JISC Information Environment Architecture

RDF Site Summary or Rich Site Summary, depending on who you ask! There are currently two flavours of RSS around. RSS 1.0 is based on the Resource Description Framework (RDF), encoded as XML. RSS 0.91 (and 0.92) is based on 'plain' XML. The JISC IE technical architecture recommends use of RSS 1.0, because of its greater flexibility, though many RSS applications will support both versions.

RSS is a simple XML metadata application. An RSS channel is just an XML file that is made available from somewhere on your Web site. The XML file consists of a list of 'items', each of which essentially consists of a title, a description and a URL of a 'news story'. The full story itself is typically a Web page that is made available separately.

To use your channel, people simply gather in the RSS XML file (using HTTP) and transform it into HTML, or WML, or whatever they want. The channel is updated by you updating the XML file, and by them re-gathering it on a regular basis, say hourly or nightly.

Create an RSS XML file and store it on your Web site somewhere. You can use UKOLN's RSS-xpress channel editor to do this if you want. The RDN maintain its central RDN news channel and Behind the Headlines channels in this way. Alternatively, if your content is stored in a database, it might be more appropriate to generate your RSS XML file directly out of the database as well.

Any filename you like - it doesn't matter. Most people use a filename with a '.xml' or a '.rdf' extension. The important thing is that you configure your Web server to serve the file using a MIME type of 'text/xml'.

Yes and no!. Yes, you are giving some of your content away as XML and letting other services present it it with their look-and-feel. However, the overall result should be that more people see your content. Furthermore, if the URL associated with each item in the channel is the URL for one of your pages, then every time someone clicks on the link in the RSS channel they will visit your site. Remember that when people visit these 'news' pages, they will be entering your site from outside - provide enough context on each page to allow people to navigate to your home page, search service(s) and other key parts of your site.

If you are very worried about giving content away, one approach is to include only a title and URL for each item in the RSS channel. That way, people have to visit your site to get any useful information.

Every RSS channel can include the name, logo and search URL of the service that made the channel available. It is reasonable to expect any service that uses your channel to display this information clearly.

On the RDN pages, a regular background Perl script is run (from cron) that pulls in the remote RSS channel (XML) and converts it to HTML. Then a Server Side Include (SSI) is used in the pages to embed the HTML into the page at the appropriate point. An alternative approach is to generate the whole page using a CGI script, part of which gathers the remote XML file dynamically and transforms it into HTML at the time the page is generated. One can do similar things with PHP, ASP or other server-side scripting languages.

Alternatively, let the Web browser retrieve and present the RSS channel using a Java applet.

A simpler approach is to make use of an external service to transform the RSS channel into HTML for you. As part of RSS-xpress, we experimentally offer something called RSS-xpress Lite, which does just this. See

http://rssxpress.ukoln.ac.uk/lite/include/

for details. You can try it by embedding the following HTML somewhere in one of your pages

<script
src="http://rssxpress.ukoln.ac.uk/lite/viewers/rss.cgi?rss=http://www.ukoln.ac.uk/rss/ukoln.xml">
</script>

and you should see a view of the UKOLN RSS channel. Replace the channel URL with one of your choice.

If you are displaying someone else's RSS channel, remember to clearly display the name, logo and a search box of the service that made the channel available.

This is a valid concern. It would be sensible to only use RSS news channels from organisations with which you have a reasonable basis for trust - both in terms of legality and appropriatness of content.

You need some software (or access to some software) that can convert the XML into HTML (see above). Your punters, as in real end-users, don't need anything special - they are just viewing normal HTML.

http://www.rdn.ac.uk/ (the news channels on the RHS)
http://www.rdn.ac.uk/news/
http://www.rdn.ac.uk/news/headlines/
http://www.ukoln.ac.uk/ (the main news section is created from an RSS channel)

See RSS-xpress for a list of UK RSS channels.

See Syndic8 for a worldwide list.

There are a number of pointers at

http://www.ukoln.ac.uk/metadata/resources/rss/

A weblog (sometimes called a blog) is a Web page where someone (sometimes called a blogger) 'logs' all the other Web pages he or she finds interesting. Weblogs are usually organised by date and are often made available in the form of an RSS channel.

A scraping service is a service that converts (scrapes) relatively unstructured HTML and/or email messages into a more structured format such as RSS.

I presume XML channels are maintained directly as RSS (XML). Scraped channels will be based on services that take HTML or email news offerings (e.g. the BBC pages) and convert them (sometimes illegally) into XML.

Probably not in many cases. Many Web sites (e.g. the BBC) explicitly state that you are not allowed to take their content and convert it into other formats.