Re: comments to references

On 28 Oct 2011, at 14:49, Tim Berners-Lee wrote:

> 
> On 2011-10 -28, at 07:07, Dominik Tomaszuk wrote:
> 
>> On 28.10.2011 11:51, Henry Story wrote:
>>> Perhaps we should have a version of this protocol as an RFC which just uses the PEM file?
> 
> No no no.  You can't just have variations of the protocol which 
> will not interoperate!  It is going to be a gargantuan challenge to get this
> to take off and everyone to adopt it, but the one thing you give adopters
> is the fact that the system will work and scale and can go viral.
> 
> If you divide the population into two incompatible halves, they will each wither away.
> 
> And if you make and RDF and a W3C spec, then please make them the same text.
> Don't let the two specs be a way for people to avoid coming to consensus and sweep
> actual differences under the rug.

Ok, I agree. Content negotiation on specs that don't quite say the same thing
could lead to misunderstandings :-)

Now, These older formats often have some sort of magic associated with them, 
by which I mean that a lot of vocal people in the security community could be able
to understand what we are doing if we allow them to pop a certificate on the hard
drive with the right mime type (and the SAN pointing to the right place), and
if they can just do cert comparison. 

We have an issue open for this 
ISSUE-6: using ASN.1 formats for WebID description

Of course then there are debates about which formats to use. And the horrible thing
is that if we allow all of them then the simple protocol, may end up becoming super
complicated! 

If enough of that community joined I would be in favour of making it required. But
I have no way of knowing if they are interested in this before hand. So one thing we
were thinking of is making the implementations vote for what is acceptable: i.e. if someone
creates a large proportion of one format out there then it becomes a requirement.

Perhaps we should list the ones that are currently supported: rdfa, n3, rdf/xml and have a separate page for others that we would support if people are serious about it.

Henry

> 
> Tim
> 
> 
>> Do you mean RFC 1421?
>> 
>> D.
>> 
>> 
> 

Social Web Architect
http://bblfish.net/

Received on Friday, 28 October 2011 13:34:55 UTC