This document is licensed under a Creative Commons Attribution 3.0 License.
Identification and privacy have been at the center of how we interact with sites on the Web. The explosion of Websites over the last decade and a half has created a point of pain for anyone that uses the Web on a regular basis. Remembering login details, passwords, and sharing private information across the many websites that people use on a daily basis has become more difficult and complicated than necessary. This specification outlines a simple universal identification mechanism that is distributed, openly extensible, improves privacy, security and control over how one can identify themselves and control access to their information on the Web.
There are a number of concepts that are covered in this document that the reader may want to be aware of before continuing. General knowledge of public key cryptography is necessary to understand how to implement this specification. WebID also uses HTTP over TLS [HTTP-TLS], X.509 certificates [X509V3], and RDFa [RDFA-CORE].
A general Introduction is provided for all that would like to understand why this specification is necessary to simplify usage of the Web.
The terms used throughout this specification are listed in the section titled Terminology.
Developers that are interested in implementing this specification will be most interested in the sections titled Authentication Sequence and Authentication Sequence Details.
This document is merely a public working draft of a potential specification. It has no official standing of any kind and does not represent the support or consensus of any standards organisation.
The source code for this document is available via Github at the following URL: http://github.com/msporny/webid-specThis section is non-normative.
The WebID specification is designed to help alleviate the difficultly that remembering different logins, passwords and settings for websites has created. It is also designed to provide a universal and extensible mechanism to express public and private information about yourself. This section outlines the motivation behind the specification and the relationship to other similar specifications that are in active use today.
This section is non-normative.
It is a fundamental design criteria of the Web to enable individuals and organizations to control how they interact with the rest of society. This includes how one expresses their identity, public information and personal details to social networks, Web sites and services.
Semantic Web vocabularies such as Friend-of-a-Friend (FOAF) permit distributed hyperlinked social networks to exist. This vocabulary, along with other vocabularies, allow one to add information and services protection to distributed social networks.
One major criticism of open networks is that they seem to have no way of protecting the personal information distributed on the web or limiting access to resources. Few people are willing to make all their personal information public, many would like large pieces to be protected, making it available only to a select group of agents. Giving access to information is very similar to giving access to services. There are many occasions when people would like services to only be accessible to members of a group, such as allowing only friends, family members, colleagues to post an article, photo or comment on a blog. How does one do this in a flexible way, without requiring a central point of access control?
Using an process made popular by OpenID, we show how one can tie a User Agent to a URL by proving that one has write access to the URL. WebID is a simpler alternative to OpenID (fewer connections), that uses X.509 certificates to tie a User Agent (Browser) to a Person identified via a URL. WebID also provides a few additional features to OpenID. These features include trust management, via digital signatures, and free-form extensibility via RDFa. By using the existing SSL certificate exchange mechanism, WebID integrates more smoothly with existing Web browsers, including browsers on mobile devices. WebID also permits automated session login in addition to interactive session login. Additionally, all data is encrypted and guaranteed to only be received by the person or organization that was intended to receive it.
This section is non-normative.
While some may say that OpenID and WebID conflict, WebID is 100% compatible with OpenID since both use a URL for identification. Therefore, WebID does not intend to replace OpenID, but can work beside OpenID just as easily as providing a complete solution. That said, there are a number of benefits that WebID achieves over OpenID:
WebID gives people and other agents a Web ID URL for identification, just like OpenId does. However, in the case of WebID, the user does not need to remember the URL, the browser or User Agent does. A login button on a WebID web site is just a button. No need to enter any identifier like one has to for OpenID. Just click the button. Your browser will then ask you what identity you wish to use. The person that is browsing does not need to remember either the WebID URL or the website password. The only password one needs to remember is the one that is used to access their collection of WebIDs in their browser.
The WebID protocol requires just one direct network connection to establish identity via the client. The server requires one connection to the client and one connection to retrieve the WebID Profile if it does not have the credential information cached. Compare this to the much more complex OpenID sequence, which requires six connections by the client to establish a login. In a world of distributed data where each site can point to data on any other site, multiple connections become costly to manage.
WebID builds on well established Internet and Web standards; REST, RDF [RDF-PRIMER], RDFa [RDFA-CORE], TLS [HTTP-TLS], and X.509 [X509V3]. By building on previous standards, it makes both explaining and implementing WebID easier on developers.
Since WebID is RESTful, you can perform basic HTTP operations to
GET
your WebID, and if you needed update it, you can use
HTTP PUT
semantics. You can also create a WebID via
POST
. This is improved from the OpenID specification, which
requires a new set of operations described in the OpenID Attribute Exchange
specification.
It is easy to extend a WebID with new attributes via RDF. The power of RDF and RDFa allows developers to add extensions to WebID by defining new vocabularies that they publish. There is no authorization process necessary and thus WebID allows for distributed innovation. Every WebID property is a URI, which when clicked, can give you yet more information about what the property means. A developer can create new usage classes by extending their vocabulary at will. A developer can add relationships to a WebID by simply adding more HTML to the developer's page. OpenID does not provide any type of distributed innovation akin to RDF or RDFa.
