Re: Murray Kucherawy's No Objection on draft-ietf-httpbis-alias-proxy-status-05: (with COMMENT)

Hi Tommy,

On Thu, Oct 26, 2023 at 8:04 AM Tommy Pauly <tpauly@apple.com> wrote:

> The main reason for being a SHOULD here (in my opinion) is that this would
> be unenforceable as a MUST — the client couldn’t tell if the order was
> different. The order also isn’t relevant for most cases, where the client
> would just be looking to see if it recognizes one of the names in the list
> (as a tracker, etc). Having it be in order is nice for consistency and for
> clarity when logging, etc.
>

I understand the reasoning, but I don't think that justifies a SHOULD
because it doesn't support interoperability.

I think you should just say what you said here: It simplifies debugging and
other operational work to have these in a particular order, so doing so is
considered the best approach.

-MSK

Received on Saturday, 28 October 2023 18:27:44 UTC