On Thu, 25 Mar 1999, Ted Lemon wrote:

> Bill, the point is that throwing in this inode change at the last
> minute like this is the wrong thing to do, not that we should or
> shouldn't use fat timestamps.   Fat timestamps are merely one of many
> examples that have been raised as to why we should have a serious
> design review before proceeding with this new code.
> 
> There seems to be a fundamental disconnect here - you want to make a
> change to a really common global structure two days before the 1.4
> code freeze, with no design review and no specification for what your
> change does other than a few messages to the mailing list.
> 
> If you were proposing this two months ago, we'd have had time to do
> the design review correctly.  You've simply chosen the wrong time to
> put forward this proposal.  Can you please stop trying to argue that
> this should go into 1.4 and instead write up a design document while
> you're waiting for the code freeze?  Then after the code freeze we can
> do a design review and get the right code committed to -current well
> before the 1.5 cutoff.

I stoped pushing for it to go into 1.4 a while ago. NAS (my boss) will be
happy if anoncvs is around, and whatever changes we need are on a 1.4-ish
tree on the anoncvs server (note there are concerns with the crypto code
and anoncvs service).

What I think I'm still arguing is that: we were proposing a large-inode
variant of the current ffs implimentation (more accuratly of the ufs
implimentation), NOT a whole new fs, and that we are providing a
framework for a lot of new stuff. As not everyting can be decided now, we
left room for future growth.

Take care,

Bill