The channel title; for RSS this is a plain-text field with encoded HTML entities while for Atom this is a Text construct that allows text or HTML.

The channel description; for RSS this is a plain-text field with encoded HTML entities while for Atom this is a Text construct that allows text or HTML.

RSS 0.91 imposes a maximum limit of 500 characters.

Specifies the HTML page this feed belongs to. Note that Atom’s link is more complex than that of RSS.

xml<link>http://example.com/doc.html</link>

xml<link rel="alternate" title="Document title"
      type="text/html"
      href="http://example.com/doc.html" />

Any Atom node, either channel’s or entry’s, may contain the xml:lang attribute.

The channel language. Language codes for RSS and Atom (RFC 3066) differ slightly but generally if you specify an ISO-639-2 (2-char long codes like «en») you’ll suit both.

xml<language>en-us</language>

xml<feed xml:lang="en">

Any Atom node, either channel’s or entry’s, may contain the xml:base attribute.

Specifies a base URL used to resolve all relative address under current node. This is useful not only feed-wise but also in elements such as entry content.
RSS doesn’t provide an alternative to this attribute and for this must always refer to addresses with their full URLs.

xml<feed xml:base="http://example.com">

This node must contain:

• url (image URL);
• title (for <img title="..." /> attribute, ≤ 100 chars for RSS 0.91);
• link (linked page, e.g. site index)

This node may contain:

• width (≤ 144, default is 88);
• height (≤ 400, default is 31);
• description (the spec says it’s meant for <a title="..."> attribute that the <img /> might be wrapped in) – must be ≤ 500 chars for RSS 0.91.

xml<image>
  <url>http://example.com/logo.png</url>
  <title>Example.com</title>
  <link>http://example.com/index.php</link>
  <width>50</width>
  <height>102</height>
  <description>An Example.com.</description>
</image>

Atom feeds may have two associated images – icon for a small glyph and logo for expanded site/channel/post image.

• icon must be square (1:1) and «suitable for presentation at a small size» (spec);
• logo must have the aspect ratio of 2:1 (twice as long as tall).

Unlike RSS Atom image elements simply contain the URL of an image file.

xml<icon>http://example.com/icon.png</icon>
<logo>http://example.com/logo.png</logo>

A copyright string; for RSS this is a plain-text field with encoded HTML entities while for Atom this is a Text construct that allows text or HTML. RSS 0.91 imposes a maximum limit of 100 characters.

Atom feeds may have none, one or more rights nodes.

For RSS this is an e-mail and/or person’s name in the suggested format: e@mail.com (Name). For Atom this is a Person construct with several different fields.

Atom feeds may have none, one or more author nodes. See also RSS webMaster node.

The format is identical to that of managingEditor above.

Atom feeds may have any number of contributor nodes including none. This is a Person construct. See also author node.

Channel rating in PICS notation (now superceded by POWDER).

The time of last channel update; RSS has this in RFC 822 format (W3C suggests RFC 1123) while Atom – in RFC 3339.

xml<lastBuildDate>
  Fri, 04 Nov 11 05:48:38 +0000
</lastBuildDate>

xml<updated>
  2011-11-04T05:48:38+00:00
</updated>

For RSS this specifies the next channel update date for feeds that update regularly (e.g. a newspaper coming out each Friday at 12:00 AM). The date is in RFC 822 format (W3C suggests RFC 1123).

For RSS these nodes specify when a reader/aggregator is safe to not to do periodical feed fetches because the feed won’t change at the specified hours and/or days. Isn’t used by most aggregators.

The following snippet states that the feed doesn’t update on Sundays and Mondays:

xml<skipDays>
  <day>Sunday</day>
  <day>Monday</day>
</skipDays>

The following snippet states that the feed doesn’t update on 0 AM and on 11 PM:

xml<skipHours>
  <hour>0</hour>
  <hour>23</hour>
</skipHours>

Unsupported by RSS 0.92; the latter may have two attributes (uri and version) and must have a text content.

xml<generator>
  UverseWiki R475
</generator>

xml<generator uri="http://uverse.i-forge.net/blog"
           version="R475">UverseBlog</generator>

For RSS this node and its children specify an arbitrary input form – a search form, commenting, etc. Doesn’t seem to be supported by most aggregators.

This node requires 4 child nodes:

• title – caption for the submit button (≤ 100 chars for RSS 0.91);
• description – an expanded form description (≤ 500 chars for RSS 0.91);
• name – the name for the input used when sending form data (≤ 20 chars for RSS 0.91);
• link – the URL of handling script.

xml<textInput>
  <title>Search Example.com</title>
  <description>Search us</description>
  <name>q</name>
  <link>htp://example.com/search.php</link>
</textInput>

A subscribing protocol on a cloud domain – see the official RSS 2.0 description. Should have 5 attributes: domain, port, path, registerProcedure and protocol (can be http-post, xml-rpc or soap).

xml<cloud domain="rpc.sys.com" port="80" path="/RPC2" protocol="xml-rpc"
       registerProcedure="myCloud.rssPleaseNotify" />

None, one or more channel categories; both are handled differently by RSS 2.0 and Atom.

For RSS 2.0 this is a node with an optional domain attribute; if it’s given then the node’s content is some kind of identifier used by that domain (e.g. a category ID); if it’s omitted then it’s a textual description separated by "/".

For Atom this is a childless node that must have term attribute and may have scheme (URL) and label (a human-readable caption if term isn’t enough).

xml<category domain="http://worldcat.com">
  100F-AA5R
</category>

<category>Personal/Vacation 2011</category>

xml<category scheme="http://example.com/cat/15"
          term="15" label="Cats" />

Unique channel identifier (must be an IRI); the Atom specification designates that it is as unique and as permanent as possible and suggests an algorithm that should be used to achieve this.

