Why Another Specification? If you have been (Web domain) following
Why Another Specification? If you have been following the development of information feeds over the last four or five years or have read the descriptions in this book of the development of the various information feed formats, one question that may occur to you is, Why another specification? This section addresses that question. Aiming for Clarity One issue that has drawn the attention of some RSS tool implementers is that the specification documents are ambiguous in places. That is an assessment we can sympathize with, having had to work through the RSS specifications to write the preceding chapters. Because of these ambiguities, developers who write aggregators have to be permissive in the structure of markup they accept. Allowing for multiple formats and, possibly, errors in document formats means that aggregators are less easy to write and may be slower than might otherwise be necessary. For the end user, the ambiguities in the specifications are unlikely to be an issue. The developers who create aggregators write the software so that it accepts almost any markup that reasonably resembles what the relevant specifications seem to say. If Atom becomes the standard information feed format then, over time, aggregators might be slimmed down so that they only accept feeds expressed in one universal format the Atom 1.0 format. At least that s what some hope. In practice, particularly in the short term, Atom is likely to add to the complexity caused by there being multiple information feed formats. Atom will simply add a little more complexity to the mix. Archiving Feeds One of the other areas where improvement is sought over RSS 0.9x and 2.0 relates to the issue of archiving information feeds and their contained items. The idea is that for some information it is important to archive the feed. If you archive a feed in what might possibly become a huge, global archive the need for a truly unique identifier for information feeds and their items becomes more of an issue. The vagueness of the RSS 2.0 specification regarding the guid element for example, becomes unacceptable, if you assume that uniqueness will need to be global and stretch over long periods of time, perhaps years or decades. The assumption that information feeds and their contents are of temporary interest only is a notion that some at the heart of Atom development reject. The notion of archiving feeds is different from the idea of storing information discussed in Chapter 5. Atom 1.0 intends to archive information about the feed itself and the items in the information feed. Chapter 5 referred to storing information that was accessed from a feed. In that context the feed itself was seen as disposable but the information resources that you could, as a user, access from the feed might be of lasting interest. The RDF Issue One of the issues that, in my opinion, means that it is unlikely that Atom 1.0 will ever be the single information feed format is the likelihood that it won t use RDF. The use of RDF in the Atom 1.0 feed format 130 Chapter 13