Dave is in fine Crazy Uncle mode on JSON: "No doubt I can write a routine to parse this, but look at how deep they went to re-invent, XML itself wasn't good enough for them, for some reason (I'd love to hear the reason). Who did this travesty? Let's find a tree and string them up. Now." It doesn't matter whether it's the war in Iraq or JavaScript Object Notation - it's a fucking travesty and you need to rioooooot! 
Mike Arrington has a post about the ethics of firing Sam Sethi, where he reveals that he often offers certain people the opportunity to have their slightly less sensible comments removed from TechCrunch. I can't remember the last time I saw something interesting on TechCrunch, and this post is a very good motivation to drop my subscription. The value of TechCrunch compared to just being subscribed to del.icio.us' popular feed for the "web2.0" tag has dropped large amounts for me.
Development diary: Del.icio.us and JSON 
Niall Kennedy has some early information about del.icio.us' new API function to get a list of the tags which people are using for their bookmarks.
Basically, you can now quite easily get hold of del.icio.us top tags for a URL in JSON format.
I'm not a big JSON fan. I prefer XML or PHP arrays. Good thing I've got PHP 5.2.0 installed on my web server. json_decode handles it without me even having to think about it.
(Also, eBay hosts this excellent XML to JSON XSLT)
So chuck anything you like at me. XML, HTML, PHP arrays, JSON. I can read it and turn it in to something useful.
I'm basically writing a tag parser at the moment - which is easy enough. You just look through and find the rel attribute and then chunk through the URL in a really ugly way to get the tag.
Only a number of popular web services don't actually follow the rel=tag microformat. Yes, that means you, Flickr, Zooomr and YouTube. I'm ignoring YouTube because the tag quality on there tends to be so low (people just tend to type in a sentence in to the tag box and so we get "i" "really" "like" "this" "video" or some equally rubbishy set of tags).
Why is it that these services aren't speaking microformats like everyone else is? They are standards for a reason...
