for Content Publishers and Webmasters
- Introducing RSS
- Choosing Content for Your Feeds
- Publishing Your Feed
- Telling People About Your Feed
- Format Versions and Modules
- Which Format Should I Choose?
- Tips for Generating Good Feeds
- Feed Tools
- More Information
- About this Document
Think about all of the information that you access on the Web on a
day-to-day basis; news headlines, search results, “What’s New”, job
vacancies, and so forth. A large amount of this content can be thought of as
a list; although it probably isn’t in HTML
the information is list-oriented.
Most people need to track a number of these lists, but it becomes difficult once there are more than a handful of sources. This is because they have to go to each page, load it, remember how it’s formatted, and find where they last left off in the list.
RSS is an XML-based format that allows the syndication of lists of hyperlinks, along with other information, or metadata, that helps viewers decide whether they want to follow the link.
This allows peoples’ computers to fetch and understand the information, so that all of the lists they’re interested in can be tracked and personalized for them. It is a format that’s intended for use by computers on behalf of people, rather than being directly presented to them (like HTML).
To enable this, a Web site will make a feed, or channel, available, just like any other file or resource on the server. Once a feed is available, computers can regularly fetch the file to get the most recent items on the list. Most often, people will do this with an aggregator, a program that manages a number of lists and presents them in a single interface.
Feeds can also be used for other kinds of list-oriented information, such as syndicating the content itself (often weblogs) along with the links. However, this tutorial focuses on the use of RSS for syndication of links.
What’s in a feed?
A feed contains a list of items or entries, each of which is identified by a link. Each item can have any amount of other metadata associated with it as well.
The most basic metadata for an entry includes a title for the link and a description of it; when syndicating news headlines, these fields might be used for the story title and the first paragraph or a summary, for example. For example, a simple entry might look like;
<item> <title>Earth Invaded</title> <link>http://news.example.com/2004/12/17/invasion</link> <description>The earth was attacked by an invasion fleet from halfway across the galaxy; luckily, a fatal miscalculation of scale resulted in the entire armada being eaten by a small dog.</description> </item>
Additionally, the feed itself can have metadata associated with it, so that it can be given a title (e.g., “Bob’s news headlines”), description, and other fields like publisher and copyright terms.
For an idea of what full feeds look like, see ‘RSS Versions and Modules’.
How do people use feeds?
Aggregators are the most common use of feeds, and there are several types. Web aggregators (sometimes called portals) make this view available in a Web page; my Yahoo is a well-known example of this. Aggregators have also been integrated into e-mail clients, users’ desktops, or standalone, dedicated software.
Aggregators can offer a variety of special features, including combining several related feeds into a single view, hiding entries that the viewer has already seen, and categorizing feeds and entries.
Other uses of feeds include site tracking by search engines and other software; because the feed is machine-readable, the search software doesn’t have to figure out which parts of the site are important and which parts are just the navigation and presentation. You may also choose to allow people to republish your feeds on their Web sites, giving them the ability to represent your content as they require.
Why should I make a feed available?
Your viewers will thank you, and there will be more of them, because it allows them to see your site without going out of their way to visit.
While this seems bad at first glance, it actually improves your site’s visibility; by making it easier for your users to keep up with your site — allowing them to see it the way they want to — it’s more likely that they’ll know when something that interests them is available on your site.
For example, imagine that your company announces a new product or feature every month or two. Without a feed, your viewers have to remember to come to your site and see if they find anything new — if they have time. If you provide a feed for them, they can point their aggregator or other software at it, and it will give them a link and a description of developments at your site almost as soon as they happen.
News is similar; because there are so many sources of news on the Web, most of your viewers won’t come to your site every day. By providing a feed, you are in front of them constantly, improving the chances that they’ll click through to an article that catches their eye.
But isn’t that giving away my content?
No! You still retain copyright on your content (if you wish to).