Ideally, this identifier must not change even between different installations of a feed generator that took place on the same resource. For this an URI for the feed generator script itself can be used.

xml<id>http://example.com/feeds/atom.php</id>

Entry title; for RSS this is a plain-text field with encoded HTML entities while for Atom this is a Text construct that allows text or HTML. RSS 0.91 imposes a maximum limit of 100 characters.

Entry synopsis; for RSS this is a plain-text field with encoded HTML entities while for Atom this is a Text construct that allows text or HTML. See also Atom content node.

xml<description>
  This node may contain
  &lt;em&gt;encoded&lt;/em&gt; HTML.
</description>

In RSS this field may contain encoded HTML that upon decoding will be used as such, as opposed to title and other fields that upon decoding will treat HTML entities as plain-text.
RSS 0.91 imposes a maximum limit of 500 characters.

Specifies the URL of the full story (a web page).

Atom’s link is more complex than that of RSS – it allows you attaching documents, providing links to different representations (HTML, PDF, etc.) and more.

xml<link>
  http://example.com/entries/174.php
</link>

xml<link rel="alternate" title="PHP tricks"
      type="text/plain"
      href="http://example.com/entries/174.txt" />

For RSS this is an e-mail and/or person’s name in the suggested format: e@mail.com (Name). For Atom this is a Person construct with several different fields.

Atom feeds may have none, one or more author nodes. The same element also exists in channel information. for details.

None, one or more channel categories; both are handled differently by RSS 2.0 and Atom. The same element also exists in channel information – look here for details.

Address of the page where comments for this story can be viewed and posted. Atom’s link can be used to mimic RSS’ <comments> element.

xml<comments>
  http://example.com/153/comments
</comments>

xml<link rel="related" type="text/html" title="Comments"
      href="http://example.com/153/comments" />

An attachment – e.g. a song or a video clip. RSS’ node has 3 required attributes (url, length and MIME type) and is childless. See also Atom’s multipurpose link element.

xml<enclosure length="1845603" type="text/plain"
           url="http://example.com/log.txt" />

xml<link rel="enclosure" type="text/plain"
      href="http://example.com/log-03.txt"
      length="1845603" title="The log" />

Unique item identifier; Atom has this element for channel node as well while RSS – for items only.

This is usually the item’s permalink although the Atom standard doesn’t specifically tell this – look here for details.

xml<guid isPermalink="true">
  http://example.com/post/20110315-2240.html
</guid>

If isPermalink is omitted it’s assumed to be true. At first glance link and guid nodes serve the same purpose but it’s not necessary:

• link is the address where the feed aggregator must go when user wants to read the full entry. This means that this page might be a synopsis of a very large article.
• on the contrary, guid (assuming it isPermalink) always refers to the full blog post/magazine article/etc.

See also the last but one paragraph in the Comments section of the RSS 2.0 spec for the explanation.

xml<id>http://example.com/post/20110315.php</id>

The time on which the story was initially posted; RSS has this in RFC 822 format (W3C suggests RFC 1123) while Atom – in RFC 3339.
See also Atom’s updated node.

xml<pubDate>
  Sat, 05 Nov 11 08:58:41 +0000
</pubDate>

xml<published>
  2011-11-05T08:58:41+00:00
</published>

The time of last significant (from the publisher’s view) entry update. Atom has this field in RFC 3339 date/time format.
See also entry publication date.

Reference to a channel that this story has come from. For RSS this is a simple node with one required attribute (url) and the text for the reference caption; for Atom this element has the same format as the entire channel – it just can’t have any <entry> nodes.

xml<source url="http://example.net/395.php">
  The Semees' Place
</source>

xml<source>
  <title>The Semees' Place</title>
  <icon>http://example.com/icon-32.png</icon>
  <rights type="html">Copyright &copy;</rights>
</source>

This Atom-only node conveys a synopsis or an except from the entry in Text construct format that allows text or HTML. See also entity description node.

In Atom you can use xml:lang and xml:base attributes in any element to set specific content language and base URL for their children.

xml<content type="text">
  Plain &apos;text&apos;
</content>

<content type="application/rtf" xml:lang="ru">
  {\rtf Некий \b RTF\b0 -документ.}
</content>

<content type="application/zip" src="posts/20110315.zip" />

This is a Text construct that in addition to standard text, html and xhtml types allows any valid MIME type – in this case:

• an extra attribute named src may be present to refer to a remote document; if src is present the content node must be childless;
• if there’s no src attribute given the content node must contain the data with all entities encoded (< as &lt;, etc.).

Atom feeds may have any number of rights nodes containing copyright strings for current story in Text construct format that allows text or HTML. The same element also exists in channel information – look here for details.

Atom feeds may have any number of contributor nodes containing information in Person construct format. The same element also exists in channel information – look here.

xml<title type="text">
  HTML chars like
  &lt; are quoted.
</title>

This node contains plain text so that even <tag>-looking strings that might happen to be encoded there are treated as-is and not as HTML or any other markup.

xml<title type="html">
  &quot;Quoted
  HTML&quot; here.
</title>

This node represents a proper HTML that the aggregator renders after unquoting this its value.

xml<summary type="xhtml">
  <div xmlns="http://www.w3.org/1999/xhtml">
    Unquoted XHTML <b>nodes</b> thanks to
    the <em>namespace</em>.
  </div>
</summary>

Since XHTML is a valid XML document its nodes can be embedded just like any other XML node.