a) The web is currently converging around web applications and mobile devices, a lot of focus is being placed on sensor networks, internet of things, and augmented reality to display information. Simply, how can these applications make use of published data readily from multiple sources if that data is not in a uniform standard?

b) The core web which people use on a daily basis is ever more silo focussed, and the size of those silo's is ever increasing - the social sector is a great example of this, and whilst there are core movements to create a more federated and distributed social web, a key blockage in the way is a lack of uniform data, often new formats are being developed, or poorly modelled application (rather than domain) specific models are making it out on to the web, and interoperability is several times harder than it could be, given the presence of uniform data. This has significant social and economic repercussions.

c) Time, a significant amount of time is invested daily by thousands (if not millions) in to re-solving the same old problems, creating a schema for this, a model for that, learning the same lessons countless people have learned before them, often the learning curve spans several years. A standard way to publish and share reusable model specific schemas (/not/ format specific like XML schema and JSON schema) would save vast amounts of developer time per annum. In addition to having significant economic impacts this would also lend to far more innovation (since more time free to innovate!) within an already important and innovative sector.

Why not "plain" RDF?

RDF has failed to be understood, adopted or loved by the general masses of the web, even many who use RDF often do not fully understand it and have many issues. Adoption has been... let's just say not good.

There are 3196 APIs on ProgrammableWeb, out of those:

• 2152 produce XML
• 1255 produce JSON
• 36 produce RDF

Perhaps more indicative though, is that those 36 are spread over 6 years, with only 1 updated so far this year, meanwhile there have been 58 new JSON based APIs in the last month alone.

Over on stack overflow, there have been 1,569,512 questions asked, 273 that's 0.017% of them, are RDF related.

The numbers are pretty clear, for all RDF's merits, and the countless benefits of the uniformity of RDF, it's just not being adopted.

To use RDF correctly requires RDF tooling, and not just tooling to parse the data (like JSON, and common usage of XML), but to use the data, to handle triples and graphs and queries, all of which requires significant investment in skills, time, and deployable technologies.

Further more, RDF data published using multiple different ontologies is difficult for people to use, the infrastructure and tooling simply doesn't exist to follow ones nose around the web and make practical use of several thousand different ontologies, that level of understanding is a good generation away, and for now all it does is serve as a blockage to adoption, and primarily as a blockage to people actually using or presenting the data. Time and time again we have seen a rallying around core ontologies, with successful mixing and matching happening more at the ontology level, than the data level. For now applications will be looking for mentions of Classes and Properties they "understand" (have a hard coded usage for).

Additionally, these difficulties in usage have lead to a second layer of centralization on the web, one which was borne from RDF, and rather ironically many of the architectural benefits of uniformity and universality are being lost. That is SPARQL, we are seeing a huge increase in SPARQL enabled datastores on the web, each of which holds a specific set of data, and each of which has key resource limitations. Practically this means that:
- clients are tightly coupled to servers
- all processing and storage weight is being handled by the servers
- data on the wire is non uniform
- clients are not using the web of data, rather they are using a datasource on the web, a datasilo.
This is a pattern which is not optimized for anybody, servers, clients, developers, data, the web, the network.

The core benefits of a web of linked data have not realized, RDF has failed to deliver them, primarily due to complexity and tooling requirements. SPARQL (positioned on the server/silo) is only compounding matters. That's not to say it cannot deliver them, or that these technologies are bad, only that they have not delivered the core benefits, yet.

Perhaps another way to put it, is that if you break things like RDBMS and Classes and Objects down you can get to triples of some sort (EAV, RDF, or to atomic relations / predicate based logic), and RDF did just this, however it was done in such a way that the data format (RDF) required a full new stack of technologies to /use/ the data, rather than being a uniform data format acting as a bridge between say classes and objects and RDBMS, a webized data model; that is to say, you can't really use "it" (RDF, the model people don't really speak of) with 95% of the deployed technology out there, you can provide an RDF view of the data from that technology, map it to RDF, but you cannot easily pull it back in and use it, and unusable data, isn't much use. There are many shades of grey between, but it's certainly more at the unusable end of the spectrum.

What can we do?

If we look at what people already do, a large proportion of web developers (most) continue to publish data via web services as XML and JSON, the common process is simple, create a schema, document it somewhere out of band (perhaps call it API documentation), publish data using that schema in some arbitrary way as XML and JSON. On the client side the same process continues, find a new API, get an XML or JSON parser, map the data as described by the API to some classes and start using it. All of this is needless work, they are showing us what works, what they can do, and how they can work with data easily. Tersely, they are missing the benefits of Uniform Data.

We can bring the benefits of uniform data to the current web 2.0, class and objects, rdbms, xml and json focussed web.

We can not only address these core issues, and bring the benefits of linked data and the semantic web to the general developer population, but we can also:
- ensure it's RDF and traditional semantic web compatible (giving "us" mountains of useful every-day data)
- provide that clear migration path to the "full" semantic web that's missing now.
- increase semantic web adoption exponentially, bringing big benefits without the high cost.

Approaches

There are two key approaches I can personally see to this:

1. Webize Classes and Objects (Java style POJOs, Data Objects, subset of UML)
2. Provide a Classes and Objects view over RDF

The first of these approaches - providing an abstract syntax for classes and objects and then defining mappings for that to XML and JSON - would bring the benefits of OWL 2 and XSD to schemas, and the benefits of "linked data" to both the schemas (/class blueprints) and instance data. It would allow data validation rules to be augmented on from sources external to the schema, it could be codified in libraries across multiple languages, it could also serve as a translation layer between Classes and Objects, NoSQL, and RDBMS, and other formats such as CSV. Additionally it would lend each schema openly being mapped to vendor specific databases, as well as vendor neutral schemas such as ANSI SQL. Furthermore, it would also lend to innovation in each layer, for example standardized queries for each kind of data could be created, with translations of those to each specific vendor or to well defined standardized languages, and even codified to work in memory in libraries (for example within instance methods or to run on GPU enabled hardware and languages). Many benefits could come from webizing what the masses already do. Other examples include providing an opportunity to refine the core datatypes on the web in a serialization agnostic way (think xsd types merged with webidl types), ensuring the correct entailments for equality are baked in to the core, providing first level support for things like lists and sets, providing a foundation upon which diff, patch, versioning can all be accomplished, providing canonicalized forms so that encryption and a data signing can be accomplished... and more I'm sure.