You also control what information is syndicated in the feed, whether it’s a full article or just a teaser. Your content can still be protected by your current access control mechanisms; only the links and metadata are distributed. You can also protect the RSS feed itself with SSL encryption and HTTP username/password authentication too, if you’d like.
In many ways, syndication is similar to the subscription newsletters that many sites offer to keep viewers up-to-date. The big difference is that they don’t have to supply an e-mail address, lowering the barrier of privacy concerns, while still giving you a direct channel to your viewers. Also, they get to see the content in the manner that’s most convenient to them, which means that you get more eyes looking at your content.
Choosing Content for Your Feeds
Any list-oriented information on your site that your viewers might be interested in tracking or reusing is a good candidate for a feed. This can encompass news headlines and press releases, job listings, conference calendars and rankings (like ‘top 10’ lists).
- News & Announcements - headlines, notices and any list of announcements that are added to over time
- Document listings - lists of added or changed pages, so that people don’t need to constantly check for different content
- Bookmarks and other external links - while most people use RSS for sharing links from their own sites, it’s a natural fit for sharing lists of external links
- Calendars - listings of past or upcoming events, deadlines or holidays
- Mailing lists - to compliment a Web-based archive of public or private e-mail lists
- Search results - to let people track changing or new results to their searches
- Databases - job listings, software releases, etc.
While it’s a good start to have a “master feed” for your site that lists recent news and events, don’t stop there. Generally, each area of your site that features a changing list of information should have a corresponding feed; this allows viewers to precisely target their interests.
For example, if your news site has pages for World news, national news, local news, business, sports, etc., there should be a feed for each of these sections.
If your site offers a personalized view of data (e.g., people can choose categories of information that will show up on their home page), offer this as a feed, so that the viewers’ Web pages match the content of their feeds.
A great example of this is the variety of feeds that Netflix provides; not only can you keep track of new releases, but also personalised reccommendations and even a listing of the movies in your queue.
Another good example is Apple’s iTunes Music Store RSS feed generator; you can customize it based on your preferences, and the views it allows match those provided in the Music Store itself.
Finally, remember that feeds are just as — if not more — useful on an Intranet as they are on the Internet. Syndication can be a powerful tool for sharing and integrating information inside a company.
Publishing Your Feed
There are a number of ways to generate a feed from your content. First of all, explore your content management system - it might already have an option to generate an RSS feed.
If that option isn’t available, you have a number of choices;
- Self-scraping — The easiest way to publish a feed from existing content. Scraping tools fetch your Web page and pull out the relevant parts for the feed, so that you don’t have to change your publishing system. Some use regular expressions or XPath expressions, while others require you to mark up your page with minimal hints (usually using <div> or <span> tags) that help it decide what should be put into the feed.
- Feed integration — If your site is dynamically generated (using languages like Perl, Python or PHP), it may have a RSS library available, so that you can integrate the feed into your publishing process.
- Starting with the feed — Alternatively, you can manage the list-oriented parts of your content in the RSS feed itself, and generate your Web pages (as well as other content, like e-mail lists) from the feed. This has the advantage of always having the correct information in the feed, and tools like XSLT make this option easy, especially if you’re starting from scratch.
- Third party scraping — If none of these options work for you, some people on the Web will scrape your site for you and make the feed available. Be warned, however, that this is never as reliable or accurate as doing it yourself, because they don’t know the details of your content or your system. Also, using third parties introduces another point of failure in the delivery process; problems there (network, server or business) will cause your feed to be unavailable.
Telling People About Your Feed
An important step after publishing a feed is letting your viewers know that it exists; there are a lot of feeds available on the Web now, but it’s hard to find them, making it difficult for viewers to utilize them.
Pages that have an associated RSS feed should clearly indicate this to viewers by using a link containing like ‘RSS feed’. For example,
<a type="application/rss+xml" href="feed.rss">RSS feed for this page</a>
where ‘feed.rss’ is the URL for the feed. the ‘type’ attribute tells browsers that this is a link to an RSS feed in a way that they understand.
