Re: Canonical configuration database order

["Randy Presuhn" <randy_presuhn@mindspring.com>]

Hi -

> From: "Andy Bierman" <ietf@andybierman.com>
> To: "Randy Presuhn" <randy_presuhn@mindspring.com>
> Cc: "Netconf (E-mail)" <netconf@ops.ietf.org>
> Sent: Friday, June 02, 2006 10:07 AM
> Subject: Re: Canonical configuration database order
...
> Hopefully, there are rules in place for canonical representation
> and sort order of UTF-8 strings, including internationalization
> and localization details.

There are lots of rules available.  We need to chose a set that
can reasonably be implemented in a way that will ensure interoperability.
I think that permitting localization to affect representation or ordering
would be a recipe for disaster, since it would require any entity potentially
needing to compare configuration data to be privy to the localization
parameters used by whatever thing(s) generated the data.

> I guess it is data type (and data modeling language) specific
> as to the exact meaning of ascending order.  I compare numbers
> as numbers, not strings, so 9.9 is less than 999 and -4 is less
> than 2.1.  Enumerations ascend in the order they are defined
> (aligns with SMIv2 and C), not alphabetic order of the enum values.
...

A consequence of doing things this way is that data could be compared
only by things with full knowledge of the schema.  That's not necessary
in the case of SNMP, and is quite useful.  This may be a bug or a feature,
but we should be aware of the implications for, for example, configuration
management systems that, in trying to identify changes in configuration,
will need to know which differences in representation are significant and
which are not.

Randy


--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>