An ideal zero-width, infinitely long, perfectly straight curve

I start at Line on Monday, 31st March.

I declare the new Youthmovies album a triumph. If you do only one thing today, listen to Youthmovies – If You’d Seen A Battlefield (Live)

I was reminded recently of a goal I’ve had for quite some time, at least since I first saw a Windows Blue Screen of Death:

Make computers work.

Over time those three words have changed what they mean to me: initially what I wanted to do was to make a computer run, reliably. Today, and certainly for the last couple of years it’s been more about people, and making machines work, consistently and as expected, for everybody.

I was reminded as I watched Ro trying to install RealPlayer on her fresh Leopard install so that she could listen to Radio 4 while pottering around the flat.

On the Mac most downloaded software falls into to categories: a .app that you drop into the /Applications folder, or an installer which on running performs some installation magic. As an aside, I’m very much against the latter of the two.

Firstly, Ro managed to miss the big yellow button labeled “Download RealPlayer 11(beta)”, and navigated her way through Real.com to the free Windows version—which of course didn’t work.

After finding her way back to the Mac version she ran what looked like an installer, correctly removing the .app to the Trash, only to find that the install didn’t work—the ‘installer’ was the whole thing, which had a few first-run setup questions—in fairness, this process got me too.

After this realisation and another download of the Mac version it was all running fine, but I was considerably annoyed with the whole process.

As far as the majority of users are concerned Real has one product—RealPlayer—and it’s not something I’ve ever known anyone to pay for. And for the same majority the only reason they would ever use it is to play the small amount of Real content on the web, which for Ro, and myself, is from the BBC.

RealPlayer has one task: allow people to watch and listen to the Real formatted content in their browser—I have never, nor have I ever wanted to do anything else with RealPlayer.

Real.com has one task: direct people to download the appropriate version of RealPlayer for their system and language—this can be done fairly accurately with a little bit of intelligence and the User-Agent string—I have never, nor have I ever wanted to do anything else on Real.com.

And why-oh-why are they offering people several download mirrors in the age of broadband, CDNs and IP-location data? Or trying to charge for RealPlayer?

So today I’m restating my “professional goal”:

Make computers work.

Edinburgh, “the greatest city in the UK”, has an award winning bus service in Lothian Buses (who knew there were annual UK Bus Awards?). And the service is reasonably good on busy roads: Leith Walk, Lothian Road, and The Bridges; and more regular services: the 22 and the 26 being the obvious ones.

Despite Lothian Buses operating about 40-50 routes in and around Edinburgh, having RFID bus-passes, digital sign-posts, and on-street ticket machines, they don’t seem to have any interesting data available through their website. Maybe it’s just me, but that data would be incredibly interesting to play with.

For me, the most intriguing devices they have are the digital sign-posts which tell you when the next bus will arrive, which could be implemented in one of two ways: the expected way; and the cool way.

The expected way

Each sign-post has the time table for the buses on that route and calculates how long it is until the next bus arrives: simple and easy. But it’s pointless, all of that information is already available at the bus-stop in the form of a large data table (OK, from a Tufte point of view it’s not pointless), but the data isn’t anything new, and is certainly not interesting.

The cool way

The sign-posts and buses are all part of a mesh network, communicating on an ad-hoc basis as they pass each other. Every bus has a GPS device which gives position and, when coupled with time gives direction. The GEO data is passed to other buses going in the opposite direction and each bus passes this information to the sign-posts to update their displays accordingly.

Not only is that a hell of a lot more technology it would enable “real-world data”, not the expected data generated from the timetable.

Now, make all of that data available as a web service and you start to enable some really cool applications: instant traffic quality; assessing real impact of road-works; bus tracking on Google Maps; and any number of others.

And that is why GEO is so cool: it’s not killer at the application level, it’s killer at the data level, so keep an eye on edinburgh2.com because I have a funny feeling there will be bus data there soon.

I’ve been a busy little bee lately, so I thought I’d share some of my work.

Brian Suda sent me a link to Hoods, which we both though was pretty awesome, so I went ahead and built a version for Edinburgh. Currently it only gathers the data, but nice maps and an API are certainly things we’ve been thinking about. All of the data will be Creative Commons once said API is written. So get on over and add your neighbourhood opinion.

As a less serious project, I built The Monday Night Whisky Club a small website. The idea being that it will grow over time and eventually to do something cool with the data: I’m thinking line-charts of regional preference over time.

And finally, I realised I was still getting a lot of spam on the older posts here on sneeu.com. So I now automatically close comments after six weeks, and to remove the existing spam I created a one-page for and a small jQuery plugin, ShiftClick, which selects (or deselects) all of the checkboxes between your last single click and a shift-click. Enjoy :)