Re: http/2 prioritization/fairness bug with proxies

On 2/4/2013 7:02 AM, Amos Jeffries wrote:
  Which goes to show why specifying an algorithm for handling opaque
> client-selected prioritization is the wrong way to go about this.
>
> You have presented a good argument for specifying a set of standardized
> priority labels with criterion for setting each label.

One possible gotcha there is that not everything is a browser. A server 
application issuing, say, web service requests may have different 
pasterns of usage than a browser client used by an typical human user.

On another hand, the example of multiple TCP connections in HTTP/1.1 and 
in particular, "download accelerators", illustrates that some client 
software will be developed to "game the system" any way it can; it's 
going to be hard to be fair.

Received on Monday, 4 February 2013 15:56:20 UTC