W3C home > Mailing lists > Public > public-sweo-ig@w3.org > April 2007

Re: Layer Cake

From: Ivan Herman <ivan@w3.org>
Date: Sun, 01 Apr 2007 08:22:04 +0200
Message-ID: <460F4F8C.3090904@w3.org>
To: kidehen@openlinksw.com
CC: public-sweo-ig@w3.org

Hash: SHA1

Kingsley Idehen wrote:
> Ivan,
> I really need some clarification about the Layer Cake:
> 1. What does URI/IRI mean or imply in the context of the layer cake?

Usage of URI-s is fundamental to the Semantic Web. It *is* the core tool
to combine graphs. That is all... But that is essential

> 2. What does XML mean or imply (likewise) such that it occupies a slot
> without N3, Turtle, or TriX?

XML is not only in terms of the syntax format. It is also the fact that,
for example, we use XML Schema datatypes or, in SPARQL, function and
operator set that come from the XML World. Also: RDF/XML *is* has a
large usage base, it cannot be completely taken out from the picture. By
making it, sort of, optional on the figure, I think the right message is

> I am very picky about this Layer Cake update because I am (and I assume
> many others are too) really picky about clarifying the Semantic Web
> vision and message.
> At the start of my participation in mailing list  I made it very clear
> that disambiguation of RDF is vital to our common goal of eradicating
> (within reason) the dire misrepresentation of the Semantic Web vision
> and RDF Data Model that permeates our industry at large.
> From my vantage point:
> 1. URIs expose Data Sources on the Web (the concept of Data Sources is
> much easier to understand than "URI" without qualification)
> 2. XML plays a vital role in RDF, it's critical middleware
> infrastructure for GRDDL and other techniques for producing RDF from non
> RDF amongst other things, I know it's the W3C's official serialization
> etc., but none of this is congruent with its placement in the Layer Cake
> while N3, Turtle, and TriX are left out.
> Ironically, XML is what is good about RDF as well as what is bad about
> RDF (from the PR and general education perspective). This point of view
> has nothing to do with the XML itself, just everything to do with
> inadvertent misrepresentation in the context of the RDF Data Model (an
> integral part of the Semantic Web vision).
> I would like to suggest the following:
> 1. [URI/IRI (Data Sources)] as an enhancement of the [URI/IRI] box that
> exists right now
> 2. [RDF Data Model] as a replacement of the current [RDF Interchange: RDF]
> Re. point 2, we need to express the role of serialization / interchange
> and the types of serializations formats. Ideally, via little boxes
> within the Data Model box (towards the top which indicates the point
> where interchange occurs).
> I think we need to resolve this matter urgently and I am hoping we can
> arrive at some consensus via this mailing list.

This also shows that everybody sees something different in a layercake.
It is the elephant of the tale... The rule of URI/IRI is the core
technology there, I do *not* see that as equivalent to data sources. I
also do not see the reason why the Data Interchange: RDF is bad; it is
the core goal of a particular piece of technology...

I must say that I am a bit uneasy restarting the discussion on the
layercake diagram. There has been a looong discussion (before SWEO came
on board) on the SW Coordination Group on how to do this without
radically changing everything and I also had a long discussion with Tim
on that. It reflects a kind of a consensus right now... This type of
discussion can drag on....


- --

Ivan Herman, W3C Semantic Web Activity Lead
URL: http://www.w3.org/People/Ivan/
PGP Key: http://www.cwi.nl/%7Eivan/AboutMe/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf
Version: GnuPG v1.4.2.1 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

Received on Sunday, 1 April 2007 06:21:46 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:28:52 UTC