How many central government websites offer RSS feeds these days? The good news is that of the 20 departments represented in the Cabinet, I could only find one that didn’t. But it was a bit of a surprise to see how few offered ‘full text’ feeds, as opposed to ‘summary only’.
I visited each of the 20 departments listed on the Parliament web page – the top result in Google for ‘UK cabinet ministers’, looking for a main RSS news feed. Here’s what I found:
- There are explicit references to RSS feeds on 18 of the 20 sites: the exceptions are the Scotland Office and Defra. There is a Defra feed if you know where to look (namely COI); but how many would know to look there? That leaves the Scotland Office as the only department completely lacking an RSS feed for departmental news. (Its Secretary of State, Jim Murphy does have a blog, but I’m not counting that here.)
- Five of the 20 fall back on the feeds produced by COI’s News Distribution Service. That leaves 14 of the 20 producing their own feeds – in most cases, in addition to the feeds at COI.
- Only one, FCO, directs people through Google’s Feedburner service.
- Only 3 of the 20 provide ‘full text’ RSS feeds – allowing people to read the full press release (etc) instantly, and opening up the possibilities for easy information re-use (ie ‘mashups’). The rest require people to ‘click through’ to a page on the originating website. This is common in commercial publishing, where on-page advertising is a key driver.
- Of the 3 offering ‘full text’, 2 are running on WordPress: Number10 and the Wales Office, both of which I admittedly had some involvement in. The other one is DECC.
- The Department of Health’s RSS feeds aren’t valid: the ‘link’ element quoted in the feed doesn’t include www.dh.gov.uk. A curious problem to have caused yourself, and a trivial one to fix. I’ve mentioned this before, in the context of Directgov; and of course, the two share a publishing platform. A broken one, in this case.
- It was a pleasant surprise to see the majority of sites have ‘autodiscovery tags‘ in the header of their homepages – a behind-the-scenes way of indicating that a site has an RSS feed, which can (for example) light up an icon in the browser interface. But 8 don’t. I’m looking at you FCO, Home Office, Defra, DFID, Cabinet Office, Defence, Transport, and DCMS. Some of them have the appropriate tags deeper into the site, to be fair… but it’s a free and instant win those sites are missing out on.
The thing is, it’s so easy to get RSS right. Ask any blogger: when executed properly, RSS feeds should be an automatic, never-even-think-about-it thing. Each time a new item becomes available on a site, it should just drop into the RSS feed, notifying people – and importantly, mechanical services – of its availability.
And the easiest way to get RSS right is to build your news website on WordPress. Out of the box, you get valid RSS feeds for virtually any view of your site’s news content. Feeds by category / press office desk / minister? By keyword tag… or combinations of keyword tags? How about infinitely customisable feeds, based on search queries? Yes, to all of those. Probably within a couple of days, if you get the right people in. (Hint hint.)
A lot of government websites are going to need a rethink following the next election. It’s the ideal opportunity to upgrade the news area, by moving to a system that’s been explicitly designed around the timely publication of short text articles, generally presented in chronological order. By which I mean, a blogging system. And specifically, WordPress.
7 thoughts on “RSS usage on Whitehall's websites”
The reason I suspect that most aren’t making use of the NDS service is that it is (was?) flakey and didn’t deliver a consistent service (and of course as you know, historically there was reluctance from COI to offer it without a charge- that has changed now thank goodness). But realistically long term there is no reason why that shouldn’t be the delivery channel for govt news as its established and has the infrastructure.
SOme time ago i pulled together a script that spewed out a report on 404 pages and autodisoverable feeds from government departments
with the report here:
Well yes BIS *is* using NDS for news, but our announcements on the home page are available as a full text, Feedburner-powered feed (and email subscription). And surely we should get some kind of bonus point for a Pipes-powered speeches feed? 😉
There are such things as Atom feeds. Please be more inclusive in the language you use.
And please don’t present wordpress as an almost unique blogging / feed generation system. It ain’t.
Apart from these grumbles, I love that you’ve shone a light on this and the rigour with which you’ve done it. Much needed, thank you.
I don’t think anyone visiting this site is left in any doubt about my preference for WordPress, Hugh… 🙂 – but I’m certainly not suggesting it’s uniquely capable in this respect. There may well be equally powerful, or more powerful tools when it comes to feed generation. But in my opinion and experience, WordPress is the easiest to install, maintain and work with.
Same goes for the Atom format, actually. I guess you could make a theoretical case that Atom’s a more powerful format in certain respects; but I’d suggest that RSS is easier to work with, easier to explain to people, and gets the job done.
I rather like Atom too, it must be said. It’s lovely and extensible, and yet as reliable to parse and subscribe to as RSS in my experience. For example, the fact that Twitter search Atom feeds include a separate element for avatar image makes it possible to do more interesting things with the feed when you display it – e.g. http://inboxlistening.com/app/
To be fair Steph, that’s nothing you can’t do with RSS: just add a custom XML namespace. For example, I’ve used the Media RSS namespace to add author pictures into the DFID Bloggers feeds, allowing DFID to pull it automatically into their corporate site homepage.
I can’t believe I’m allowing this to become a format battle. It absolutely doesn’t matter. One single globally agreed format would have been nice, but two is bearable. And hey, at the end of the day, it’s all XML.
Comments are closed.