RSS technology revisited

This week, RSS tools! First we have a Windows RSS aggregator or reader (the terms are used interchangeably) called Feedreader for reading your favorite RSS feeds. RSS stands for Rich Site Summary, Really Simple Syndication, or RDF Site Summary – no one seems to know for certain.

This week, RSS tools! First we have a Windows RSS aggregator or reader (the terms are used interchangeably) called Feedreader for reading your favorite RSS feeds. RSS stands for Rich Site Summary, Really Simple Syndication, or RDF Site Summary – no one seems to know for certain.

If you haven’t come across RSS feeds before, check out our previous foray into the topic and XML.com story.

Feedreader was released with a Gnu General Public License (GPL) and was free. We say “was” because it is no longer under development. That said, it has a feature we really like that we haven’t found in another reader: A built-in Web server that lets us integrate its output with our intranet.

Feedreader can understand RSS 0.9, 0.91, 0.92, 1.00 and 2.0, plus the Dublin Core and Slashback extensions, and supports Outline Processor Mark-up Language (OPML).

There are actually nine versions of RSS with all sorts of technical issues that limit backward-, forward- and sideways-compatibility. See the “Dive into Mark” blog  for an interesting and rather critical discussion.

The second one, the Dublin Core, from the Dublin Core Metadata Initiative, is interesting because it is a set of metadata standards “that support a broad range of purposes and business models,” and has an RFC.

Dublin Core has been described as “a metadata pidgin for digital tourists who must find their way in this linguistically diverse landscape. Its vocabulary is small enough to learn quickly, and its basic pattern is easily grasped” (quoted from “A Grammar of Dublin Core” by Thomas Baker of the German National Research Center for Information Technology).

The most useful document concerning RSS is Expressing Simple Dublin Core in RDF/XML by Dave Beckett, Eric Miller and Dan Brickley.

Regarding “Slashback extensions,” we wish we could point you to some background but we could find nothing.

Finally, Outline Processor Mark-up Language, according to Dave Winer, the godfather of RSS and OPML, is “a file format that can be used to exchange subscription lists between programs that read RSS files.” What is particularly interesting about OPML is that it describes, as its name implies, outlines, which means that it is applicable to all sorts of tasks where structured lists are required (more on the spec).

The only negative about Feedreader – apart from it now being “abandonware” – is that the formatting of its Web output is “compiled in” the executable code. As we are not Delphi 7 programmers, moving the embedded formatting to external templates that would be loaded on start-up is not going to happen. We think that a bit of dynamic HTML hocus-pocus might work: Load the server output into an in-line frame (an iframe) in another Web page, then find the content of the iframe by exploring the document object model and applying a cascading style sheet. If you feel the desire for everlasting Gearhead fame, feel free to send us your solution.

If you want something that isn’t abandonware, check out the free GPL’ed SharpReader, which is still under development. This software requires the Microsoft .Net Version 1.1 framework and supports all RSS versions: ATOM (a competitor to RSS), Dublin Core, content:encoding (this means that rich in-line content in HTML such as graphics is supported), and xhtml:body (the big boys’ XML version of content:encoding).

Another free, GPL’ed RSS utility worth checking out is Syndirella. Based on the .Net Version 1.0 framework, Syndirella supports all the variants of RSS and OPML import, and can even scrape Web sites that do not offer a news feed and treat the data as if it were RSS.

Syndirella also addresses one of the big problems with RSS usage: needless feed downloads. Let’s say that you like to read the Gibbs blog (no, it doesn’t exist yet) and its RSS feed is, say, 50K bytes. If you check that feed every hour then you will be downloading something in excess of 600K bytes per day. And if you are doing that in concert with, say, 20,000 other people, the Gibbs server will be delivering 1.2G bytes of content every day. But as the feed might only update a couple of times per day, that’s a lot of wasted bandwidth on everyone’s part – particularly the Gibbs server. But we’ve run out of column bandwidth.

Next week: Down with wasted bandwidth. But waste a little writing to gearhead@gibbs.com.