Cover | Table of Contents

Table of Contents

Chapter 1: What Makes Up a Feed

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

For all the discussion given to syndication feeds, their format and use is really quite simple. Each feed contains a certain number of individual items, usually with a title, content, category, author, and date. The overall feed also has information such as source title, last update (time), update frequency, site owner, and so on. Within these simple fields, there are various options that can lead to a great deal of variety from feed to feed. To better understand these, we'll look at each of the three major feeds to see which fields are shared, and which are unique to the specific format.

The feed "container" is information about the overall site that provides the feed, consisting of several fields that are listed once for the entire feed. Included within this container are the following fields that repeat across most syndication feeds:

link

This is the link for the site's URL, such as the weblog's URL.

title

This is the title for the site, such as the weblog's title.

description

This is a description of the weblog or site, and usually contains whatever subtitle is given for the site. In Atom, this is the subtitle field.

author

In Atom, author is a structure containing the author's name and email, or just the author's email. In RSS 2.0, this field is replaced by webMaster and managingEditor, which are email addresses. In RSS 1.x, this is replaced, usually, by dc:creator (which may be either a structure or a simple value).

date

In RSS 1.x, this field, usually dc:date, is the date the feed was updated. In Atom, this is the updated field; in RSS 2.0, this is the lastBuildDate.

generator

In RSS 2.0, this is the tool used to generate the syndication feed. The Atom feed also uses generator, while RSS 1.x uses generatorAgent.

copyright

Copyright information.

language

What language the text of the document is in.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Container Elements

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

The feed "container" is information about the overall site that provides the feed, consisting of several fields that are listed once for the entire feed. Included within this container are the following fields that repeat across most syndication feeds:

link

This is the link for the site's URL, such as the weblog's URL.

title

This is the title for the site, such as the weblog's title.

description

This is a description of the weblog or site, and usually contains whatever subtitle is given for the site. In Atom, this is the subtitle field.

author

In Atom, author is a structure containing the author's name and email, or just the author's email. In RSS 2.0, this field is replaced by webMaster and managingEditor, which are email addresses. In RSS 1.x, this is replaced, usually, by dc:creator (which may be either a structure or a simple value).

date

In RSS 1.x, this field, usually dc:date, is the date the feed was updated. In Atom, this is the updated field; in RSS 2.0, this is the lastBuildDate.

generator

In RSS 2.0, this is the tool used to generate the syndication feed. The Atom feed also uses generator, while RSS 1.x uses generatorAgent.

copyright

Copyright information.

language

What language the text of the document is in.

id

This field is specific to Atom, and provides a unique identifier for the site.

image

An icon or image representing the feed or site for both RSS 1.x and RSS 2.x. In Atom, this would be icon, though logo can be used for a logo representation. In RSS 1.1 and RSS 2.0, image is a structure pointing to the image's URL, title, and link to site. In RSS 1.x, the image should be 88x31. In Atom, the icon "should have an aspect ratio of one (horizontal) to one (vertical), and should be suitable for presentation at a small size." In RSS 2.0, the image should have a maximum width of 144, and maximum height of 400; default is 88x31.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Entry Elements

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

Each individual entry at a site has its own entry within the feed, up to the number designated for each feed (this number does change, but usually ends up being between 5 and 15). In Atom, the entries are designated by entry elements, while in RSS 1.x and RSS 2.0, they're listed as item.

Examining the more comment elements within each entry:

title

The entry's title, if any.

link

The link or URL for the individual item.

guid

The unique identification of the item. In Atom, this is id.

description

In RSS 1.1 and RSS 2.0, the description is a text description or summary of the item. In Atom, the field for this value is summary.

content

The content is the encoded full content for the item, including any HTML or XHTML markup. The field could also be called content:encoded with RSS 1.x. If you specify both content and description, then the description (or summary) should be an abbreviated copy of the text. If you specify just the summary or description, and don't want to provide full content, you should not include the content field.

author

Author of item in Atom. This is dc:creator in RSS 1.x and RSS 2.0.

pubDate

Publication date of item. This is issued in Atom and dc:date in RSS 1.x.

category

The category of the item.

Examples of entries from all three feed types follow. First, Atom:

  <entry>
  <title>The WebKit Open Source Project</title>
  <link rel="alternate" type="text/html" href="http://feeds.feedburner.com/
  oreilly/radar/atom?m=520" />
  <id>http://webkit.opendarwin.org/</id>
  <created>2005-10-13T08:34:02Z</created>
  <modified>2005-07-03T09:06:11Z</modified>
  <author>
  <name>rael</name>
  </author>
  <dc:subject>apple</dc:subject>
  <dc:subject>browser</dc:subject>
  <dc:subject>affordances</dc:subject>
  <dc:subject>web</dc:subject>
  <dc:subject>etech</dc:subject>
  <dc:subject>etech06</dc:subject>
  <dc:subject>safari</dc:subject>
  <summary type="text/html" mode="escaped">By rael
  WebKit, the embeddable Web browser behind Mac OS X's Safari, Mail.app,
  Dashboard, and third party apps like NetNewsWire and TextMate, now makes
  nightly builds available....</summary>
  <content type="text/html" mode="escaped" xml:lang="en" xml:base="http://radar.
  oreilly.com/">By rael
  WebKit, the embeddable Web browser behind Mac OS X's Safari, Mail.app,
  Dashboard, and third party apps like NetNewsWire and TextMate, now makes
  nightly builds available....
  <img src="http://feeds.feedburner.com/oreilly/radar/atom?g=520"/></
  content>
  <feedburner:origLink>http://webkit.opendarwin.org/</feedburner:origLink>
  </entry>

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Chapter 2: Publishers: Industry Support

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

Though syndication feeds are becoming widespread at many different kinds of sites, they achieved their level of popularity today through their use in web—logging. If you haven't been exposed to weblogging, this is a type of site where entries—most often created by one author—are listed out as they are created in reverse chronological order. Rather than entire articles, weblog entries tend to be personal jottings or updates of information, which may or may not include links to other online material.

Weblogs and syndication feeds tend to mirror each other in basic functionality. In fact, it's difficult to say which was the driving force behind dated, time—limited, reverse-ordered entries: weblogging or feeds. Regardless, all weblogging tools provide built-in functionality to generate feeds in one or more formats. Taking a look at several of these tools:

Blogger:

The granddaddy of hosted weblogging environments, Blogger still accounts for a significant percentage of weblogs. It originated as a separate entity, but is now owned by Google. Blogger provides only Atom as a built—in feed type, though if you're a Blogger Pro account holder, you can also select RSS 2.0. At this time, RSS 1.0 is not supported.

Six Apart:

Six A part currently supports three weblogging tools: the self-hosted Movable Type, with default templates supporting Atom and RSS 2.0; the hosted high-end weblogging tool TypePad, which supports RSS 2.0 and Atom; and the very popular LiveJournal, which also supports only RSS 2.0 and Atom. Unfortunately, the company references RSS 2.0 as just "RSS," which can be confusing, considering there are older legacy uses of RSS 0.9x still in use—not to mention RSS 1.0. However, when you see RSS in regard to any of these three products, know that this is RSS 2.0.

WordPress:

WordPress provides support for all three major syndication feed formats—RSS 1.0, RSS 2.0, and Atom—in addition to some legacy formats, such as RSS 0.9x.

ExpressionEngine:

pMachine's ExpressionEngine provides support for all three major syndication feeds.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Chapter 3: Discovering Feeds

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

Returning to the beginning of this article, those links or buttons labeled Atom, RSS, or XML all lead to a syndication feed. Clicking on them will open that feed in your browser (or whatever application you've defined to handle the feed). Either your publication tool adds this to your site, or you can add it easily—just create a hypertext link with the appropriate label.

To make your site easier to subscribe to—after all, not every reader is going to know what to do with this XML file that opens—you can also include buttons and links that specific aggregators provide, such as Bloglines, NewsGator, and so on. These should be marked clearly at the site, or instructions provided in the tool. Again, your publication tool may also provide this.

The best approach of all to publicize a feed is to use autodiscovery. Autodiscovery involves putting a simple line into the HEAD section of each web document, and pointing the reference to the location of your feed. You also provide some information about the type of feed so that aggregators know how to find it.

An autodiscovery link looks as follows, depending on your publication tool:

  <link rel="alternate" type="application/rdf+xml" title="Entries in RSS 1.0"
          href="http://yourwebsiteloc.com/index.rdf" />

In this, the feed is RSS 1.0, located at http://yourwebsiteloc.com/index.rdf, and the type of file is application/rdf+xml. You can change the title to what you would prefer.

Your publication tool may provide much of this, like the following, in which case you'll want to use caution before changing the text:

  <link rel="alternate" type="application/rdf+xml" title="Entries in RSS 1.0"
          href="<?php echo get_settings('siteurl'); ?>/index.rdf" />

And you can provide multiple feeds, such as different formats, or a feed specifically for your comments, such as the following:

  <link rel="alternate" type="application/rss+xml"
          title="Recent comments in RSS 2.0"
          href="http://yourwebsiteloc.com/commentsrss2.php" />

The application type for RSS 1.x is application/rdf+xml; the one for RSS 2.0 is

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Chapter 4: Subscribing to and Reading Feeds

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

Just as with publication tools and syndication feed formats, you have many choices about what kind of feed aggregator you use. You can use a browserbased tool, such as Firefox, in which case the feed shows up in a manner similar to a bookmark. Or you can use any number of desktop or web-based aggregators.

