Re: [OPSAWG] AD Review of draft-ietf-opsawg-smi-datatypes-in-xsd-05.txt

["Natale, Bob" <[e-mail]>]

Hi Juergen,

You were extremely helpful in getting to the correct XSD representation of IpAddress, so I hate to respond critically to your comment below...but I feel I must on two counts:

1. The requirements list for the selection of mappings also says this:
- - - - -
   R4.  The XSD datatype specified for a given SMI datatype MUST
        represent any special encoding rules associated with that SMI
        datatype.

   R5.  The XSD datatype specified for a given SMI datatype MUST include
        any restrictions on values associated with the SMI datatype.
- - - - -
So, the selected mappings for IpAddress and ObjectIdentifier go to "string" to enable the proper restrictions on values in a reasonable way.

2. I believe that tossing "marketing" into the ring is either a red herring or a low blow or both (nothing like getting smacked below the belt with a red herring :).  I personally consider it a worthy engineering goal when doing coding translations to be as faithful (in the sense of fidelity to the original) and direct (in the sense of the minimum necessary "extras" to achieve the true meaning of the original.  It's reasonable to suggest more accurate synonyms, but the objective is engineering quality, regardless.

Man, this is a tough place to complete a simple task. :(

Or, maybe I'm just too sensitive and/or in too much of a rush...after all, what's a couple of years for what should be a straightforward mapping of a few datatypes...?

Cheers,
BobN

-----Original Message-----
From: Juergen Schoenwaelder [mailto:[e-mail]] 
Sent: Wednesday, June 17, 2009 12:42 AM
To: David Harrington
Cc: Natale, Bob; 'Romascanu, Dan (Dan)'; [e-mail]
Subject: Re: [OPSAWG] AD Review ofdraft-ietf-opsawg-smi-datatypes-in-xsd-05.txt

On Wed, Jun 17, 2009 at 12:17:53AM +0200, David Harrington wrote:
 
> Personally, I think fidelity and faithful are the right words.
> 
> From Merriam-Webstre Dictionary:
> 1 a: the quality or state of being faithful b: accuracy in details :
> exactness
> 2: the degree to which an electronic device (as a record player,
> radio, or television) accurately reproduces its effect (as sound or
> picture) 
> 
> I think that Bob could state how this word was used to describe "the
> most exact reproduction of the SMIv2 datatype possible with XSD". He
> could also quote from the Merriam-Webster definition.
> 
> "most direct" also says ""the most exact reproduction of the SMIv2
> datatype possible with XSD"
> 
> Of course, you could use exactness:
> 1 : exhibiting or marked by strict, particular, and complete
> accordance with fact or a standard 
> 2 : marked by thorough consideration or minute measurement of small
> factual details 
> 
> #1 applies when you can do an exact translation, and #2 applies when
> you need to find the "most direct" almost-equivalent.

Well, I could easily claim that the "most exact" or "most direct" or
"faithful" translation of OCTET STRING (SIZE(4)) would be a hexBinary
of the same length. Still, we translate IpAddress into a textual
string containing a dotted number representation of the 4 octet value.

I am not questioning this translation of IpAddress here - but I do
share Dan's concerns about a marketing like language that claims that
this translation is the "most exact" or "most direct" or "as faithful
as possible" translation.

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: [phone]         Campus Ring 1, 28759 Bremen, Germany
Fax:   [phone]         <http://www.jacobs-university.de/>