W3C home > Mailing lists > Public > public-xg-webid@w3.org > October 2011

serialisation formats - Re: comments to references

From: Henry Story <henry.story@bblfish.net>
Date: Fri, 28 Oct 2011 13:23:50 +0200
Cc: WebID Incubator Group WG <public-xg-webid@w3.org>
Message-Id: <F1C078DA-7CE3-4274-B28F-B649AED78A44@bblfish.net>
To: Dominik Tomaszuk <ddooss@wp.pl>
Perhaps what we can do in order to list all the serialisation formats without looking weird is to explain that what we wish to do is bring them all together. We can list all the serialisation formats for X509 (PEM, ...)
then PGP, then the new JSON format by BrowserID, then the different formats people want to embed them in :
Html5, rdfa, n3 (for writing by hand), rdf/xml for Tim Bray, etc...

Doing this we can explain that all these formats are assertions about a public key it's relation to an agent, and some information about that agent. These can then be signed. So the semantics of these formats that interest us are these few relations.

Then we can show how people could publish each of these formats at the same URL with their mime type using content negotiation. 

I think we could then call this BableFish Identity, in honour of Douglas Adams.

  Of course as we all know this will cause greater and more destructive wars than ever, as each one group discovers that it's syntactic holy cow was just an arbitrary way of encoding a few relations.

	Henry


On 28 Oct 2011, at 11:51, Henry Story wrote:

> 
> On 28 Oct 2011, at 11:24, Dominik Tomaszuk wrote:
> 
>> My proposals to references in WebID spec:
>> 1. section 2.1:
>> "Alternate RDF serialization formats, such as N3 [N3] or Turtle [TURTLE], may be supported by the mechanism providing the WebID Profile document."
>> Most N3 documents is compatible with the Turtle (Turtle is subset of N3). These simple solutions like WebID most used Turtle. I believe that there is no need to list here N3, especially when it is not a complete list of serialisations.
> 
> I agree that sadly there are not many complete N3 parsers, so we should use turtle.
> 
>> Additionally, RDF Working Group concludes that they will not standardize N3 (they focus on Turtle).
>> 2. section 2.1 - the same paragraph
>> I propose to add RDF serbialization based on JSON. It is more
>> It is often used in web environments. I propose RDF/JSON [1]. RDF/JSON is created by RDF Working Group.
> 
> Just above you say that we should not making a list of all serialisations. Now you want one more, that is not standard. As we have only very few people whole have implemented the EARL reports for their endpoint, I also find it difficult for the moment to tell where we are standing.
> 
> http://www.w3.org/2005/Incubator/webid/wiki/Test_Suite
> 
> Having so many serialisations is probably starting to look a bit silly, especially for people coming from the hard nosed security world.
> 
> Before adding json I would even prefer to add plain PEM files for that community.
> 
> Perhaps we should have a version of this protocol as an RFC which just uses the PEM file?
> 
> 
> 
>> 3. I suggest to change Turtle references to more actual [2]
> 
> that's a good idea
> 
>> 
>> [1] https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-json/index.html#
>> [2] http://www.w3.org/TR/turtle/
>> 
>> Best,
>> Dominik 'domel' Tomaszuk
> 
> Social Web Architect
> http://bblfish.net/
> 

Social Web Architect
http://bblfish.net/
Received on Friday, 28 October 2011 11:24:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 28 October 2011 11:24:29 GMT