Wikipedia has an extensive list of news aggregators at en.wikipedia.org/wiki/ List_of_news_aggregators. Depending on your machine and operating system, you can choose from tools like FeedReader or RSS Bandit for Windows; NetNewsWire, NewsMac, and so on for the Mac; and Snownews or Olive for Unix. There are also a number of cross-platform tools from which to choose.

Some tools will be able to install as is; some may need to have other software installed. Follow the tool's installation instructions.

A desktop aggregator works by configuring how often you want it to check for syndication feed updates. A common choice is to check for updates hourly—to do so more frequently is considered "bad manners," as it can add unnecessary burden to the feed provider's bandwidth.

Once the tool is installed and configured, it's just a matter then of subscribing to sites as you find them—from personal weblogs to major publications like the New York Times.

Another option is to use a web-based aggregator tool. The benefits of these are that you don't have to install the tool, and if you have many machines, you can access the aggregator from all of them. The disadvantage, of course, is that you don't have access to the feeds when you're offline, traveling on the train or plane, or the like.

Still, many of us like the ease of an online aggregator, especially if we access our subscriptions from many machines. There are not as many online feed services as there are desktop tools, but there is a good assortment.

One popular tool is Bloglines, found at bloglines.com. You can sign up for a free account and add subscriptions immediately, or even follow the service's recommendations for adding new subscriptions. The service sends out an automated bot to check for updates, and highlights subscriptions that have new entries. You can then read the entry in the right column of the web page.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

New Syndication Challenges

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

For being a simple concept, syndication feeds have generated a great deal of intense interest surrounding in the last few years. One reason is the introduction of podcasting. Though beyond the scope of this article, it's worth your time to provide a quick overview of podcasting, and how you can get started in this if you're interested.

Podcasting is making an individual recording in the nature of a radio broadcast, and then posting this recording in your weblog. When you include the link in your weblog to the file (usually in MP3 format), most tools will generate an enclosure statement in the feed for the item.

Other tools that are capable of playing the podcast will read through the syndication feed and pull out the enclosed sound file, storing it for playback. One popular playback device is the Apple iPod, hence the name podcasting.

(For more on getting set up as a podcaster, see the Endgadget tutorial at engadget.com/entry/5843952395227141/, the About.com how-to at radio.about. com/od/podcastin1/a/aa030805a.htm, and other tutorials at Wikipedia's podcast page, en.wikipedia.org/wiki/Podcasting.)

Podcasting started out slow enough, and then exploded when some major players became interested in the concept. Soon, radio and television networks were providing podcasts of some or even all of their shows. Some popular amateur podcasters took on a star quality comparable to well-known radio personalities. What caused a major explosion, though, was when Apple turned its corporate eye to the phenomenon.

In 2005, Apple announced support for podcasts in iTunes, and provided the ability for anyone to register their podcasts with the iTunes central server. Now, not only could you find podcasts through sites such as audio.weblogs.com and ipodder.org, you could also search for new 'casts using iTunes.

The iTunes rollout wasn't without some controversy. For instance, Apple only supported one syndication feed format: RSS 2.0. This choice was particularly controversial because there is confusion about how many enclosures RSS 2.0 supports—it's not specified in the RSS 2.0 specification. So, if you have a podcast linked in your weblog post (and they are linked, just like web pages), and you have a second in the same post, some tools will create two enclosures, while others will only create one for the first they find.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Chapter 5: Deciding Which Feed Works Best for You

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

All three syndicationfeeds work with the major aggregators, and are usually supported by most tools. Not all, though. Google, which owns Blogger, a very popular hosted weblogging tool, supports Atom and RSS 1.0 but not RSS 2.0. WordPress, a very popular open source tool, provides files to create all three syndication feed types, but comes automatically enabled for RSS 2.0 and Atom, and not RSS 1.0.

Do you need all three types? No. In fact, to support three types of syndication feeds is to cause some confusion as to how many people are subscribed to you in some centralized tools. For instance, Bloglines will give a count of subscriptions for each feed, not each site. If you provide all three feeds, then the counts are split across feed type.

If you don't provide all three, then which you pick depends on your needs and what your tool supports. As mentioned, tools like Blogger support only a subset of the tools, and others may support only one. If you're planning on podcasting then you probably should look into supporting RSS 2.0 at least, primarily because this is currently the only version that Apple supports. However, you don't have to pick RSS 2.0 to take advantage of Microsoft's new changes because most of these are internal to tool use, and not necessarily syndication feed-specific. The one Microsoft tool that would be a syndication feed consumer, IE or Internet Explorer, should support all three types.

