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

Comments to Semantic Web FAQ [2/2]

From: Uldis Bojars <uldis.bojars@deri.org>
Date: Tue, 17 Apr 2007 03:01:04 +0100
To: <public-sweo-ig@w3.org>
Message-ID: <E1Hdd0I-0002bR-DE@maggie.w3.org>

Hi All,
 
Here are the remaining comments to the Semantic Web FAQ.
 
> New questions:
 
These questions, while maybe mentioned in other answers, deserve a FAQ
entry.

 >> What formats can RDF data be encoded / represented in ?
 
RDF itself is a set of simple statements (triples) used to describe some
data - information about resources and relations between them. 
 
These statements can be encoded in a number of different formats - usually
text-based (Turtle, Notation 3, ...) or XML-based (RDF/XML). Generally it
does not matter which of these formats (or serialisations) is used to
express data - the actual information is represented in RDF triples and the
particular format is only the "syntactic sugar" which can be chosen
depending on implementation preferences. Data can be freely converted from
one format to another while keeping all the information described.
 
Answer to "What is RDF?" (4.1.) says that information described in RDF forms
a directed, labeled graph. This graph structure can be represented visually
as a "nodes and arrows" diagram and is often used in examples as an
easy-to-understand visual explanation. 
 
"RDF Primer" [ http://www.w3.org/TR/rdf-primer/ ] is a good material for
further reading on RDF.
[ and contains examples in RDF/XML, text notation and their graphical
representation ]
 
The link in 4.5. with link text "only one of the possible serialization" can
then link to this answer. 
 
>> How can I know if RDF data I am looking at is valid ?
 
RDF data must be valid in order to correctly process them. 
 
This is very much like XML which must be valid in order to correctly parse
and process the information. For data represented in RDF/XML it must be
valid XML with some additional requirements on top of that. You can use the
W3C RDF Validator [ http://www.w3.org/RDF/Validator/ ] to check the validity
of RDF/XML data and also render its graphical representation.
 
You can also use any of the well known RDF frameworks (such as Jena or
Redland) to parse and validate both text-based and RDF/XML representations
of RDF. Redland, for example, contains a command-line tool called "rapper" [
http://librdf.org/raptor/rapper.html ] which can parse various
serialisations of RDF (and validation report errors if found) and convert
them from one to another. 
 

> Existing questions:
 
>> 1.2.  What are the major building blocks of the Semantic Web? 

As already mentioned in previous email the 1st paragraph does not really
list the fundamental building blocks, at least some of them are mentioned as
differences.

Suggestion: can leave most of the paragraph as is, and create a bullet list
after it with 2 items (fundamental building blocks):
 - "RDF, which is one of the fundamental building blocks of the Semantic
Web, gives a formal definition for that interchange."
 - URIs (add descriptive text here, can use some of the text in 1st
paragraph )

Link the "RDF" in first bullet to answer "4.1.  What is RDF?"
Link the "URIs" in second bullet to answer "4.3.  Where is the "Web" in the
Semantic Web?" in the formulation as suggested by Sandro.

>> 1.4.  Will I "see" the Semantic Web in my everyday browser? 

Add a paragraph:

You can, however, see the "tip of an iceberg" or a human-readable display of
RDF data attached to webpages by using tools such as PiggyBank and Semantic
Radar. They are Firefox browser extensions which make use of RDF data which
can be found on web pages (for example LiveJournal user pages). While end
users will not have a need to see Semantic Web data (instead they will
benefit from better information systems built on top of it) it may be
helpful to developers to be aware of Semantic Web data present on web pages
so that they can use this information in their applications.

>> 2.7. . microformats 
 
The last part of the 1st sentence "and must be agreed upon by a community"
does not distinguish microformats from ontologies in any way. Both must be
agreed upon by a community or some body and for more complex ontologies can
take quite some time. Suggest to remove this part. 
 
Practical eRDF [ http://keithalexander.co.uk/erdf-article/ ] article has
some answers which we may use or reference to make this FAQ entry "sharper".

See "When do I use Microformats, and when do I use eRDF?" and "use
Microformats when / use eRDF when / use both at once!" entries.
This article also has a FAQ entry about RDF: "But what is RDF anyway? and
isn't it really complicated?"
( agree to what it's saying that fundamentally RDF is very simple )

Data described in microformats each address a specific problem area (and
they are usually used in XHTML when both human and machine readable data are
included on the page). RDF can represent any information including that
extracted from microformats present on the page. This is where microformats
can benefit from RDF - as a universal format to store and query data coming
from different microformats. 

RDF as a separate "data page" linked to from a web page is also useful when
the data on a particular web page and the machine readable data have
different scope. E.g., when the first web page contains only a part of data
available in a machine readable form (see "friends" list on many social
software sites where the friends list is paged, but a pure machine-readable
page such as FOAF may contain the full list of friends at once).

>> 3.4.  Are the SW tools as robust and as ubiquitous as, say, the xerces
XML parser? 

"On the open source domain Jena [and Redland], for example, can easily be
compared to xerces" - add a reference to Redland.

>> 3.5.  How do I put RDF into my (X)HTML Pages? 

Make the sentence "The best solution is to store the RDF separately and use
the URI-s to refer to the XHTML page and the link element in the XHTML page
to refer to the RDF content." a separate paragraph. Add to it: "This
technique is often called an RDF autodiscovery link."

[ http://lists.w3.org/Archives/Public/semantic-web/2006Nov/0146.html ]
contains links to more resources about RDF autodiscovery.

May also add an example from the FOAF Specification (if needed): "The
linking markup is as follows: <link rel="meta" type="application/rdf+xml"
title="FOAF" href="http://example.com/people/~you/foaf.rdf" />"

The link in 1st sentence with link text "usage of the meta and the link
elements in the header" is broken.

>> 4.6.  What role do ontologies and/or rules have on the Semantic Web? 

Add hyperlinks to information about rules on the web. 
Notation3 can be mentioned as a language that one can write rules in.

>> 4.12. Will W3C be standardizing any particular ontologies? 
 
Add SIOC [ http://rdfs.org/sioc/spec/ ] to the list of community-developed
ontologies in the 1st paragraph.

---

That's all of the comments.

Uldis

[ http://captsolo.net/info/ ]
Received on Tuesday, 17 April 2007 02:01:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:17:37 GMT