Re: Encouraging a healthy HTTP/2 ecosystem

In message <735E34EE-70B3-4140-8EF4-03BA19D00F9E@mnot.net>, Mark Nottingham writes:

>> For instance the $BIGWWW focus on running over TLS means that other
>> high-performance applications, such as News, TV and porn gets
>> saddled with a small and horribly inefficient (2^n-1, really ?)
>> framesize.
>
>Making smaller, more targeted proposals might help; e.g., a well-aligned 
>but still small frame size.

I did.

It was called jumbo frames.

It got shot down.

You just did so again a minute ago in another email.

>Unfortunately, you've left it until very late to get involved in 
>detail, which makes getting such changes in more and more difficult.

Come of it Mark...

You know perfectly well that I already from the start objected to the
accelerated timeframe, because it only left room to gold-plate SPDY
with all its warts.

Some of us tried to squeeze out a competing draft nontheless, only to
be steam-rolled with the "no time for that..." argument and none of
the points made in that draft ever revisited systematically and whenever
we raise them, they get shut down.

Given the appearance of a pre-ordained result and no willingness
to even accept the legitimate concerns of proxy implementors, I had
better things to do.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.

Received on Wednesday, 2 July 2014 11:48:21 UTC