From Wikipedia, the free encyclopedia

No-www is an initiative to make all websites accessible from both the http://www.example.com/ and http://example.com/ forms of their names.

The reason behind it is to standardize domain names providing web content and to avoid typing that is not necessary. Technically, domain names, including third level domains, are not meant to identify the type of content provided, but rather a host, which may provide different types of content. That is not a problem, since the content type is requested using the corresponding protocol. In the case of web pages, that is the HyperText Transfer Protocol ("http://").

Its website offers help with getting rid of the www prefix, validation for No-www conformance, a count of validated sites, and banners for sites that want to support them.

Contents 1 Compliance Classes 1.1 Class A 1.2 Class B 1.2.1 Apache example 1.3 Class C 2 Counter Arguments 3 See also 4 External link

No-www defines three levels of conformance to their philosophy.

The summary table of Compliance Classes

Class	example.com	www.example.com
A	Access	Access
B		Redirect
C		Unreach
FAIL	Unreach	Access
NA		Unreach
FAIL	Redirect	Access

• Access means that the website can be accessed by using the domain in corresponding column, without redirection.
• Unreach means that the website cannot be reached by using the domain in corresponding column.
• Redirect means that the website can be accessed by using the domain in corresponding column, with redirection to the domain in opposite column.

For other situations, no-www.org defines NA.

A Class A domain name is one where the same website is returned whether or not the "www." is included. No redirection occurs. Many web hosts configure domains to behave like this by default.

Note: This is not recommended as search engines such as google will treat both addresses as two separate websites, rather than one.

A Class B domain name is one where going to the website including the "www." prefix will perform an HTTP redirect to send the user to the canonical unprefixed URL.

Using the Apache HTTP server, this can be done with a simple mod rewrite rule. A file called .htaccess should contain following lines:

RewriteEngine On
RewriteCond %{HTTP_HOST} ^www\.example\.com$ [NC]
RewriteRule ^(.*)$ http://example.com/$1 [R=301,L]

A Class C domain name is one where "www." simply does not exist, being absent from the domain name's DNS records. This is rarely implemented, and probably inadvisable, because most web users assume domain names to include a "www." prefix. Certain web browsers, when unable to resolve a domain name starting with "www.", will automatically try it without the "www.", but this is not a common feature.

The no-www.org website makes a flawed analogy between web hosting and email handling. The reason one addresses email to recipient@example.com rather than recipient@mail.example.com is that there is an MX record for example.com that contains a prioritised list of mail servers which handle email for example.com (none of which need to be at example.com themselves). However, something similar may be trivially implemented for web hosting (or, in fact, any other service) using SRV records.

Of course there can be any number of hosts at example.com (mail.example.com, www.example.com, sirius.example.com) but the crux of the matter is that the DNS system allows for example.com to be a host as well as a domain. What no-www.org are suggesting is that people treat example.com as a host and point it to a web server which serves the example.com domain. One problem is that there may be multiple web servers at example.com (www1.example.com, www2.example.com etc) - which one should be accessible at http://example.com/? Or perhaps you would like to be able to type "ssh example.com" and get to a different machine than the one on which the example.com web server (if such a thing exists) is running? Or perhaps you are one of the many admins who doesn't like to confuse hosts with domains and prefers not to have any A or CNAME records associated with the bare domain.

Another interesting point is that a web server which was available only as http://www2.example.com/ would be considered Class C according to no-www.org, whereas one that was available only as http://www.example.com/ would be considered non-compliant. This seems a retrograde step in that it requires treating the hostname "www" as a special case, different from every other legal hostname.

Another source of confusion is the "Class A, Class B, Class C" nomenclature itself. Most Internet savvy users when seeing the expression "Class C domain name" would scratch their heads and wonder what ties a particular domain name to being on a Class C network.

One of the biggest concerns for dropping the www. subdomain is that your average user may be unable to identify that it is in fact a website, unless its specifically mentioned. For example, in the media: www.example.com can be easily identified it is a website, example.com is identified as a domain, however example.it may not be identified by some users as a real website, yet www.example.it would. The only way to fully qualify a URL as a valid url would be to also add the protocol, eg: http://example.it/ - this defeats the object of removing the www subdomain.

• URL normalization

• No-www