URIs URNs and URLs: What's the difference?

From The Digital Classicist Wiki
Jump to navigation Jump to search

URI

A Uniform Resource Identifier is a globally unique string that identifies a resource or entity. It should include a part that identifies the issuer or namespace of the identifier (for instance a web domain or scheme prefix), along with a part that identifies the entity within that scheme. There are various types of URI, some of which we talk about on this page.

HTTP URI

A HTTP (Hypertext Transfer Protocol) URI is a URI that begins with the characters "http://" followed by a Web domain, and then other text. It may, but need not, be the address of an information resource on the World-Wide Web.

URL

A Uniform Resource Locator is a HTTP URI that serves as the address of a web-page or other resource on the Web. It may also serve as a proxy referrent for either (a) the information resource that sits at that web address, or (b) the subject of that web page: e.g. <https://en.wikipedia.org/wiki/Muhammad_al-Idrisi> is potentially both the identifier of an online encylopaedia article created in 2011 and with some hundreds of authors, and of a twelvth-century Arab geographer about whom biographical and prosopographical statements can be made. The fact that it is also a link to said article about said geographer is a convenience for its use as identifier for either.

How is a URI different from a URL?

The short answer is that it isn't necessarily. All URLs are URIs, but not all URIs are URLs. URI stands for "Uniform Resource Identifier", meaning a string (sequence) of characters that can uniquely identify anything on the internet. URL stands for 'Uniform Resource Locator', meaning a string of characters that indicate where something can be found.

In Linked Open Data, it is a best-practice to use a URL as the URI for any publicly available web-based resource that can be easily re-used by others. Meaning, the string 'http://opencontext.org/subjects/6A48C92B-E9DB-4CA0-1F9E-2F3ECEFC1C7C' is both the permanent identifier for that page describing a coin found at Petra, and the location where that page can be found. OpenContext goes a long way to making this clear by offering a 'Suggested Citation' that includes the URI. In general, combining the roles of URI and URL, which requires a commitment to the permanence of URLs, enables many of the benefits of Linked Open Data. In this context, 'URI' is often the preferred term within Digital Humanities and Linked Open Data. This usually means that in casual conversation, you can replace with 'URL' with 'URI' whenever somebody is talking about permanent, high-quality web-based resources.

Linked Data sites will often use URIs to refer to things that can't be retrieved on the internet, such as actual people or places. For example, the Virtual International Authority File (VIAF) URI for the Roman emperor Claudius is 'http://viaf.org/viaf/87172361'. When it is dereferenced, a redirect is issued by the server to 'http://viaf.org/viaf/87172361/' (note the trailing slash), which is VIAF's page about Claudius. Pleiades, on the other hand, uses 'http://pleiades.stoa.org/places/697725#this' (note the '#this' on the end of the URI) to refer to Petra the place, as opposed to the Pleiades article about Petra. This distinction is important to keep in mind, because in certain circumstances you may want to avod conflating web pages with real people or places.

See also Very clean URIs.

URN

A Uniform Resource Name is a kind of URI that provides a unique name for digital resource, including online resources. Unlike URLs, URNs don't specify a way to retrieve an instance of the resource. To find the resource named by a URN, you have to use some sort of resolver service, such as CTS (Canonical Text Services), which uses URNs to refer to its resources. A URN might refer to a particular author-text-citation combination, while different resolvers might turn that into URLs that point to different instances of that fragment of text online (in different repositories, formats, languages, etc.).