Shared publicly  - 
The Semantic Web Arrived Unannounced!

Hashtags have basically completed the self-annotating journey towards a Web where links bear fine-grained entity relationship semantics. It's all happening via Hashtags across Web 2.0 platforms using crowd-sourcing patterns. 

Here are some excerpts from a circa 2005 blog post I wrote about the Semantic Web's eventual materialization. 

"Today we are generally excited about "tagging" but fail to see its correlation with the "Semantic Web", somehow? I have said this before, and I will say it again, the "Semantic Web" is going to be self-annotated by humans with the aid of intelligent and unobtrusive annotation technology solutions. These solutions will provide context and purpose by using our our social essence as currency."_

"The annotation effort will be subliminal, there won't be a "Semantic Web Day" parade or anything of the like. It will appear before us all, in all its glory, without any fanfare. Funnily enough, we might not even call it "The Semantic Web", who cares? But it will have the distinct attributes of being very "Quiet" and highly "Valuable"; with no burden on "how we write", but constructive burden on "why we write" as part of the content contribution process (less Google/Yahoo/etc juice chasing for more knowledge assembly and exchange)."  

Original post is at: .

#WayBackPost   #Web30   #SemanticWeb   #LinkedData   #Web20  
phil jones's profile photoBill Seitz's profile photoDaniel Lemire's profile photoKingsley Idehen's profile photo
+Kingsley Idehen I would like to agree with your pronouncement but until the majority of metaspace social interaction occurs off of the  #siloed   #Web , the annotated data will remain hostage to a few players. This is not the vision of the #SemanticWeb I hold dear. I think it is fair to say that the Semantic Islands have arrived, not the Semantic Web.
What you are describing is not the Semantic Web. The Semantic Web is RDF + OWL. Look up "Semantic Web Stack". This ship has sunk. Where is the RDF, where is the OWL?

The difference is not just a nuance. There are fundamental differences between the formal ontological approach of the Semantic Web and what has actually emerged.

But people are so insistent that the Semantic Web cannot fail, the same way expert systems could not fail, that they will twist the words around... claiming credit for things they ridiculed years ago because they have nothing else to hold on to.
I disagree with +Daniel Lemire. As far as I understand it, the Semantic Web is a concept that is not necessarily dependend on a specific technology like RDF or OWL or any other in the Semantic Web stack. In the end, the Semantic Web Stack only gives suggestions on how to implement the concept of the Semantic Web (implementation-wise I actually think that those technologies are not the best...).
However, I also don't quite agree with the statement that hashtags complete the set of self-annotating tools in the Semantic Web. A hashtag only provides the term, but never the semantics. To put it simple: With a (hash)tag you don't know whether the content is about #Jaguar , the car or #Jaguar , the animal.
PS: Click on the my hashtags to convince yourself :-)
+Seb Paquet I agree wrt to the broad sense of the word SW, and +Daniel Lemire I think that's what +Kingsley Idehen is trying to convey. But, my question to Kingsley is ; wouldn't a simple hyperlink on a web page be considered as well to be part of the SW? At that point, it is a moot point to discuss whether SW is arriving or not, because sure, it was here all along. So, it might be more productive to have higher expectations from the machines such that they are more or less capable of differentiating the meaning of one hash tag from another, and applying it in cases where it is appropriate (and here, I wouldn't expect them to do any better than the humans, and that says a lot). After all, in the case of hash tags, the way we humans use them is subject to time and space, and not just flat out a string.

I'll leave you with this:
Semantic Web is too big to fail.  Secretly, everyone knows it. 

Imagine that the semantic web was to become the next big "wave".  Can anyone honestly say that they didnt see it coming?
+Sarven Capadisli -- a link on the Web is a predicate with basic (coarse-grained) semantics i.e., two entities are related.  The semantics of the aforementioned relationship between two denoted things (typically web docs) isn't immediately comprehensible to Web users and their user agents. 

As people continue to proliferate hashtags via crowd-sourcing, combined with the APIs provided by social media oriented data space providers, the whole thing starts to change i.e, virtuous cycles start to emerge en route to ultimate fusion.

Example: Facebook's 1 Billion profiles are all accessible in Linked Data form via their Graph API feature. Your "Likes" and other data that you curate for Facebook are all there and accessible in Linked Data form.  The same goes for G+, Twitter, LinkedIn etc.., although in these cases transformation to Linked Data is required which is what we deliver via URIBurner which then gets pumped into the Linked Data cloud. 
+Jeff Sayre -- the pursuit of silos that are based on hyperlinks ultimately increases the quality of entity relationship semantics across Web content. It's a positive "unintentional consequence" that arises from being myopic within a deftly architected system that always had open data access at its core :-)
+Kingsley Idehen Yes, I agree with that premise. However, as the vast majority of socially annotated content currently resides in a few silos, the Web of Data, as it were, is not healthy.