WebID is built on RDF and thus enables all of the advanced semantic web concepts that RDF enables. For example, a developer may perform machine reasoning with a WebID. One can construct machine-executable statements like "If this WebID claims to be a friend of one of our partner WebIDs that is trusted and the relationship is bi-directional, trust the WebID." While OpenID attempts to support this use case by mapping OpenID to RDF, it's far easier to do with WebID because WebID is natively RDF-aware.
Implementing WebID is easier than OpenID because all of the basic technologies have been working and integrated into Web browsers for many years. There were already three interoperable implementations of WebID before this specification was written.
WebID is truly decentralized - with WebID you get a web of trust. OpenID only supports the Web of Trust model if you indirectly trust the OpenID provider. In other words - OpenID is not truly decentralized. In OpenID you must trust OpenID providers. With WebID you only have to trust the people and the organizations with which you are communicating. In other words, you don't have to ask anyone whether or not you can trust your friends. You can query people that you trust directly to see if someone is trustworthy or not. There is no need for a central WebID authority.
WebID is fully distributed, anyone can setup a WebID by placing a single file on a web server of their choosing. There is no need for a special OpenID-like provider service. The only thing anyone that wants a WebID needs is a web account where you can post your WebID file, ideally on your own domain name. You can also use a WebID hosting provider, but it's not necessary for WebID to work. While it is possible to run an OpenID server, other OpenID applications may not trust you and thus you won't be able to fully utilize your private OpenID credentials. The reason that there are a few large OpenID providers and very few small OpenID providers is because of this trust design issue related to OpenID.
WebID does not require HTTP redirects. Redirects are are problematic on many cell phones, because telecoms heavily rely on proxys, which selectively block redirects.
A WebID provider is 100% compatible with an OpenID provider and thus can inter-operate with OpenID-powered networks.
This section is non-normative.
OAuth and WebID are mutually beneficial when used together. WebID can be used to provide RSA parameters to the RSA-SHA1 signature method required by OAuth 1.0. WebID can also be used to establish the consumer_key and HTTPS connection that will be used to transmit OAuth Tokens in OAuth 2.0.
Subject Alternative Name
field pointing to a URL that is
dereference-able and results in a document containing RDF data. For example
the certificate would contain http://example.org/webid#public
,
known as a WebID URL, as
the Subject Alternative Name
:
X509v3 extensions:
...
X509v3 Subject Alternative Name:
URI:http://example.org/webid#public
Subject Alternative Name
field of the
Identification Certificate that identifies a
WebID Profile document.The following steps are executed by Verification Agents and Identification Agents to determine if access should be granted to a particular resource.
Subject Alternative Name
field of the
Identification Certificate. The WebID Profile document
must be dereferenced and all triples pertaining to the public key associated
with the WebID URL must be extracted.
The Identification Agent may re-establish a different identity at any time by executing all of the steps in the Authentication Sequence again. Additional algorithms, detailed in the next section, may be performed to determine if the Verification Agent can access a particular resource after the last step of the Authentication Sequence has been completed.
This section covers details about each step in the authentication process.
This section will detail how the TLS connection process is started and used by WebID to create a secure channel between the Identification Agent and the Verification Agent.
This section will detail how the certificate is selected and sent to the Verification Agent.
A server responding to a WebID Profile request must support returning an
XHTML+RDFa [XHTML-RDFA] document with either a text/html
or
application/xhtml+xml
MIMEtype. A server may support HTTP content
negotiation and return a document that conforms to N3 [N3], Turtle
[TURTLE], or RDF/XML [RDF-SYNTAX-GRAMMAR].
This section will explain how a Verification Agent extracts semantic data describing the identification credentials from a WebID Profile.
The Verification Agent may use a number of different methods to extract the public key information from the WebID Profile.
The following SPARQL query outlines one way in which the public key could be extracted from the WebID Profile:
PREFIX cert: <http://www.w3.org/ns/auth/cert#>
PREFIX rsa: <http://www.w3.org/ns/auth/rsa#>
SELECT ?modulus ?exp
WHERE {
?key cert:identity <http://example.org/webid#public>;
a rsa:RSAPublicKey;
rsa:modulus [ cert:hex ?modulus; ];
rsa:public_exponent [ cert:decimal ?exp ] .
}
This section still needs more information.
This section will explain how a Verification Agent may use the information discovered via a WebID URL to determine if one should be able to access a particular resource. It will explain how a Verification Agent can use links to other RDFa documents to build knowledge about the given WebID.
This section is non-normative.
2010-07-11 Initial version.
This section is non-normative.
The following people have been instrumental in providing thoughts, feedback, reviews, criticism and input in the creation of this specification: