - Windows 7 license (home edition is enough but please go to 64 bits)
 - Visual Studio C++ (I use the free Express version but if we want to
  build bind9 too a paying (i.e., expensive) version is required.
  The VS shell stuff is included in paying versions too so IMHO it is
  better to install more than Express)
>From these and an (admin) account on the VM I can finish.

> After that, I guess it's a matter of how attractive the Windows
> support is compared to other features to be added later this year.

=> I am not a WIN32 addict at all but ignoring 91% of the world is never
a good idea in the long term, so anyway we have to do it.

> Once we agree on the time for prioritizing the Windows support, I
> think your proposal makes sense.

=> my idea is to exchange motion speed against priority. As we have a way
to avoid regression it should not badly interfere with other new features,
i.e., we should not go backward on either WIN32 port or new features.

Regards

Francis Dupont <fdupont at isc.org>