- From: Adam Rice <ricea@google.com>
- Date: Thu, 21 Aug 2014 13:50:03 +0900
- To: Mark Nottingham <mnot@mnot.net>
- Cc: Greg Wilkins <gregw@intalio.com>, HTTP Working Group <ietf-http-wg@w3.org>
- Message-ID: <CAHixhFqbw00FrGSrCRS1rK_HqEj8osRXtXpj+DtYmU=tqFyBkQ@mail.gmail.com>
As a client, why would I add a header to my request that is going to cause the proxy to block it? What is the benefit to the user? On 20 August 2014 10:12, Mark Nottingham <mnot@mnot.net> wrote: > > On 20 Aug 2014, at 11:09 am, Greg Wilkins <gregw@intalio.com> wrote: > > > > > Mark, > > > > thanks for those links. > > > > I think the document itself needs to be a bit stronger on the intended > usage of the header. Currently it reads that this header field can be sent > by a client and that it can be ignored by the proxy. > > That's the intent. > > > Perhaps it should be little bit stronger and say that a proxy MAY > (SHOULD?) consider this header when deciding to create a tunnel or not. > > We can't retroactively require proxies to pay attention to a header; it's > up to them. This header is merely enabling those proxies who choose to act > upon the header. > > (Yes, that corresponds to MAY, but we try to avoid overusing it, else our > documents get filled with MAYs. It's not really a conformance requirement, > it's just a statement.) > > Cheers, > > > > > > cheers > > > > > > > > > > > > > > > > > > On 20 August 2014 10:45, Mark Nottingham <mnot@mnot.net> wrote: > > Greg, > > > > See: > > > https://httpwg.github.io/wg-materials/ietf90/IETF90_draft-hutton-httpbis-connect-protocol.pdf > > > https://github.com/httpwg/wg-materials/blob/gh-pages/ietf90/minutes.md#draft-hutton-httpbis-connect-protocol > > > > Cheers, > > > > > > On 20 Aug 2014, at 10:40 am, Greg Wilkins <gregw@intalio.com> wrote: > > > > > Mark, > > > > > > Is there a specific use-case motivating this additional header? ie > are there situations that a proxy can use this to do more than just > log/debug a tunnel? > > > > > > I'm certainly not opposed to having the additional information that > this header provides, but I'd like to know what advantage there is for a > client to include the header. If there is none, then it is not likely to > be sent. > > > > > > cheers > > > > > > > > > > > > > > > > > > -- > > > Greg Wilkins <gregw@intalio.com> > > > http://eclipse.org/jetty HTTP, SPDY, Websocket server and client that > scales > > > http://www.webtide.com advice and support for jetty and cometd. > > > > -- > > Mark Nottingham https://www.mnot.net/ > > > > > > > > > > -- > > Greg Wilkins <gregw@intalio.com> > > http://eclipse.org/jetty HTTP, SPDY, Websocket server and client that > scales > > http://www.webtide.com advice and support for jetty and cometd. > > -- > Mark Nottingham https://www.mnot.net/ > > >
Received on Thursday, 21 August 2014 04:50:37 UTC