The second of these approaches has less wide scale benefits, but would provide a more usable abstraction layer on top of RDF, which is currently (dare I say painfully) missing. This would ultimately make working with data more familiar, a codified example could be:

var person = new Class('foaf:Person');      // external class definitions loaded from the web
person.load('http://example.org/bob#me');   // instance data loaded
print(person.name);                         // simple access to pre-known properties
person.validate();                          // in built validation from OWL 2
                                            // and XSD data type restrictions

// work with a schema class at a time..

var man = new Class('gender:Male');         // different class for different data
man.load('http://example.org/bob#me');      // same data
print(man.wife);                            // different, domain specific properties
man.expand();                               // full entailment regimes support to get
                                            // the most from schema definitions 

The best approach will become clear as time progresses, for now I'm keen and happy to work on either or both.

Just some musings..

So, here's a simple-ish summary of the problem - disclaimer, all IMHO of course:

Outline:

    each URI <u> is bound to a thing T by a set of agents SA (this is the naming process)
    <u> refers to T

    some URIs are bound to a set of representations SR over time by the dereferencing process (i.e. GET a URI)
    <u> refers to T, SR

Let's assign the name XT to this class of URIs which are bound to both a T and an SR. (things which name something, and which you can GET content+meta by dereferencing)

Where T == SR this forms a subclass of XT which we'll call YT (just means that the URI is used to refer to the thing you GET, like a web page or an image)

The opposing views:

1. for all <u> in XT, T == SR
   (all members XT are members of YT, doesn't account for T != SR - this is an information resource theory)

2. SR is bound to T not <u>
   (means T == SR && T != SR - this is the content+meta gives information about the thing theory, slash uris name anything)

3. <u> is bound to T, and T != SR
   (SR is unbound to any name, or bound to "some other name" - this is the <u> :retrieved_from "u" approach, or content-location = graph uri)

4. <u> is bound to SR, and T != SR
   (T is unbound to any name, or bound to "some other name" - can't see what it equates to but it's a theory)

5. <u> is bound to T, SR and T != SR
   (can't use deref URIs as names - URI collision, or the chimera theory, this is where information about <u> consists of info about both the information and the real world thing named. Practically it's the same as view 2)

Summary:

View 1, is the httpRange-14 solution (accounting for view 3 with the 303 solution too). View 2 is implied by the REST dissertation (actually so is 1 and 3..). View 3 is the "we don't normally talk about the document" view. All of them have issues for somebody.

This is quite a dry subject, and I could be wrong (in fact in some ways I want to be proved wrong, this is how we learn), but do read on if you're interested.

Example

Over the past few months I've hit on a number of occasions where I was convinced I needed Named Graphs in order to address the task at hand.

A notable example is the scenario where using WebAccessControl and the ACL ontology, a system would have to figure out just who should be given access to a resource, and who should be denied.

In this example I'll cover the notion of ACL for "groups" in a linked data world.

The task at hand is to allow access if:
the graph serialized within the document obtained by dereferencing the URI of the group states the <webid#me> is a member.

Otherwise written as:
if we dereference <groups#admin> does the graph returned include the following { <groups#admin> sioc:has_member <webid#me> }

Or in SPARQL:

ASK
GRAPH <groups> {
  <groups#admin> sioc:has_member <webid#me>
}

In this example we *do not* want to dereference the users webid to see if the graph returned specifies that { <webid#me> sioc:member_of <groups#admin> } , or indeed consider the open world possibilities that another yet unknown graph could assert that the user is a member of our admin group, as that would breach security.

The ACL

To proceed with the example, consider the following ACL:

[] a acl:Authorization ;
	acl:accessTo <https://example.org/sensitive> ;
 	acl:agentClass :mygroup ;
 	acl:mode acl:Read .

:mygroup owl:equivalentClass [
 	a owl:Restriction ;
 	owl:hasValue <groups#admin> ;
 	owl:onProperty [ owl:inverseOf sioc:has_member ];
 	] .

The Problem

The problem proposed by this ACL is that any of the following four sets of triples would infer that <webid#me> would qualify as an instance of :mygroup (or a member of <groups#admin> if you prefer).

• <webid#me> sioc:member_of <groups#admin> .

• <webid#me> _:x <groups#admin> .
  _:x owl:inverseOf sioc:has_member .

• <groups#admin> sioc:has_member <webid#me> .

• <groups#admin> _:y <webid#me> .
  _:y owl:inverseOf sioc:member_of .

In other words, the ACL does not specify a "Named Graph" to query, and at the moment, no way exists to specify with (OWL or RDF) which "Named Graph" to query / trust.

This, point in case, is one example where I saw the need for Named Graphs in RDF and OWL.

Another way of looking at it

You will have noticed the notion of "Named Graphs" creeping in above, seems like a logical thing to say, especially when you consider that to process this ACL and grant access you'd probably use SPARQL, and specify a Named Graph to query over. However, much of what follows arose because I'd decided not to use SPARQL, and rather to code an ACL processor in my preferred language.

If you consider the situation, the ACL processor which decides if access should be granted or not, must implicitly "trust" the document which contains the serialized ACL graph. That is to say, that it must by extension trust any resources pointed to by said ACL, and if it doesn't then the ACL isn't fit for the purpose.

