Not two weeks after being pleased with myself that I could subtly rejig the design without only a few lines of CSS, I decided on Friday to completely redo this site.

Not only did I change the layout but I've made some major changes under the hood, too. I decided to have my first attempt at an HTML5 page. Granted, it might just fall apart at any moment in any given browser but...hey, it might not.

On the subject of HTML5, Mark Pilgrim (he of the 'Dive into...' series) brought up an interesting point in the WHATWG Blog last week on the topic of whether XHTML was actually a good idea in terms of enforcing XML syntax on an HTML document:

It provides no perceivable benefit to users. Draconianly handled content does not do more, does not download faster, and does not render faster than permissively handled content. Indeed, it is almost guaranteed to download slower, because it requires more bytes to express the same meaning -- in the form of end tags, self-closing tags, quoted attributes, and other markup which provides no end-user benefit but serves only to satisfy the artificial constraints of an intentionally restricted syntax.

And, I guess, it is a good point that a well-formed XHTML document will be larger than the equivalently well-formed HTML document. If, however, developers are given a strict set of rules and a strict validator and told "make your page according to these rules, this alarm will go off if you've done it wrong", they're less likely to fall into bad habits than if they are told "These are mostly rules but sometimes suggestions, this alarm will only go off if you got things very very wrong". Mark Pilgrim is, quite rightly, focusing on the user's point of view but it just seems to me that users will also benefit from more maintainable, better structured code.

Of course, none of this actually matters yet and won't for the next five years or so. It probably won't matter then, either. It is only the interwebs, after all.

I've just updated trulyinnovative.com and www.trulyinnovative.co.uk.

I randomly thought it was about time for a bit of a refresh but writing it, I was amazed at just how much has changed in the 3 years between the first and the second. Despite the fact that they're both supposed to be tongue-in-cheek, they are quite accurate descriptions of two different aspects of the people who 'do web'. Must remember to update them again in 2012.

"And several months later, our intrepid hero returns to the village to find it full of people he doesn't know..."

A quick update. It's now over a year since Noodle launched and about 4 months since Wwwitter launched and I seem to have started a whole new niche of web-conversation sites. Most of them, I won't mention here but there are two notable examples: Convotrack and Tweetboard. If only I'd figured out a way to make money out of it before others did...oh well.

As I've said many times before, I need some kind of marketing genius to take all these little projects of mine and find out where the money-making potential is. I'll quite happily sit in my room repeatedly coming up with the Next Big Thing if someone will then go off and sell it. On a related note, if anyone fancies marketing MonkeyTV, let me know...

Those who know me will know that I have - using the parlance of the modern kids - mad skillz when it comes to video encoding and transcoding. I'll happily admit to wasting far too many hours learning the finer points of ffmpeg, mencoder and vlc, combining them with all manner of shell scripts, web interfaces, cron jobs and the like to set up my own mediacentre, video RSS feeds and shared video chatrooms (currently in maintenance mode).

So imagine my combination of frustration ("it should just work") and elation ("ooh, a challenge") when I discovered that my shiny new Mylo is extremely temperamental when it comes to video. I had assumed PSP-friendly video would have been fine but it turns out I was wrong. Any slight variation in frame-rate, bitrate, frame-size, aspect ratio, codec or container and I'd get a lovely "Sorry, the Mylo doesn't support this format" error (but in Japanese).

Anyway. In case you're interested, here's a shell script:

#!/bin/bash
ffmpeg -i "$1" -y -threads 2 -map 0.0:0.0 -f mp4 -vcodec xvid -b 768 -aspect 4:3 -s 320x240 -r ntsc -g 300 -me epzs -qmin 3 -qmax 9 -acodec aac -ab 64 -ar 24000 -ac 2 -map 0.1:0.1 -benchmark "$1.MP4"

call this from the command-line with the path to the file you want converted and 10 minutes later, your Mylo-ready file will be sitting next to the original.

I'm really only posting this here because in about 6 months, I'll have forgotten all about this and, given my current luck with technology, all my computers and all their backups will have simultaneously formatted themselves.