Re: [Doh] DoH client-server interoperability vs. strict HTTP parameter checking

[Patrick McManus <[e-mail]>]

On Thu, Jun 6, 2019 at 1:14 PM Vladimír Čunát <[e-mail]>
wrote:

>
> On 6/1/19 11:22 PM, Patrick McManus wrote:
>
> 8484 does not, and really cannot, proscribe the error handling for whether
> the rest of the http server fixes the issue
>
> I fail to see reasons why an RFC can't (or shouldn't) specify behavior
> around unknown parts in the URL.  In this case I suspect that a clear
> decision will be better than leaving it unspecified, *regardless* of
> which way it will be decided.  I believe the robustness principle is about
> handling messages that do *not* conform to such a specification.
>

The DoH service is scoped by the URI template.. things that match that
template are defined by 8484, but HTTP servers certainly can implement
other services (such as the hypothetical doh fixup gateway) on other URIs
and 8484 cannot define them. This independence is an important part of
HTTP. So this particular boundary is a really hard place to put a hard
error requirement on.