Tomorrow I might register some dot-com domain names. I'm probably going to use GoDaddy - they're cheaper than most of the UK registrars and they've been drummed in to me by PodShow (although Steve Gillmor's six-minute ad waffles didn't do it, that's for sure). My only problem with GoDaddy is that their site is very tough to navigate. What am I going to register? Well, that would be telling. I may register the domain for a startup idea I have, and I probably will register the domain for a new home for my blog and all the various OPML trimmings that I've been working on and thinking about. You're going to love the name of that, although certain busybodies may possibly disapprove. 
Lisa posted this link the other day - Elmer at Teknoids: "In case you hadn't noticed, the OPML Editor is becoming an open source version of Radio". Indeed, I was screwing around the other week trying to produce a hacky alternative to the code that Dave is going to be releasing soon to mirror OPML blogs. That's a project I can strike from my list of things to do.
Today, I've been working on some PHP for Grazr. The other day, I was scratching my head trying to figure out how to solve a particular problem. This may sound somewhat context-free since I'm not supposed to tell you what the problem is nor what it's for. But, preg_replace_collapse (and pregi_replace_collapse by extension) are superbly useful functions. preg_replace with the /e modifier is useful, but the collapse functions make all sorts of things possible that otherwise are very hard to achieve. When I've got a little bit more time, I'm going to be experimenting with using the collapse function in combination with classes and also pre-existing array structures and so on. Collapse also has the benefit of being able to test for optional components of required code which is useful if you are writing code that does any kind of simple interpreting. It is also yet another reason to avoid ereg. PHP's PCRE functions have so many uses and are very easy to get in to.
I can't find the Animal Traps video (if you can find it online, please point me to it), but YouTube has a great clip from Bullshit! about the World Trade Centre, and here's a great video about blood - real and fake - directed to former A-G Janet Reno.
I've been interrupted every minute of the day today. I just wanna develop some software in peace. I ought to get a sign that says "A one minute interruption means a fifteen minute disruption. Are you worth it?" - kind of a demotivational poster to stop people trying to pull me away from my computer. And if that doesn't work, I'll set up a bear trap outside my door (anyone remember Penn & Teller's wonderful "Animal Traps" routine?). No doubt, that means that they'll switch to email. And with 385 unreads sitting in my Inbox, they can wait - and they can be filtered away from "stuff I have to do for work" in to "stuff I have to do when YouTube isn't working". At least during the weekend, everyone around me is a little less prone to interruption. I may have to switch to a night shift permanently, since during the day, it's nearly impossible to do anything because of this heat.
Podcast Directory 
Dave has launched podcasting.opml.org. Some thoughts: (1) make sure that it's valid OPML - the old indiepodder/iPodder directory contained hundreds of instances of OPML where people had linked to an RSS feed but they'd linked it as if it were HTML not RSS - they link differently, and OPML browsers like Grazr don't work if they aren't fed the right type of OPML. The tools that are being built here are going to be essential in OPML's development. I've spoken to so many people who don't quite get what OPML does. Then I pop open a Grazr and show them and they get it immediately. (Usual disclaimer applies here, folks).
(2) Make sure that the subject root for the directory has it's own OPML file. The other stuff is useful, but it's a side-serving.
(3) Split files as frequently as possible - it means that we can recombine them in more interesting ways. I'm hoping to release some code soon that will do this - so you can feed it a URL like splitter.php/$url/4 or splitter.php/$url/Philosophy and Religion/ and it'll dynamically extract an OPML subset.
(4) Make it so that we can do interesting stuff with the files, like splitting them up in to smaller pieces, by liberally licencing them.
(5) Let's have fun. This is the most important part.
