Re: *Req and conformance

Bob Braden <braden@isi.edu> Wed, 16 March 1994 22:06 UTC

Date: Wed, 16 Mar 1994 13:38:16 -0800
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Bob Braden <braden@isi.edu>
Message-Id: <199403162138.AA28151@zephyr.isi.edu>
To: rreq@rice.edu, barns@cove.mitre.org
Subject: Re: *Req and conformance

  *> From barns@cove.mitre.org Wed Mar 16 11:59:24 1994
  *> To: rreq@rice.edu
  *> Cc: barns@cove.mitre.org
  *> Subject: *Req and conformance
  *> Date: Wed, 16 Mar 94 14:52:54 -0500
  *> From: barns@cove.mitre.org
  *> Content-Length: 817
  *> X-Lines: 16
  *> 
  *> 1. Multi-tiered specifications for conformance purposes make sense to me.
  *> 2. Multi-tiered implementor's (and debugger's) guides don't make sense to me.
  *> 
  *> These documents fill both roles.  (And that's why HR RFC is useful even if
  *> nobody ever builds a conformant host.  It does influence what gets built!)
  *> I guess we have to decide which goal is more important (or more worthy of
  *> being mentioned in the title).  The first goal is (so to speak) closer to
  *> my head, but the second is closer to my heart.
  *> 
  *> Publication of the current draft (perhaps with tweaks), be it as
  *> informational, historic, or whatever, is more relevant to the second goal.
  *> There is wisdom in there that isn't written down in any extant RFC.  Maybe
  *> we should change the title or something, but let's put the content out
  *> where people can use it.
  *> 
  *> /Bill
  *> 

Well said.

Bob Braden