Re: [hybi] Proposed way forward for WebSockets

["Anne van Kesteren" <[e-mail]>]

On Thu, 29 Jul 2010 10:08:17 +0200, Martin J. Dürst  
<[e-mail]> wrote:
> On 2010/07/29 16:35, Anne van Kesteren wrote:
>> "Unrecognized fields can be safely ignored, and are probably either the
>> result of intermediaries injecting fields unrelated to the operation of
>> the WebSocket protocol, or clients that support future versions of the
>> protocol offering options that the server doesn't support."
>
> So, just to take an example, let's say it turned out that browsers sent  
> the User-Agent: header field, and some servers actually looked at it.
>
> Would your understanding of extensibility mean that we can explain this  
> as a (as of yet) not standardized future version/feature of WebSockets?  
> And that if it turned out that this were reasonably prevalent, and  
> remotely useful, somebody could write an RFC (as part of this WGs future  
> charter or as an individual submission) that would 'legalize' this  
> header.
>
> I think it would help at least my understanding of your extensibility  
> model if you could confirm/deny the above.

Yes, though preferably sending additional headers from the client is  
discussed with other parties beforehand on a relevant forum, such as  
[e-mail].


-- 
Anne van Kesteren
http://annevankesteren.nl/