It's also important to note that "trust" is context specific, in this case we trust the resources pointed to by the ACL for the purpose of WebAccessControl.

One could then pretty quickly conclude that in this scenario the ACL processor already know's how to process the ACL, it must only use resources it trusts, therefore it must only allow access if the graph serialized within the document obtained by dereferencing the URI of the group states the <webid#me> is a member.

(because <groups#admin> is specified in the ACL, and thus by extension, trusted)

Named Graphs in SPARQL

The aforementioned logic would also apply if I was using SPARQL to process the ACL, it would equate to the ACL processor asking:

ASK
GRAPH <groups> {
  <groups#admin> sioc:has_member <webid#me>
}

But again this is very context specific to the example, let's consider for a moment that the URI for the group could have been a non-fragment URI, <groups/admin> for example.

This leads us to an important problem, when we dereference <groups/admin> it would have to 303 See Other through to a different URI, let's say <data/groups/admin> - which would then mean that the Named Graph to be used was <data/groups/admin> - this URI, you may note, we do not know when we are writing our ACL; so if we ASKed the above SPARQL, the results would always come back negative, since their is no GRAPH <groups>.

The URI of the Named Graph issue is compounded by modern web servers and publishing practises, because <data/groups/admin> could easily be content negotiated (or rewritten), thus giving various final URI's of <data/groups/admin> or <data/groups/admin.rdf> or <data/groups/admin.ttl> or <data/groups/admin.n3> and so forth. One could quite easily (and often does) end up with the same Graph repeated multiple times within a quad store, all under "different" "Named Graphs".

I'll expand on a possible way of addressing this problem further on.

Directionality

Previously I mentioned that the ACL processor didn't have a problem with the above ACL, because it by nature trusted all resources which were mentioned in the ACL graph. However, again this is very context specific.

Let's consider for a moment an inverted ACL, where we want to allow access if:
the graph serialized within the document obtained by dereferencing the URI of the users webid states that <webid#me> is a sioc:member_of <groups#admin>.

We don't know the users webid ahead of time when we write the ACL, so again we have no way of writing how to trust a resource - it is critical to note that even if RDF(2) did support the concept of Named Graphs, it still wouldn't address the situation because we wouldn't know the Named Graph ahead of time, in order to trust it!

If we now consider the following ACL:

[] a acl:Authorization ;
	acl:accessTo <https://example.org/sensitive> ;
 	acl:agentClass :mygroup ;
 	acl:mode acl:Read .

:mygroup owl:equivalentClass [
 	a owl:Restriction ;
 	owl:hasValue <groups#admin> ;
 	owl:onProperty sioc:member_of;
 	] .

The outcome of our previous logic concludes that again we should be querying the "trusted" resource <groups#admin>, which gives us another problem, that's not the resource we want to be asking in this scenario.

The only thing that remains, and I'll later argue the only thing that ever matters in a web of linked data, is direction.

If we analyse the first ACL closer, we can see that we ultimately used the direction inferred by the presence of owl:inverseOf to place <groups#admin> in the subject position, rather than the value/object position it could have been in, indicated by the presence of owl:hasValue. (bare with me).

In this example, we can use the strong semantics of owl:hasValue (and lack of owl:inverseOf) to place <groups#admin> in the value/object position, and thus our ACL processor can come to the outcome we want, which is to look for the a triple with the meaning { <webid#access> sioc:member_of <groups#admin> }, and that means dereferencing the URI in the subject position, in other words asking the graph serialized in the document returned by GETting <webid> if it contains such a triple.

I've applied some understanding to OWL that quite simply isn't there though, as I earlier stated both ACL examples could easily equate to looking for any one of those four sets of triples.

However, this is the point - machine understanding of data is in the domain of the machine, the application doing the processing. And "truth" or "trust" is entirely context specific.

I'm increasingly convinced that the combined context of the data in a graph and the context under which that graph is being queried, specifies or infers in which direction you want to be reading, and directionality can be determined with linked data by dereferencing whichever uri you place on the left / in the subject position.

I recently found that Tim Berners-Lee wrote about this in a blog post entitled Backward and Forward links in RDF just as important:

One meme of RDF ethos is that the direction one choses for a given property is arbitrary: it doesn't matter whether one defines "parent" or "child"; "employee" or "employer". This philosophy (from the Enquire design of 1980) is that one should not favor one way over another. One day, you may be interested in following the link one way, another day, or somene else, the other way.

Key here is the sentence "One day, you may be interested in following the link one way, another day, or somene else, the other way.", and that is exactly what all these examples are doing, following a link one way, or the other way.

To conclude this part, in every scenario thus far where I've thought I needed Named Graphs, it turns out that I in-fact needed directionality - and because I'm dealing with Linked Data, whatever I place in the subject position defines the URI which I need to dereference, and ultimately the Graph(s) which are considered when resolving the answer to the question being ASKed.

I'd thus suggest that "Named Graphs", do not exist in a web of data, they are needed in N3 and when using rules, because all data is often in a single file, however that is not the case for Linked Data, where we dereference.

Back to SPARQL and Named Graphs

Previously I mentioned the complications with the way we currently use named graphs in SPARQL and in our quad stores, where the URI we end up using could literally be, anything; and often we get duplicate data under different graphs.

To address this, I'd suggest that what we should be storing as the graph ?g value, is not some made up "named graph" but rather: the dereferenced URI which we initially requested.

• in the case of <group#admins> this would be <group>.
• in the case of <group/admins> this would be <group/admins>

To clarify, *never* the URI that a GET request finally resolves to, and *always* the initial dereferenced URI we requested.

The above ensure that we'd never have duplicate data in our quad stores again, that SPARQL queries including a FROM clause always dereferenced, that publishers and web server administrators were free to relocate and restructure their data, and ultimately make for a much nicer, healthier web of data.