Additionally, some programs look for a link in the <head> section of your HTML. To support this, include a <link> tag;
<head> <title>My Page</title> <link rel="alternate" type="application/rss+xml" href="feed.rss" title="RSS feed for My Page"> </head>
These links should be placed on the Web page that is most similar to the feed content; this enables people to find them as they browse.
Note that Atom feeds should use
application/atom+xml rather than
application/rss+xml in both styles of use.
Finally, there are a number of guides and registries for RSS feeds that people can search and browse through, much like the Yahoo directory for Web sites; it’s a good idea to register your feed; see More Information.
Format Versions and Modules
There are a number of different versions of the RSS format in use today, but the main choices are RSS 1.0 and RSS 2.0. Each version has its benefits and drawbacks; RSS 2.0 is known for its simplicity, while RSS 1.0 is more extensible and fully specified. Both formats are XML-based and have the same basic structure.
There’s one more choice; Atom is an effort in the IETF (an Internet standards body) to come up with a well-documented, standard syndication format. Although it has a different name, it has the same basic functions as RSS, and many people use the term “RSS” to refer to RSS or Atom syndication.
This section presents a quick overview of each; for more information, see their specifications and supporting materials.
RSS 2.0 is championed by UserLand’s Dave Winer. In this version, RSS stands for “Really Simple Syndication,” and simplicity is its focus.
Included in 2.0.1 - the
latest stable version of this branch — are channel metadata like
allows you to specify a thumbnail image to display with the feed);
managingEditor, to identify who’s
responsible for the feed, and
lastBuildDate, which shows when
the feed was last updated.
Items have the standard
well as other, more experimental facilities like
which allows attachments to be automatically downloaded (don’t expect these
features to be supported by all aggregators, however). Finally, items can have
guid element that identifies the item uniquely; this allows some
advanced functionality in some aggregators.
Here’s an example of a minimal RSS 2.0 feed:
<?xml version="1.0"?> <rss version="2.0"> <channel> <title>Example Channel</title> <link>http://example.com/</link> <description>My example channel</description> <item> <title>News for September the Second</title> <link>http://example.com/2002/09/01</link> <description>other things happened today</description> </item> <item> <title>News for September the First</title> <link>http://example.com/2002/09/02</link> </item> </channel> </rss>
In the RSS 2.0 roadmap, Winer states that this branch is, for all practical purposes, frozen, except for clarifications to the specification.
However, exensions to the format are allowed in separate modules, using XML Namespaces to avoid conflicts in their names. For example, if you had an ISBN module to track books, it might look like this;
<item xmlns:book="http://namespace.example.com/book/1.0" rdf:about="http://www.amazon.com/exec/obidos/tg/detail/-/0553575376"> <title>Excession</link> <link>http://www.amazon.com/exec/obidos/tg/detail/-/0553575376</link> <book:isbn>0553575376</book:isbn> </item>
Generally, though, you should look for available RSS Modules, rather than defining your own, unless you’re sure that what you need doesn’t exist.
RSS 1.0 stands for “RDF Site Summary.” This flavor of RSS incorporates RDF, a Web standard for metadata. Because RSS 1.0 uses RDF, any RDF processor can understand RSS without knowing anything about it in particular. This allows syndicated feeds to easily become part of the Semantic Web.
RSS 1.0 also uses XML Namespaces to allow extensions, in a manner similar to RSS 2.0.
RSS 1.0 feeds look very similar to RSS 2.0 feeds, with a few key differences;
- The entire feed is wrapped in
</rdf:RDF>elements (so that processors know that it’s RDF)
rdf:aboutattribute that usually, but not always, matches the
<link>; this assigns an identifier to each item
- There’s an
<items>element in the channel metadata that contains a list of items in the channel, so that RDF processors can keep track of the relationship between the items
- Some metadata uses the
rdf:resourceattribute to carry links, instead of putting it inside the element.
