Re: [netconf] crypto-types fallback strategy

[Kent Watsen <[e-mail]>]

Hi Juergen,

> Perhaps it is best to not look too much behind the curtain and just
> talk about using existing names, avoiding to dive into details what
> these names really are and where they come from. Perhaps we need to
> move to a more opaque type that can hold names of crypto algorithms
> and we point to well-known names (i.e., defined in some IANA
> registries) that apply to certain protocol contexts.

I use thinking that we'd use numbers only, or more precisely, the ASN.1 OID structure.   An OID's common "name" value is only "needed" for presentation; though it makes using `genarate-asymmetric-key" and `generate-symmetric-key` more complicated, as both actions need to take an "algorithm" parameter.

Kent // co -author