Cool URIs don't change, and they wouldn't, just because the final document serializing a graph may move to a different URI, doesn't mean the original URI has to change.

Conclusion

Apologies for the length of the post, but I figured everything needed covered, in context. Simply put we need to focus less on Named Graphs (which IMHO aren't needed) and focus more on directionality. Every problem I've encountered thus far is covered by what Tim said years ago: "One day, you may be interested in following the link one way, another day, or somene else, the other way."

Comments?

I'm assuming that if you are reading this, you know what linked data is, and REST as per the dissertation of Roy T. Fielding. If not go do some reading :)

Interface Constraints

REST is defined by four interface constraints:

1. identification of resources
2. manipulation of resources through representations
3. self-descriptive messages
4. hypermedia as the engine of application state.

From here I'll look at each of these four constraints and build up the approach as I go.

What a resource is

Quoting extensively from REST 5.2.1.1 Resources and Resource Identifiers:

The key abstraction of information in REST is a resource. Any information that can be named can be a resource: a document or image, a temporal service (e.g. "today's weather in Los Angeles"), a collection of other resources, a non-virtual object (e.g. a person), and so on. In other words, any concept that might be the target of an author's hypertext reference must fit within the definition of a resource...

A resource is a conceptual mapping to a set of entities, not the entity that corresponds to the mapping at any particular point in time...

The values in the set may be resource representations and/or resource identifiers...

A resource can map to the empty set, which allows references to be made to a concept before any realization of that concept exists...

The only thing that is required to be static for a resource is the semantics of the mapping, since the semantics is what distinguishes one resource from another...

What a representation is

Again, quoting extensively from REST 5.2.1.2 Representations:

... A representation is a sequence of bytes, plus representation metadata to describe those bytes. Other commonly used but less precise names for a representation include: document, file, and HTTP message entity, instance, or variant...

If the value set of a resource at a given time consists of multiple representations, content negotiation may be used to select the best representation for inclusion in a given message...

The data format of a representation is known as a media type

Identification of resources

To do this properly I need to identify some resources, so for this I'm going to work with "Something" :)

• "Something" - a resource, a non-virtual object

At any point in time I have a description of Something which has multiple representations in different mediatypes, all semantically matching or equivalent:

• "something.rdf" - representation of Something with mediatype RDF+XML
• "something.n3" - representation of Something with mediatype RDF+N3
• "something.en.html" - representation of Something, in english, with mediatype text/html
• "something.de.html" representation of Something, in german, with mediatype text/html

Each one of those representations is also a resource because they can be the target of a hyperlink. Of course by resource I mean a conceptual mapping to each of the things listed, and I haven't assigned URIs but will..

To be able to make this set of representations manageable and to indicate they are in a set, I'm going to add in another resource which is a collection of resources, which can be considered a set of these equivalent representations of Something at a fixed point in time. For the purpose of this exercise, that point in time is today.

• "Something-20100311" - a resource which is a collection of equivalent representations of Something on the 11th March 2010.

Additionally, for the sake of argument, I'm going to say that a new set of representations (or version) is added every day - to handle this I then need one more resource, a collection of resources, where each resource in the collection is itself a collection of resources (one of the aforementioned and including the example "Something-20100311"). This will give me a conceptual mapping which covers time, and therefore everything I could need.

• "Somethings" - a resource which is a collection of resources, see above for full description!

Finally, I'm going to add in two shortcut resources which have no representation and are simply conceptual maps to the first and most current sets of representations.

• "first" - a resource which always maps to the first collection of representations of Something.
• "latest" - a resource which maps to the most recent collection of representations of Something.

Giving the resources URIs

Now to assign some URIs for this use case, there is no set structure and I'm not going to define one because it is up to each server (or manager of) to control it's own URI space, but for the sake of this exercise I'll define mine as follows:


base: http://data.webr3.org
...
/d/Something
/rg/Somethings
/rg/Somethings/first
/rg/Somethings/latest
/rg/Somethings/Something-20100311
/rg/Somethings/Something-20100311/something.rdf
/rg/Somethings/Something-20100311/something.n3
/rg/Somethings/Something-20100311/something.en.html
/rg/Somethings/Something-20100311/something.de.html
...
/rg/Somethings/Something-20100305
/rg/Somethings/Something-20100305/something.rdf
/rg/Somethings/Something-20100305/something.n3
/rg/Somethings/Something-20100305/something.en.html
/rg/Somethings/Something-20100305/something.de.html
...


From the above you can see that every possible representation has its own URI, in addition every collection of equivalent representations has its own URI, as does the collection of all those collections; and so does "Something" our non virtual object.

Also we've exposed multiple resources which could also be RESTful CRUD access points operating on an atompub style protocol. Small sentence, big potential, will cover approaches and protocols in later posts.

The Key resource

The most important thing, which I haven't yet covered, is that we've exposed a key resource, namely /rg/Somethings. This is a resource at the top of the representation chain which can be used to expose content negotiation, be it server or agent driven (or a mix of both), and regardless of the mappings and levels of collection further down the line this can always be a single point of entry to get representations.

I'll cover just how in a moment, but for now something important.

Important

I've had to give a fixed example just to make some progress, but we have to remember that every system has different needs, in some cases it may be that there is only a single fixed representation for a resource, whilst in others each strand of representation (like something.de.html) may take it's own versioning / temporal path. This could indicate that a structure such as the following may be in order:

...
/d/Something
/rg/Somethings
/rg/Somethings/first
/rg/Somethings/latest
/rg/Somethings/Something-20100311
/rg/Somethings/Something-20100305
/rg/Somethings/Something-rdf
/rg/Somethings/Something-rdf/20100311.rdf
/rg/Somethings/Something-rdf/20100305.rdf
/rg/Somethings/Something-html-en
/rg/Somethings/Something-html-en/20100311.html
/rg/Somethings/Something-html-en/20100305.html
/rg/Somethings/Something-html-de
/rg/Somethings/Something-html-de/20100308.html
/rg/Somethings/Something-html-de/20100303.html


The above highlights that whilst we may have added more resources, the core resources are still the same; remember that they are "conceptual maps", meaning that Something-20100311 may "map" to the version of en-html on the 11th and de-html on the 8th, because the de version was written first, then translated to english and from there rdf and so forth, but they are all semantically equivalent, containing the same information even though they were created at different times.

