Re: [Json] [Technical Errata Reported] RFC8259 (7603)

[Tim Bray <[e-mail]>]

The discussion around this report made me feel bad and as partial reaction,
Paul and I cooked up https://datatracker.ietf.org/doc/draft-bray-unichars/
which with luck will help future draft authors deal with these issues.

On Aug 17, 2023 at 11:18:31 AM, Carsten Bormann <[e-mail]> wrote:

> On 2023-08-17, at 17:37, Tim Bray <[e-mail]> wrote:
>
>
> Yes, you’ve been clear about what you want, but obviously we can’ t do
> that because JSON has been clear for years that naked surrogates are
> allowed (obviously this makes me unhappy) and we can’t retroactively assert
> that all these formerly compliant JSON documents are no longer valid. Not
> without chartering another WG.
>
>
> Tim,
>
> RFC 4627 has been published in 2006.
> For the 17 years and the two follow-on documents RFC 7159 and RFC 8259,
> the phrase that we are discussing based on errata report 7603 hasn’t
> changed, including over 5 years after publication of the “Internet
> Standard” RFC 8259.
>
> This is an approved document, on the highest level of approval we have in
> the IETF.
>
> The consensus that this approval is based on includes quite a few people
> who know what the phrase means and how it is imperfect as agreed, but
> expressing a delicate balance.
> Throwing the table on this now would send a signal that I don’t want to
> have to explain.
>
> This is all I’d need to say.
>
> A few more cents on the technical side:
>
> Note that the term you use above, “valid”, isn’t even used in RFC 8259.
> JSON as ECMA sees it, is a syntax only (this is even in the title of
> ECMA-404).
> The JSON specifications that we have, have no concept of what a “valid”
> text would be — Section 8 of 8259 is just emergency surgery around the
> issues created by this.
>
> I admire the charter artist who wrote for the jsonbis charter that the
> main task of the work from 7159 forward was to “fully acknowledge the
> syntax definition in ECMA-404” [1] — which RFC 8259 undoubtedly does.
>
> Anyway, I’ll shut up now (*), and I expect this errata report to become a
> textbook example for RSWG and friends for how we need a better
> understanding of the errata process.
>
> Grüße, Carsten
>
> [1]: https://datatracker.ietf.org/wg/jsonbis/about/
> (*) Unless someone is wrong on the Internet [2], of course.
> [2]: https://xkcd.com/386/
>
>