Re: [Json] On flat vs nested JSON encoding style

[Anders Rundgren <[e-mail]>]

On 2016-02-05 08:41, Carsten Bormann wrote:
> Anders Rundgren wrote:
>> 18 months
>
> JSON has been around for more than a decade.  It would take another
> decade for a change that goes so deep into the fundamentals to become
> widespread (and sufficiently debugged* on enough platforms to become
> reliable).

I have successfully tested Chrome, Firefox, Safari, My Java library, and a Python
patch against a file with 100M of random and selected IEEE 64-bit values.


> Wake me up when, say, ArduinoJson supports your form of canonicalization.

There are different markets with different needs and traditions.  Must they
all use the same standard(s)?

An  acquaintance of mine working for a major IoT vendor (who is very active
in the IETF/JSON space), claims that the IoT market rather will go for CBOR.

FWIW, I'm claiming that the Payment industry won't buy into JOSE, they will
probably build something of their own and ES6/V8 serialization makes that a
viable option.

PHB also seems to be experimenting with ways to get around the Base64URL dogma:
https://mailarchive.ietf.org/arch/msg/acme/hLmlH1lEEQ2GgkhgDnl5bbhhNXQ

Standards are fun...the more the merrier! :-)
https://xkcd.com/927/

Regards
Anders

>
> Grüße, Carsten
>
> *) Decimal number representation code is hard.  (I'm too lazy to look up
> the CVEs to provde this.)  Even sorting text is not exactly easy.
>