The Conceptual Maps are as follows, from what I can tell this should always cover any use-case, no matter how complex.


Thing 1-1 CollectionOfCollections
CollectionOfCollections 1-* CollectionOfEquivalentRepresentations
CollectionOfEquivalentRepresentations 1-* Representation


aside:At times like this I wish I'd had a chance to study computer science so that I could express these things formally, so you'll have to make sense of it as best you can :( sorry.

Exposing via Content Negotiation

In my research so far, I've been able to figure out how to expose all of the aforementioned via HTTP, RESTfully using content negotiation in a manner which seems to be transparent to existing web browsers, but exposes all the information needed in a manner that is visible to machines; without using any additional extensions headers. As follows:

1 The client does a normal GET request on our "Something", notice that no content negotiation is happening yet, we are simply asserting via a 303 "that the requested resource does not have a representation of its own that can be transferred by the server over HTTP."

#Request
GET /d/Something HTTP/1.1
Host: data.webr3.org
Accept: text/html;q=0.5, application/rdf+xml


#Response
HTTP/1.1 303 See Other
Location: http://data.webr3.org/rg/Somethings


2The client does a GET on the URI we specified in the Location field, namely to our key resource that can be used for content negotiation over all the representations.

#Request
GET /rg/Somethings HTTP/1.1
Host: data.webr3.org
Accept: text/html;q=0.5, application/rdf+xml


#Response
HTTP/1.1 300 Multiple Choices
Location: http://data.webr3.org/rg/Somethings/latest
Content-Type: application/xhtml+xml
Content-Length: 17400


<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML+RDFa 1.0//EN"
"http://www.w3.org/MarkUp/DTD/xhtml-rdfa-1.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"
xmlns:foaf="http://xmlns.com/foaf/0.1/"
xmlns:dc="http://purl.org/dc/elements/1.1/"
version="XHTML+RDFa 1.0" xml:lang="en">
...


Here's where it gets interesting and clients can take different routes; first the route of the typical user agent:

User Agent Route


#Request
GET /rg/Somethings/latest HTTP/1.1
Host: data.webr3.org
Accept: text/html;q=0.5, application/rdf+xml


#Response
HTTP/1.1 307 Temporary Redirect
Location: http://data.webr3.org/rg/Somethings/Something-20100311


#Request
GET /rg/Somethings/Something-20100311 HTTP/1.1
Host: data.webr3.org
Accept: text/html;q=0.5, application/rdf+xml


#Response
HTTP/1.1 302 Found
Vary: Accept
ETag: W/"xyzzy"
Last-Modified: Wed, 11 Mar 2010 12:45:26 GMT
Content-Type: application/xhtml+xml
Content-Length: 17400
Content-Language: en
Content-Location: http://data.webr3.org/rg/Somethings/Something-20100311/something.en.html


<!DOCTYPE html...


First you can see that the user agent simply goes straight through to the most recent content and what they expect to see; which is nice, with additional Server driven content negotiation.

Further, we can see that full cache control is in there which as we know speeds up the net, and further still we have a rather nifty "weak" entity tag; this entity tag is shared by all representations which are semantically equal, and asserts they are equal via the entity tag. It's also worth noting that you could add this entity tag to your RDF graphs and further assert provenance which could come in very handy down the line for POST and PUT implementations.

To recap, common user agents just go straight through to the expected resource via server driven content negotiation and can take full advantage of cache / control data.

The Machine Route

Back at 2 the server returned a 300 Multiple Choices as soon as /rg/Somethings was requested. All important was that the entity returned was XHTML+RDFa (although this could have been Atom or similar..), which means we can give both a human and machine readable list of all our various representations, the "machine" can then select which one it finds most fitting. The choices could be expressed using any suitable ontology; and further both Alternative and Link headers could be added if publishers wished.

I think that covers it all, if there are any errors or things I've missed please do let me know asap; but for now that'll do me - it's verbose, but I like verbose - prove it works then optimise it later :)

I'm primarily documenting this journey for my own benefit, for reference and to unload it from my brain; but hopefully it'll be of use to the wider community and any feedback will be massively appreciated.

Here goes, I'll start by analysing the web thus far:

The Web till now

The power and the success of the web so far, in my opinion, has come from four main things:

• The URI
• Hyperlink
• The Resource
• HTTP and it's RESTful design.

More importantly it's the combination of the four working together that makes the web so great, because, they let you cut out everything else and go straight to the resource you want. This is a point that we need to concentrate on for a minute.

Going straight to the resource

Back at the start of the web, this allowed people to (for the first time) jump from a resource in the bowels of one companies hierarchy straight to another resource in a different companies hierarchy - very much like reading a book, looking at the references and suddenly having the book or paper it references right there in your lap - amazing to say the least.

Skip forwards a few years and we have the search engines, suddenly simply by typing a few keywords you can jump right to a resource (page/image/...) anywhere on the web. Fast forward some more and we get to web 2.0 where again people are amazed every time direct access to a resource is exposed. Yup.. all your web 2.0 is just this simple principal..

