[ZenoArrow]

Thanks for the clarification. Out of interest, did you evaluate building a Lisp DSL? If so, any pros and cons you identified from that approach? I would've thought it would have been a quicker option to get going, though if you've reached version 1 you've put in the groundwork required now.

[camoberg]

I know that parts of the original design team for YANG are old lisp-fans, but I can't recall the option of developing a lisp-based DSL ever being evaluated. I would personally suggest that the fit would not be very good, since YANG is fundamentally a modeling language that "models the hierarchical organization of data as a tree in which each node has a name, and either a value or a set of child nodes." (https://tools.ietf.org/html/rfc7950#section-4.1).

But I happen to personally think that a datalog- or lisp-based policy language for YANG-modelled data sources would be a very interesting concept.

[ZenoArrow]

> "models the hierarchical organization of data as a tree in which each node has a name, and either a value or a set of child nodes"

That matches the structure of Lisp. Lisp code is basically set out as an AST. A Lisp node basically either contains a function and a list, or a function and simple value(s). The former can be thought of as a parent node, the latter can be thought of as a child node.

[camoberg]

So what you're saying then is that YANG could be a data modeling language for Lisp ;-)