If Google Plus, Twitter, or Facebook ever shut down, a major portion of all the human-powered work of annotating data would go dark. The Web of Data is not as much of a healthy interconnected web right now -- the original vision of the Web -- as it is a few social islands that control the majority of social interaction. Whereas they allow access to some of that data, they can at any moment decide to limit access. Imagine an ecological food chain that had only a few participants. That would be a system dangerously out of balance.
+Stephanie Stroka -- Yes! The Semantic Web is really about machine / human discernible entity relationship semantics across Web content. RDF is an EAV (Entity Attribute Value + Classes & Relationships) data model enhanced with explicit (rather than implicit) semantics. OWL and RDFs are mechanisms for describing and defining entity relationship semantics. 

Everything on the Web is loosely coupled. The underlying architecture is implicitly dexterous. Thus, HTTP, RDF, RDFS, or OWL are options re. exploitation of the machine/human comprehensible entity relationship graph aspect of the Web :)  
+Stephanie Stroka If you look back at what the SemWeb people were saying in the early days, they were indeed insisting that the SemWeb was DIFFERENT from ad-hoc machine readable data formats (eg. XML, CSV, RSS2.0, hashtags etc.) because of the principled data modelling. It wasn't exactly about RDF but it was totally about the idea of concepts being formally identified with URIs and modelled as collections of triples ( )

Those of us who were SemWeb sceptics said this was all nonsense. That didn't mean we were against having machines publishing data that other machines could read. It just meant that we thought that design-by-committee was a pretty lousy way to come up with data formats and protocols (cf the entire history of computer science) and that the idea of formally defined ontologies (ie. collections of assigned URIs to meanings) was likely to tend in this direction. We also thought that trying to formally assign fixed, context-independent "meanings" to symbols in advance of their use was a very bad misunderstanding about where real semantics come from.

In a sense, I totally agree with +Kingsley Idehen Hashtags ARE an interesting convention which machines can easily tap into to extract some useful "meaning". I just want to deny that that has ANYTHING at all to do with what the "Semantic Web (TM)" people were thinking of or advocating. It's not just that they didn't think of it. It's that that kind of thing is what they were proposing the SemWeb in OPPOSITION to.

That kind of semantics of machines exchanging data has been part of the web since the very beginning. It's there with the first mime-types, when browsers knew that one chunk of data was a picture and another chunk of data was text. It's the kind of semantics that computers have always had. The meaning of a piece of data emerges not from a committee or an ontology but from how the programs in the wild actually treat it. Just as words in English get their meaning not from the dictionary but from the way they're used in everyday speech and writing. 
+Rick Dane, +phil jones : RDF, OWL and all that are indeed a core component of the Semantic Web. You can go back to the original 1998 paper, or the more recent 2006 paper ( for a description of what is the Semantic Web and it is not an abstract concept or just the idea of "metadata on the Web". It is a very specific set of standards. Let me quote wikipedia here: "The Semantic Web stack builds on the W3C's Resource Description Framework (RDF)." There is absolutely no doubt that RDF is a foundational technology for the Semantic Web stack.  If you throw away RDF and classical AI, and still talk about the Semantic Web... well, then I can call my dog "Semantic Web" and claim that the idea is very much alive.
Useful overview :

It's kind of sad to see the SemWeb people still think that the problem with RDF is the syntax, so they keep making simpler and easier ways of writing and embedding it. (I admit RDFa is kind of cute) But ultimately the syntax isn't the biggest problem, it's the whole stupid idea which is flawed, so I don't suppose even this will help much.
Arguing about whether something is the SemanticWeb is a waste of time. You need to pick specific scenarios you want to execute...

The problem with the "Semantic Web Project" is that its comprehension has been hijacked (unfortunately) by technical gobbledegook. RDF, RDFS, OWL etc.. are important, but they are also gobledegook vectors.

A more Semantic Web is simply about the Web of Documents  evolving into a fine-grained Entity Relationship Graph constrained by an Entity Attribute Value (EAV) model endowed with explicit human/machine discernible semantics. 

RDF is a data model derived from EAV that adds explicit machine/human entity relationship semantics to the mix. Where EAV is ultimately ambiguous RDF is precise. 

RDF Schema (RDFS) and OWL (Web Ontology Language) are all about expressing entity relationship semantics.

None of what I've mentioned so far is specific to an actual notation delivered by a markup language. By this I mean: the semantics in question aren't inextricably associated with a specific markup syntax.

Adding Semantics to the Entity Relationship Graph that is the World Wide Web, is all about creating documents with structured content that represent entity relationship graphs.

The graphs in question have existed since day one. What's evolving right now are the entity relationship semantics which becoming fine-grained where they used to be coarse-grained. 

Hashtags are an example of fine-grained semantics being injected into the Web via crowd-sourcing patterns :-)


1. -- Entity Attribute Value Model . 

#LinkedData   #SemanticWeb   #Web30   #EAV  
+Daniel Lemire - barring any inadvertent typos on my part, I am sure you know that I know that :-)

RDF is a Data Model associated with a variety of markup syntaxes.

You have RDF documents comprised of content. The content is expressed using a variety of syntaxes. Said content can be serialized across the wire using a variety of formats. 
Add a comment...