Re: Push and Caching

[Martin Thomson <[e-mail]>]

On 25 August 2014 15:58, Mark Nottingham <[e-mail]> wrote:
>
> On 26 Aug 2014, at 2:44 am, Martin Thomson <[e-mail]> wrote:
>
>> On 22 August 2014 10:30, Mike Bishop <[e-mail]> wrote:
>>> "While the stream identified by the promised stream ID is still open" - meaning that as long as the client has asked for it before the server has finished sending it?  That's a fairly small amount of time, particularly if the resource is very small, but sounds like a good starting point.
>>
>>
>> How would people feel if I removed that clause.  They can be
>> considered validated.  The period over which that validation applies
>> is then no different to a regular request/response exchange, which
>> suffers all of the same sorts of thorny and ambiguous validity issues.
>> (i.e., By avoiding an attempt to define validity expressly, we're not
>> making it worse, even if we're not tackling the issue fully.)
>
> You mean removing the entire sentence, or just the clause quoted above?


I was thinking just the quoted clause.  I think that we need this to
cover the no-cache case (and must-revalidate), but I don't think that
we need to fix the problem.

BEFORE:
Pushed responses are considered successfully validated on the origin
server (...) while the stream identified by the promised stream ID is
still open.

AFTER:
Pushed responses are considered successfully validated on the origin
server (...).

How about the past tense instead?

TAKE TWO:
Pushed responses are considered successfully validated on the origin
server (...) at the time that the response is generated.

Does this avoid the unended implication?