Re: The robustness principle, as view by user agent implementors (Re: Working Group Last Call: draft-ietf-httpbis-content-disp-02)

Adam Barth wrote:
> 
> (There's a little more to the story to explain why user agents also
> want a complete semantic theory, one that defines the semantics of all
> messages, but I can explain that in another message.)
> 

I already understand your concerns.  But you still haven't answered my
questions about why the concerns of browsers apply to all user agents,
or anyone else implementing HTTP for any other purpose.  I don't see
the relevance, and the rationale for standardizing nonconformant syntax
still eludes me.

HTTP isn't meant as a guideline for building user agents.  Unlike HTML,
the concerns of browser vendors aren't of paramount importance; and in
this case, not even in-scope.  I doubt HTTPbis would get IETF approval
if it were based on dictating how user agents interpret nonconformant
syntax instead of referring to other RFCs where conformant syntax is
defined, justified by "IE and Chrome do it that way."

-Eric

Received on Sunday, 3 October 2010 19:48:10 UTC