We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.

Leveraging Information and Intelligence

David Linthicum

The Joys of Blogging RDF

user-pic
Vote 0 Votes

Yesterday I received a Twitter message from Kingsley Uyi Idehen, who asked why his comments were not posted on the blog. I told him that I saw one comment posted, but no text.

Today I see a response to my post on his blog, I'm assuming thinking that he was somehow censored. His blog is posted the OpenLink Software site, he's the CEO.

Just to be clear I never censor comments unless they are spam. Moreover, I've never deleted a comment here on ebizq.net. So, I wanted to make sure that I pointed out his response specifically, which I thought was thoughtful, insightful, albeit a bit passive aggressive.

Here goes.

"Inaccuracy #1:

Resource Description Framework (RDF), a part of the XML story, provides interoperability between applications that exchange information.

Correction #1:

RDF and XML are not inextricably linked in any way. RDF is part Data Model (EAV/CR style Graph) with associated markup and data serialization formats that include: N3, Turtle, TriX, RDF/XML etc."

He's right there, providing a specific definition, but not sure that makes my first statement inaccurate.

"RDF, According to W3C specifications and Mozilla's documentation":

"The RDF format allows to mix together two XML documents into a single one, describing the relations in the data,"

Did not mean to imply they are "inextricably linked." In fact, did not say "linked." The point I was attempting to make is that RDF provides a method for conceptual description or modeling of information, and really more if it's used within the domain of data integration, to be more clear.

"Inaccuracy #2:

RDF uses XML to define a foundation for processing metadata and to provide a standard metadata infrastructure for both the Web and the enterprise.

Correction #2:

RDF/XML is an XML based markup and data serialization format.As a markup language it can be used for creating RDF model records/statements (using Subject, Predicate, Object or Entity, Attribute, Value). As a serialization format, it provides a mechanism for marshaling RDF data across data managers and data consumers."

Again, he's not wrong, providing some additional details, but neither am I. See RDF and Metadata," by Tim Bray. "Resource Description Framework, as its name implies, is a framework for describing and interchanging metadata."

"Inaccuracy #3:

The difference between the two is that XML is used to transport data using a common format, while RDF is layered on top of XML defining a broad category of data.

Correction #3:

See earlier corrections above."

See my earlier comment.

"Inaccuracy #4:

When the XML data is declared to be of the RDF format, applications are then able to understand the data without understanding who sent it.

Correction #4:

You do not declare data to be of RDF format. RDF isn't a format it is a data model (as stated above). You can "up lift" or map data from XML to RDF (hierarchical to graph model mapping). Likewise you can "down shift" or map data from RDF to XML (example: SPARQL SELECT query patterns "down shift" to SPARQL Results XML, which isn't RDF/XML, while keeping access to graphs via URIs or Entity Identifiers that reside within the serialization)."

"...lightweight multipurpose extensible metadata description and syndication format." From W3C. However, data model is a better description of RDF.

Not sure what the objective of Kingsley Uyi Idehen was here. He seems to know his RDF stuff, and I suspect is immersed in it more so than I. If he wanted to expand on my blog post, fine. Not sure of the value of taking each thing out of context, providing his own take on it, and calling me inaccurate.

I would have been happy to clarify or correct things, was I asked. I was not.

I suspect there will be follow up posts around this. This is not an RDF blog, so I'm going to end it here.


4 Comments

David,

In my original comment which never surfaced, I opened by saying: It's been a while.

I am the same "Kingsley Idehen" that you would have stumbled across in the early '90s during the early days of Open Database Connectivity (ODBC).

I am not a "Semantic Web" zealot (a zillion miles from that).

If you know me well (and I think you do or at least google up etc..). I am about Open Data Access and Data Integration. It so happens that my passion and vision happen to intersect with the "Linked Data" meme.

Bearing in mind I've known you (at least your commentary) since the '90's and I have a sense of your audience, I felt compelled highlight the inaccuracies in your post as best I could.