An RSS feed, well it lets you read a resource (a post) outside the context of the website and inside lets say google reader. You can rip a resource (video) out of youtube and embed it anywhere you like. You can interact with a web application super fast thanks to targeting a resource directly with say ajax and only updating that resource rather than updating the entire view (page). You can interact with web applications using desktop clients because they let you access one resource at a time; and so on and so forth, virtually every improvement you see on the web comes down to that one thing, directly accessing a resource (and creating resources at a more granular level).

How we made the web faster

Negating the rather obvious upgrades in technology over the years, there is one primary thing that speeds up the web and virtually everything computer oriented, the cache.

Before all the web 2.0 stuff, resource caching was at an all time high and was making the web faster for all of us; caching at the resource level is enabled by HTTP and its RESTful design. Control Data allows us to limit how much information needs transferred over the web, request an image once and it gets transferred, request it a second time and thanks to caching and HTTP odds are very high that it won't get transferred again. When you consider that the average web page can easily have 30, 50, 300 images and static files embedded in it this is a huge speed increase, and frankly one we could not live without.

Skipping forwards to web 2.0 and the present day again, we've gone wild with caching; anybody who's been involved with a high traffic site will tell you that the only way to do it is to cache everything you can; from data in memory, through to code and op code caches. But this is only half the story.. a strange this has happened..

How we made the web slower

Simply, we forgot HTTP and a RESTful web somehow - that all important web caching whether it be at intermediate servers or in a web browser, it's forgotten.

To illustrate, if you view an image and then view it again, it'll be there instantly - why? because last-modified, etag and other control data is sent by great web servers like apache for static files, until you force a refresh or the file changes on the server you'll simply get a 304 response telling whatever cache down the line to use it's own copy instead. Now, try jumping on to a web page, even this one and you'll find the whole thing is reloaded, every time. I'd estimate that circa 80% of all pages you visit are fully reloaded every single time you see them, if not more.

Here's the reason - most pages are generated by scripts now, and something that goes unnoticed by most developers is that the web server (like apache) hand over *full* control to the language runtimes, and in turn to the developer. In other words, unless developers are calculating, receiving and sending control data for each response, and validating every http message in to their scripts, then most of the benefits of HTTP and RESTful design are completely lost; especially caching.

Here's a fact, whether you agree with it or not, to me it is a fact: the web has to be RESTful for it to work properly, whether it's a web of documents, or a web of data, or both.

Looking at the current state of Linked Data

Linked Data is amazing because it takes the big four I mentioned earlier (URI, Hyperlink, Resource, HTTP) to a new level; we create resources at the most granular level possible, assign them URIs, link them together with typed hyperlinks then expose them via HTTP.

Notice I didn't mention REST in there? that's because I (and I'm not the only one) don't feel that Linked Data is currently RESTful. And as we can learn from web 2.0, unless this is addressed we'll face major problems down the line. In addition, because of this lack of RESTful-ness I feel like the data isn't linked; simply using URIs from different datasets on both sides of a triple does not link those datasets, well not from a client perspective anyway.

To expand and refining the issues:

SPARQL Silos

Issue one, is that SPARQL and the servers with RDF stores which power it are positioned at the wrong side of the client / server relationship imho. Because each major dataset effectively has it's own server and access point (SPARQL interface) it means that when you query it, it can only return the Linked Data which it stores. This leaves us three options at the minute:

• let that server pull in remote Linked Data and store it too (which makes the server fill up and slow down, and turns it in to a silo).
• use one great big server that tries to store all the linked data (which feels like a silo all over again to me, not distributed at all).
• Run our own server and only store limited data in it (limited.. and again a silo I guess)

If we moved SPARQL to the client side however, then all it would need is a starting point from which it could traverse the web of data, only pulling in what it needed for a query. This may sound slow but if all data was exposed as resources like it should be, and with control data so it could be cached, this slow down would soon disappear; lesson from web 1 and 2!

With regards the caching, this could happen at traditional intermediate caches within the internet and at ISPs, locally in client side triples stores (like a browsers cache) or the existing big servers that attempt to store all the linked data could be repositioned as linked data caches.

For example a small RDF document could simply delegate seeAlso http://datacac.he/http://subject.uri and that linked data cache could return back all the information it knows about the subject by returning the RDF results of a SPARQL describe. This alone would be a HUGE speed up, prevent silos and create a real web of data.

In addition, this would keep all linked data transferred through the web in RDF format, and thus machine readable and typed. At the minute we have lots and lots of SPARQL queries, which essentially are just untyped junk data that a machine couldn't possibly understand - SPARQL results remove all the goodness from RDF and give us something that is domain and developer specific, not re-usable. Think about that for a moment..

Clarification: I'm not saying SPARQL + RDF stores shouldn't be on the server side, they should as they are needed in most cases, I'm simply saying that the primary interface to linked data shouldn't be SPARQL over HTTP to a remote SPARQL endpoint. Rather we should be accessing RDF documents, or entities if you like via HTTP.

RESTful RDF

Issue two, the focus has been on getting data on the web, finding ways to link it, access it, store and query these vast datasets; and the work done thus far is amazing! But now that's handled it's time to go back to basics and find ways of both getting and publishing Linked Data RESTfully, at a granular per resource level.

This means handling RDF like ATOM, and essentially making atompub all over for RDF (as many are thinking and working on). I feel that regardless of what's implemented behind the interface, and whether triples stores and SPARUL are used, we still need to manage RDF / Linked Data in terms of documents and entities for it to be RESTful.

An additional issue raised by this is loosing the notion of a quad, g s p o, Named Graphs are vastly important to linked data, but we need to get named graphs in to triples and out of quads so we are always working with RDF through HTTP.

Also worth noting that temporal, provenance, multi-language, multiple representations etc will all need handled too; without using any HTTP extensions; no point half baking it or making the solution dependant on drafts - needs to work with the Universal Interface!

