Uniform Resource Locator
Domaining Guide
Uniform Resource Locator
Home
Up
Uniform Resource Locator (URL) is a technical,
Web-related term used in two distinct meanings:
- In popular usage, it is a widespread synonym for Uniform Resource
Identifier (URI) — many popular and technical texts
will use the term "URL" when referring to URI;
- Strictly, the idea of a uniform syntax for global identifiers of
network-retrievable documents was the core idea of the World Wide Web. In
the early times, these identifiers were variously called "document names",
"Web addresses" and "Uniform Resource Locators". These names were
misleading, however, because not all identifiers were locators, and even for
those that were, this was not their defining characteristic. Nevertheless,
by the time the
RFC 1630 formally defined the term "URI" as a generic term best suited
to the concept, the term "URL" had gained widespread popularity, which has
continued to this day.
URI/URL syntax in brief
Every URI (and therefore every URL) begins with the scheme name that defines
its namespace, purpose, and the syntax of the remaining part of the URI. Most
Web-enabled programs will try to dereference a URI according to the semantics of
its scheme and a context-vbn For example, a Web browser will usually dereference
a http://example.org/ by performing an HTTP request to the host example.org, at
the default HTTP port (see Port 80). Dereferencing URI mailto:bob@example.com
will usually open a "Compose e-mail" window with the address bob@example.com in the "To" field.
"example.com" is a
domain
name; an IP address or other network address might be used instead.
URLs as locators
In its current strict technical meaning, a URL is a URI that, “in addition to
identifying a
resource, [provides] a means of locating the resource by describing its
primary access mechanism (e.g., its network ‘location’).”[1]
Clean URLs
"Clean" and "cruft-free" describe URLs which are:
- Not tied to technical details, such as the software used or whether the
resource comes from a file or a database - so that a change in the
technology will not break existing links to the resource. e.g.
/cars/audi/ is preferable to /cars/audi/index.php or
/myprog.jsp?page=cars/audi/.
- Not tied to internal organisational structure, such as the current
editor or department that created the document - so an internal
reorganisation will not cause existing links to the document to break. e.g.
/recommendations/2007/xyz/ is better than
/~users/jane/current-work/xyz/ or /xyz-team/recommendations/.
- Consistent with other URLs in the same site in terms of hierarchy. This
is desirable so a user can see where they are in the structure of the site,
and can predict where to find what they are looking for. e.g.
/cars/audi/ and /cars/ford/, instead of
/cars/audi/ but /ford-cars/.
- Consistent with other URLs in the same site in terms of action. This is
desirable so a user can predict other, similar URLs on that site, e.g. if
/blogs/andrea/feed/ shows a
feed of Andrea's
blog, then appending /feed/ to any another blog on the same
site should show a feed for that blog.
- A single location for a single resource. The same resource should not be
available from multiple URLs, as this results in both confusion (Are they
the same resource, or is one a copy of the other? Which is the 'right' one?
Is one new and the other due to be removed?) and technical difficulties,
e.g. counting links to a particular resource, or caching content to speed up
access but not being able to show the cached content when the resource is
accessed using a different URL.
An example of the difference between "clean" and "standard" URLs could be
seen as:
Standard:
http://example.com/index.php?section=articles&subsection=recent
Clean:
http://example.com/articles/recent/
or
http://example.com/articles/2007/
Clean URLs with web services
Web services have been created that allow users to create short URLs which
are easier to write down, remember or pass around. They are also more suitable
for use where space is limited, for example in an IRC conversation, email
signature, online forum or fixed width document (eg. email). A sample of current
web services are provided below:
- TinyURL.com - probably the most widely used due to its memorable name.
Example: http://www.tinyurl.com/qvqqo
- doiop.com - one of the early services which offers keywords as opposed
to random URLs. Example: http://doiop.com/keyword
- dtmurl.com
- gu.ma
- notlong.com - lets you choose your own sub-domain. Example:
http://your-choice.notlong.com/
- SnipURL.com (synonyms: snurl.com, snipr.com)
- shorl.com
- URLStrip.com
of third-party clean URLs
Ultimately these services hide the final destination from a web user. This
can be used to unwittingly send people to sites that offend their sensibilities,
or crash or compromise their computer using browser vulnerabilities. To help
combat such abuse, TinyURL allows a user to set a cookie-based preference such
that TinyURL stops at the TinyURL website, giving a preview of the final link,
when that user clicks TinyURLs. Substituting http://preview.tinyurl.com for
http://tinyurl.com in the URL is another way of stopping at a preview of the
final link before clicking through to it. Opaqueness is also leveraged by
spammers[2], who can use such links in spam (mostly blog spam), bypassing URL blacklists.
Furthermore, this approach creates dependency on a third-party service that
may change, go away, or maintain
privacy-compromising
logs of user activity indefinitely.
Address bar
Address Bar in Microsoft Internet Explorer.
URLs are typically entered into the address or
location bar of a web browser. To the right is a standard Microsoft Internet
Explorer address bar. Address bars may of course vary in appearance depending on
which web browser it is displayed in, and which skin is in use.
External links
Home
Up
This guide is licensed under the GNU
|