If there are any items you want to clear up. I will invest the time to comment until we are in sync :-)

Hi, David --

Maybe I can help make this a bit clearer...

You said, #1, "Resource Description Framework (RDF), a part of the XML story, provides interoperability between applications that exchange information."

RDF is not part of the XML story, nor vice versa. RDF has no direct connection to XML. RDF data *may be expressed* ("serialized") as XML, but that is not the same thing at all. SQL data may *also* be expressed as XML -- and I think you'll agree, SQL is also not "a part of the XML story"...

You said, #2, "RDF uses XML to define a foundation for processing metadata and to provide a standard metadata infrastructure for both the Web and the enterprise."

This sentence would be much more accurate with a small change -- "XML may be used to define ..." -- as RDF doesn't use XML for anything. The Web and the Enterprise may use either or both RDF and XML for many things.

You said, #3, "The difference between the two is that XML is used to transport data using a common format, while RDF is layered on top of XML defining a broad category of data."

XML is often used to translate data between other formats, because many formats may be (relatively easily) translated into XML, and vice versa, while there may not be a direct translation between those other formats. For instance, XML may be used to help translate from RDF to SQL and vice versa.

RDF is not "layered on top of XML" -- though RDF data is sometimes *expressed* as XML, typically using the RDF/XML standard -- but RDF/XML is not the only way that XML can be used to express RDF, nor is XML the only way to express RDF! If anything, you might say "XML is layered on top of RDF", as the XML expression is typically substantially larger than (for instance) the N3 expression of the same RDF data.

RDF/XML is a very unfriendly format, for most computer users -- who are not programmers, have little to no experience with Web or other markup, and generally just want to learn more about something they were just reading about. N3, Turtle, and other serializations of RDF, on the other hand, can be read much like English, such as <Thing> <hasCharacteristic> <Character>, e.g. (to steal an example from Tim Berners-Lee's N3 Primer), --

  <#pat> <#knows> <#jo> .

The relative friendliness of these serializations isn't all that important -- except when they are being used to explain the concepts that lie beneath. As that does seem to be a key intent of your recent posts, I think it worth taking a look at this other page showing how to express the same bits of RDF-structured information in a number of formats.

In short -- RDF can be discussed accurately and completely without ever mentioning XML -- and this generally makes for a much clearer presentation. For good or ill, bringing XML into the discussion tends to make people's eyes glaze over, and that effect seems to me worth avoiding.

Be seeing you,

Ted

David,

In a nutshell, RDF/XML a Blessing and Course re. both RDF and the Semantic Web vision as a whole.

Inextricably linking RDF and XML perpetuates a painful misconception that ultimately makes the core concepts of RDF incomprehensible.

This matter is too important for anyone to sit by and overlook, especially if myths are inadvertently perpetuated, bearing in mind your target audience i.e., enterprises mired in the woes of disparate data integration.

btw - my post about the end of "RDBMS primacy" should shed light on my background, should my identity remain ambiguous etc..


Kingsley


Guys,

That is why I posted again to clear things up. The issue here was that I put up a very general blog entry promoting RDF as something that may be of interest for those doing data integration. You guys hammered in on the semantics, which is fine, but to call it “inaccurate� was a bit of a stretch. Perhaps asking for me to be clearer would have been a better approach.

I think I cleared up the who XML meets RDF thing above, and the way you took my statement was not really what I meant to say, or perhaps did not convey what I meant correctly. They way you respond to these things is important.

Not sure this was a productive way to address
this topic.

Dave

Industry expert Dave Linthicum tells you what you need to know about building efficiency into the information management infrastructure

David Linthicum

David Linthicum is the CTO of Blue Mountain Labs, and an internationally known distributed computing and application integration expert. View more

Subscribe

 Subscribe in a reader

Recently Commented On

Categories

Microsoft,

Monthly Archives

Blogs

ADVERTISEMENT