Re: [netmod] float vs. decimal (fwd)

[Juergen Schoenwaelder <j.schoenwaelder at jacobs-university.de>]

On Mon, Oct 26, 2009 at 03:27:34PM +0100, David Spakes wrote:

> Between March 11 and June 9, there was a lengthy discussion on this
> list that resulted in a proposal to add a derived type called 'real'
> to draft-ietf-netmod-yang-types-04.  The proposal was submitted in
> its final form (below) before WGLC for draft-ietf-netmod-yang-types-03
> ended on June 10.
> 
> The draft-ietf-netmod-yang-types-04 document is out now and I see that the
> new derived type was not included.  Juergen's alternative proposal to make
> decimal64's fraction-digits substatement be optional (15-May) is not
> reflected in draft-ietf-netmod-yang-08 either.
> 
> Was this omission done intentionally, and if so, why?

Thanks for bringing this up again. Reviewing the mailing list
discussion thread "float vs. decimal", it is difficult for me to
determine a concensus position. What I saw were opinions to

a) add the proposed 'real' typedef
b) make fraction-digits optional, more closely following the XSD approach
c) leave things as they are (not clear the 'real' is needed urgently
   enough to add it now)
d) remove a bunch of other builtin types from YANG and make them typedefs
   instead

So the omission was done intentionally to the extend that I as the
editor do not understand the edits the WG agrees to (but the truth is
that this issue slipped through).

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: +49 421 200 3587         Campus Ring 1, 28759 Bremen, Germany
Fax:   +49 421 200 3103         <http://www.jacobs-university.de/>