While the question of WHERE the models finally get submitted is reasonable (a fact I made to the preliminary discussion around creating NETCONF/Yang models in the ONF for everything the wire protocol didn't easily cover), the fact that the models are being developed (as mentioned below) is more accidental than planned, IMHO.
That's the fundamental point - for me. Without THIS WG's mandate to do create such models and do so in an agreed upon and cohesive manner (which is what Jan is going on about - "here's the BASE model, and examples of how to extend it ... that's topology, let's use this as a method for the other shite on the list"), you're going to have either a very incoherent/potentially-chaotic result OR an eventual return to the entropy that has stifled innovation for years prior. So, IMHO, I'm happy to hear that work is going on in other WGs but worried about whether it is being tracked toward a common goal and (ultimately) forms a cohesive model in the whole - be it for network/topology or device. I'd add as well that other SDO-like organizations are considering doing some of this modeling in the gold-rush to create a "NorthBound API" for SDN, motherhood and apple-pie. So, if we keep on our current vector (providing no oversight and expecting things to organically blossom where interest lies) we may end up with a different result than any of us are contemplating. ________________________________________ From: [email protected] <[email protected]> on behalf of Juergen Schoenwaelder <[email protected]> Sent: Thursday, November 07, 2013 9:53 AM To: Russ White Cc: 'Eric Osborne (eosborne)'; [email protected]; 'Nikolay Milovanov'; 'Alia Atlas' Subject: Re: [i2rs] topology info model - what makes it a "network" model vs. a "device" model On Thu, Nov 07, 2013 at 09:21:38AM -0500, Russ White wrote: > > > If we have a device-centric model showing interfaces and so on, then > there's > > not a good way to express the learned IGP topology. Would we then need a > > different IM - perhaps as part of an IGP-specific IM - to communicate the > > topology learned via the IGP? Would that be preferable? > > Yes, you are going to need different network models for different protocols, > services, etc. There's not going to be any way to combine such models into a > "coherent whole." > Data models for different protocols such as OSPF or BGP have so far been done in WGs that care about those protocols and this has generally worked well as far as I can tell. We are now moving towards YANG models for configuration and state data and a general framework for YANG routing models has been defined in the NETMOD WG [1] (the next update of this document will go to WG last call). We expect that BGP, OSPF, ... specific extensions of this core routing model will be produced and we envision that this work takes place in the routing area, e.g., in WGs maintaining these routing protocols. Of course, we first need concrete proposals to start from. I think what I am saying is that (a) there is work going on outside of I2RS and we better avoid overlapping activities and (b) I like to remind you that work can be split and it is not necessary that I2RS creates all data models on its own. /js [1] http://tools.ietf.org/html/draft-ietf-netmod-routing-cfg-11 -- 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/> _______________________________________________ i2rs mailing list [email protected] https://www.ietf.org/mailman/listinfo/i2rs _______________________________________________ i2rs mailing list [email protected] https://www.ietf.org/mailman/listinfo/i2rs