First Step

The above means one of the first challenges and things I'll try to tackle, is to find a way to fit a RESTful RDF publishing and exposing protocol in to the shared http space on a web server; taking in to account things like content negotiation, multiple representations, versions / time varying representations, and backwards compatibility with the current web.

note: I'm not going to define the protocols, plenty of more intelligent people than me are working on these things, just leverage a space where a full RESTful protocol can work in unison with the way we currently do things so that it's transparent to browsers and visible to linked data clients. This should then allow a stable test environment to try out different ways of doing things and test that the current web doesn't break.

To be continued.. often and frequently. I'm blocked on my current, v important project, and need to address these things.

I've grouped all the links in to two main sections, and then sub-grouped by how they make sense in my head! :)

Web, HTTP and REST

Roy T. Fielding Dissertation - Architectural Styles and the Design of Network-based Software Architectures Of particular relevance and note are chapters 4-6 (many only ever read chapter 5 and miss the context + summary *needed* in chapters 4 and 6!)

• Chapter 4 - Designing the Web Architecture: Problems and Insights
• Chapter 5 - Representational State Transfer (REST)
• Chapter 6 - Experience and Evaluation

Roy T. Fielding - REST APIs must be hypertext-driven
Discussion on HTML5 and RESTful HTTP in browsers
Discussion on URIs Resources and Switching content types w/ REST angle (v good)

RFC 2616 HTTP/1.1 and the HTTPbis Working Group HTTP/1.1 update in parts:

1. Messaging
2. Semantics
3. Payload
4. Conditional
5. Range
6. Cache
7. Authentication

Linked Data and the Semantic Web

Linking Open Data Community Project
Linked Data Applications
Equivalence Mining and Matching Frameworks
Linked Data Browsers, Mashups and other Client Applications

Dataset Dynamics - On the Dynamics of Linked Datasets
Realizing a write-enabled Web of Data
Web Access Control (WAC) - a decentralized system for allowing different users and groups various forms of access to resources where users and groups are identified by HTTP URIs.
Discussion of the WAC vocabulary
Socially Aware Cloud Storage Design Note
Distributed Social Networking through Socially Aware Cloud Storage from TimBL
AWWSW - "Architecture of the World Wide Semantic Web" Task Force

SPARQL 1.1 Uniform HTTP Protocol for Managing RDF Graphs
A Linked Data Frontend for SPARQL Endpoints
RAP - RDF API for PHP V0.9.6
Inav the Terrible - An idea for posting RDF through HTTP.

Talis Changesets
Triplify Update Vocabulary

skos as atom

RFC 4287 - The Atom Syndication Format
RFC 5023 - The Atom Publishing Protocol
AtomPub Tombstones - The Atom "deleted-entry" Element
RFC 5005 - Feed Paging and Archiving
Versioning Link Relations - Link Relation Types for Simple Version Navigation between Web Resources

Named Graphs, Provenance and Trust
Accessing Site-Specific APIs Through Write-Wrappers From The Web of Data
Provenance Information in the Web of Data - LDOW 2009 paper
Using Reification To Extend RDF (historical reification approach)
RDF Policy-based URI Access Control for Content Authoring
The Open Provenance Model Core Specification (v1.1)
W3C Provenance Incubator Group

History of the Web 1945, 1980 through 1997 on W3
LEIRI - Legacy extended IRIs for XML resource identification The type of "URI" used in xml:base
CSHALS 2010 W3C Semanic Web Tutorial
Mindswap online RDF Converter
W3 online RDF Validator

I've chosen the rather old RFC 2616 (1999!) at this time rather than the upcoming HTTPbis because I feel it's important to know where you are coming from, and whilst many things about the Web of Data feel new, they are really age old principals and technologies which have never been used to their full potential. Further you won't be able to appreciate the refinements in HTTPbis if you don't know what it's refining.

Virtually everything from here on is just a snippet/quote or paraphrase of the RFC. Let's start with a simple one:

Why use HTTP?

HTTP is an application-level protocol for distributed, collaborative, hypermedia information systems. ... HTTP allows an open-ended set of methods and headers that indicate the purpose of a request. ... HTTP is also used as a generic protocol for communication between user agents and proxies/gateways to other Internet systems ... HTTP allows basic hypermedia access to resources available from diverse applications. source

I do fully recommend reading the entire RFC and the new HTTPbis, most questions can be answered by returning to these documents and reading what they say (it's all in the detail); here's some more info gleaned from the RFC:

The difference between POST and PUT, URIs as Identifiers, and URIs to identify more than just documents.

The fundamental difference between the POST and PUT requests is reflected in the different meaning of the Request-URI. The URI in a POST request identifies the resource that will handle the enclosed entity. That resource might be a data-accepting process, a gateway to some other protocol, or a separate entity that accepts annotations. In contrast, the URI in a PUT request identifies the entity enclosed with the request -- the user agent knows what URI is intended .. source

Using POST RESTfully for more than just form data.

The POST method is used to request that the origin server accept the entity enclosed in the request as a new subordinate of the resource identified by the Request-URI in the Request-Line. POST is designed to allow a uniform method to cover the following functions: Annotation of existing resources; ... Extending a database through an append operation. The actual function performed by the POST method is determined by the server and is usually dependent on the Request-URI. The posted entity is subordinate to that URI in the same way that ... a record is subordinate to a database. source

What to do if something is created as a result of a POST request.

If a resource has been created on the origin server, the response SHOULD be 201 (Created) and contain an entity which describes the status of the request and refers to the new resource, and a Location header. source

When to use a PUT request?

The PUT method requests that the enclosed entity be stored under the supplied Request-URI. source

How to handle a PUT

If the Request-URI refers to an already existing resource, the enclosed entity SHOULD be considered as a modified version of the one residing on the origin server.

