[core] draft-ietf-core-links-json-01 review

["FOSSATI, Thomas (Thomas)" <[e-mail]>]

Hi Carsten,

we wanted to use the format in JSON link I-D for HTTP-CoAP proxy discovery.

So I took a look at it, and this is my brief review.

=-=-=-=

Both Abstract and Introduction say “[…] may be useful to represent these
collections in JSON”.  Perhaps worth adding some text to better
substantiate its useful-ness over RFC5988 serialisation?  I guess one of
the main motivations for preferring JSON over Link is the increased "web
API friendliness" (i.e. no added parsing cost for extracting link
information, format homogenisation, etc.).  Is it correct?

On the same topic: I find the argument given in the second para of the
Introduction -- ie. the "proactively try to stop formats' proliferation"
function of this spec — not very compelling.   But this might just be me
:-)

Also I think that it'd be nice to give the reader enough context to
understand how JSON link differs from other similar proposals (JSON
formats JSON-LD, JSON Reference, HAL, etc.) and why it is a better choice
than any of those for our use case.

Re: the implicit vs explicit rel="hosts" in Section 2, I'd be inclined to
add it explicitly because in the wider web it can’t be assumed, but would
love to hear from you and/or  others for reasons why it shouldn't.

In section 2.1, it'd be nice to have examples that show the less obvious
bits, e.g. multi-valued attribute to array, escape chars, etc.

=-=-=-=

That said, the doc is in pretty good shape!  Any plan for an updated
version?

Cheers, t