For formats that, for legacy reasons, support multiple encodings, the
benefit is that iäthe BOM unambiguously signals UTF-8. For UTF-8-only
formats, the benefit of not treating the BOM as an error is to allow
authoring with tools designed for the kind of formats where the BOM
actually signals UTF-8 relative to other possibilities.

On Mon, Aug 19, 2019, 04:08 Tony V E <tvaneerd_at_[hidden]> wrote:

> Why bother then?
>
> What's the benefit of a BOM for UTF8?
>
>
> Sent from my BlackBerry portable Babbage Device
> Original Message
> From: Tom Honermann
> Sent: Sunday, August 18, 2019 8:06 PM
> To: Henri Sivonen
> Cc: ben.boeckel_at_[hidden]; unicode_at_[hidden]
> Subject: Re: [SG16-Unicode] BOM in JSON (was: Re: SG16 meeting summary for
> July 31st, 2019)
>
> + Ben.
>
> Thank you, Henri! This is very helpful!
>
> Ben, the context here is whether we’re ok with producers of the dependency
> format you specified producing a UTF-8 BOM. It looks like we should be ok
> to allow them to (optionally) do so given that both of the specs below
> allow consumers to remove one.
>
> Tom.
>
> > On Aug 15, 2019, at 2:12 PM, Henri Sivonen <hsivonen_at_[hidden]> wrote:
> >
> >> On Thu, Aug 15, 2019 at 5:16 AM Tom Honermann <tom_at_[hidden]>
> wrote:
> >> - Are we ok with allowing a BOM (JSON doesn't permit one)?
> >
> > Consuming JSON from a byte source in the Web Platform only supports
> > UTF-8 but removes the BOM if there is one. There is no corresponding
> > authoring conformance requirement in the Infra Standard, but the
> > practical effect is that the BOM does not fail to parse but doesn't
> > signal anything.
> >
> > https://infra.spec.whatwg.org/#parse-json-from-bytes
> >
> > The IETF wording requires producers to use UTF-8 without a BOM but
> > allows consumers to remove the BOM if it's there, so the Infra
> > Standard language and the IETF RFC are compatible on this point.
> >
> > https://tools.ietf.org/html/rfc8259#section-8.1
> >
> > (Apologies if this distinction between producer and consumer
> > conformance requirements was already made in the meeting.)
> >
> > --
> > Henri Sivonen
> > hsivonen_at_[hidden]
> > https://hsivonen.fi/
> > _______________________________________________
> > SG16 Unicode mailing list
> > Unicode_at_[hidden]
> > http://www.open-std.org/mailman/listinfo/unicode
>
> _______________________________________________
> SG16 Unicode mailing list
> Unicode_at_[hidden]
> http://www.open-std.org/mailman/listinfo/unicode
>