RE: draft-housley-two-maturity-levels

["Tony Hain" <[e-mail]>]

Bob Braden wrote:
> >
> > We really should get serious about the term 'proposed', and note that
> the
> > referenced document is under development. It is not an end state in
> itself,
> > just aging on the shelf to meet a process check mark.
> >
> > Tony
> 
> Tony,
> 
> That would not work, would it?  The driving force behind most WG
> efforts
> is the community of vendors, who want to produce products ASAP. They
> certainly don't want to wait for a document to "develop". For their
> purposes, Proposed Standard and product out the door (if not earlier,
> at
> the I-D stage).
> 
> Presumably the WG process takes so long because different vendors have
> to work hard to reach a consensus. Once it is reached, they don't want
> a
> bit to change in the spec before the end of their product cycle.
> 
> In this environment, the only thing that seems to make sense is for WGs
> to start usually at Experimental (someone else suggested this, I
> apologize for not recalling who it was).

It would work if the overall process were more efficient. Now we effectively
go WG I-D to full IS, which is what your eloquent overview of the driving
force notes. If we truncated WG I-D at the common points people could agree
to start implementing, and have PS actually document the evolution of the
implementations, we would get back closer to when the IETF was productive. 

Fundamentally the problem is that the process has evolved to state that you
don't get an RFC number until the work is done. Instead of requiring
'perfection', we need to get back to 'good enough to start', then be
proactive about progressing PS to IS to document 'done'. 

Tony