What follows is a write-up of my journey through the rabbit hole that is the Reboot contest. I can only apologise that I didn't get to spend as much time as I'd imagined on my mock-up - and half of what I allowed myself was spent squashing bugs. It's likely that you'll encounter such disturbances on my mash up, so I ask any viewers to first be patient while the RSS caches (if you've been unlucky enough to catch it while it's doing so), and simply refresh if any problems occur. Deleting the session cookies might also be a more extreme option. Perhaps that was a little too pessimistic for an opening paragraph, but I prefer to play safe. If needs be, you can contact me at james dot willock at gmail. Thanks for your time.
My philosophy heading into the design process was for a modular homepage - as I feel this best represents the amount of content that can be served to the average user. For aesthetic purposes I went with cool blue gradients with silver accompaniments, though technically even the colour could be user defined if sufficient time was allocated. Overall, I am happy with the finished visual and am convinced it does not detatch itself too much from the current bbc.co.uk design, which I feel is important for usability's sake.
One of the things I am most happy with is the ability for a user to select the news and television relevant to their self. There was debate earlier about the BBC not being fit to challenge as a live homepage - but with user controlled content the comparisons are inevitable. In a real mockup the range of content would be far, far greater - but having no access to the BBC's APIs my reach extended as far as the advertised feeds. Bookmarks, messages and conversations would all be modules that I would look at in a real-world situation.
One of the trends in web standards based design last year was to try and level the playing field between complete user-agents and less privileged ones (no JavaScript, Flash etc.) However, thoughts have drifted and methods changed. JavaScript is now used to aid those with access to it rather than punish those without. This is evident in the ability to shift and remove modules on the mockup design before you. Moving page elements is useful for workflow and overall personalisation - but it's not mandatory. Neither are stylesheets, mind. Or images. The important thing is that JavaScriptless agents are not affected.
The livesearch was one of the things I was most keen to implement. Its usefulness on a site where search terms will be repeated oft is immeasurable. Coupling this with the popular searches could result in a very happy user group. Currently the livesearch is just fed off a JS array rather than a database - so try just typing in 'BBC' or 'News' and waiting half a second.
Using the del.icio.us API, I created a small area on the page that randomly selects certain items from the del.icio.us feeds and displays them to the user. As it stands they are merely BBC-related, but could possibly be user-defined and hierarchical. I feel this may be something to counter the inherent seriousness of the news.
I started off with a different module for each news section, but found these quickly added up and caused havoc. Tabs are very useful for one purpose - showing some data whilst hiding the rest. This is why tabs are useful for displaying different categories of news - but not, say, television listings which need to be compared on-screen. Because the data is preloaded the tabbing is instantaneous and all the better for it - too much too soon can overwhelm even the wiliest of users.
Realistically, the iPlayer would be much more of a grand beast than a Flash MP3 player - but I felt that given the resources available to me the daily NewsPods were very relevant. Hidden away in the radio site they're done no justice - quick audio access to the day's events is merely a click away.
The A-Z was a purposeful admission from the homepage. In its current form it has no place there other thank to take up precious real-estate and wildly guess at a user's destination. A JavaScript solution would fail on a non-JS user agent and leak massive amounts of data onto the page - so there are a few solutions. Simply copy the current A-Z page to the homepage with a few style changes and hope that someone uses it over the search function - or merely place a link in its place. The A-Z is useful, but those who would use it over the search know where it is. And those who don't, don't need it.
The building blocks are here - the only limitation is in what the BBC make available for us to work with. I've tested the mockup in IE6 and FF1.5 only, so beware of rendering errors on more exotic browsers. Thanks for your time.