If the Request-URI does not point to an existing resource, and that URI is capable of being defined as a new resource by the requesting user agent, the origin server can create the resource with that URI. If a new resource is created, the origin server MUST inform the user agent via the 201 (Created) response. If an existing resource is modified, either the 200 (OK) or 204 (No Content) response codes SHOULD be sent. source

if extra headers were sent?

Unless otherwise specified for a particular entity-header, the entity-headers in the PUT request SHOULD be applied to the resource created or modified by the PUT. source

and what if I want to save it somewhere other than the URI specified by the client?

If the server desires that the request be applied to a different URI, it MUST send a 301 (Moved Permanently) response. source

and if the PUT can't be done..

If the resource could not be created or modified with the Request-URI, an appropriate error response SHOULD be given that reflects the nature of the problem. source

can I use PUT with server side versioning?

A single resource MAY be identified by many different URIs. For example, an article might have a URI for identifying "the current version" which is separate from the URI identifying each particular version ... a PUT request on a general URI might result in several other URIs being defined by the origin server. source

how would I let a client know I implement server side versioning when they PUT?

If an existing resource is modified, either the 200 (OK) or 204 (No Content) response codes SHOULD be sent.. source

200 indicates a message body in the response ;)

and DELETE?
well it's short so you may as well read it all..

The DELETE method requests that the origin server delete the resource identified by the Request-URI. This method MAY be overridden by human intervention (or other means) on the origin server. The client cannot be guaranteed that the operation has been carried out, even if the status code returned from the origin server indicates that the action has been completed successfully. However, the server SHOULD NOT indicate success unless, at the time the response is given, it intends to delete the resource or move it to an inaccessible location.

A successful response SHOULD be 200 (OK) if the response includes an entity describing the status, 202 (Accepted) if the action has not yet been enacted, or 204 (No Content) if the action has been enacted but the response does not include an entity. source

note that a response can be 200, meaning you can return a response message (like i have X other versions here [list] or delete them all by clicking here [form input which POSTs to a service] ), or an RDF response that can be interpreted by a client to do the aforementioned :)

but can't I tunnel all actions through GET?

Safe Method .. GET .. SHOULD NOT have the significance of taking an action other than retrieval! source

edit: removed small section about URI vs URL! Do however see the comment from Michael which links to more information on the subject.

Thanks for reading :)
There is much more information in the RFC, but those were some nicer points I found useful and relevant to current Web of Data topics & discussions.

To me, the draw of linked data and the semantic web has always been exploration; the notion that even the most unskilled of publishers should be able to enrich their content via semi-automated software to the standards of a near perfect wikipedia article has always been the driving force. Additionally, content classification, relation, linkage, data centralization and the like are all major benefits which will make a vast difference to the usability of the web.

Search will always be a major part of the internet, at the moment we use search to find content on a specific subject, then search again to find more, and search again to find related or expanded info, help, facts, answers, whatever; however, in the future I hope to see search move to a less prominent role, one where we use search to find the most suitable "entry point" in to the web of linked data - and from there every other piece of related / expanded information is either on page, or a click (link) away.

Some major hurdles need to be jumped before we can get to that stage though, both through lack of organization and lack of appropriate software. Personally I have a mental blueprint / overview of what's needed (imho), and some very specific ideas on the software side, with any luck I'll get a chance to contribute + build some of this, we'll see.

Some thoughts of what's needed from my little brain.

Linked Data Ping
A central service API which is pinged by all software as it publishes information with machine accessible content. (Needed way before (x)HTML+RDFa takes off). Provides a stream of all recent pings to be consumed (xmpp pub-sub?).

Clustered Servers holding a centralized data GUID lookup and proxy.
In essence all resources on the net should be a linked pair of GUID to endpoint, each endpoint should contain a reference to the GUID, and each GUID should be a URI which redirects to the endpoint, endpoints change GUID/URI stays unique. In an ideal scenario when somebody creates a link to X resource or Y document, the publishing/controlling software should replace the endpoint with the GUID instead. This would also enable multiple other services such as centralized pingback, references, statistics etc.

Machine Readable Data Cache.
Together with the aforementioned services a high availability database of cache'd information should exist; in principal this would work by reading the stream of "Linked Data Pings", getting the GUID for the content and then retrieving all machine readable data and caching it. Much like the RDF data exposed through dbpedia, however for everything. Even if only a predefined subset of the common rdf vocabularies was stored and exposed it'd be enough to start, from there all other domain specific ontology could be retrieved by reading the endpoint itself.

Semantic CMS
Ideally we need a new breed of CMS, one that not only has simple FOAF and Dublin Core (~Drupal 7), but also support for full content enrichment using the aforementioned machine services; and provides a simple UI for manually exposing entities, events, facts etc. (Think highlight name in text, mark as Person with Name, system finds guid and builds relevant RDFa and we have another triple of linked data.)

The possibilities from this point are endless; if you're reading this document after all this has been made, then you'd see a whole host of in text links through to more information on each keyphrase, person, entity etc; you'd be aided by auto injection of sources, related reading, comments, further documents discussing the content here, in short you'd be exploring the net one click at a time, linked data all the way; not searching.

In summary (and very much imho), Linked Data is not for searching, it's for linking data - search was invented to address the issue that everything isn't linked, when it is then the link takes precedence again.

My only worry in all of this, is the idea that all rdf triples are fact, and true - already the major search engines are exposing rdfa data in summaries, 5* ratings on products and suchlike, the room for abuse will only get worse.

Thanks Georgi for placing the spark that clarified my current thoughts.

Finally, this isn't a biased opinion in anyway, or an endorsement, but to me openlink virtuoso, dbpedia, zemanta and open calais are leading the way and enabling all of this; together with the hard working folks contributing to the various linked W3C projects and specs. If only dbpedia/zemanta/calais would unify there uri's/guids/endpoints we'd be a lot further along.. (well I would ;).

Regards!