On Wed, Aug 19, 2015 at 11:40 AM, Juergen Schoenwaelder < [email protected]> wrote:
> On Tue, Aug 18, 2015 at 09:38:04PM -0400, Lou Berger wrote: > > > > Why is this a meta-model? > > > because the aim to to show how other models inter-relate rather than > > define the details of all models. As stated in the intro: > > > > This document aims to provide an extensible structure that can be > > used to tie together other models. It allows for existing, emerging, > > and future models. The overall structure can be constructed using > > YANG augmentation and imports. > > > > You can easily extract relationships between YANG data models out of > YANG language constructs such as import relationships, leafref path > restrictions, augment references, etc. > > - Why do you think relying on language constructs does not work? > > - Why do you believe that relying on a certain hierarchy that several > independent SDOs and vendors have to augment and fill consistently > is going to be lead to a robust solution? > > If YANG adoption continues as it looks right now, we will count > modules in the thousands in a couple of years originating from many > different organizations. I would trust tools that are able to > interpret YANG language constructs to scale up to such a scenario. I > doubt a simple hierarchy of containers (which will essentially become > unmaintainable once it is used by several standards) will be of much > value in 10+ years. > > Andy's early work on YANG packages that provide machine readable > meta-data is a useful direction to explore further. > > I took the conformance stuff out of the YANG packages draft because there were concerns before that it conflicted with YANG module conformance. It seems like the "augment with mandatory nodes" type of problem indicates we need a conformance model that is capable of describing an API that spans more than one YANG module. > /js > > PS: As a spare-time Unix system administrator, I recall several fancy > proposals back in the 1990's how to best organize your /usr/local > to make software installation and sharing easy. It seems none of > these approaches that are essentially relying on naming > conventions succeeded in the long run. Instead, we got software > package systems that have sufficient machine readable > meta-information to make software management easy and reliable. > > -- > 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/> > Andy
_______________________________________________ rtgwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/rtgwg
