Re: ietf-nntp Three proposals

[Ben Polk <[e-mail]>]

At 04:39 PM 12/17/96 -0500, Ofer Inbar wrote:

>If we "rename" XOVER to OVER, nothing will break.  

Nothing will break today.  This question shouldn't be decided based
on what will happen today, where it is obvious that everyone would
have to support both if we rename the X commands.  The decision
should be on what will happen over the next few years as we try
to replace XOVER with OVER.  If someone can lay out a description
of how they would expect OVER to really replace XOVER without 
causing a lot of disruption I'll drop my objection.  But I claim
that the only way to make this happen is to at some point have
XOVER no longer be a manditory synonym for OVER.  When that 
happens some new implementations will not support it.  And once 
that happens there will surely be things breaking.

That is not an absolute reason not to do this, it is the cost
we need to consider before making the decision.

I guess another way to say it is how do we go from having everyone
support XOVER to having no one support it without a lot of pain?

I think this is what the "Remove the X" group is proposing:

Today:     XOVER defacto manditory
Tomorrow:  OVER manditory, XOVER manditory but depricated
Day after: OVER manditory, XOVER optional but depricated
Future:    OVER manditory, XOVER prohibited

The breakage will occur starting the "Day after" when people
start implementing servers without XOVER.