Re: splitting HTTP section or not?

On Thu, 16 Jun 2011 13:50:53 +0200, Silvia Pfeiffer  
<silviapfeiffer1@gmail.com> wrote:

> On Thu, Jun 16, 2011 at 6:33 PM, Philip Jägenstedt <philipj@opera.com>  
> wrote:
>> On Wed, 15 Jun 2011 23:40:00 +0200, Jack Jansen <Jack.Jansen@cwi.nl>  
>> wrote:
>>
>>> I've been thinking a bit about whether or not we should split the  
>>> section
>>> on HTTP implementation
>>
>> I would support splitting out or removing the HTTP section, on the basis
>> that we do not intend to implement it. When/if we implement MF it will  
>> be
>> client-side only, relying only on byte range requests.
>
> But that's what 5.2.1 describes, so why remove it?
>
> Also, the spec is not just for browsers - it is for all sw that
> intends to support MF URIs.
>
> Why remove things now just before the spec gets into the hands of
> implementers? It's just a CR and not a PR.

OK, to clarify:

I support dropping section 5.2.2 Server mapped byte ranges because of the  
reasons stated above.

I also support dropping section 5.2.1 UA mapped byte ranges, because:

1. What it describes is mostly not specific to MF, it's something that  
generally applies to byte range requests.

2. Dealing with byte range requests and caching for <video> is quite  
messy, a high-level description like this doesn't help me as an  
implementor. The HTML spec says nothing about how to seek to a particular  
time offset using HTTP, and I don't think MF needs to either.

In short, I support removing it because it's not useful to me, and  
probably not to any browser implementor. If other implementors step  
forward and say that they find it valuable, then I will also not object to  
keeping it.

-- 
Philip Jägenstedt
Core Developer
Opera Software

Received on Thursday, 16 June 2011 12:27:09 UTC