Paul Hoffman <[email protected]> wrote: > > Of course. For the RDATA part of a JSON structure, you can be very > JSON-y, or very masterfile-y like draft-levine-dnsextlang has gone. I > would prefer being JSON-y because the rest of the data we are talking > about is already in that format, and we would not be reliant on anyone > understanding the new proposed format from draft-levine-dnsextlang.
I don't see draft-levine-dnsextlang as being particularly masterfile-y. What it gives you is a sequence of type + name pairs. These can be used to serialize to/from wire format or presentation format. They should be equally useful for serializing to/from JSON. Or to/from a web form for user-friendly editing. (Several RR types are too complicated so will need special handling; but you can just view this as extending the set of RDATA field types that the description language knows about.) A key advantage to having machine-readable RDATA descriptions is that they should be general-purpose so you don't have to keep reinventing the wheel each time you want a new representation: you can re-use the mapping between RR types and seqences of fields, and just add a new mapping between fields and your new representation. Tony. -- f.anthony.n.finch <[email protected]> http://dotat.at/ Forties, Cromarty: East, veering southeast, 4 or 5, occasionally 6 at first. Rough, becoming slight or moderate. Showers, rain at first. Moderate or good, occasionally poor at first. _______________________________________________ DNSOP mailing list [email protected] https://www.ietf.org/mailman/listinfo/dnsop
