Re: [NGO] NETCONF Data types

Juergen Schoenwaelder wrote:
> On Fri, Dec 07, 2007 at 03:28:23PM +0100, tom.petch wrote:
> 
>> On the other hand, it does seem a silly little implementation
>> restriction (now that I am not coding an assembler) to have ..16,
>> ..32 and ..64; BER.1 did it much better with its any length up to
>> approx 2**127 bits.
> 
> No size limit means that every implementor chooses his own limit (only
> very few people will go and implement arbitrary precision arithmetic
> on network devices). I think this very much contradicts the very
> purpose of a standard.
> 

I strongly agree with Juergen.
Both the agent and manager implementors want fixed sized numbers
to save time and memory, and not have to support more than one
way of encoding a given data type.

Unbounded integers are really not okay for data models.
This is one difference between an info model and a data model,
something Juergen knows all about.  We standardize data models for
NM for SNMP, and we should do the same thing for NETCONF.
The value of a standard is in its deployment, after all.

> /js
> 

Andy

_______________________________________________
NGO mailing list
NGO@ietf.org
https://www1.ietf.org/mailman/listinfo/ngo