Re: protocols and W3C

On 6 Aug 2010, at 14:16, Dave Raggett wrote:

> On Fri, 2010-08-06 at 13:13 +0200, Thomas Roessler wrote:
>> On 4 Aug 2010, at 15:09, Dave Raggett wrote:
>> 
>>> The IETF is well known and highly regarded for its work on transport
>>> layer protocols, including HTTP, and provides an effective venue for
>>> work on HTTP, and related work such as streaming HTTP and Web sockets.
>> 
>> An observation that I got about this text from somebody around the IETF
>> crowd is that HTTP is a "transfer", not a "transport" protocol.
> 
> What is the difference between transporting and transferring something?
> However, I don't think we need to be too precise here, and suggest that
> that would actually be counterproductive.  If you think about a stack of
> protocols, then W3C is more likely to be interested in ones that sit on
> top of lower level ones that are defined by the IETF. 


Right. So, perhaps the language could read:

"The IETF is well known and highly regarded for its work on transport layer protocols, and on transfer protocols like HTTP.  It provides an effective venue.... blah blah"

>> 
>>> W3C is an appropriate venue for protocols at a higher level and with a
>>> strong focus on Web data formats and conformance requirements for Web
>>> user agents. This builds upon the skill sets of participants in W3C
>>> Working Groups.
>>> 
>>> Strong standards benefit from extensive review and implementation
>>> experience. This is why it is valuable for W3C Working Groups to review
>>> and coordinate with related work at the IETF, and vice versa. In some
>>> cases there will be multiple proposals, and these will have to play out
>>> in the market place, but solid peer review from a broad range of
>>> perspectives will benefit the market whichever solution wins out.
>> 
>> 
> 
> -- 
> Dave Raggett <dsr@w3.org> http://www.w3.org/People/Raggett
> 
> 
>

Received on Friday, 6 August 2010 12:21:18 UTC