RSS 1.0 is developed and maintained by an ad hoc group of interested people; see their Web site for more information about RSS 1.0 and RSS Modules. See below for an example of an RSS 1.0 feed.
Dublin Core Module
The most well-known example of an RSS 1.0 Module is the Dublin Core Module. The Dublin Core is a set of metadata developed by librarians and information scientists that standardizes a set of common metadata that is useful for describing documents, among other things. The Dublin Core Module uses these metadata to attach information to both feeds (in the channel metadata) and to individual items.
This module includes useful elements like
associating dates with items,
dc:subject, which can be useful
for categorizing items or feeds, and
dc:rights, for dictating
the intellectual property rights associated with an item or a feed.
Here’s an example of a minimal RSS 1.0 feed that uses the Dublin Core Module:
<?xml version="1.0"?> <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns="http://purl.org/rss/1.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" > <channel rdf:about="http://example.com/news.rss"> <title>Example Channel</title> <link>http://example.com/</link> <description>My example channel</description> <items> <rdf:Seq> <rdf:li resource="http://example.com/2002/09/01/"/> <rdf:li resource="http://example.com/2002/09/02/"/> </rdf:Seq> </items> </channel> <item rdf:about="http://example.com/2002/09/01/"> <title>News for September the First</title> <link>http://example.com/2002/09/01/</link> <description>other things happened today</description> <dc:date>2002-09-01</dc:date> </item> <item rdf:about="http://example.com/2002/09/02/"> <title>News for September the Second</title> <link>http://example.com/2002/09/02/</link> <dc:date>2002-09-02</dc:date> </item> </rdf:RDF>
As you can see, RSS 1.0 is a bit more verbose than 2.0, mostly because it needs to be compatible with other versions of RSS while containing the markup that RDF processors need.
Both RSS 1.0 and 2.0 are informal specifications; that is, they aren’t published by a well-known standards body or industry consortium, but instead by a small group of people.
Some people are concerned by this, because such specifications can be changed at the whim of the people who control it. Standards bodies bring stability, by limiting change and having well-established procedures for introducing it. To introduce such stability to syndication, a group of people established an IETF Working Group to standardise a format called Atom.
Atom is functionally similar to both branches of RSS, and is also an XML-based format.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <title>Example Feed</title> <link href="http://example.org/"/> <updated>2003-12-13T18:30:02Z</updated> <author> <name>John Doe</name> </author> <id>urn:uuid:60a76c80-d399-11d9-b93C-0003939e0af6</id> <entry> <title>Atom-Powered Robots Run Amok</title> <link href="http://example.org/2003/12/13/atom03"/> <id>urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a</id> <updated>2003-12-13T18:30:02Z</updated> <summary>Some text.</summary> </entry> </feed>
As you can see, Atom has a
feed element that contains both the feed-level
metadata as well as the
entrys (analogous to RSS’
entry can contain similar metadata, such as
id (instead of RSS 1.0’s
rdf:about or RSS 2.0’s
guid), and a short textual
summary (instead of RSS’
Generally, Atom isn’t as widely supported as RSS 1.0 or 2.0 right now, because it’s relatively new. However, it should catch up quickly, because of the broad base of vendors supporting the standardisation effort.
Which Format Should I Choose?
One of the most confusing and unfortunate problems in syndication is the large number of formats in use. In addition to those listed above, there are many other formats (e.g., RSS 0.9, 0.91, 0.92) that are commonly encountered on the Web.
For better or worse, the decision isn’t as critical as you might think. Most aggregators and other software use syndication libraries which abstract out the particular format that a feed is in, so that they can consume any popular syndication feed.
As a result, which format to choose is a matter of personal taste. RSS 1.0 is very extensible, and useful if you want to integrate it into Semantic Web systems. RSS 2.0 is very simple and easy to author by hand. Atom is now an IETF Standard, bringing stability and a natural community to support its use.
