Alex,

Thanks for the pointer to RFC3444; I'd read it about 6 months ago to
clarify in my own mind the difference between an information model and a
data model.  I think it's very useful.

On Wed, Jan 23, 2013 at 5:00 PM, Alexander Clemm (alex) <[email protected]>wrote:

>  For an overview of information and data models (and the distinction
> between them), there is also RFC 3444 http://www.ietf.org/rfc/rfc3444.txt
> ****
>
> ** **
>
> Just as a comment, I like the pyang-ish data tree as a summary of the
> model structure as well. A short-hand for some of the “cornerstones” of the
> model, then more specifics can be found in the more formal definition.  I
> would think what is currently there should provide more or lesswhat is
> needed, hence I’d hope the group would be able to just reuse that (building
> any extensions on top if and when needed, but not introducing a new
> variation).
>

Agreed - I'd prefer to avoid a variation.

Alia


>
> Kind regards****
>
> --- Alex****
>
> ** **
>
> ** **
>
> *From:* [email protected] [mailto:[email protected]] *On Behalf
> Of *Alia Atlas
> *Sent:* Wednesday, January 23, 2013 1:44 PM
> *To:* [email protected]
> *Subject:* [i2rs] format for information models****
>
> ** **
>
> As part of  the proposed i2rs charter, the WG would need to produce
> information models.  I know there's****
>
> lots of interest in this and it would be great to agree on a format that
> we'd like to see the information models****
>
> in.****
>
> ** **
>
> From my perspective (learning all the time), an information model is to
> describe the information and purpose****
>
> of the data needed in the model, without all the details that make the
> reasoning and structure hard to understand****
>
> (and also take lots of time to develop).  Of course, there's Wikipedia's
> definition (http://en.wikipedia.org/wiki/Information_model)****
>
> and various RFCs that include information models (RFC 5102, 3585, etc).
>  Frequently, the information model****
>
> can just be in English.****
>
> ** **
>
> Based on some side conversations, I'd like to see if we can discuss and
> agree on a format to try for the information****
>
> models.  What I'd suggest is basically a data-tree (similar to the
> TreeOutput from pyang - seen at ****
>
> http://code.google.com/p/pyang/wiki/TreeOutput) and then information per
> data item.  The required information would ****
>
> be a description, a type, and, if appropriate, a reference (to where else
> the concept is defined).****
>
> ** **
>
> What do you all think?  Any better/different ideas?****
>
> ** **
>
> Alia****
>
> _______________________________________________
> i2rs mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/i2rs
>
>
_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs

Reply via email to