Single Page RSS Feeds – So What? So this…
Having posted about Single Item RSS Feeds on WordPress blogs: RSS For the Content of This Page, it struck me that whilst this facility might be of interest to a very, very select few, most people would probably have the response: so what?
To answer that question, it might help if I let you into a little secret: I’m not really that into content, open educational or otherwise. What I am interested in is how content can flow around the web, and how it can be re-presented in different ways and different places around the web by different people, all pulling on the same source.
So if we consider single page RSS feeds, what this means is that I can re-present the content of any of my WordPress blogged posts anywhere that accepts RSS. So for example, I could view just that post as a Wordle generated word cloud, or subscribe to the RSS version of single blog post on a Netvibes page (maybe along with other related posts):
and view the post in that location:
(At the moment not many other platforms appear to offer single page RSS feeds. I was hopeful that the Guardian might, because they have quite a well developed feed platform, but I couldn’t find a way to grab a single page feed trivially from a page URI:-(
To see why that might be useful, you need to know another of my little secrets. I don’t really think of RSS feeds being used to transport new content, such as the latest posts from the many blogs I still subscribe to. For sure, they can be used for that purpose, and a great many RSS readers are set up to accommodate that sort of use (only showing you feed items you haven’t already read, for example), but that is a special case. The more general case is simply that feeds are used to transport content that has quite a simple structure around the web. And this content might be fixed, static, immutable. That is, the content of the feed might never change once the feed has been created, as in the case of OpenLearn course unit full content RSS feeds.
AS AN ASIDE… I generally think of RSS feeds as providing a way of transporting simple content “items” around where each item has a quite simple structure:
If you think of a blog post or news article as an item, the title is hopefully obvious (the title of the post/article), the description is the content “body” of the item (e.g. the text content of the news article) and the link is the URL of where that post or article can be found on the web. The other elements are optional: what I refer to as annotations correspond to things like latitude and longitude co-ordinates that can be used add geographical information to the item so that it can b plotted on a map for example; and what I term a payload would be something like an audio file that gets delivered when you subscribe to an RSS podcast feed from somewhere like iTunes or IT Conversations.
Once you start viewing RSS feeds as a general transport mechanism, then you start to see the world in a slightly different way… So for example: the a href=”http://ouseful.wordpress.com/2009/07/08/single-item-rss-feeds-on-wordpress-blogs-rss-for-the-content-of-this-page/”>Single Item RSS Feeds post reveals how to create single item RSS feeds from the URL of a blog post hosted on WordPress. Now if I bookmark a series of WordPress hosted blog posts to somewhere like the delicious.com social bookmarking site, and tag them all in the same way, I can get an RSS feed out that contains a list of posts that can be obtained in XML form (that is, as single item RSS feeds).
So maybe if I find a series of posts from WordPress blogs all over the world on a particular topic, I can create my own custom RSS feed of those posts that I can use as the basis of a reading list, for example, or to feed a Netvibes page on a particular topic, or even to feed an RSS2PDF service*?
* these needn’t be really horrible and divisive… For example, the Feedjournal service will take in an RSS feed and produce a rather nice looking newspaper version of your feed… ;-)
Now it just so happens, I’ve prepared one of these earlier. In particular, I’ve posted a small collection of blog posts on the topic of WordPress from a variety of (WordPress) blogs at http://delicious.com/psychemedia/singlefeeddemo:
You’ll notice that I can get an RSS feed of this list out too: from http://delicious.com/rss/psychemedia/singlefeeddemo in fact.
Now the links I’ve bookmarked are links to the original HTML page version of each blog post; but all it takes is the simple matter of rewriting those URLs by adding ?feed=rss2&withoutcomments=1 on to the end of them to get the RSS version of each post.
Hmm… Yahoo Pipes, where are you? Let’s just pull in the RSS feed of those WordPress hosted blog post bookmarks, and rewrite the URLs to their single item RSS feed equivalent:
Now we can loop through each of those items, and replace it with the actual content of those single item RSS feeds:
The output of the pipe is then a real RSS feed that contains items that correspond to the content of WordPress blog posts that I have bookmarked on delicious.
Now just think about this for a moment: most RSS feeds are transitory – the content that appears in the feed on a blog post is a reverse chronological list of the 10 or 20 most recent items on the blog (or in a particular category on a particular blog). The feed we are pulling in to this pipe may be fixed (e.g. if we create a list of bookmarks tagged in a particular way, and then don’t tag any more bookmarks in that way) and used to create a very specific a list of blog posts from all over the web. By rewriting the URLs to get the RSS version of each bookmarked post, we can create our own full RSS feed of those list items. (Actually, that isn’t quite true – if the blog is configured to only emit partial RSS feeds, we’ll only get a partial version of a post, typically the first sentence or two.)
(Pipes’ homepages only show preview versions of a feed description, even if the full description is available.)
Just to recap, here’s the whole pipe:
We take in a list of bookmarked URLs that correspond to bookmarked WordPress blog posts, and generate the single item RSS feed URL for each post. We then use these URLs to pull in the content for each post, and this create out own, full content custom RSS feed. The pipe itself emits RSS, so w can take the RSS feed from the pipe and feed it into any service that consumes RSS, such as Feedjournal:
Alternatively, I could subscribe to the pipe’s output feed in somewhere like Netvibes (or even a VLE) and then view the contents of my customised feed in that location. Or I could import that feed into a new WordPress blog. And so on…
Now of course I appreciate that many people will still say: so what? But it’s a start… a small step towards a world in which I can declare an arbitrary list of links to content spread all over the web and then pull it into a single location where I can consume it, or process it further, such as converting it into a PDF (which is a preferred way of consuming large chunks of content for many people) or even delivering it in drip feed fashion over an extended period of time as a serialised RSS feed, for example.
An exercise for the interested reader: clone the pipe and modify it so that it will accept as user input an RSS URL so that the pipe can be used to consume any social bookmarking service RSS feed.
Note: as the pipe stands, the order of items in the feed will correspond to the order in which they were bookmarked. It is possible to tag each bookmark with its desired position in the RSS feed, but that is a rather more advanced topic. (See a soon to be(?!)* deprecated solution to that problem here: Ordered Lists of Links from delicious Using Yahoo Pipes.
* If @hapdaniel hasn’t already published a more elegant solution to this problem using YQL Execute somewhere, I’ll try to do so when I get a chance…
PS ho hum, maybe we don’t need RSS after all: Instapaper, Del.icio.us, Yahoo! Pipes and being Slack (via @mediaczar)