Re: http/2 and "extensions"

On 16 July 2014 04:04, Martin Thomson <martin.thomson@gmail.com> wrote:

>
> Implicit in any discussion of making anything an extension is that it
> is optional-to-implement.
>
>
The corollary​ to this is that extension authors have to consider, and
probably explicitly code into their extensions, how to deal with the
inevitable failure modes.

My feeling, coming from my encoded data thing, is that the more "real" the
extension (i.e. not just communicating nice-to-know metadata, but actually
doing something semantic), the more design and text is going to be
dedicated to dealing with unsupportive or broken peers.


-- 
  Matthew Kerwin
  http://matthew.kerwin.net.au/