On Thu, Nov 07, 2013 at 04:49:19PM +0000, Jan Medved (jmedved) wrote:
> 
> I2RS is not chartered to develop data models  - it's chartered to develop
> *information models*. Not that I agree or disagree, I am just pointing out
> a subtle difference, which however has a practical impact: for example,
> for the topology model, we first developed a yang data model which was
> presented in the netmod WG. Then we also created a corresponding
> information model which was presented in the i2rs WG. For the I2RS RIB
> information model, there is a corresponding RIB yang data model, where we
> worked with Lada and aligned it with [1]. We would want to present it at
> the next IETF.
> 
> So, there are no overlapping activities - the way the I2RS charter is
> defined, the WG operates at a different level. It has yet to decide what
> data modeling language should be adopted for i2rs, or whether data models
> would be developed in the WG.
> 

Yes, I am aware of the difference. The point, however, is that
creating consistent data and information models that provide a
reasonable coverage is a huge task and it either requires very tight
coordination between WGs running at different speeds or a kitchen sink
WG that tries to get all done in one place. Both options are very
difficult to implement in the IETF. At the end of the day, it is not
useful for running networks if information models and data models
developed within the IETF are not consistent.

So from a larger perspective, there is an interesting scoping and
coordination challenge here that we need to manage somehow.

/js

-- 
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

Reply via email to