- From: Mark Nottingham <mnot@mnot.net>
- Date: Thu, 10 Sep 2026 11:12:35 +1000
- To: "Roy T. Fielding" <fielding@gbiv.com>
- Cc: ietf-http-wg@w3.org, draft-ietf-sidrops-publication-server-bcp.all@ietf.org, last-call@ietf.org, sidrops@ietf.org
> On 10 Sep 2026, at 10:18 am, Roy T. Fielding <fielding@gbiv.com> wrote: >> A resource that supports one content coding also needs to send Vary, as it also needs to support the identity encoding. > > Nope; identity is one content coding. A resource that is only available in > the gzip encoding, for example, is always going to be in a message with > a Content-Encoding of "gzip", and thus does not need to send Vary because > the representations do not vary (no matter what advisory info a client might > send in the Accept-Encoding request field). > > Yes, it's confusing to talk this way about identity, but it is precise. ;-) Resources that are only available with a compressed coding but not identity are an anomaly -- bordering on pathological. Much more common (and problematic) is the case that a resource supports both but doesn't send Vary correctly -- hence the suggested text. Cheers, -- Mark Nottingham https://mnot.net/
Received on Thursday, 10 September 2026 01:12:46 UTC