Tips for Generating Good Feeds
RSS and Atom are easy to work with, but like any new format, you may encounter some problems in using them. This section attempts to address the most common issues that arise when generating a feed.
- Distinct Entries — Make sure that aggregators can tell
your entries apart, by using different identifiers in
guid(RSS 2.0) and
id(Atom). This will save a lot of headaches down the road.
- Meaningful Metadata — Try to make the metadata useful
on its own; for example, if you only include a short
<title>, people may not know what the link is about. By the same token, if you shove an entire article into
<description>, it’ll crowd people’s view of the feed, and they’re less likely to stay interested in what you have to say. Generally, you want to put enough into the feed to help someone decide whether they should follow the link.
- Encoding HTML — Although it’s tempting, refrain from
including HTML markup (like
<p>) in your RSS feed; because you don’t know how it will be presented, doing so can prevent your feed from being displayed correctly. If you need to include a a tag in the text of the feed (e.g., the title of an entry is “Ode to <title>”), make sure you escape ampersands and angle brackets (so that it would be “Ode to <title>”).
- XML Entities — Remember that XML doesn’t predefine
entities like HTML does; therefore, you won’t have
©and other common entities available. You can define them in the XML, or alternatively just use an character encoding that makes what you need available.
- Character Encoding — Some software generates feeds using Windows character sets, and sometimes mislabels them. The safest thing to do is to encode your feed as UTF-8 and check it by parsing it with an XML parser.
- Communicating with Viewers — Don’t use entries in your
feed to communicate to your users; for example, some feeds have been
known to use the
<description>to dictate copyright terms. Use the appropriate element or module.
- Communicating with Machines — Likewise, use the
appropriate HTTP status codes if your feed has relocated (usually,
301 Moved Permanently) or is no longer available (
404 Not Found).
- Making your Feed Cache-Friendly — Successful feeds see a fair amount of traffic because clients poll them often to see if they’ve changed. To support the load, Web Caching can help; see the caching tutorial.
- Validate — use the Feed Validator to catch any problems in your feed; it works with RSS and Atom. Also, don’t just run it once; make sure you regularly check your feed, so that you can catch transient errors.
This is an incomplete list of tools for creating feeds and checking them to make sure that you’ve done so correctly. Note that there are many more libraries that help parsing feeds; these haven’t been included here because this tutorial focuses on the Webmaster, not consumers of feeds.
- xpath2rss — Tool for scraping Web sites using XPath expressions (a method of selecting parts of HTML and XML documents).
- Site Summaries in XHTML — Online service (also available as an XSLT stylesheet) that uses hints in your HTML to generate a feed.
- myRSS — An online, third-party automated scraping service. Doesn’t require any special markup.
- RSS.py — Python library for generating and parsing RSS.
- ROME — Java library for parsing and generating RSS and Atom feeds, as well as translating between formats.
- XML::RSS — Perl module for generating and parsing RSS.
- Online Validator - Check your RSS 1.0, 2.0 and Atom feeds.
- Syndicated content — Good list of best practices for creating an RSS feed.
- Syndic8 — A community effort to gather, validate and search feeds with lots of other information.
- RSS Workshop — A well-regarded introduction to publishing RSS feeds, from the state of Utah Online Services division.
- RSS Devcenter — O’reilly’s Web portal for all things RSS.
About this Document
If you do mirror this document, please send e-mail to the address above, so that you can be informed of updates.
All trademarks within are property of their respective holders.
Although the author believes the contents to be accurate at the time of publication, no liability is assumed for them, their application or any consequences thereof. If any misrepresentations, errors or other need for clarification is found, please contact the author.
The latest revision of this document can always be obtained from http://www.mnot.net/rss/tutorial/
Translations are available in: Brazilian Portuguese.
Version 0.91 — September 7, 2005