Feeds, Readers, and Feedburner

October 13, 2007

I finally got round to figuring out how Feedburner works, and have added a subscription button (“chicklet”) to my sidebar. Now you can read me three ways. (1) Just drop in every so often; (2) Read the content through your favourite feedreader; (3) Get an email when I post something new. Some of you are already subscribed to me through some feedreader or other. I can’t easily get the stats on these; but if you unsubscribe and then re-subscribe (using your same favourite feedreader) then you will get the same stuff routed through the feedburner server, and I will know you are there ! (For bloggers who run the WP software themselves, feeds can be automatically re-directed, but for blogs hosted at wordpress.com you can’t do this – see below).

I have cleared my head by explaining below what I now understand… hope it helps somebody else !

What is a feed ?

I know that I have geeky readers who will say “you are explaining what a feed is ? why ?“, and other readers who will say “eh wot pardon ?” – so this is for them.

Its like Reuters for the Web. Modern web pages can change quite often. Its annoying to have to keep checking whats new. This is true of blogs a fortiori of course. So the technique is to collect new material and make it available (syndication) through a standardised document (feed) ; an application (feedreader) can then subscribe to the feed and display for you. Your reader can collect lots of different feeds in one place (aggregation) and so you can check all your favourites for updates in one visit rather than forty.

The syndication stage is done automatically for me by the WordPress software. (Many other types of software, like Wikis for example, also automatically construct a feed.) The feed itself is an XML document that is basically a set of links to material, plus typically some extracts from the material. There are (sigh) at least two standards for the format – RSS and Atom – but luckily the feedreaders usually figure this out. You can learn more about all that at this wikipedia article, and at this O’Reilly article. (There is nice general article by mezzoblue here.) For my blog, at https://andyxl.wordpress.com, the standard feed document provided by WP is at

https://andyxl.wordpress.com/feed

and my new Feedburner feed document is at

http://feeds.feedburner.com/TheE-astronomer.

You can click on these to take a look, and do “view source” to see how the XML works.

So how do you set about subscribing and reading ?

First, you can just read feeds in a standard web-browser like Firefox and IE – they understand the XML. “Subscribing” is just a matter of collecting the bookmarks as usual. Often (but not always), FF or IE knows when a website contains a feed and adds a wee symbol in the address bar. If you click on this, you can see the feed and bookmark it; but you may also be given a list of web services through which you can subscribe to the feed. (Depends how your browser is set up)

Second, you can register with a web service, like Bloglines or Newsgator or Google Reader. You subscribe to each feed you want, their server trawls the feed docs, and you read them at their web site. I use Bloglines myself.

Third, you can install a client application on your own PC, like NewzCrawler or FeedDemon . These work like Outlook Express – they suck down the new stuff from your subscribed feeds, and save it on your hard disk for reading at your leisure.

There is friendly explanation of this stuff in a BBC news site article.

Why this change to feedburner now ?

Until some months back, WordPress used to provide stats on how often your blog was read through a feedreader, rather than by viewing the actual page – but they withdrew this, for reasons I never quite understood. As a result its quite hard to know what one’s actual readership is. From the old stats, it looked like I was getting roughly equal numbers of feed reads and page reads. Of course some of those page views were people clicking through to the real thing if something looked interesting (I do this all the time), and the feed-reads were just the services collecting updates, which doesn’t prove anybody read them (likewise, guilty). But even this dodgy info disappeared when WP withdrew the FeedStats feature.

From the point of view of the reader, Feed Burner is just the same as using the original feed. (If you compare the feed docs linked above you can see they are the same; its just that one is sitting on the WP server and one on the FB server.) But the publisher gets the stats; and because it is all routed through the FB server regardless of which reading method your readers are using, you don’t miss anything.

Sounds fun, how do I do this too ?

I won’t wade through this in detail, but rather point you at some other web pages I found handy.

If you want to just put an icon on your blog linking to your regular WP feed, this is explained in the WP FAQ on this page.

I found a general article article about setting up FeedBurner by Marshall Kirkpatrick, and another specifically about Feedburner and WP blogs by HarisTV; however FeedBurner itself has a very clear Word Press Quick Start page.

Key issue that I gradually realised : there are two types of WordPress blog. You can install the WP software on your own server (self-hosted) or you can set up and run your blog using the WP server and its interface (WP-hosted). Some things can only be done for self-hosted blogs. Example one is improved auto-discovery – at the moment Firefox seems to know my feed exists, but IE doesn’t. You can fix this by putting some extra lines of code in your blog template, which FB give you. But if you are WP-hosted, you can’t edit your blog template. Example two is automatic re-direction. As I explained earlier, new subscribers will be routed through Feedburner whichever reader they use, but Feedburner won’t know about people who already subscribed to my existing WP feed. However, somebody terribly clever called Steve Smith wrote a plug-in which would automatically re-direct readers. This has now been officially absorbed by Feedburner and called FeedSmith. However, once again this works by plonking the right bit of code in your WP installation – so can’t be done by WP-hosted bloggers.

Phew. Think I am done.

And in closing…

As John Ebdon used to say, if you have been, thanks for listening.