I personally have the capability of supporting all three types, but I only support RSS 1.0. The main reason for this is that I use RDF/XML for all my other applications, and I want to be consistent. And since I don't podcast, Apple iTunes needs are not an issue. I've not yet found an aggregation tool that can't work with RSS 1.0, and I've been using this syndication type primarily for well over a year.

Your mileage may vary, though. See what your tools support, chat with others, and assess your needs. You may find that providing all three feed types is your best option.

You would think that providing a feed is enough to satisfy most needs, but it goes much further. For instance, one option you'll be given by most tools that generate feeds is whether to provide an excerpt or to provide full content. When you provide full content, this usually includes any photographs or other files you have embedded in your post.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Other Issues

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

You would think that providing a feed is enough to satisfy most needs, but it goes much further. For instance, one option you'll be given by most tools that generate feeds is whether to provide an excerpt or to provide full content. When you provide full content, this usually includes any photographs or other files you have embedded in your post.

If you provide full content, chances are most people will read your material in their aggregators rather than necessarily go to your website. You have to decide for yourself if you want to give people this option, or provide an excerpt only. Another issue is that if you do embed photographs, the aggregators may access the photos from your site, and this can be costly in terms of bandwidth.

Another point on full content versus excerpts is that many people like to get unread posts in their laptop-based aggregator and then read the posts while offline. They might read the posts on the train or plane, or in some other location. In addition, some people may read content using tools such as small PDA-like devices, which can handle the unformatted feed syntax.

Also, it's easier for people to read posts in their aggregators. Some folks are so stubborn, they won't even read a weblog or other site that doesn't provide full content. Most people, though, base their decision on the material rather than the technology, so I wouldn't let this attitude impact on your decision too much.

Another possibility with full-content feeds can and should impact on your decision, and that is that some sites republish feeds, and even provide their own comment and other systems so that these republished works are a little different than the original posting. If they are reputable, these sites will also note that the material is republished, and provide links to the original site. If they are less reputable, well, you may only find out about it by doing a search.

However, if you don't provide full content, any links within your content may not get picked up by aggregator tools such as IceRocket, Feedster, and so on. This means that people may not be aware that you've linked to their writing, and your link may not get included in their link counts—a sometimes important issue in environments such as weblogging.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Here I Am, World

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

You've found a tool to generate your web content, and you've made the decision about full content versus excerpts, feed type, and whether you're going to be podcasting or not. You know where to look for feeds to subscribe to, and you've picked your aggregator to manage your subscriptions. Now, how do you let people know about your feed?

Again, as mentioned earlier, you or your tool should provide an autodiscovery link for each feed you support. With this, most tools can find and subscribe to your site immediately. You can also provide a link in your sidebar (or individual post for comments), though this approach is falling out of favor; after all, when you click on the link, you get XML, which is not humanly readable.

But how do you let people know you've updated? Well, many aggregators will automatically test your feed once every specified time period, usually no more than once an hour, no less than once a day. But most aggregators are dependent on being notified, and that's where the concept of pings comes in.

Popular ping services such as weblogs.com and blo.gs have web services that can be invoked by your tool to add your weblog to those in the recently updated list. An advantage to this is that any tool that monitors these services then knows you've updated. A downside, though, is that nefarious types such as weblog spammers use these lists to troll for innocent weblogs in which to dump spurious and usually offensive "comment spam."

Other services, usually blog search engines, provide ping capability and then use this information to access your feed to update their databases.

There is a master ping service, Ping-o-matic, at pingomatic.com, which can be customized to ping only specific services, and then can be invoked manually or included in your tool. Most popular tools, such as Movable Type, TypePad, WordPress, and others, automatically ping Ping-o-matic for you.

Most importantly, you need to ensure that your feed is working properly. Again, anytime you make a change in feed options, run your feed against the Feed Validator at feedvalidator.org. This will validate any of the three types: just pass in the URL of the feed.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Summary

Content preview·Buy PDF of this chapter|Buy reprint rights for this chapter

When you enter the world of syndicated content, you'll be faced with many people telling you the "proper" way to do syndication. It is important to listen to their advice, especially if they tell you your feed is broken. However, one person's "broken" is another person's option, so always remember to validate your feed to verify that it is truly broken.

More importantly, your syndication feed is an extension of your site, and hence of yourself. It should reflect your interests, your concerns, and your choices. After several years of working with syndication feeds, one thing I can tell you that's truth: you will not please everyone.

Once your feed is validated, your options picked, your content published (excerpt or full content), and your services set up to be pinged, you are ready to forget about your syndication feed and focus instead on what's really important at your site: you, and the gifts you give of your writing, recordings, and art.

Additional content appearing in this section has been removed.
Purchase this book now or read it online at Safari to get the whole thing!

Return to What Are Syndication Feeds