Re: New Version Notification for draft-troan-6man-universal-ra-option-03.txt

[Ted Lemon <[e-mail]>]

On Oct 7, 2020, at 4:01 PM, Brian E Carpenter <[e-mail]> wrote:
>   The Concise Binary Object Representation (CBOR) is a data format
>   whose design goals include the possibility of extremely small code
>   size, fairly small message size, and extensibility without the need
>   for version negotiation.
> 
> That's why future embedded systems are highly likely to include a CBOR
> encoder/decoder anyway, and as anyone who has used them knows, cbor
> loads() and dumps() make life very easy for the programmer.

This is great marketing copy. That said, having recently re-implemented a protocol that previously used a JSON-like data structure for transport as a TLV protocol, my experience has been that the TLV protocol is definitely less work. The bulk of the work is not encapsulation/decapsulation, but validation. It’s really easy to validate a TLV-based encapsulation because any spurious data is a syntax error. Validating a JSON dictionary that contains the same data requires you to make sure that all the keys are there, that there are no extra keys, that each data element is the right type, and so on. It winds up being quite a lot of code, more like an XML validator than a wire format parser.

The reason you want a JSON-like encapsulation for data is that you want to be able to send arbitrary data. If your protocol does not need the ability to send arbitrary data, a JSON-like encapsulation is always going to be more work, and more prone to security issues when the validator isn’t quite right.