How about WordPress blogXML?

Weblogs have become more sophisticated than an index page that spits out the 10 most recent posts - people are even using them for real corporate ventures… and they want control over their data.

With the advent of WordPress 2.5, I got to thinking about blogs, posts, XML and the present state of things. Now, there’s galleries and enhanced shortcode tags, excellent in themselves, but necessarily ad hoc improvements, bolted on where feasible.

As a practical example - let’s say you wanted an index page with some static content, the most recent post in its entirety and then excerpts from preceding posts - no real problem - but what if you didn’t want any images in the excerpts, just to keep them nicely in their little box, or you wanted thumbnails of images …

You might mess around with regexs, but it would always be messy and unreliable. At the moment, you haven’t got that level of control. At risk of sounding completely fascist/neurotic, you really do need control.

(Aside: The 5th law of programming - If you really do need a regex, you’ve probably gone wrong somewhere).

blogPostXML

Strictly, what we’re talking about here is blogPostXML - structured content data, to be consumed in any fashion the end user desires. RSS goes some way towards this - it structures the data and metadata of a post; title, author, timestamp etc., but still leaves the important part, the content, as an inflexible lump of XHTML.

So blogPostXML would go something like this:-

<post>
<content>
<text>Some text here, with an <image id=”1″ /> here, then more text.</text>
<image id=”1″>
<title>My dog</title>
<loc>http://www.mysite.com/images/my_dog.jpg</loc>
<caption>etc etc<caption>
</image>
<image id=”2″>
… more here
</image>
</content>
</post>

Similarly, a post can have associated media files, becoming podcasts (as, again, RSS already does), author information, etc etc, another step towards the supposed nirvana of complete free interchange.

While we’re at it, some other possibilities, global to the post, not just the postContent:-

• <related> - related posts, local and remote - also part of a series of posts etc.
• <rights> - relicensing, syndication conditions - for what good that might do…
• <tags> - well, obviously.

Any other ideas?

This is microformats territory, and I’ve no doubt a whole slew of specifications, formats, proposals, already exists - for example, here - ancient, tentative and incomplete. The adoption of XML types is nearly always like the VCR/Betamax/Phillips format war - somebody wins out, their namespace becomes the standard.

Specifically, it’s intimately connected with the principles of RDF - Resource Description Framework - but people’s heads start to spin a bit when you mention the Semantic Web, Web Ontology Languages, XSLT, SPARQL, so another time…

So when?

Now you may well say, what’s the point? Too complicated, overkill, geeks in ivory towers… and yes, it would be a big jump in the programming, with the sheer grief a complication layer always brings. And yet this could be something available as required, wrapped up and hidden from the average punter.

(If I ever get the time), it wouldn’t take much hacking around a WordPress install to see what gains could be made, and, of course, where the problems lie.

Formats like FOAF, XFN had their arrival, their struggle, their decline. So I can see why there hasn’t been much done in this area yet - equally, I wonder whether its time is not too far away.